TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk

ENBRIDGE NORTHERN GATEWAY PROJECT

FINAL

Northern Gateway Pipelines Inc.

April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

Table of Contents

1 Introduction ...... 1-1 1.1 Objective ...... 1-2 1.2 Existing Documentation and Project-Specific Baseline Studies ...... 1-2 2 Project and Site Description ...... 2-1 2.1 Project Overview ...... 2-1 2.2 Distribution of Communities along the Carrier Route (3.15.4) ...... 2-1 2.3 Relevant Climatic / Environmental Considerations ...... 2-7 2.3.1 Waves, Wind and Current ...... 2-7 2.3.2 Visibility ...... 2-8 3 Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk ...... 3-1 3.1 Design Ship Safety Features ...... 3-1 3.1.1 Hull and Cargo Tank Components ...... 3-1 3.1.2 Navigational Equipment ...... 3-2 3.1.3 Fire Prevention and Fire Fighting ...... 3-2 3.1.4 Oil Tanker Vulnerability to Shipboard Incidents ...... 3-2 3.1.5 Future Oil Tanker Requirements ...... 3-2 3.2 Marine Terminal Safety and Monitoring Equipment ...... 3-3 3.2.1 Gas Monitoring ...... 3-3 3.2.2 Fire Detection and Fire Fighting ...... 3-4 3.2.3 Control Room Monitoring ...... 3-4 3.2.4 Quick Release and Load Monitoring of Mooring Lines ...... 3-5 3.2.5 Metocean Monitoring System ...... 3-5 3.2.6 Docking Monitoring System ...... 3-5 3.3 General Terminal Measures to Avoid or Minimise Spills ...... 3-6 3.4 Vessel Operations and Environmental Protection ...... 3-7 4 Marine Shipping Network Risk Analysis ...... 4-1 4.1 Shipping Route Description ...... 4-3 4.2 Hazard Identification ...... 4-4 4.3 Probability of Hydrocarbon Spills along the Shipping Routes ...... 4-5 4.3.1 Incident Statistics ...... 4-6 4.3.2 Conditional Probabilities of a Spill ...... 4-11 4.3.3 Risk of a Spill from a Tanker...... 4-13 4.3.4 Consequence Analysis (Spill Size) ...... 4-17 5 Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal ...... 5-1 5.1 Terminal Description ...... 5-1 5.2 Probability of Hydrocarbon Spills ...... 5-4 5.2.1 Probability of Spills associated with Cargo Transfer Operations at the Kitimat Terminal ...... 5-5

April 2010 FINAL Page i

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

6 Properties and Weathering of Liquid Hydrocarbons ...... 6-1 6.1 Physical Properties ...... 6-1 6.2 Chemical Properties ...... 6-2 6.3 Weathering and Fate of Spilled Hydrocarbon ...... 6-3 6.3.1 Spreading ...... 6-3 6.3.2 Evaporation ...... 6-4 6.3.3 Dissolution ...... 6-4 6.3.4 Oil-in-Water Emulsification ...... 6-4 6.3.5 Dispersion ...... 6-4 6.3.6 Sedimentation ...... 6-4 6.3.7 Oxidation ...... 6-5 6.3.8 Biodegradation ...... 6-5 6.3.9 Estimates of Weathering Processes ...... 6-5 7 Incident Prevention and Response ...... 7-1 7.1 Canadian Coast Guard, Transport Canada and the Pilotage Authority ...... 7-2 7.2 Response Approaches and Capabilities ...... 7-3 7.3 Communications ...... 7-7 7.4 Response Objectives and Strategies ...... 7-7 7.5 Equipment and Personnel ...... 7-11 7.5.1 Escort Tugs ...... 7-13 7.6 Protection of Sensitive Areas ...... 7-13 7.6.1 Environmental Sensitivity Atlas ...... 7-14 7.6.2 Key Sensitive Areas...... 7-14 7.7 Shoreline Response ...... 7-16 7.8 Financial Responsibility ...... 7-16 7.8.1 Response to Marine Spills ...... 7-17 7.8.2 Ship Owner Liability ...... 7-17 7.9 Compensation ...... 7-17 7.9.1 International Oil Pollution Compensation Fund ...... 7-17 7.9.2 Ship-Source Oil Pollution Fund ...... 7-18 7.9.3 The International Supplementary Fund ...... 7-18 8 Setting for the Open Water Area...... 8-1 8.1 Continental Slope ...... 8-4 8.2 Queen Charlotte Sound, Hecate Strait and Dixon Entrance ...... 8-8 8.3 Queen Charlotte Strait ...... 8-9 8.4 North Coast Inlets and Fjords ...... 8-9 8.5 Vancouver Island Shelf ...... 8-10 8.6 Uncommon Oceanographic Features ...... 8-10 9 Setting for the CCAA ...... 9-1 10 Effects of Hydrocarbons on the Biophysical Environment ...... 10-1 10.1 Approach ...... 10-1 10.1.1 Geographic Extent of Study Area ...... 10-3 10.1.2 Assessment Data ...... 10-4

Page ii FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

10.2 Air ...... 10-5 10.3 Water and Sediment...... 10-5 10.4 Plankton ...... 10-5 10.4.1 Potential Effects of Hydrocarbons on Plankton ...... 10-6 10.4.2 Mitigation Measures ...... 10-7 10.5 Marine Vegetation ...... 10-7 10.5.1 Baseline Conditions ...... 10-7 10.5.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Vegetation...... 10-9 10.5.3 Potential Effects of Condensate on Marine Vegetation ...... 10-11 10.5.4 Mitigation Measures ...... 10-11 10.5.5 Follow-up Monitoring ...... 10-11 10.6 Marine Invertebrates ...... 10-12 10.6.1 Baseline Conditions ...... 10-12 10.6.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Invertebrates ...... 10-18 10.6.3 Potential Effects of Condensate on Marine Invertebrates ...... 10-20 10.6.4 Mitigation Measures ...... 10-20 10.6.5 Follow-up Monitoring ...... 10-20 10.7 Fish, Fish Habitat and Marine Fisheries Management ...... 10-20 10.7.1 Baseline Conditions ...... 10-22 10.7.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Fish and Fish Habitat ...... 10-32 10.7.3 Potential Effects of Condensate on Fish and Fish Habitat ...... 10-37 10.7.4 Mitigation Measures ...... 10-37 10.7.5 Follow-up Monitoring ...... 10-38 10.8 Marine Birds ...... 10-38 10.8.1 Baseline Conditions ...... 10-39 10.8.2 Potential Effects of Diluter Bitumen or Synthetic Oil on Marine Birds ...... 10-46 10.8.3 Potential Effects of Condensate on Marine Birds ...... 10-50 10.8.4 Mitigation Measures ...... 10-50 10.8.5 Follow-up Monitoring ...... 10-51 10.9 Marine Mammals and Turtles ...... 10-52 10.9.1 Baseline Conditions ...... 10-52 10.9.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Mammals ...... 10-63 10.9.3 Potential Effects of Condensate on Marine Mammals ...... 10-66 10.9.4 Mitigation Measures ...... 10-67 10.9.5 Follow-up Monitoring ...... 10-67 10.10 Terrestrial Mammals ...... 10-68 10.10.1 Baseline ...... 10-68 10.10.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Terrestrial Wildlife ...... 10-69 10.10.3 Mitigation Measures ...... 10-70 10.10.4 Follow Up and Monitoring ...... 10-71

April 2010 FINAL Page iii

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

10.11 Summary of Effects on the Biophysical Environment ...... 10-72 11 Effects of Hydrocarbons on the Human Environment ...... 11-1 11.1 Approach ...... 11-1 11.2 Heritage Resources ...... 11-1 11.2.1 Baseline Conditions ...... 11-1 11.2.2 Potential Effects of a Hydrocarbon Release on Heritage Resources ...... 11-2 11.2.3 Mitigation Measures ...... 11-3 11.2.4 Follow-up Monitoring ...... 11-3 11.3 Traditional Marine Resource Use ...... 11-4 11.3.1 Baseline Conditions ...... 11-5 11.3.2 Other Communities of Interest ...... 11-8 11.3.3 Potential Effects of a Hydrocarbon Release on Traditional Marine Use ..... 11-8 11.3.4 Mitigation Measures ...... 11-8 11.3.5 Follow-up Monitoring ...... 11-9 11.4 Non-traditional Marine Use ...... 11-9 11.4.1 Baseline Conditions ...... 11-10 11.4.2 Potential Effects of a Hydrocarbon Release on Non-Traditional Marine Use ...... 11-19 11.4.3 Mitigation Measures ...... 11-21 11.4.4 Follow-up Monitoring ...... 11-21 11.5 Socio-Economic ...... 11-21 11.5.1 Baseline Conditions ...... 11-22 11.5.2 Potential Effects of a Hydrocarbon Release on Socio-economic Conditions ...... 11-25 11.5.3 Mitigation Measures ...... 11-26 11.5.4 Follow-up and Monitoring ...... 11-27 11.6 Summary of Effects on the Human Environment ...... 11-27 12 Mass Balance Examples for Response Planning ...... 12-1 12.1 Introduction ...... 12-1 12.2 General Spill Response ...... 12-1 12.3 General Mitigation ...... 12-2 12.4 Selection of Mass Balance Examples for the CCAA and OWA ...... 12-3 12.4.1 Description of a Mass Balance Example ...... 12-6 12.5 Example 1: Wintertime Spill at Emilia Island ...... 12-7 12.5.1 Spill Characteristics ...... 12-7 12.5.2 Mitigation ...... 12-9 12.5.3 Potential Effects on Key Resources at Risk ...... 12-11 12.6 Example 2: Summertime Spill at Principe Channel ...... 12-13 12.6.1 Spill Characteristics ...... 12-13 12.6.2 Mitigation ...... 12-15 12.6.3 Potential Effects on Key Resources at Risk ...... 12-17 12.7 Example 3: Summertime at Wright Sound ...... 12-18 12.7.1 Spill Characteristics ...... 12-19 12.7.2 Mitigation ...... 12-21

Page iv FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

12.7.3 Potential Effects on Key Resources at Risk ...... 12-24 12.8 Example 4: Wintertime Spill at Ness Rock in Caamaño Sound ...... 12-26 12.8.1 Spill Characteristics ...... 12-26 12.8.2 Mitigation ...... 12-28 12.8.3 Potential Effects on Key Resources at Risk from a Spill in Caamaño Sound ...... 12-30 12.9 Example 5: Summer Spill at Butterworth Rocks in North Hecate Strait ...... 12-32 12.9.1 Spill Characteristics ...... 12-32 12.9.2 Mitigation ...... 12-34 12.9.3 Potential Effects from a Spill at Butterworth Rocks ...... 12-36 12.10 Example 6: Medium Size Diluted Bitumen Spill at Kitimat Terminal ...... 12-38 12.10.1 Spill Characteristics ...... 12-38 12.10.2 Mitigation ...... 12-40 12.10.3 Potential Effects on Key Resources at Risk from a Diluted Bitumen Spill at the Terminal ...... 12-42 12.11 Example 7: Medium Size Condensate Spill at the Terminal ...... 12-44 12.11.1 Spill Characteristics ...... 12-44 12.11.2 Mitigation ...... 12-46 12.11.3 Potential Effects on Key Resources at Risk ...... 12-47 12.12 Summary...... 12-49 13 Risk Assessment Related to Hydrocarbons in the Marine Environment ...... 13-1 13.1 Introduction ...... 13-1 13.1.1 Hypothetical Example ...... 13-1 13.1.2 Assessment Area...... 13-2 13.2 Ecological Risk Assessment...... 13-2 13.2.1 Methodology for the ERA ...... 13-3 13.2.2 Exposure Assessment ...... 13-8 13.2.3 Hazard Assessment ...... 13-10 13.2.4 Results of the ERA ...... 13-10 13.2.5 Prediction Confidence ...... 13-19 13.3 Human Health Risk Assessment ...... 13-20 13.3.1 Methods for the HHRA ...... 13-20 13.3.2 Results of the HHRA ...... 13-21 13.3.3 Prediction Confidence ...... 13-24 13.4 Follow-up and Monitoring ...... 13-29 14 Vapour Cloud Simulations and Risk Assessment ...... 14-1 14.1 Hazard scenario definition ...... 14-1 14.2 Dispersion, Fire and Explosion Analysis ...... 14-2 14.3 Ignition Frequency Evaluation ...... 14-3 14.4 Risk Analysis ...... 14-4 15 Mitigation and Amelioration ...... 15-1 15.1 Shipping Risk Mitigation ...... 15-1 15.1.1 Escort Tugs ...... 15-3 15.1.2 Improvements to Navigational Aids ...... 15-3

April 2010 FINAL Page v

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Table of Contents

15.1.3 Electronic Chart Display and Information System (ECDIS) ...... 15-4 15.1.4 Improvements to Vessel Traffic Service (VTS) ...... 15-4 15.1.5 Traffic Separation ...... 15-5 15.2 Terminal Risk Mitigation ...... 15-5 15.2.1 Closed Loading (with Vapour Return System) ...... 15-7 15.2.2 Ignition Source Prevention and Suppression ...... 15-7 16 Conclusions and Recommendations ...... 16-1 17 References ...... 17-1 17.1 Literature Cited ...... 17-1 17.2 Internet Sites ...... 17-25 17.3 Personal Communication ...... 17-33 Appendix A Figure References ...... A-1 Appendix B General Marine Mammmal and Marine Bird Species Lists for the Open Water Area ...... B-1

Page vi FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk List of Tables

List of Tables

Table 2-1 Coastal Settlements Proximate to the Proposed Tanker Routes ...... 2-4 Table 2-2 Summary of Wave, Wind and Surface Conditions in the Assessment Area ...... 2-7 Table 2-3 Visibility at Locations along the Shipping Routes ...... 2-9 Table 3-1 International Regulations That Will Apply in the Near Future ...... 3-3 Table 3-2 Fire Suppression Operating Parameters ...... 3-4 Table 4-1 Route Distances and Approximate Sailing Times ...... 4-3 Table 4-2 Assumed distribution of ship traffic to and from the Kitimat Marine Terminal ...... 4-4 Table 4-3 Base/Worldwide Frequencies for Main Incident Types for Oil Tankers per Nautical Mile ...... 4-6 Table 4-4 Summary of Scaling Factors ...... 4-7 Table 4-5 Scaling factors for incidents considered along the marine tanker routes ...... 4-8 Table 4-6 Total unmitigated and scaled incident frequency per nautical mile for each incident type for each route segment ...... 4-9 Table 4-7 Risk reducing effect of using escort tugs/tethered ...... 4-10 Table 4-8 Material Damage from Accidents and Conditional Probability of a Spill ...... 4-12 Table 4-9 Mitigated Return Period (Years) of an incident resulting in a release of cargo (including oil, condensate or bunker) per route segment, based on average forecast traffic ...... 4-14 Table 4-10 Oil spill return periods for forecasted route choices with different use of tugs ...... 4-15 Table 4-11 Return period of a spill associated with the tanker traffic for the Northern Gateway Project ...... 4-15 Table 4-12 Potential Spill Size Distribution for Unmitigated / Mitigated Spill Risk ... 4-17 Table 4-13 Summary of Mitigated and Unmitigated Return Periods for Spills occurring during tanker operation along the preferred marine routes ... 4-20 Table 5-1 Mitigated Return Period of a Spill from a Tanker at Berth by Cargo Type and Release Volume ...... 5-5 Table 6-1 Physical Properties of Project Hydrocarbons in the Marine Environment ...... 6-1 Table 6-2 Chemical Properties of Liquid Hydrocarbons ...... 6-2 Table 6-3 Hydrocarbon Evaporation Estimates ...... 6-6 Table 7-1 Typical Sequence of Initial Response Actions ...... 7-5 Table 7-2 Examples of Response Objectives and Strategies ...... 7-8 Table 7-3 Overview of OSRP Equipment ...... 7-12

April 2010 FINAL Page vii

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk List of Tables

Table 7-4 Proposed Hydrocarbon Recovery Capability ...... 7-13 Table 7-5 Examples of Sensitive Shoreline Areas...... 7-15 Table 8-1 Ecosections in the OWA ...... 8-1 Table 10-1 Overview of Effects of Hydrocarbons on Marine Biota and Terrestrial Wildlife ...... 10-2 Table 10-2 Marine Invertebrates Associated with Ecosections in the OWA ...... 10-14 Table 10-3 Important Areas for Fish in the OWA and CCAA ...... 10-22 Table 10-4 Important Bird Areas Associated with Ecosections in the OWA and CCAA ...... 10-41 Table 10-5 Marine Mammals and Turtles Associated with Ecosections in the OWA and CCAA ...... 10-55 Table 11-1 Communities, Indian Reserves and Rural Areas included in the OWA ...... 11-23 Table 11-2 Income Dependencies (After Tax) of Communities and Regions in the OWA, 2006 ...... 11-25 Table 11-3 Summary of Areas and their Uses within the Open Water Area ...... 11-28 Table 12-1 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill off Emilia Island during Winter ...... 12-8 Table 12-2 Response for Emilia Island Example ...... 12-9 Table 12-3 Mass Balance Estimates for a Hypothetical 10,000 m3 Diluted Bitumen Spill in Principe Channel during Summer ...... 12-14 Table 12-4 Response for Principe Channel Example ...... 12-15 Table 12-5 Mass Balance Estimates for a Hypothetical 36,000 m3 Diluted Bitumen Spill at Wright Sound during Summer ...... 12-20 Table 12-6 Response for Wright Sound Example ...... 12-21 Table 12-7 Mass Balance for a Hypothetical 10,000 m3 Diluted Bitumen Spill at Caamaño Sound during Winter ...... 12-27 Table 12-8 Response for Caamaño Sound Example ...... 12-28 Table 12-9 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill at Butterworth Rocks during Summer ...... 12-34 Table 12-10 Response for Butterworth Rocks Example ...... 12-34 Table 12-11 Mass Balance for a Hypothetical 250 m3 Diluted Bitumen Spill at the Terminal during Summer ...... 12-39 Table 12-12 Emergency Response for Diluted Bitumen ...... 12-40 Table 12-13 Mass Balance Estimates for a Hypothetical 250 m3 Condensate Spill at the Terminal during Summer ...... 12-45 Table 12-14 Emergency Response for a Condensate ...... 12-46 Table 13–1 COPCs from a Hypothetical Diluted Bitumen Example to the Marine Environment in Wright Sound ...... 13-5 Table 13–2 Receptors for the Marine Environment in Wright Sound ...... 13-12

Page viii FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk List of Tables

Table 13-3 Fisheries Closures Following Recent Oil Spills ...... 13-22 Table 13-4 Analysis of Assumptions Used in the HHRA ...... 13-25 Table 14-1 Hazard scenario summary ...... 14-2 Table 14-2 Fire and explosion analysis results summary ...... 14-3 Table 14-3 Summary of individual risk assessment results ...... 14-5 Table A-1 Oceanographic Summary Figure References...... A-3 Table A-2 Benthos Summary Figure References ...... A-5 Table A-3 Marine and Anadromous Fish Summary Figure References ...... A-7 Table A-4 Marine Birds Summary Figure References ...... A-8 Table A-5 Marine Mammals and Turtles Summary Figure References ...... A-9 Table A-6 Human Use Summary Figure References ...... A-11 Table B-1 Marine Birds Associated with Ecosections in the OWA ...... B-3 Table B-2 Marine Mammals and Turtles with Ecosections in the OWA ...... B-27

April 2010 FINAL Page ix

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk List of Figures

List of Figures

Figure 2-1 Study Area for TERMPOL Study 3.15 ...... 2-2 Figure 2-2 Socio-Economic Region used for the TERMPOL Study 3.15 ...... 2-3 Figure 4-1 Shipping Routes and Corresponding Segments ...... 4-2 Figure 4-2 Mitigated spill return frequencies per segment for tankers transporting Crude Oil and Condensate respectively ...... 4-16 Figure 4-3 Accumulated frequency of spills exceeding a certain size; Unmitigated / Mitigated ...... 4-18 Figure 4-4 Return Period for a Total Loss Event ...... 4-19 Figure 5-1 Proposed location of the Kitimat Terminal ...... 5-2 Figure 5-2 Proposed layout of the Kitimat Terminal ...... 5-3 Figure 5-3 Proposed Turning Basins, Navigational Clearances and Vessel Manoeuvres (TERMPOL 3.10) ...... 5-4 Figure 5-4 Comparison of unmitigated and mitigated spill return periods for releases during cargo transfer at the marine terminal...... 5-6 Figure 7-1 Example of Contingency and Response Plan Integration ...... 7-4 Figure 7-2 Sample Coastal Sensitivity Map ...... 7-6 Figure 7-3 Typical Response Activities for Hydrocarbons that Reach the Marine Environment ...... 7-10 Figure 8-1 Ecosections within the Open Water Area ...... 8-3 Figure 8-2 Winter Currents in the Open Water Area ...... 8-5 Figure 8-3 Summer Currents in the Open Water Area ...... 8-6 Figure 8-4 Important Ecological and Biological Areas in the Open Water Area ...... 8-7 Figure 10-1 Summary of Biologically Important Areas for Marine Benthos in the OWA ...... 10-13 Figure 10-2 Summary of Biologically Important Areas for Ground, Pelagic and Anadromous Fishes in the OWA ...... 10-25 Figure 10-3 Summary of Biologically Important Areas for Marine Birds in the OWA ...... 10-40 Figure 10-4 Summary of Biologically Important Areas for Marine Mammals in the OWA ...... 10-54 Figure 11-1 Summary of Human Use in the Open Water Area ...... 11-11 Figure 11-2 Recreational Areas in the CCAA and on Nearby Land and Water ...... 11-12 Figure 11-3 Conservancies, Marine Parks, Parks and Protected Areas in and Near the CCAA ...... 11-18 Figure 12-1 Mass Balance Example Locations ...... 12-4 Figure 12-2 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill off Emilia Island during Winter ...... 12-8

April 2010 FINAL Page xi

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk List of Figures

Figure 12-3 Mass Balance Estimates for a Hypothetical 10,000 m3 Diluted Bitumen Spill in Principe Channel During Summer ...... 12-14 Figure 12-4 Mass Balance Estimates for a Hypothetical 36,000 m3 Diluted Bitumen Spill at Wright Sound during Summer ...... 12-20 Figure 12-5 Mass Balance for a Hypothetical 10,000 m3 Diluted Bitumen Spill at Caamaño Sound during Winter ...... 12-27 Figure 12-6 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill at Butterworth Rocks during Summer ...... 12-33 Figure 12-7 Mass Balance for a Hypothetical 250 m3 Diluted Bitumen Spill at the Terminal During Summer ...... 12-39 Figure 12-8 Mass Balance Estimates for a Hypothetical 250 m3 Condensate Spill at the Terminal during Summer ...... 12-45 Figure 13-1 Conceptual Exposure Model for Marine Key Indicators ...... 13-9 Figure 13-2 Acute Hazard Index at the Water Community-Level for a 36,000 m3 Hypothetical Diluted Bitumen Example in Wright Sound during Summer...... 13-18 Figure 13-3 Chronic Hazard Index at the Subtidal Sediment Community-Level for a 36,000 m3 Hypothetical Diluted Bitumen Example in Wright Sound during Summer ...... 13-18 Figure 13-4 HQ Estimates for Benzene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer ...... 13-26 Figure 13-5 ILCR Estimates for Benzene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer ...... 13-27 Figure 13-6 HQ Estimates for C>16-C21 Aromatics – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer ...... 13-28 Figure 13-7 ILCR Estimates for Benzo(a)anthracene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer ...... 13-29 Figure 14-1 Event tree ...... 14-4 Figure 14-2 Terminal collective risk spectrum ...... 14-6 Figure 16-1 Accumulated frequency of spills exceeding a certain size; Unmitigated / Mitigated ...... 16-2

Page xii FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Abbreviations

Abbreviations

AIR ...... Average Individual Risk ADEC ...... Alaska Department of Environmental Conservation AIR ...... Average Individual Risk ALARP ...... As Low as Reasonably Practicable BC ...... British Columbia BCMEC ...... British Columbia marine ecological classification BTEX ...... Benzene, toluene ethylbenzene, and xylenes CANUSDIX ...... Canadian US Dixon Entrance CCAA ...... Confined channel assessment area CCG ...... Canadian Coast Guard CEA Act ...... Canadian environmental Assessment Act CLC ...... Civil liability convention CMT ...... Command management team COPC ...... Contaminant of potential concern COSEWIC ...... Committee on the status of endangered wildlife in canada COTP ...... Captain of the port (uscg) CP ...... Contingency plan CRIMS ...... Coastal resources information management system CRW ...... Condensate Blend CRW-824 CRW ...... Condensate Blend CRW-824 CSA ...... Canada Shipping Act, 2001 CSRF ...... Collective Specific Risk Factors DNV ...... Det Norske Veritas EBSA ...... Ecologically and biologically significant area EC ...... Environment Canada ECDIS ...... Electronic chart display and information systems ERA ...... Ecological risk assessment FL ...... Flammability Limit FMEA ...... Failure Modes and Effects Analysis FMO ...... Federal monitoring officer (CCG) FOSC ...... Federal on scene coordinator (USCG) FSC ...... Food, social and ceremonial GOSRP ...... General oil spill response plan GRP ...... Geographic response plan HADD ...... Harmful alteration, disruption rr destruction of fish habitat HAZID ...... Hazard Identification IAs ...... Important areas ICS ...... Incident command system IMO ...... International maritime organization (united nations) IOPC ...... International oil pollution compensation (fund)

April 2010 FINAL Page xiii

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Abbreviations

IR ...... Individual Risk IRI ...... Individual Risk Intensity. IRPA ...... Individual Risk per Annum Isopleth ...... A line of constant value of a variable such as a thermal isopleth of 12.5 kW/m2. ISR ...... Individual Specific Risk. ITOPF ...... International tanker owners pollution federation KI ...... Key indicator LFL ...... Lower Flammability Limit LOMA ...... Large ocean management areas LRFP ...... Lloyd‘s Register Fairplay LRIT ...... Long range identification and tracking information MAH ...... Monocyclic aromatic hydrocarbons MARPOL Convention ...... International convention for the prevention of pollution from ships MCTS ...... Marine communication and traffic services MCTS ...... Marine communication and traffic services MEPC ...... Marine environment protection committee MLA ...... Marine Liability Act MoE ...... Ministry of environment (bc) MSC ...... Maritime safety committee MSQM ...... Marine water quality model MSQM ...... Marine sediment quality model NEB ...... National Energy Board NGPP ...... Northern Gateway Pipeline Project NRC ...... National Research Council OMA ...... Oil-mineral aggregates OPEP ...... Oil pollution emergency plan OPRC...... International convention on oil pollution preparedness, response and co-operation OSC ...... On scene commander OSCP ...... Oil spill contingency plan OSR ...... Oil spill response OSRP ...... Oil spill response plan OWA ...... Open water area P and I ...... Protection and indemnity (club) PAH ...... Polycyclic aromatic hydrocarbon PAR ...... Primary area of response PHC ...... Petroleum hydrocarbon PNCIMA ...... Pacific north coast integrated management area QCB ...... Queen charlotte basin QCI ...... Queen Charlotte Islands (Haida Gwaii) QRA ...... Quantitative risk assessment

Page xiv FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Abbreviations

QRA ...... Quantitative Risk Analysis REET ...... Regional environmental emergencies team RFP ...... Request for Proposal RMS ...... Response management system (CCG) RO ...... Response organization ROO ...... Ratio of Occurrence RP ...... Responsible party SARA ...... Species at Risk Act SCADA ...... Supervisory control and data acquisition SER ...... Socio-economic region SERVS ...... Ship escort response vessel system (Alyeska - Alaska) SMT ...... Spill management team SOLAS ...... International convention for safety of life at sea SOPEPs ...... Ship oil pollution emergency plans SOPF ...... Ship-source oil pollution fund TC ...... Transport canada TDR ...... Technical data report TERMPOL ...... Transport Canada technical review process for marine terminal systems and transshipment sites the Project ...... Enbridge northern gateway project TOM ...... Terminal/ transshipment site operations manual TPH ...... Total petroleum hydrocarbon TRACE ...... A multi-purpose consequence analysis software model developed by DuPont and sold by Safer Systems. TRP ...... Termpol review process UC ...... Unified command UFL ...... Upper Flammability Limit USCG ...... United States Coast Guard VDR ...... Voyage data recorder VEC ...... Valued environmental component VLCC ...... Very large crude carrier VOC ...... Volatile organic compound

April 2010 FINAL Page xv

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 1: Introduction

1 Introduction An unmitigated liquid hydrocarbon spill could have acute (immediate) and chronic (months to years) effects on the health of wildlife, fish and humans and reduce habitat quality. This study quantitatively determines the risk of hydrocarbon spills occurring during transport of hydrocarbons to and from the Kitimat Terminal and during transshipment at the terminal, and identifies the potential effects on the biophysical and human environments. It is expected that such incidents would be unlikely (see Section 4 and 5) because the probability is low, and risk is further reduced by the accident prevention and mitigation measures implemented by Northern Gateway. This study identifies, predicts, evaluates effects, and summarizes the risks and mitigation of effects on the relevant biophysical and social components potentially affected by the proposed project. Several examples were developed to evaluate potential effects for spill response planning. Mass balance examples examine how hydrocarbons move into compartments such as air, water, and shoreline. These examples were developed for specific locations along the tanker approaches, as well as the types of accidents that could result in hydrocarbons entering the marine environment. Assessment of the examples will aid in developing response plans and mitigating potential environmental effects. An ecological risk assessment (ERA) and human health risk assessment (HHRA) are also included, using the example of a spill of diluted bitumen from a vessel collision in Wright Sound during summer. An ERA developed for a spill of diluted bitumen and of condensate at the Kitimat Terminal describes potential effects on the environment and is generally applicable to a small spill along the tanker approaches. The effects of hypothetical groundings or collisions are discussed, as these encompass the effects of hydrocarbons entering the marine environment. Although the probability of such incidents is small, the scenario examples provide realistic descriptions of: key design and operational features to avoid incidents appropriate mitigation (or contingency) measures to limit the occurrence of incidents the behaviour of hydrocarbons in the environment the probability of an incident and further measures to reduce probability effective response plans and equipment potential biophysical, social and economic effects of hydrocarbons in the environment The hypothetical mass balance examples consider the following: behaviour and fate of released hydrocarbons at three locations in the confined channel assessment area (CCAA) and two locations in the open water area (OWA). Selection of these locations was based on potential navigational hazards and environmentally sensitive areas approximate maximum volume of a spill due to grounding or collision of a laden double-hulled tanker (with complete loss over time of hydrocarbons from multiple storage compartments). physical and chemical characteristics of the three types of hydrocarbon products to be transported in vessels (diluted bitumen, synthetic oil and condensate)

April 2010 FINAL Page 1-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 1: Introduction

a range of environmental conditions (tide and winds) at the time of an incident that reflects seasons during which effects could be greatest valued environmental components (marine and human) likely to be affected

Regulatory guidance and requirements are considered in development of the examples of incidents. Transport Canada has established standards and guidelines for oil tankers. The International Convention for the Prevention of Pollution from Ships (MARPOL Convention), to which Canada is a party, sets out criteria for oil tankers. Small spills from vessel operations (e.g., through ballast or bilge water discharges) in the OWA are unlikely to occur, given adherence to these policies and procedures. An Oil Pollution Prevention Plan (OPPP) and an Oil Pollution Emergency Plan (OPEP) will be prepared well in advance of operational start-up and would be implemented to manage risks related to hydrocarbon spills that might occur during vessel transport in Canadian waters (TERMPOL 3.18).

1.1 Objective In accordance with the TERMPOL Code, TP473E 2001, the objective of TERMPOL Study 3.15 is to characterize the risks associated with accidental release events either en route or at the marine terminal site. This is accomplished by providing information on: key measures to prevent accidental vessel and marine terminal based hydrocarbon spills the likelihood of accidental hydrocarbon spills (quantitative risk assessment) emergency response planning, response and recovery characteristics of hydrocarbons that will be handled during marine transportation, including their fate in the marine environment (i.e., evaporation, dissolution and emulsification) potential effects on the biophysical and human environments in the unlikely event of a spill, including the need for and approaches to follow-up and monitoring credible worst case examples of hydrocarbon spills to develop emergency response planning.

1.2 Existing Documentation and Project-Specific Baseline Studies Literature reviews included assessments of other projects near the Project and similar projects elsewhere; geographic and technical databases; and scientific literature to identify relevant issues. In addition, Northern Gateway commissioned a number of studies to characterize the marine environment and risks of shipping hydrocarbon products; these studies, compiled in Technical Data Reports (TDRs), were used to write this volume: Marine Physical Environment TDR (ASL 2010a) Weather and Oceanographic Conditions at Sites in the CCAA and in Queen Charlotte Sound, Hecate Strait and Dixon Entrance TDR (ASL 2010b) Marine Fish and Fish Habitat TDR (Beckett and Munro 2010)

Page 1-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 1: Introduction

Marine Mammals TDR (Wheeler et al. 2010) Marine Birds TDR (d'Entremont 2010) Risk Assessment for Accidental Hydrocarbon Spills at the Kitimat Terminal and in the Confined Channel Assessment Area TDR (Stephenson et al. 2010a) Marine Ecological Risk Assessment for Kitimat Terminal Operations TDR (Stephenson et al. 2010b) Properties and Fate of Hydrocarbons from Hypothetical Spills in the Confined Channel Assessment Area and at the Marine Terminal TDR (SL Ross 2010a) Properties and Fate of Hydrocarbons Associated with Hypothetical Spills at Three Sites in Open Water TDR (SL Ross 2010b) Wind Observations in Douglas Channel, Squally Channel and Caamaño Sound TDR (Hay & Company Consultants 2010) Coastal Operations and Sensitivity Mapping of the Open Water Area TDR (Polaris 2010a) Coastal Operations and Sensitivity Mapping of the Confined Channel Assessment Area TDR (Polaris 2010b) Non-traditional Land Use TDR (Karki and Hamelin 2010) Quantitative Risk Analysis (Det Norske Veritas [DNV] 2010) Vapour Cloud Analysis (Bercha Engineering Limited 2010) TDRs highly relevant to TERMPOL 3.15 are included in Volume 2, additional TDRs will be made available by Northern Gateway, either by request, or through the Northern Gateway website.

April 2010 FINAL Page 1-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

2 Project and Site Description

2.1 Project Overview Northern Gateway Pipelines Limited Partnership (Northern Gateway) proposes to construct and operate the Enbridge Northern Gateway Project (the Project), which will comprise the following major facilities: an oil (diluted bitumen and synthetic oil) export pipeline a condensate import pipeline (for the purposes of this Application, condensate is defined as low volatility hydrocarbon equivalent to diluent) a tank terminal and marine terminal near Kitimat, British Columbia (referred to as the Kitimat Terminal). The marine terminal will accommodate loading oil tankers and discharging of condensate tankers. Northern Gateway will construct and operate the pipelines and the Kitimat Terminal. Third parties will operate the tankers. Northern Gateway considers accident prevention and environmental protection to be key commitments for the Project, and will implement specific protection measures during engineering design, and through operational plans, to reduce the potential for, and effects of, accidents and malfunctions, including liquid hydrocarbon spills. Northern Gateway has committed to assessing environmental effects of a potential hydrocarbon spill in the confined navigation channels of Kitimat Arm, Douglas Channel and northern and southern vessel routes into Douglas Channel (the Confined Channel Assessment Area or CCAA) and along the shipping routes in the open water area (OWA), as shown in Figure 2-1. When the term assessment area is used it is referring to both the CCAA and OWA. The assessment focuses on potential spills during the operations phase of the Project.

2.2 Distribution of Communities along the Carrier Route (3.15.4) In general, the shipping routes coincide with four census divisions/subdivisions, shown in Figure 2-2. the Skeena-Queen Charlotte Regional District (Census Division), which includes Prince Rupert and Port Edward on the mainland and Masset and Queen Charlotte City on Haida Gwaii Subdivision D of the Kitimat-Stikine Region District (Census Subdivision), the District Municipality (DM) of Kitimat and Kitamaat Village (Kitamaat 2 Reserve) Central Coast Regional District (Census Subdivision), which includes Bella Bella and Bella Coola Mount Waddington Regional District (Census Subdivision) which includes Port Hardy, Port McNeill, Port Alice and Alert Bay on Vancouver Island and various small communities on the mainland

April 2010 FINAL Page 2-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Figure 2-1 Study Area for TERMPOL Study 3.15

Page 2-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Figure 2-2 Socio-Economic Region used for the TERMPOL Study 3.15

April 2010 FINAL Page 2-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

These areas are termed the Socio-Economic Region (SER) and include 27 Aboriginal reserves located on coastal areas or inland channels. Nine of these reserves, all in the Mount Waddington RD, do not have a resident population. Settlements along the proposed route from Kitimat to Dixon Entrance are summarized in Table 2-1.

Table 2-1 Coastal Settlements Proximate to the Proposed Tanker Routes

Approximate Adjacent Name Location Population Comments to Route Kitimat head of Kitimat Arm 8,987 an industrial centre that provides ~10 km deep sea port facilities for Methanex, Alcan and Eurocan Haisla 11 km south of Kitimat 514 on original human settlement of the ~ 5 km Nation in Kitimat Arm Reserve area, home to the Haisla Nation (Kitamaat 1,639 Total Village Council) Kemano southeast of Kitimat off - accessible only by helicopter, ~62 km the Gardener Canal seaplane or boat; produces extension of Douglas electrical power for Alcan and Channel northwest BC Gitga’at at the mouth of the 157 on- accessible only by air and water ~3km Nation Douglas Channel, 140 Reserve (Hartley km southeast of Prince 671 Total Bay) Rupert and 80 km southwest of Kitimat Barnard southern end of Whale - location of King Pacific Fishing ~10 km Harbour Channel Lodge Gitxaala 70 km southwest of 384 on- accessible only by air and water; ~13 km Nation Prince Rupert on Reserve community has developed an (Kitkatla) Dolphin Island, near 1,776 Total active aquaculture industry. Browning Entrance Prince 721 km west of Prince 12,815 western Canada’s largest northern ~47 km Rupert City George on Highway 16 deep sea port, with multi-billion dollar annual exports of coal, grain, lumber and potash; and growing with several proposals for port expansion. Currently, the Fairview Terminal is being modified to accept container vessels. Port Edward a few km east of Prince 577 declared a National Historic Site on ~47 km Rupert its 100th anniversary in 1989; site of BC’s oldest surviving salmon cannery

Page 2-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Table 2-1 Coastal Settlements Proximate to the Proposed Tanker Routes (cont’d)

Approximate Adjacent Name Location Population Comments to Route Metlakatla 6 km northwest of 109 On- accessible only by air and ~76 km Prince Rupert Reserve water;S1/2 Tsimpsean 2 Indian 796 Total reserve, includes the Annette Island Packing Company a 100% aboriginal owned packing plant which focused on specially local products Lax 56 km north of Prince 760 On- accessible only by air and water, ~48km Kw’alaams Rupert Reserve located in Port Simpson, primary 3,257 Total industry is related to fishing and fish processing. During the summer of 2008 the fish plant employed 70 people. Georgetown situated in Big Bay, - a small community supported by ~45km Mills between Port Simpson logging and sawmills and Metlakatla Skidegate centrally located in the 697 On- Skidegate Landing Ferry Terminal ~80 km Band Haida Gwaii Reserve is on Graham Island 1,466 Total Queen on Skidegate Inlet, 948 houses provincial and federal ~ 85km Charlotte southern shore of government offices, a Royal Graham Island (Haida Canadian Mounted Police (RCMP) Gwaii) station, general hospital, and a variety of businesses, shops, services and accommodations Sandspit on the northeastern tip 387 the only settlement on Moresby ~67 km of Moresby Island, 15 Island. There are some abandoned km east of Alliford Bay sites on Moresby Island: Moresby (Haida Gwaii) and Aero Camps are abandoned logging camps and the Haida settlement of Cumshewa, which was formerly abandoned due to smallpox, is now being resettled. The abandoned Haida villages of New Clew and Skedans on Louise Island south of Cumshewa Inlet also may be inhabited again, for at least part of the year. Lawnhill midway between - salmon hatchery at Lawnhill ~73 km Skidegate and Tlell (Haida Gwaii)

April 2010 FINAL Page 2-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Table 2-1 Coastal Settlements Proximate to the Proposed Tanker Routes (cont’d)

Approximate Adjacent Name Location Population Comments to Route Tlell Southeast corner of 187 headquarters for Naikoon ~75 km Naikoon Provincial Provincial Park, also a very active Park, 40 km north of artisan community the Skidegate ferry terminal (Haida Gwaii) Old Massett northern end of 686 on- the largest town on Haida Gwaii, ~50 km Village Graham Island, in Reserve with logging and fishing the main Council Dixon Entrance (Haida 2,698 Total industries in Massett, and Gwaii) increasingly important tourism industry Bella Bella Campbell Island, north 1,066 on- accessible only by air and water; ~125 km (Heiltsuk of Port Hardy located Reserve Port used by BC Ferries, Nation) on the mainland side of 2,192 Total ecotourism operators, and the local Queen Charlotte processing plant Bella Coola Sound Fisheries Ltd. Port Hardy Northeast tip of 3,822 Southern terminus of BC Ferry ~180 km Vancouver Island route to Prince Rupert

SOURCE: BC Tourism website, Statistics Canada

Other communities of interest include: Butedale –in Princess Royal Channel, about 30 km southwest of Wright Sound, was a salmon cannery, abandoned in the 1950s; there is still a small resident population and the area is popular with scientists who study the Kermode Bear Dodge Cove – across the water from Prince Rupert with a population of 42 Hunts Inlet –30 km south of Prince Rupert on the north coast of Porcher Island, with several ocean- side properties on the inlet, with boat landings and private beaches. Town of Port McNeill, Village of Port Alice, Village of Alert Bay and nine Indian reserves (Tsulquate 4, Alert Bay 1, Alert Bay 1A, Kippase 2, Quatsino Subdivision 18, Quaee 7, Gwayasdums 1, Hope Island 1, and Fort Rupert 1 – these communities, with a combined location of 5,588 in 2006, are located on the east and west sides of the northern end of Vancouver Island, or on islands in the north end of Queen Charlotte Strait.

Page 2-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

2.3 Relevant Climatic / Environmental Considerations Climatic conditions in the study area are discussed below, with a focus on aspects relevant to the quantitative risk assessment.

2.3.1 Waves, Wind and Current Table 2-2 summarizes data (maximum and average wave height, wind speed, surface current speed) from various weather stations (ASL Environmental Sciences 2009). Strong winds are more common during winter than at other times of year. The operational wind speed limit for berthing and deberthing worldwide is normally in the range of 25 to 40 knots. A preliminary wind speed limit in the range of 30 to 35 knots for berthing tankers is proposed in TERMPOL 3.13, in light of the fact that the prevailing winds and currents in the area of the Marine Terminal run parallel to Douglas Channel and the marine berths. Maximum environmental operating limits will be determined in consultation with pilots and pending detailed operational mooring analyses which will be conducted during the detailed design phase of this project.

Table 2-2 Summary of Wave, Wind and Surface Conditions in the Assessment Area

Measurement Meteorological and ocean Location period parameter Max Mean Queen Charlotte Sound 1989-2008 Significant wave heighta 14.5 m 2.7 m Wind speedb 25.3 m/s 7.2 m/s 1981-1982 Surface current speed 0.9 m/s 0.2 m/s 1990-1991, 1995 Dixon Entrance 1991-2008 Significant wave height 11.2 m 1.6 m Wind speed 24.0 m/s 6.7 m/s 1984-1985, 1991 Surface current speed 1.2 m/s 0.3 m/s Hecate Strait 1984-2008 Significant wave height 10.2 m 1.3 m 1991-2008 Wind speed 25.1 m/s 7.1 m/s 1997, 1983-1984 Surface current speed 1.1 m/s 0.3 m/s South Hecate Strait 1991-2008 Significant wave height 14.3 m 1.8 m Wind speed 28.0 m/s 6.6 m/s Nanakwa Shoal 1989-2009 Significant wave height 2.3 m 0.1 m Wind speed 28.0 m/s 4.6 m/s

NOTES: a Significant wave heights -Is the average height (trough to crest) of the one-third highest waves valid for the indicated 12 hour period b Higher wind speeds have been recorded at the Cape St James Station

SOURCE: ASL Environmental Sciences 2010

April 2010 FINAL Page 2-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Wind levels in the Project area were compared with those at similar operations in Norway (Norwegian Meteorological Institute 2009), and found to be similar. Wind should not constitute an unmanageable risk to oil tankers in the Project area, given that the wind limitations will be followed and that tethered escort tugs will be used in the CCAA. Significant wave heights are similar for all areas outside the CCAA (Table 2-2). Mean waves heights are limited to approximately 2 to 3 meters in the CCAA (e.g., at Nanakwa Shoals). Tankers, such as the types that will be accepted at the marine terminal, are constructed for world trade and regularly sail in areas with ocean wave conditions. In addition weather stations and weather forecasts will provide early warning of weather conditions that may exceed maximum environmental operating conditions and enable scheduling of ship movements to avoid excessive conditions. Maximum surface currents of up to 1 m/s, or 2 knots are found throughout the routes to and from the marine terminal. In the CCAA the surface currents will predominately run in the longitudinal direction of the channel and do not pose a challenge to navigation. Wind and currents may make controlling an emergency situation more challenging. Currents have a greater influence on laden tankers than tankers in ballast due to the larger draft, or portion of hull underwater and exposed to current. The reverse is true for wind. Surface currents are not assessed to constitute an increased risk to tanker operations compared to other areas in the world, such as terminals in western Norway. British Columbia Coast Pilots have intimate knowledge of the local currents and are able to safely guide tankers into Kitimat.

2.3.2 Visibility Reduced visibility renders navigation more difficult making it more difficult to judge the correctness of sound, distances, and movements. However, is should be noted that the use of modern navigation technology including AIS, DGPS, ECDIS and radar alleviates these difficulties. Generally visibilities lower than 1 nm are regarded as problematic for navigation and are reflected in safety limitations for operations at terminals. The visibility at locations near the proposed marine routes is shown below in Table 2-3. The operational limit for safe berthing and de-berthing manoeuvres will be in the range of 1 to 2 nm. Modern navigation technology including AIS, DGPS, ECDIS and radar will be used for transiting the CCAA and OWA which alleviates navigational difficulties associated with reduced visibility. The actual operational limit will be confirmed during detailed design and development of safe operating criteria with pilots. May to August is the period with the poorest visibility conditions.

Page 2-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 2: Project and Site Description

Table 2-3 Visibility at Locations along the Shipping Routes

Percentage of time below Location Measurement period 2 km (~1nm) visibility Queen Charlotte Sound Jan – Mar 9.2% (Cape St James ) Apr – Jun 8.6% Jul – Sep 14.4% Oct – Dec 10.4 Queen Charlotte Sound Jan – Mar 4.8% (Cape Scott) Apr – Jun 5.5% Jul – Sep 14.4% Oct – Dec 6.0 Dixon Entrance Jan – Mar 2.6% Apr – Jun 1.7% Jul – Sep 6.7% Oct – Dec 2.4% Hecate Strait 1 Jan – Mar 2.7% (Sandspit) Apr – Jun 1.0% Jul – Sep 1.3% Oct – Dec 2.0% Hecate Strait 2 Jan – Mar 3.4% (Bonilla Island) Apr – Jun 2.5% Jul – Sep 8.2% Oct – Dec 5.3% Hecate Strait 3 Jan – Mar 1.8% (Triple Island) Apr – Jun 1.5% Jul – Sep 5.8% Oct – Dec 1.3%

SOURCE: ASL Environmental Sciences 2010

April 2010 FINAL Page 2-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3 Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk Northern Gateway will implement an integrated approach during the planning, execution and operation of the Kitimat Terminal, shipping, and associated facilities to identify and address all applicable environmental, regulatory and statutory requirements. Northern Gateway will also monitor and document compliance with response requirements. These commitments will be documented in an OPPP and OPEP, which form part of the overall oil spill response plan (OSRP). During the operational life of the Project, incidents have the potential to occur because of accidents or malfunctions (e.g., ship grounding, ship collision), human error, vandalism, third-party damage or natural events such as severe weather. The potential for, and effects of, spills will be reduced through measures such as implementing modern tanker specifications, tanker operational plans and emergency response plans. Detailed versions of the operational plans will be prepared before the commissioning and operations of the marine terminal, and for ships calling on the Kitimat Terminal. The following sections describe some key operational measures to prevent project-related spills, and the contingency response plans that would be implemented.

3.1 Design Ship Safety Features In TERMPOL Study 3.9, Ship Specifications (TERMPOL 3.9), the designs of tankers that will be accepted at the marine terminal are described. In the following section these designs are assessed with respect to key safety features. The features which have been assessed are: Hull and cargo tank components Navigational equipment (early warning of potential collision or grounding situations) Fire prevention and fire fighting

3.1.1 Hull and Cargo Tank Components The tankers specified in TERMPOL 3.9 will be of double hull design. As per international guidelines to which Canada is a party, single hull tankers will be phased out of service worldwide by the time the Northern Gateway Project commences operation. The most common outcome of a collision or grounding involving a double hull tanker is a breach of the outer hull and no breach of the inner hull that contains the liquid cargo. The cargo tank arrangements described in TERMPOL 3.9 include the three most common tank arrangements for tankers. The arrangements vary in number of cargo tanks and some arrangements have cargo tanks that extend the width of the tanker (minus the ballast tanks). DNV recommends that cargo tank arrangements extending the width of the tanker (minus the ballast tanks) not be accepted. Northern Gateway has accepted this recommendation and tankers calling at Kitimat Terminal should be equipped with tanks with at least one centre partitions. Increasing the number of tanks reduces the volume per tank and the cargo spilled if the inner hull is penetrated.

April 2010 FINAL Page 3-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3.1.2 Navigational Equipment A vessel registered in a country that has ratified the International Maritime Organisations (IMO) International Convention for Safety of Life at Sea (SOLAS) is fitted with the navigation equipment and systems required in SOLAS, Chapter V, Regulation 19, and will also satisfy the navigation bridge visibility requirements in Regulation 22. The SOLAS convention's general purpose is to ensure that a ship is fit for the service for which it is intended. TERMPOL 3.9 states that all tankers calling at the marine terminal must satisfy SOLAS requirements.

3.1.3 Fire Prevention and Fire Fighting As described in TERMPOL 3.9 tankers will be equipped with fire prevention and firefighting systems per international rules and regulations. All tankers operating around world must meet the same standard.

3.1.4 Oil Tanker Vulnerability to Shipboard Incidents TERMPOL 3.9 specifies that the oil tankers will have a double hull design. In the event of a collision or grounding, the double hull will, in most low impact cases, lead to breaching of only the outer hull and damage to the external (ballast water) tanks between the outer hull and cargo tanks The ship specification (TERMPOL 3.9) describes the three most common cargo tank arrangements for oil tankers. These vary in number of cargo tanks and orientation. While some specified designs permit the cargo tank (but not ballast tanks) to span the width of the oil tanker, this type of construction will not be accepted for tankers calling on the Kitimat Terminal, as the limited number of tanks with this arrangement increases the potential amount of oil that could be spilled if a cargo tank is penetrated (DNV 2010).

3.1.5 Future Oil Tanker Requirements The international rules and requirements, including those applicable for oil tankers, are under constant improvement. New requirements for all ship types, especially oil tankers, are ratified at regular intervals. New requirements are most often implemented over a number of years after they are introduced to allow time for ship owners and designers to adjust to new equipment and design requirements for both existing and new vessels. Table 3-1 lists new requirements that have been approved and will come into force in the near future. Many of these will be in place at the time the Kitimat Terminal and pipeline become operational. In addition to these requirements, single hull oil tankers are scheduled to be phased out prior to 2015, proposed commencement of operations, leaving only double hull oil tankers available for the Project. In addition to the requirements listed above it is worth mentioning that all single hull tankers are scheduled to be phased out of operation by 2010 leaving only double hull tankers in the worldwide tanker fleet. TERMPOL 3.9 describes most modern tankers currently trading internationally. All vessels to be accepted at the Kitimat Terminal will meet IMO regulations and classification society rules. As such, the vessels accepted at the Kitimat Terminal will be fit to carry cargo and transit the waters off the BC coast and the open ocean.

Page 3-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

Table 3-1 International Regulations That Will Apply in the Near Future

Date of entry Subject Description into force VOC Volatile Organic Compounds (VOC) Management Plan shall be 2010-07-01 management approved VDR fitting Final date for fitting a Voyage Data Recorder (VDR), which may be an 2010-07-01 SVDR (simplified voyage data recorder). ECDIS Draft regulations to make mandatory the carriage of Electronic Chart 2015-07-01 Display and Information Systems (ECDIS) and Bridge Navigational Watch Alarm System (BNWAS), under SOLAS chapter V Safety of Navigation, were agreed by the Sub-Committee on Safety of Navigation (NAV) when it met for its, 54th session. The proposed new regulations were submitted to the Maritime Safety Committee (MSC) for approval at its 85th session in November-December 2008, and adopted by MSC 86 in May 2009. Ballast water After this date, all vessels with more than 5,000 m3 ballast water 2016-01-01 treatment capacity must install ballast water treatment to ensure that any organisms in the ballast water are killed before water is pump out of the ballast tanks. A number of designs are awaiting approval.

It is noted that TERMPOL 3.9 specifies that vessels are to be of less than 20 years of age and have double hull construction. As discussed in the ship specification most tankers will from 2010 and onwards meet those requirements.

3.2 Marine Terminal Safety and Monitoring Equipment The marine terminal will be equipped with the latest in safety and monitoring systems to help ensure controlled and safe loading and discharge operations. Specific design measures to prevent spills will include: designing the marine terminal to comply with applicable codes and standards developing materials specifications for components to be used in construction of the marine terminal developing an engineering plan during the early stages of detailed design that describes the design process, specifies the need for constructability and operability assessments, and outlines the methods for design review and verification, including the process for checking produced documents (i.e., the engineering quality control process) A brief description of some of the safety systems are provided below. The systems are described in detail in TERMPOL 3.10 and TERMPOL 3.11.

3.2.1 Gas Monitoring

Gas alarms to detect H2S and other vapours will be installed throughout the terminal to detect gas well before an explosive condition develops.

April 2010 FINAL Page 3-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3.2.2 Fire Detection and Fire Fighting Fire fighting systems will be provided at both tanker berths to extinguish a fire within the area of the berth platforms and the immediate vicinity of the loading arm connection to the ship‘s manifold. Firefighting equipment includes water and foam monitors located on the main loading platforms. Fire hydrants and monitors will be located on the berth structures and throughout the site as detailed in TRP Study 3.11, Section 3.4. The fire suppression operating parameters are summarized, as shown in Error! Reference source not found..

Table 3-2 Fire Suppression Operating Parameters

Item Metric Units Imperial Units Fire Water Flow Rate 17,010 L/min. 4,400 US gpm (including Foam Solution) Fire Water Pump Discharge 1896 kPa 275 psig Fire Water Pond Volume 5,500 m3 1,400,000 US gal. Total Foam Solution Volume 90 m3 24,000 US gal. Total Foam Concentrate Volume 3.8 m3 1,000 US gal.

The fire water pond will be sized to provide a minimum of four hours of supply. There will be one electric-motor-driven water pump and one back-up water pump driven by a diesel engine in case of power failure. Given the proximity of the firewater pond to the foreshore area, additional stand-by firewater pumps that direct-draw seawater are proposed. One pump is proposed for each tanker berth. These additional pumps will serve as an emergency firewater source back-up system in the event the feed from the firewater pond is interrupted. The pumps will tie into the main firewater distribution system at each tanker berth.

3.2.3 Control Room Monitoring The Kitimat Control Centre is located onshore near the water. It will monitor and control the cargo transfer operations at the marine terminal, as well as the associated terminal equipment. This control centre will also provide primary oversight of the entire terminal including the reception of all operating data, the capability of controlling valves and the monitoring of all security systems. A system redundancy back-up plan will be evaluated during detailed design. It is proposed that an alternate centre will be tied in as a fully redundant back-up control centre that is separate from the primary site. Ship-to-shore communications will be maintained throughout the entire cargo transfer operation.

Page 3-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3.2.4 Quick Release and Load Monitoring of Mooring Lines Quick release hooks are standard equipment for marine oil terminals, providing a safe means of securing a vessel alongside the berth and, in an emergency situation, can allow for the rapid release of mooring lines even with the mooring lines under load. Mooring line load monitoring equipment will be installed at the marine terminal to measure the load on the mooring lines in real time and warn terminal operators if loads are increasing to an unsafe level, the ship‘s duty officer will be advised and if required tug boats readied for support..

3.2.5 Metocean Monitoring System Meteorological and oceanographic monitoring equipment will be installed at the marine terminal and at other points along all the ship routes. These sensors will provide real time data on wind speed, wind direction, barometric pressure, temperature, and visibility. In addition, the terminal equipment will monitor tidal changes, wave height, wave direction, current speed, and current direction. The information gathered by the weather sensors will be used to guide decisions by ship masters, pilots and terminal operations personnel.

3.2.6 Docking Monitoring System Kitimat Terminal will install a docking monitoring system to assist in docking and undocking tankers. This system provides feedback information to the pilot and ship‘s crew in order to facilitate the safe berthing of the vessel. The docking system is used by the pilots and terminal operators to assist in vessel berthing over the final 200 to 300 metres of approach. Laser sensors measure the vessel‘s approach speed, distance and angle with respect to the berth structures. The vessel‘s distance and speed data are typically displayed on a large outdoor display board located on the berth structures. The data can also be transmitted and displayed to the pilots and ship personnel in real time via carry-on laptops or hand-held monitors. The system improves the safety of the berthing operation by helping the pilot and ship‘s crew manage the vessel‘s speed and approach vectors and verify that the approach procedure is within the specified terminal limits. The system can be designed to perform three major functions including: Monitoring the vessel as it approaches and is manoeuvred towards the berth; Monitoring the vessel‘s approach immediately prior to docking as it makes contact with the fender(s); and, Monitoring the drift movements and position of the vessel while it is moored at the berth. All sensor information is sent to the control centre for display and logging.

April 2010 FINAL Page 3-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3.3 General Terminal Measures to Avoid or Minimise Spills The marine terminal will be designed to minimize and mitigate environmental effects during construction and operations. Particular attention will be paid to implementing industry best practices and mechanical measures to avoid spills. During operations, the following measures will be employed to avoid or reduce the risk of spills from the marine terminal: operational procedures for unloading and loading liquid hydrocarbons, including using trained personnel to monitor tanker loading and unloading operations fail-safe valves for connecting loading arms to the oil tanker‘s manifold secondary containment for loading and unloading arms and associated fittings oil spill containment boom at each tanker berth, deployed during all oil loading operations, extending out from shore, around the tanker and back to shore. Because condensate dissipates quickly, the containment boom will not be used during condensate off-loading electronic sensors on arms to monitor position and engage the automatic shut-off valve, if the oil tanker moves beyond the safety limit of the loading arm envelope during transfer of liquid hydrocarbons an alarm system that provides a warning and shutdown if a spill is detected use of adequate number and capacity of tugs during berthing and unberthing of oil tankers exclusion zone or navigational restriction while oil tankers are at berth wind speed monitoring during loading and unloading operations, with cargo transfer stopped when weather limitation criteria are reached The following measures will be implemented for detection, containment and cleanup of spills: leak detection systems continuous system monitoring with the Supervisory Control and Data Acquisition (SCADA) system routine visual inspection routine checks of valves conformance to design codes emergency response plans capability to immediately deploy containment and response equipment at the marine terminal emergency response equipment emergency response training establishment of trained response crews trained at the Kitimat Terminal Additional details about these procedures and equipment are provided in TERMPOL Study 3.10 ―Site Plans and Technical Data‖ and TERMPOL Study 3.11 ―Cargo Transfer Systems‖. Measures to prevent and responded to hydrocarbon releases are detailed in TERMPOL Study 3.18 ―Contingency Planning‖. Prevention of and response to spills from oil handling facilities is strictly regulated in Canada. An OPEP, pursuant to the Canada Shipping Act, will be in place for the marine terminal.

Page 3-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

3.4 Vessel Operations and Environmental Protection Northern Gateway is committed to confirming that tankers transporting condensate, diluted bitumen and synthetic oil to and from the Kitimat Terminal adhere to high safety standards and operate in an environmentally responsible manner. The safe passage of marine traffic will be achieved through a comprehensive strategy that brings together highly trained professionals, technology and planning. The strategy includes the following: All vessels navigating within Canadian waters and calling at the Kitimat Terminal must be in full compliance with all relevant shipping regulations and safety standards required under the Canada Shipping Act. Tankers calling on the Kitimat Terminal will have double hulls, double bottoms and separate (segregated) tanks for ballast to prevent ballast seawater from coming in contact with hydrocarbon products. Ships officers and crew will be certified in accordance with the IMO International Convention on Standards of Training, Certification and Watchkeeping for Seafarers, 1978 and the 1995 amendments to this convention. Full bridge simulations have be done, and will be continued with pilot and tug crew training to identify navigational risks and assess potential emergency situations (such as engine or steering failures), as well as recovery from these situations (e.g., with assistance from escort tugs). As ships approach or leave the CCAA, experienced coastal pilots with local knowledge will board the tankers at designated locations. Pilots will be equipped with independent navigational systems and will provide guidance to the ship‘s master during the voyage into and out of the Kitimat Terminal. All laden tankers in the CCAA will be accompanied by two escort tugs with one tug attached (tethered) to the tanker, and the second tug in close escort. Ballasted tankers within the CCAA will be accompanied by a close escort tug. In the OWA, all tankers (laden and ballasted) will be accompanied by one close escort tug between the pilot boarding station (at Triple Island, or at potential sations at Caamano Sound and Browning Entrance) and the CCAA. Escort tugs will also be available for emergency (ocean rescue) response outside the CCAA. All tugs will be equipped with emergency response equipment and would act as the first response to a hydrocarbon spill. Safe transit speeds for tankers and tugs underway will be identified in Northern Gateway‘s Port Information Book to limit shoreline disturbances and reduce the likelihood of navigational incidents. Speed restrictions would also allow a tethered escort tug to arrest a ship in the event of a complete engine failure. Radar will be installed along important sections of the ship routes to monitor all marine traffic and provide additional information to the MCTS centre and to pilots. Operational safety limits will be established to cover visibility, wind and sea conditions.

April 2010 FINAL Page 3-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 3: Operational and Design Measures to Prevent Hydrocarbon Spills and Reduce Risk

Northern Gateway is also committed to environmental protection measures to safeguard coastal resources from potential maritime accidents, as demonstrated through the following examples: Tanker vetting criteria for all tankers calling on the Kitimat Terminal will confirm adherence to the highest standards of tanker construction, maintenance, onboard navigation and communication equipment and response capability. Operational environmental limits will be identified in Northern Gateway‘s Terminal/Transhipment Site Operations Manual (TOM) for tanker and cargo handling at berth. Terminal personnel will be highly skilled and trained to deal effectively with operational incidents including the use of an Emergency Shutdown System. Crews of the escort tugs and other support vessels (e.g., harbour tugs) will have extensive training in emergency response. The escort tugs and harbour tugs will carry response equipment that meets the requirements for a Tier 1 response (see Section 5.5). This equipment will include containment boom, anchors, skimmer system and temporary storage tanks for recovered oil. All tankers calling on the Kitimat Terminal will be required to have a Shipboard Oil Pollution Emergency Plan (SOPEP) under Regulation 26 of Annex I to the International Convention for Prevention of Pollution from Ships (MARPOL 73/78). All tanker operators are required to have a contractual agreement with a certified Oil Spill Response Organization, and to carry suitable insurance coverage under the Civil Liability Convention. Response equipment will be stored at several locations within the CCAA and OWA. Locations will be determined as part of the response planning process and through discussions with participating groups and the federal and provincial government. Locally based staff will be trained in the effective deployment of response equipment. Northern Gateway will, in consultation with participating groups, define areas of particular ecological sensitivity and assess whether installation of permanently located quick deployment first response systems is viable for the most critical of these locations. Each tanker berth will be equipped with a containment boom, which will be deployed during all loading operations for oil. The containment boom will extend from shore, out around the tanker and back to shore. The containment boom will not be used during condensate off-loading, both because condensate dissipates quickly, and to reduce the risk of ignition. In addition to the design features and operational measures, facilities that would allow for exclusion booming of sensitive areas would be a key environmental protection measure to limit the spread hydrocarbons into sensitive areas and facilitate cleanup. Other environmental protection measures include ballast management and caches of response equipment and available trained personnel to respond quickly to any marine-related emergency.

Page 3-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4 Marine Shipping Network Risk Analysis This section discusses the probability, or likelihood, of a hydrocarbon spill occurring along the routes shown in Figure 4-1 as ships travel to and from the marine terminal. Calculations are based on a quantitative risk analysis (QRA), (DNV 2010). The QRA is a statistical analysis of the probability of an incident occurring, the conditional probability of a release occurring given an incident, and an evaluation of the potential volume released. The QRA results reflect risk of shipping incidents during passage to and from the Kitimat Terminal, and of incidents at the marine terminal, areas of increased risk along the route and a model sensitivity analysis. The QRA follows international best practice from the IMO definition of a Formal Safety Assessment (IMO 2002, Internet site). The QRA for marine shipping accounts for: the physical system in which vessels will be operating, including current and projected traffic volumes, navigation systems and weather conditions identification of hazards along the proposed shipping routes to and from the proposed Kitimat Terminal, using input from British Columbia pilots, other marine professionals and local stakeholders to capture local conditions and relevant causes of potential incidents (e.g., collision and groundings) global frequencies of incidents in combination with local scaling factors (e.g., wind, bathymetry current and ship traffic) to provide regional-specific information associated with vessel operations in the CCAA and the OWA. The frequency assessment calculates the probability of incidents that could result in the breaching of the ship‘s cargo containment system. It considers the potential magnitude and likely outcome, and the output is represented by the return period, which is the estimated average recurrence interval (in years) between incident events. assessment of incidents on ship passages to and from the Kitimat Terminal and incidents at the terminal for various kinds of events (such as collision, groundings, foundering etc),based on ship size, ship traffic density and meteorological and oceanographic conditions to calculate the potential quantity of oil that might be spilled for each type of incident. Outcomes of an incident can range from minor damage without any spill to total loss of vessel. the risk of a spill as it relates to navigational difficulty. For example, the tanker approaches are divided into segments (Figure 4-1), with each segment evaluated for risk of a spill. The risk is lower in relatively straight sections compared with sections that require greater navigational input. Navigational difficulty, density of ship traffic, and/or exposure to environmental forces influences the probability of grounding, collision, foundering, mitigation measures taken to reduce potential risk and consequences of spills; for example, incidents might occur which may or may not result in loss of cargo containment, but contingencies can limit or avoid hydrocarbon spills into the environment.

April 2010 FINAL Page 4-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Figure 4-1 Shipping Routes and Corresponding Segments

Page 4-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

The QRA results reflect the risk of incidents during passage from the OWA to and from the terminal, incidents at the terminal, areas of increased risk along the route and a model sensitivity analysis for validating the applied scaling factors. Findings suggest overall risk levels are comparable to that of existing shipping routes and terminals with similar meteorological, oceanographic and physical conditions (areas with fjords, fog and wind). Compared to other world operations the sailing distance from open water to the terminal is comparatively longer, however, shipping traffic along the coastal shipping route are substantially lower. In accordance with TERMPOL Code, TP473E 2001 this section highlights: probabilities of credible incidents that result in the breaching of the ship‘s cargo containment system risks associated with navigational and operational procedures the risk of an incident becoming ―uncontrollable‖ Risks associated with berthing, loading and discharge at the marine terminal are discussed in Section 5.

4.1 Shipping Route Description There are three proposed routes between the open ocean and the Kitimat Terminal: two transit Queen Charlotte Sound south of Haida Gwaii and one transits Dixon Entrance north of Haida Gwaii (Figure 4-1). The three routes have been divided into segments based on variances in bathymetry, traffic and meteorological and oceanographic conditions. Segments are used in the risk analysis to differentiate areas or risk along the route. The segment breakdown of the shipping routes is provided in Table 4-1.

Table 4-1 Route Distances and Approximate Sailing Times

South Route via Caamaño South Route via Browning North Route Sound Entrance average average average sailing sailing sailing Segment length speed time length speed time length speed time (nm) (kn) (h) (nm) (kn) (h) (nm) (kn) (h) 1 45 10 4.5 45 10 4.5 45 10 4.5 2 15 10 1.5 15 10 1.5 15 10 1.5 3 56 10 5.6 56 10 5.6 4a 25 13 1.9 4b 45 13 3.5 5 65 13 5.0 6 20 10 2.0 7 35 10 3.5 8 75 13 5.8 75 13 5.8 9 68 13 5.2 Total 251 22.0 190 17.3 259 22.6

April 2010 FINAL Page 4-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

The proposed routes are existing shipping routes and have been selected to provide route options when entering or departing from Kitimat Terminal. A detailed route analysis is located in TERMPOL Study 3.5/12. A summary of each segment, including existing vessel traffic is included in the QRA. Risk calculation where preformed based on the current vessel forecast and routing (Table 4-2).

Table 4-2 Assumed distribution of ship traffic to and from the Kitimat Marine Terminal

VLCC SUEZMAX AFRAMAX TOTAL Oil Cond. Oil Cond. Oil Cond. Oil Cond. North Route 45 0 28 0 0 0 73 0 South Route via 4 0 30 44 27 13 61 57 Caamaño Sound South Route via 1 0 7.5 10.5 6.5 3.5 15 14 Browning Entrance Total 50 120 50 220

4.2 Hazard Identification A HAZID workshop was held in Vancouver on April 27th 2009, British Columbia with local maritime experts to identify local hazards and discuss their influence on the risk to marine transportation to and from the Kitimat Terminal. In addition to the formal HAZID workshop in Vancouver, the following meetings were held: On April 29 2009, DNV was escorted by boat from Prince Rupert to Kitimat to view portions of the North and South Routes. The purpose of this trip was to provide DNV with an opportunity to view, firsthand, sections of the North and South Routes. Key personnel from DNV also participated on a second boat trip from Prince Rupert to Kitimat on June 17, 2009. A number of interviews with local stakeholders were held in Prince Rupert (April 28, 2009), Kitimat (April 30, 2009) and Vancouver (May 1, 2009). The overall conclusion from the hazard identification process was that the hazards presented were manageable, particularly when the risk mitigation systems proposed by Northern Gateway are taken into account. No hazards were discovered that would indicate that the area of the British Columbia examined is more challenging than other areas of the world with similar marine terminal and tanker operations. Detailed results of the HAZID are included in the QRA (DNV 2010).

Page 4-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4.3 Probability of Hydrocarbon Spills along the Shipping Routes The risk of spills associated with project tanker traffic along the three proposed routes has been quantified. The information presented here is a summary of the QRA prepared by Det Norske Veritas (DNV). Risks associated with berthing, loading and unloading at the terminal are discussed in Section 5. Incidents included in the analysis of the tanker transit include: Ship grounding, both powered and drift Two ship collision Fire and / or Explosion Foundering The incident frequency for world wide tanker incidents is based on the Lloyds Register Fairplay (LRFP) marine incident database. LRFP is generally considered as the most comprehensive incident database in the world, recording incidents since 1978. For the QRA, statistics over the period 1990 – 2006 were utilized for vessel incidents and adapted to assess tanker transits to and from the Kitimat marine terminal since vessels operating and incidents occurring after 1990 are considered to be more representative of modern tanker operation than pre-1990. The incident frequencies derived from the LRFP data are considered to be valid for all three tanker classes forecast to call at the Kitimat Terminal. Tanker incident frequencies are influenced more by the specific shipping route, than the type of tanker. The materials and equipment as well as hull and tank configurations do not vary significantly between classes. Few shipping incidents involving large bulk carriers or tankers have occurred off the BC coast, and statistically valid incident frequencies could not be developed based on local events. Therefore, worldwide statistics were used and qualitatively scaled to the BC coast and Kitimat area. The sources of worldwide casualty and spill statistics are discussed in TERMPOL 3.8. Incidents in LRFP are divided into the following three damage categories: Minor damage - any event reported to LRFP and included in the database, not being categorized as major damage or total loss. Major damage - breakdown resulting in the ship being towed or requiring 3rd party assistance from ashore; flooding of any compartment; or structural, mechanical or electrical damage requiring repairs before the ship can continue trading. In this context, major damage does not include total loss. Total loss - where the ship ceases to exist after an incident, either due to it being irrecoverable (actual total loss) or due to it being subsequently broken up (constructive total loss). The latter occurs when the cost of repair would exceed the insured value of the ship. The overall risk of oil spills related to marine shipping was determined by evaluating hazards or local conditions associated with shipping routes (scaling factors) and estimating the probability of an incident occurring the likelihood of a spill occurring in the event of an incident (conditional probability) the likely volume of oil spilled

April 2010 FINAL Page 4-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4.3.1 Incident Statistics Probability estimates for each segment were calculated for powered grounding, drift grounding, collision, foundering, fire and/or explosion during the voyage. The QRA was designed using a per voyage methodology to allow for forecast annual tanker frequency and routes to be evaluated independently. The LRFP incident frequencies were converted to incident per nautical mile using conversion calculations specific to each incident type. In general, the total nautical miles per ship year and percentage of time that a ship may travel in an area exposed to hazard was calculated. For example, a grounding incident is only possible while a ship is in coastal waters, so only that portion of the annual sailing miles is included. The resulting incident frequencies calculated per nautical mile for the QRA are shown in Table 4-3.

Table 4-3 Base/Worldwide Frequencies for Main Incident Types for Oil Tankers per Nautical Mile

Incident type Frequency Powered grounding frequency (worldwide) 0.598 per million nm in coastal waters Drift grounding frequency (worldwide) 0.0996 per million nm in coastal waters Collision frequency (worldwide) 0.454 per million nm in coastal waters Foundering frequency (worldwide) 0.000504 per million nm Fire and/or explosion frequency (worldwide) 0.0326 per million nm

SOURCE: converted from LRFP 2007

Worldwide frequencies were used as the basis for estimating frequencies for individual segments in the OWA and CCAA by applying scaling factors using this formula:

FrequencyBC coast = FrequencyGlobal * KTotal

Scaling factors appropriate for the OWA and CCAA were developed during the HAZID process and a peer review by experts from the marine risk consultant that completed the QRA (DNV 2010). The local scaling factors (K) used to calculate each local frequency are listed in Table 4-4. In general, a scaling factor of 1.0 represents the average condition worldwide for a given variable. Scaling factors can then be increased if local conditions are considered to be more hazardous, or decreased if conditions are considered to be less hazardous, than the average global condition. For example, for the navigational route factor, values greater than 1.0 were assigned for segments in the CCAA, while values less than 1.0 were assigned to the open water segments. Table 4-4 provides a summary of the scaling factors considered. Detailed scaling factors for each segment are included in Table 4-5.

Page 4-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-4 Summary of Scaling Factors

Scaling Incident type Factor (K) Application Assigned Value Powered navigational corrects for the number of course ≥ 1.0 for CCAA segments grounding route changes and amount of room for <1.0 for OWA segments navigational error (distance to shore/ shallow water) mitigation considers the use of pilots with ≤ 1.0 for segments where local measures for good knowledge of the local pilots are assigned to the vessel powered navigational conditions grounding navigational accounts for visibility, currents, ≥ 1.0 for all segments except difficulty marking of the passage and segment 3 disturbance from other vessels Drift distance to distance to shore, wind and current ≥ 1.0 for segments of the CCAA, grounding shore direction determine whether and at which is narrow, with a distance to what speed a vessel will drift shore typically less than 2 nm toward shore. The closer to shore an oil tanker is at the time it starts drifting, the more likely it is to hit shore before it can regain engine power or receive assistance. emergency emergency anchoring can be used > 1.0 because in the CCAA and anchoring to avoid drift grounding. OWA, deep water (≥ 100 m) and rapid drop-off away from shore means there are very few or no emergency anchoring possibilities Collision traffic density generally, traffic volumes in the < 1.0 (highest for segment 2, assessment area are lower than in Wright Sound, which has higher areas where collisions normally traffic volume than elsewhere in occur (e.g. English Channel and the assessment area, but are still off the coast of Japan). lower than in the global dataset) navigational accounts for visibility, currents, ≥ 1.0 for all segments except difficulty for marking of the passage and segment 3 collision disturbance from other vessels mitigation considers the use of pilots with ≤ 1.0 for segments where local measures for good knowledge of the local pilots are assigned to the vessel collision navigational conditions and effectiveness of vessel traffic services Foundering weather accounts for wind, waves and < 1.0 (well below) for CCAA conditions currents; harsh weather increases (severe weather but limited wave the probability of foundering height) except segment 7 ≥ 1.0 for OWA (weather and waves can become severe) Fire and/or none no scaling factors applied to none explosion frequency of fire and/or explosion

April 2010 FINAL Page 4-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-5 Scaling factors for incidents considered along the marine tanker routes

Powered Grounding Drift Grounding Collision Foundering

Route Knavigational Kmeasur Knavigational K Kdistance to Kem- K KTraffic Knavigational Kweather Segment route es difficulty total shore anchoring total density KMeasures difficulty K total conditions 1 1.5 0.9 1 1.35 1.3 1.2 1.56 0.2 0.9 1 0.18 0.01 2 2.1 0.9 1 1.89 1.3 1.2 1.56 0.6 0.9 1 0.54 0.01 3 1.5 0.9 0.9 1.22 1.3 1.2 1.56 0.4 0.9 0.9 0.32 0.01 4a 0.6 1 1.2 0.72 1.1 1.2 1.32 0.2 1 1 0.2 1 4b 0.6 0.9 1 0.54 1.1 1.2 1.32 0.2 1 1 0.2 1.2 5 0.001 1 1 0.001 0.05 1.2 0.06 0.01 1 1 0.01 1.5 6 1.8 0.9 1.2 1.94 1.3 1.2 1.56 0.2 0.9 1.2 0.21 0.01 7 1 0.9 1.5 1.35 1.3 1.2 1.56 0.4 0.9 1.5 0.54 1.5 8 0.001 1 1.2 0.001 0.01 1.2 0.01 0.01 1 1.2 0.01 1.5 9 0.1 1.1 1 0.11 0.5 1.2 0.6 0.01 1 1 0.01 1.3

Page 4-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-6, below, shows the scaled incident frequency per nautical mile for each incident type and each route segment. The incident frequency is calculated by applying the local scaling factors (Table 4-4) to the global incident frequencies (Table 4-3).

Table 4-6 Total unmitigated and scaled incident frequency per nautical mile for each incident type for each route segment

Segment 1 2 3 4a 4b 5 6 7 8 9 powered 8.07E- 1.13E- 7.26E- 4.30E- 3.23E- 5.98E- 1.16E- 8.07E- 7.17E- 6.57E- grounding 07 06 07 07 07 10 06 07 10 08 drift grounding 1.55E- 1.55E- 1.55E- 1.31E- 1.31E- 5.98E- 1.55E- 1.55E- 1.20E- 5.98E- 07 07 07 07 07 09 07 07 09 08 collision 8.17E- 2.45E- 1.47E- 9.08E- 9.08E- 4.54E- 9.81E- 2.45E- 5.45E- 4.54E- 08 07 07 08 08 09 08 07 09 09 foundering 5.05E- 5.05E- 5.05E- 5.05E- 6.06E- 7.58E- 5.05E- 7.58E- 7.58E- 6.57E- 12 12 12 10 10 10 12 10 10 10 fire and 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- 3.26E- explosion 08 08 08 08 08 08 08 08 08 08 total 2.13E- 1.03E- 1.21E- 2.50E- 3.80E- 4.22E- 6.83E- 1.02E- 8.97E- 6.44E- 02 02 02 03 03 04 03 02 04 04

To reduce the frequency of incidents, Northern Gateway will use escort tugs extensively. The effectiveness of escort tug on reducing incidents is based on previous DNV studies (DNV 2002). Typical causes of grounding and collision incidents were studied by DNV to ascertain how an escort tug might help a tanker avoid an incident, or minimize the damage if the incident was to occur. The tug escort plan for Northern Gateway will be as follows: All laden tankers will have a close escort tug between the pilot boarding stations at Triple Islands, Browning Entrance and Caamaño Sound and the Kitimat Terminal (Segments 1, 2, 3, 4a, half of 4b, 6 and 7). In addition all laden tankers will have a tethered escort tug between Browning Entrance and Caamaño Sound and the Kitimat Terminal (i.e., throughout the CCAA, Segments 1, 2, 3, 6 and 7). All tankers in ballast will have a close escort tug between the pilot boarding stations at Triple Islands, Browning Entrance and Caamaño Sound and the Kitimat Terminal (Segments 1, 2, 3, 4a, half of 4b, 6 and 7). The predicted effect on lowering the frequency of incidents is provided in Table 4-7. In total this gives a reduction of the total incident frequency by some 65 %. Together with the frequency reduction (preventing groundings and to a lesser extent collisions from occurring altogether) the escort tug can also have a positive effect on reducing the consequence should a grounding or collision actually occur.

April 2010 FINAL Page 4-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-7 Risk reducing effect of using escort tugs/tethered

Tug escort effect on Segments where tug escort and Incident frequency Vessel reducing the frequency mitigation effect is to be mitigated condition frequency of incidents applied Powered grounding Laden 80 % Tethered and close escort 1, 2, 3, 6, 7 and close escort 4a & half of 4b Powered grounding Ballast 80 % Close escort 1, 2, 3, 6, 7, 4a & half of 4b Drift grounding Laden 80 % (close escort Close escort 4a & half of 4b effect) Drift grounding Laden 90 % (tethered escort Close and tethered escort 1, 2, 3, 6, 7 effect) Drift grounding Ballast 80 % Close escort 1, 2, 3, 6, 7, 4a & half of 4b Collision Laden 5 % Tethered and close escort 1, 2, 3, 6, 7 and close escort 4a & half of 4b Collision Ballast 5% Close escort 1, 2, 3, 6, 7, 4a & half of 4b

An escort tug can reduce the consequences by reducing the speed at the time of impact thereby reducing the damage to the tanker and the potential volume of cargo or bunker spilled. The exact consequences in the case of a grounding or collision will depend on many parameters, such as wind and the type of sea bottom. It is conservatively assumed for the purposes of this report that an escort tug will not reduce the imminent consequence of grounding in terms of the volume of cargo or bunkers spilled. Tugs escorting the tanker in the case of a spill will remain and assist the tanker during the oil spill response. All escort tugs will carry a complement of oil spill response equipment. Once the tug(s) have control the of tanker, one of the escort tugs will be available to assist in the oil spill response. These plans are discussed in more detail in TERMPOL 3.18. In addition to the use of tethered tugs, several other mitigation measures will be employed to further reduce incident return periods (Section 3). One of the navigational mitigations is the use of Electronic Chart Display and Information Systems (ECDIS) on the tankers, which will be mandatory on new tankers built after 1 July 2012 and on older tankers after 1 July 2015, and is currently used by some tankers. All tankers calling on the Kitimat Terminal will be required to have ECDIS. For an individual tanker otherwise without ECDIS installed, the installation of ECDIS is expected to reduce the probability of powered grounding by some 30 %, and the total probability of an oil spill by some 15 to 20 % (DNV 2010). In addition it is proposed that pilots have hand carried units which can be operated independently of the ship ECDIS. In most cases, an incident will not result in a hydrocarbon spill. The conditional probability of spills resulting from the rare incidents described above is discussed in Section 4.3.2.

Page 4-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4.3.2 Conditional Probabilities of a Spill The QRA report (DNV 2010) identifies the types of incidents that may result from marine shipping. Incidents such as grounding do not always result in a spill. For the purposes of this section, consequence refers to the vessel damage per the three categories described below: Minor damage: Any event not categorized as major damage or total loss. Very minor damage involving little repair. Hull and cargo tanks assumed not to have been punctured Major damage: Does not include total loss. A breakdown resulting in the ship being towed or requiring third party assistance from ashore Flooding of any compartment Structural, mechanical or electrical damage requiring repairs before the ship can continue trading. For double hull tankers not all major damage results in a spill, or a breach of the outer and inner hulls. Total loss: The ship ceases to exist after an incident, due to it being irrecoverable (actual total loss) or being subsequently broken up (constructive total loss). Constructive total loss occurs when the cost of repair would exceed the insured value of the ship. It is conservatively assumed for both laden and tankers in ballast that all cargo/bunker will be lost. This is a conservative assumption given some oil may be recovered from some vessels or spills, most notably in cases involving a constructive rather than actual total loss. Conditional probabilities are based on damage reported in the LRFP casualty database. DNV has estimated the conditional probability of a spill (the probability of a spill, given an incident has already occurred) for each incident type and damage category. The conditional probability provides an estimate of the likelihood of an oil spill, given the level of damage sustained. For all incident types except foundering, there is less than a 10% conditional probability of a spill from a vessel in ballast. Given that half of the vessels calling on the terminal will be either transiting to or from the terminal in ballast, the probability of a spill occurring as a result of an incident is reduced.

April 2010 FINAL Page 4-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-8 Material Damage from Accidents and Conditional Probability of a Spill

Damage category Probability of a spill by accident Type of Major Minor type and laden vs Accident Accident Characteristics Total Loss Loss Damage ballast Grounding Frequency distributiona 2.4 % 40.4 % 57.2 % NA

Conditional probability of 100 % 75 % 0 32.7 % spill – vessel laden

Conditional probability of 100 % 10 % 0 6.4 % spill – vessel ballastb Collision Frequency distributiona Negligible 25.5 % 74.5 % NA Conditional probability of 100 % 75 % 0 19.1 % spill - vessel laden Conditional probability of 100 % 10 % 0 2.6 % spill – vessel ballastb Foundering Frequency distributiona 100 % NA NA NA Conditional probability of 100 % - - 100 % spill - vessel laden Conditional probability of 100 % - - 100 % spill – vessel ballastb Fire or Frequency distributiona 2.8 % 48.4 % 48.8 % NA Explosion Conditional probability of 100 % 50 % 0 27 % spill - vessel laden Conditional probability of 100 % 10 % 0 7.6 % spill – vessel ballastb NOTES: a Source: LRFP 2007 b Ballast indicates spill of vessel fuel plus slop tanks, engine oil, bilge water and other, but not cargo c The conditional probability of a spill occurring in the event of an incident involving a ballasted taker is lower because potential oil loss would be from the ship’s bunker fuel tanks, the cargo tanks are empty. NA = not applicable

Page 4-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4.3.3 Risk of a Spill from a Tanker Based on the local scaling factors, the use of escort tugs, and the conditional probability of a spill occurring spill probabilities where calculated for each segment (Table 4-9). The effect of using escort tugs has been calculated by multiplying incident frequency for each relevant segment (i.e. 1, 2, 3, 4a, 4b, 6 & 7) with the effect of tug escort from Table 4-7. The effect on oil spill return periods for the applicable segments is shown in Table 4-9, based on the forecast transits per year through each segment as discussed in Table 4-2. The overall risk of hydrocarbon spills were calculated by applying segment specific scaling factors to global incident frequencies and evaluating the conditional probability of a spill occurring, given a specific incident. Return periods for oil spills (i.e., spills of bunker, crude oil or condensate) have been calculated for three different tug plans for each segment for the forecasted distribution of vessels over the different routes (ref. Table 4-1). The results were calculated to show the variance in scenarios were tugs are used versus operations without (or with only partial) use of tugs. The results are shown in Table 4-10. As can be seen the use of escort tugs is predicted to have an important effect on reducing the overall spill frequency. The implementation and proper operation of escort tugs more than triples the return period for oil spills, from 78 to 250 years. Given the very different properties of oil and condensate, the overall risk of spills has been evaluated by product type (Table 4-11). Spill here refers to any spill of cargo or bunkers from ―oil tankers‖ (laden or in ballast), and similarly ―condensate spill‖ to any spill of cargo or bunkers from ―condensate tankers‖ (laden or in ballast). The forecast traffic to and from the Kitimat Terminal consists of 71 tankers carrying condensate and 149 tankers carrying crude oil every year. Mitigated spill return frequencies per segment for tankers transporting condensate and crude oil respectively are shown in Figure 4-2.

April 2010 FINAL Page 4-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-9 Mitigated Return Period (Years) of an incident resulting in a release of cargo (including oil, condensate or bunker) per route segment, based on average forecast traffic

Segment 1 2 3 4a 4b 5 6 7 8 9 powered grounding 1,600 3,400 3,000 16,000 4,000 900,000 4,600 3,800 320,000 19,000 drift grounding 14,000 42,000 24,000 53,000 9,800 90,000 59,000 34,000 190,000 21,000 collision 5,900 5,900 5,700 29,000 15,000 210,000 20,000 4,700 76,000 510,000 foundering >1,000,000 >1,000,000 >1,000,000 540,000 250,000 130,000 >1,000,000 150,000 59,000 380,000 fire and explosion 8,900 26,000 15,000 48,000 26,000 18,000 37,000 21,000 8,000 44,000 total 1,000 1,900 1,600 7,300 2,200 12,000 3,200 1,800 6,100 8,000

Page 4-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-10 Spill (including oil, condensate or bunker) return periods for forecasted route choices with different use of tugs

With tug escort only in Without tug escort With tug escort laden condition Segment 1 240 1,000 680 Segment 2 530 1,900 1,300 Segment 3 450 1,600 1,100 Segment 4a 2,100 7,300 5,200 Segment 4b 1,400 2,200 2,000 Segment 5 12,000 12,000 12,000 Segment 6 770 3,200 2,100 Segment 7 540 1,800 1,300 Segment 8 6,100 6,100 6,100 Segment 9 8,000 8,000 8,000 Total 78 250 180

Table 4-11 Return period of a spill associated with the tanker traffic for the Northern Gateway Project

Return Periods (Years) of an Accidental Hydrocarbon Spill Associated with Project Tankers Location Oil Cargo ((bunker included) Condensate Cargo (bunker included) (Years) (Years) Segment 1 1,500 3,100 Segment 2 2,800 5,900 Segment 3 1,900 12,000 Segment 4a 7,300 NA1 Segment 4b 2,200 NA1 Segment 5 13,300 NA1 Segment 6 6,300 6,700 Segment 7 3,500 3,700 Segment 8 12,300 13,200 Segment 9 15,700 16,800 Total 350 890

NOTE: 1Condensate tankers are not forecast to use the North Route

April 2010 FINAL Page 4-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Figure 4-2 Mitigated spill return frequencies per segment for tankers transporting Crude Oil and Condensate respectively

Page 4-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

4.3.4 Consequence Analysis (Spill Size) Monte-Carlo simulations, following IMO‘s MARPOL regulations (MEPC 49/22/Add.2), were conducted as part of the QRA to evaluate spill volumes for grounding and collision incidents. The analysis indicates a conditional probability of a spill of 10 000 m3 or more provided a laden VLCC has grounded and a spill has resulted of approximately 5.5 %, while the probability of spill exceeding 25 000 m3 and 40 000 m3 is approximately 1 % and 0.2 % respectively (DNV 2010). A collision has a higher conditional probability of a spill and higher expected spill volume compared to grounding. The conditional probability for spills associated with collision of a laden VLCC are 25%, 10% and 0.3% for spill volumes exceeding 10,000 m3, 20 000 m3 and 50 000 m3, respectively (DNV 2010). The overall spill size distribution, calculated based on all incident types is included in Table 4-12. Together with the frequency reduction (preventing groundings and collisions from occurring altogether) the escort tug can also have a positive effect on reducing the consequence should a grounding or collision actually occur. An escort tug can reduce the consequences by reducing the speed at the time of impact thereby reducing the damage to the tanker and the volume of cargo or bunker spilled. The exact volume spilled in the case of a grounding or collision will depend on many parameters, such as wind and the type of sea bottom. It was conservatively assumed in the QRA that an escort tug will not reduce the imminent consequence in terms of the volume of cargo or bunkers spilled. Since escort tugs are most effective at preventing grounding incidents, which are typical associated with smaller spill volumes the spill distribution appears to shift towards larger spills. Figure 4-3 is effective at showing how mitigation reduces both the frequency and spill size.

Table 4-12 Potential Spill Size Distribution for Unmitigated / Mitigated Spill Risk

Distribution Distribution Potential spill volume (Unmitigated) (Mitigated) (m3) < 5000 58 % 53.2 % 5,000 – 10,000 23 % 18.5 % 10,000 – 20,000 14 % 19.2 % 20,000 – 40,000 3.7 % 7.4 % > 40,000 0.8 % 1.7 %

The expected annual frequency of spills exceeding a certain size is illustrated in Figure 4-3. The term ―accumulated‖ in the figure refers to the fact that when a frequency of 0.001 per year is read from the graph for a spill volume of 10 000 m3, it means that the total frequency of all spills of 10 000 m3 or more is 0.001 per year, or once per 1,000 years. The mitigated return period of spills of 5,000 m3 and above has been estimated to be of the order of 550 years, while the return period for spills of 20,000 m3 will be some 2,800 years. The return periods for extremely large oil spills, exceeding 40 000 m3, have been estimated to more than 15,000 years.

April 2010 FINAL Page 4-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Accumulated frequency of spills exceeding a certain size

0,014

0,012

0,01

0,008 Unmitigated Mitigated

0,006

0,004 Annual frequency [per year] frequency Annual

0,002

0 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Spill volume [m3]

Figure 4-3 Accumulated frequency of spills exceeding a certain size; Unmitigated / Mitigated

4.3.4.1 Risk of an Event Becoming Uncontrollable In the unlikely event of an incident, there is a risk that the incident could become uncontrollable. For the purpose of the TERMPOL, an uncontrollable event is considered an incident that leads to a total loss, which is defined as the ship being no longer navigable or operational after an accident. This could be due to it being irrecoverable (actual total loss) or to the cost of repair exceeding the insured value of the ship (constructive total loss). Total losses are accidents where there is potential for the entire hydrocarbon cargo and bunker fuel to be lost. This is a conservative estimate because emergency response operations include lightering and recovery of oil from undamaged tanks, especially in cases of economic loss and not an actual total loss. The return periods for an uncontrollable event are described in Figure 4-4. The frequency of a total loss is moderate in the CCAA, most notably for Segment 1. The return periods for a total loss on the remaining segments are large enough for the risk to be considered negligible. Segments 5, 8 and 9 are not displayed and have mitigated return periods of 57, 000 26,000 and 57,000, respectively

Page 4-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Figure 4-4 Return Period for a Total Loss Event Key return periods and the effect of risk mitigation measures are summarized in 4-13 below. Table 4-13 includes the effect of full tug escort (tankers laden and in ballast) and was calculated using the forecast distribution of traffic across all route segments. By default the spills referred to in 4-13 also include the release of bunkers, as do all oil or condensate spill return periods in this QRA. It should be noted that the return periods below do not include the risk from berthing and passing vessels. These have been calculated as small risks and do not affect the totals in a significant way. The use of an appropriate placed and sized escort tug fleet can more than triple the overall estimated return period of an oil or condensate spill. In the relative comparison it was determined that the North Route had the higher risk compared to the other two routes (due primarily to the length of this route). However, using a forecast traffic distribution of traffic on each route the actual tanker risk was found to be highest on the South Route via Caamano Sound. However, the use of tug escort also had the greatest risk reduction effect on this route.

April 2010 FINAL Page 4-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 4: Marine Shipping Network Risk Analysis

Table 4-13 Summary of Mitigated and Unmitigated Return Periods for Spills occurring during tanker operation along the preferred marine routes

Unmitigated Return Mitigated Return Scenario Period Period (years) (years) any size oil or condensate spill 78 250 any size oil spill 110 350 any size condensate spill 250 890 any size spill (oil or condensate) along the South 150 550 Route via Caamano Sound any size spill (oil or condensate) along the South 640 1900 Route via Browning Entrance any size spill (oil or condensate) along the North 210 620 Route spills exceeding 5,000 m3 200 550 spills exceeding 20,000 m3 1,750 2,800 spills exceeding 40,000 m3 12,000 >15,000

The overall levels of risk for the Northern Gateway project, even without mitigation measures, are within the level of risk accepted in many other international operations. With mitigation the Project tanker transport and marine terminal incident rates are calculated to be well below world average. A ―total overall scaling factor‖ is estimated to be 0.94 for the unmitigated total incident frequency and 1.06 for the unmitigated spill frequency. Therefore without mitigation measures in place the Project is predicted to have a slightly lower incident frequency, and a slightly higher spill frequency, compared to the worldwide average. ―Overall scaling factors‖ were estimated to be 1.31 for powered grounding, 1.48 for drift grounding and 0.31 for collisions. Therefore without mitigation measures in place, the Project is predicted to have a higher frequency of powered and drift grounding, and a lower frequency of collisions, compared to the world average. With the use of risk mitigation measures (mainly tug escorts) the ―overall scaling factors‖ for powered grounding, drift grounding are reduced from greater than 1.0 to approximately 0.3. Tug escorts have a significant effect on reducing the frequency of groundings. The ―overall scaling factor‖ remains close to 0.3 for collisions with mitigation measures in place. Therefore with the risk mitigation measures the frequencies of powered grounding, drift grounding and collisions associated with the Project are estimated to be about one third the world average. With mitigation measures in place ―total overall scaling factor‖ for the total incident frequency and spill frequency are both approximately 0.3. Therefore the Project is predicted to have an overall incident frequency and spill frequency that is one third the world averages.

Page 4-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

5 Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

5.1 Terminal Description The Kitimat Terminal will be located approximately 13 km south of the residential area of the town of Kitimat, near the head of Kitimat Arm. The proposed location is shown in Figure 5-1. The land was logged in the past and a logging road currently provides access to the site. The Kitimat Terminal will accommodate tankers at two berths for loading and offloading operations, as shown in Figure 5-2. Oil products can be loaded at both berths simultaneously, but only one berth at a time can be used for the discharge of condensate. Water depths at the marine terminal drop off rapidly from the intertidal zone leaving sufficient water depth (greater than 27 meters) for a fully laden VLCC with the largest draft. Kitimat Arm is 1.2 to 1.5 nm wide at the Kitimat Marine terminal site. There is ample space for the tankers to manoeuvre both during arrival and departure. As mentioned in TERMPOL 3.10 the turning basin outside the terminal exceeds TERMPOL requirements. The turning basin and typical tanker berthing manoeuvres are shown in Figure 5-3. All tankers berthing at the marine terminal will be assisted by 2 to 4 tugs to and from the berth. At very low speed, tankers of the proposed size have limited maneuverability and therefore need to be assisted during berthing and de-berthing by tugs pushing directly on the tanker hull or pulling on lines fixed to the tanker deck. As tankers contact the berthing structure fendering units, mooring lines will be fixed from the tanker to moorings located on shore. Only after the vessel is moored to the satisfaction of the ships master will the tugs return to standby at their moorage at the utility berth, north of the two tanker berths. Oil and condensate will be loaded and discharged using marine loading arms. Loading arms are special components of the cargo transfer system, designed to be connected to a pipe manifold on the tanker deck. The arms are assembled from articulated pipe assemblies that give the arm the ability to swivel in all directions and accommodate the vertical and horizontal movements of a moored ship. Loading arms are standard equipment for most marine oil terminals around the world. For a further description of the loading operation and equipment see TERMPOL 3.11. The Kitimat Terminal will be equipped with several safety measures and systems to ensure controlled and safe operation during loading and offloading operations, described briefly in Section 3.1 and in detail elsewhere (TERMPOL 3.10 and TERMPOL 3.11).

April 2010 FINAL Page 5-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

Figure 5-1 Proposed location of the Kitimat Terminal

Page 5-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

Figure 5-2 Proposed layout of the Kitimat Terminal

April 2010 FINAL Page 5-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

Figure 5-3 Proposed Turning Basins, Navigational Clearances and Vessel Manoeuvres (TERMPOL 3.10)

5.2 Probability of Hydrocarbon Spills The probability, or likelihood, of a hydrocarbon spill occurring at the terminal was calculated based on a quantitative risk assessment (QRA) completed by DNV. Terminal frequency data used were based on both LRFP data as well as DNV‘s extensive research of terminals worldwide and on the Norwegian west coast (DNV 2010). Incidents included in the transhipment analysis at the marine terminal include: Tanker struck at berth by passing vessel Tanker struck by Harbour Tug Tanker striking pier during berthing Release during Loading / Discharging. Through the QRA it was determined that the risk contribution from a tanker being struck by a passing vessel at the terminal, struck by a harbour tug, or a tanker striking the pier during berthing is so small that whether or not it is included, the effect on the total risk will be insignificant.

Page 5-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

5.2.1 Probability of Spills associated with Cargo Transfer Operations at the Kitimat Terminal A release associated with cargo transfer operations (loading oil or offloading condensate) at the Kitimat terminal may result from different events. Table 5-1 provides a summary of estimated return periods associated with different events.

Table 5-1 Mitigated Return Period of a Spill from a Tanker at Berth by Cargo Type and Release Volume

Return Period Return Period Return Period Small1 Medium2 Return Period Medium2 Oil Condensate Condensate Event Small1 Oil spill spill Spill Spill (years) (years) (years) (years) Release from loading arm 140 1,300 300 2,700 Failure in equipment - 1,300 - 2,700 Failure in the vessels piping 1,000 9,300 2,100 19,000 system or pumps Human failure 1,000 9,300 2,100 19,000 Mooring failure - 1,700 - 3,700 Overloading of cargo tank Negligible Negligible Negligible Negligible Total 110 430 230 910 NOTE: In line with the International Tanker Owners Pollution Federation Limited (ITOPF), spills are generally categorized by size: 1small (less than 7 tonnes ~ 10 m3) 2medium (7-700 tonnes ~ 10 – 1000 m3) large is used to describe those greater than 700 tonnes (~ 1000 m3).

The Kitimat Terminal will be designed with a closed loading system incorporating vapour emissions control system in order to collect the vapours that are displaced from the cargo tanks during loading. Modern cargo tank systems have sensitive tank gauging as well as high level alarms, and if the cargo tanks were to be accidentally overloaded the excess oil would be captured by the vapour control system, mitigating the risk of a release. The size of a spill resulting from a loading or offloading incident is a function of the type of failure (e.g., loading arm, pumping systems, human error), the cargo transfer rate, the detection time, and the emergency valve shut down time. The actual rates, as well as expected detection time and shut down time will depend on the detailed design. For both oil loading and condensate discharge the maximum expected volume of a major failure is approximately 250 m3 assuming one loading arm has failed at the full transfer rate. (DNV 2010). The QRA determined that although the medium category is characterized as ranging from 10 to 1,000 m3, it is unlikely that a terminal spill during loading or offloading will be greater than 250 m3 (DNV 2010). As seen in Table 5-1 small releases have lower return periods than larger incidents.

April 2010 FINAL Page 5-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 5: Safety Analysis for Berthing/Unberthing and Cargo Transfer Operations at the Kitimat Terminal

Figure 5-4 shows the risk reducing effects of using the closed loading and vapour return systems at the terminal.

Figure 5-4 Comparison of unmitigated and mitigated spill return periods for releases during cargo transfer at the marine terminal

Page 5-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons

6 Properties and Weathering of Liquid Hydrocarbons The physical and chemical characteristics of the hydrocarbons to be transported to and from the Kitimat Terminal are provided in the Properties and Fate of Hydrocarbons from Hypothetical Spills in the CCAA and at the marine terminal TDR (SL Ross 2010a), and the Properties and Fate of Hydrocarbons Associated with Hypothetical Spills at Three Sites in the Open Water TDR (SL Ross 2010b). Three types of liquid hydrocarbons will be shipped, diluted bitumen and synthetic oil (exported) and condensate (imported). Diluted bitumen and synthetic oil are viscous fluids that persist in the environment, whereas condensate is a low-viscosity fluid that evaporates and disperses rapidly.

6.1 Physical Properties Samples of the liquid hydrocarbons were sent for laboratory analysis to better understand their weathering properties and predict their possible behaviour in the marine environment (SL Ross 2010a). Weathering processes include evaporation, photo-oxidation, dispersion and biodegradation, with evaporation typically accounting for the majority of weathering for light hydrocarbon fractions. Over time, weathered hydrocarbon characteristics differ. Characteristics for both 3 to 4 hours and 24 hours after a spill are shown in Table 6-1. However, the actual amount weathered depends on oceanographic and meteorological conditions at the time the oil is released. Of the three hydrocarbons analysed, diluted bitumen would be the most persistent in the environment (13% evaporated after 24 hours) and condensate the least persistent (75% evaporated after 24 hours), with synthetic oil intermediate (30%) in persistence (see Table 6-1).

Table 6-1 Physical Properties of Project Hydrocarbons in the Marine Environment

Amount Density at Viscosity at Emulsion Parameter Evaporated 15°C 15°C Pour Point Tendency (% of volume) (g/cm3) (mm2/s) (°C) Weathered (3 to 4 hours)1 Diluted Bitumen 9 0.965 1427 < -23 unlikely Synthetic Oil 24 0.918 46.9 -24 unlikely Condensate 57 0.810 7.5 < -22 unlikely Weathered (24 hours)1 Diluted Bitumen 13 0.970 2652 -18 unlikely Synthetic Oil 30 0.926 87.6 -21 likely Condensate 75 0.852 7.3 < -23 unlikely

NOTE: 1 Weathering characteristics vary with meteorological and oceanographic conditions (SL Ross 2010a(CCAA), b(OWA)).

April 2010 FINAL Page 6-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons

6.2 Chemical Properties Chemical analyses of the three hydrocarbons are useful in describing the probable behaviour of these hydrocarbons and their relative potential effects on the biophysical and human environment. A complete chemical characterization of the three hydrocarbons is provided in the Marine Ecological Risk Assessment for Kitimat Terminal Operations TDR (Stephenson et al. 2010b) and summarized in Table 6-2.

Table 6-2 Chemical Properties of Liquid Hydrocarbons

Detection Diluted Synthetic Limit Bitumen Oil Condensate Chemical Parameter Concentration (mg/kg) Total Petroleum Hydrocarbon Fractions F1 (C6-C10) 0.01 41,000 104,000 374,000 F2 (C10-C16) 0.05 136,000 246,000 80,900 F3 (C16-C34) 0.01 394,000 467,000 59,900 F4 (C34-C50) 0.05 51,000 62,000 20,400 Polycyclic Aromatic Hydrocarbons 1-methylnaphthalene 2.5 6.6 99 190 2-methylnaphthalene 2.5 4.2 79 110 Naphthalene 5.0 < 5.0 85 86 Fluorene 5.0 < 5.0 14 15 Phenanthrene 5.0 5.3 12 25 Pyrene 5.0 6.1 30 < 5.0 Benzo(a)anthracene 5.0 7.9 < 5.0 < 5.0 Benzene, Toluene, Ethylbenzene and Xylenes Benzene 0.01 280 1,100 13,100 Toluene 0.05 990 3,300 25,300 Ethylbenzene 0.01 360 1,200 2,900 Total Xylenes 0.05 1,500 4,200 21,000 Other Volatile Organics 1,3,5-Trimethylbenzene 100 200 430 1,800 1,2,4-Trichlorobenzene 100 610 1,600 3,400 NOTE: Only parameters that were detected in at least one of the three hydrocarbon samples are listed.

Page 6-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons

Oils and condensates are complex hydrocarbon mixtures containing both organic compounds and inorganic trace elements. The main constituents are aliphatic and aromatic hydrocarbons. Total petroleum hydrocarbons (TPH), other volatile compounds (e.g., benzene, toluene ethylbenzene, and xylenes, known as BTEX) and polycyclic aromatic hydrocarbons (PAH) were measured in samples of the three hydrocarbons. For the analysis, TPHs are divided into four main fractions (F1, F2, F3, and F4) according to molecular weight, with the F1 fraction containing compounds of lowest molecular weight and highest volatility. PAHs comprise a relatively small amount of both oil and condensate. PAHs can persist in sediments and soils. They are considered toxic to many aquatic and terrestrial species. Relative proportions of volatile and less volatile compounds govern, in part, their physical properties and behaviour in the environment. Among the three hydrocarbons, condensate has the highest concentrations of volatile compounds (e.g., F1 fraction, BTEX), and diluted bitumen has the lowest (see Table 6-2). This contributes to the rapid weathering (primarily evaporation) rates for condensate shown in Table 6-1. PAH concentrations provide an indication of persistent toxic components, due to the longer time needed for natural degradation in the environment. Condensate contains the highest concentrations of parent PAHs among the three hydrocarbons tested (see Table 6-2). Total parent PAH concentrations were 30 mg/kg for diluted bitumen, 319 mg/kg for synthetic oil and 426 mg/kg for condensate.

6.3 Weathering and Fate of Spilled Hydrocarbon When liquid hydrocarbons are exposed to the physical elements, their physical and chemical properties begin to change immediately as the result of processes including spreading, evaporation, dissolution, water-in-oil emulsification, dispersion, sedimentation, oxidation and biodegradation (described below). Collectively, these processes are referred to as weathering. Although these processes usually act simultaneously, their relative importance varies with time and determines the fate and behaviour of hydrocarbons. Spreading, evaporation, dissolution, dispersion and oil-in-water emulsification are most important hydrocarbons first enter the environment. Sedimentation, oxidation and biodegradation are slower and longer-term processes that determine the ultimate fate hydrocarbons. Important factors determining the fate and effects of hydrocarbons in the marine environment include physical and chemical properties of the hydrocarbons, volume and prevailing climatic and sea conditions. When these weathering processes act on diluted bitumen, synthetic oil and condensate in the marine system, it is expected that condensate would have effects that are much more limited in space and time than those associated with bitumen or synthetic oil. Additional information on weathering in provided in the following subsections.

6.3.1 Spreading Spreading is one of the most important processes when hydrocarbons first enter seawater. The rate of weathering increases as spreading increases. The rate of spreading depends on the volume, viscosity, pour point and density of the hydrocarbon and the surface tension of the water. Low viscosity hydrocarbons

April 2010 FINAL Page 6-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons spread more quickly than those with a high viscosity. At temperatures below their pour point, hydrocarbons rapidly solidify and surface oil can remain many centimetres thick and spread very little.

6.3.2 Evaporation The rate of evaporation is determined primarily by the amount of volatile components in the hydrocarbon, ambient temperature and wind speed. Generally, components with a boiling point below 200°C would evaporate within 24 hours at 13°C. Hydrocarbon remaining on the sea surface after evaporation would increase in density and viscosity, which would affect subsequent weathering processes.

6.3.3 Dissolution Dissolution increases the bioavailability of hydrocarbon components. The rate and extent of dissolution depends on hydrocarbon composition (water-soluble fraction), degree of spreading, water temperature, turbulence and degree of dispersion. The most soluble components are low molecular weight aliphatic and aromatic hydrocarbons. The concentration of dissolved hydrocarbons in seawater under surface oil rarely exceeds 1 mg/L. Dissolution is a much slower process than evaporation, so it is not effective in removing hydrocarbons from the sea surface.

6.3.4 Oil-in-Water Emulsification Under moderate to rough sea conditions, hydrocarbons take up water to form oil-in-water emulsions. Stable emulsions can contain up to 80% water and the resultant viscous ―chocolate mousse‖ slows other weathering processes, particularly evaporation and natural dispersion, leading to enhanced persistence of the hydrocarbon. Breaking waves and wind speeds greater than 5 cm/s are generally regarded as necessary conditions for formation of stable oil-in-water emulsions. Stability of the emulsion usually increases with decreasing temperature.

6.3.5 Dispersion Turbulence from breaking waves mixes surface hydrocarbon into the upper water column as droplets of varying size. The larger droplets have sufficient buoyancy to re-join the surface oil. Others rise more slowly to form a sheen. Some droplets that are sufficiently small are kept in suspension and never re- surface, and thus become naturally dispersed. The amount of dispersion depends on the type of hydrocarbon. For example, light oils, which do not form stable oil-in-water emulsions, have a higher rate of natural dispersion than those hydrocarbons that form stable oil-in-water emulsions.

6.3.6 Sedimentation When weathered hydrocarbons interact with inorganic and organic particulates in seawater, the hydrocarbon can be adsorbed onto the particles. It is also possible for fine particulates to adhere to the dispersed droplets. The hydrocarbon/particulate complex accumulates in the benthic environment when the density of the complex exceeds the density of seawater. When oil and fine particles interact they typically form a buoyant emulsion referred to as oil-mineral aggregates (OMA). This process is usually most important when hydrocarbons enter particle-rich waters such as estuaries or shallow coastal waters.

Page 6-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons

6.3.7 Oxidation Hydrocarbons in oil react with oxygen in the presence of sunlight to produce soluble products and, over time, persistent aggregates referred to as tar balls. Degradation of oil by photo-oxidation is not a major process from a mass balance point of view but it can influence other weathering processes such as emulsification due to the production of polar compounds with surface-active properties.

6.3.8 Biodegradation Microbial degradation usually only occurs at a hydrocarbon–water interface. Factors influencing the rate and extent of this process include characteristics of the hydrocarbon, temperature, availability of oxygen and nutrients (primarily nitrogen and phosphorus) and the hydrocarbon surface area available for the microbes. The available surface area increases as oil and fine mineral particles interact, as does the rate of microbial degradation. Microbial degradation is a relatively slow process but, over time, can remove a large amount of introduced hydrocarbons from the marine environment.

6.3.9 Estimates of Weathering Processes Weathering rates of diluted bitumen, synthetic oil and condensate were estimated in a wind tunnel to simulate conditions (SL Ross 2010a). Results were provided as an input to the mass balance models developed for response planning purposes (Section 12). Results presented in Table 6-3 are representative of conditions after 6, 12, 18 and 24 hours and 3 days at sea. This information might then be used to assess further the fate of the hydrocarbons in seawater. Table 6-3 provides estimates of evaporation, the process that accounts for most of the weathering that would occur in the first three days after hydrocarbons enter seawater. In the conditions modelled, bitumen weathering occurs at a slower rate than synthetic oil. Condensate evaporates quickly, greater than 50% in the first 30 minutes. Predicting where hydrocarbons might end up is important for emergency response planning. The most volatile hydrocarbons typically evaporate immediately and the remainder spread over the water surface, mix into the water column and reach the shore. PAHs, which form a small percentage of the hydrocarbons to be transported, are among the most persistent compounds, and typically settle in sediment (Table 6-2). One way to predict the fate of hydrocarbons is to calculate the mass balance. Mass balance calculations take into account the physical properties of the specific hydrocarbons, which would influence how they are broken down in the environment. Mass balance calculations also incorporate how hydrocarbons weather in the environment, given a set of meteorological and oceanographic conditions at the time of the incident. The output is an estimated volume of how much hydrocarbon would likely end up in a given environmental compartment without mitigation or clean up intervention. The relative proportions of diluted bitumen and its fate in the environment of each of these environmental compartments is modelled (Section 12) for periods ranging from four hours to 15 days after an incident, depending on the volume of the hydrocarbons that enter the environment. Unmitigated hypothetical hydrocarbon spills and the respective fates in the environment are described in mass balance examples, based on meteorological and oceanographic conditions for a winter or summer event. The examples show how hydrocarbons might behave in the environment at specific locations, including interactions with shoreline areas.

April 2010 FINAL Page 6-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 6: Properties and Weathering of Liquid Hydrocarbons

Table 6-3 Hydrocarbon Evaporation Estimates

% Total Evaporated Hydrocarbon Type 6 hr 12 hr 18 hr 24 hr 48 hr 72 hr Diluted bitumen1 1 3 -- 7 -- 11 Synthetic oil2 7 14 19 20 22 22 Condensate3 >50* NOTES: 1 Based on an example for Principe Channel, in summer, involving a 10,000-m3 spill. 2 Based on an example for Emilia Island, in winter, involving a 10,000-m3 spill. 3 Based on an example for the Kitimat Terminal, in summer, involving a 250-m3 spill. *51% evaporated in the first 30 minutes and the remaining dispersed within the first hour.

SOURCE: Wind Observations in Douglas Channel, Squally Channel and Caamaño Sound TDR (Hay and Company Consultants 2010).

The mass balance calculations assume that none of the hydrocarbons are contained or recovered. However, in reality, recovery, containment (booming around the tanker) and exclusion booming to protect sensitive shoreline habitat would reduce the volume of hydrocarbons reaching the shore or highly sensitive areas. Therefore, the mass balance calculations include the maximum amount of hydrocarbons that could interact with the environment in each example. Estimates of where hydrocarbons might be found in the environment are based on results from mass balance analyses developed for response planning. Results of the mass balances, hydrocarbon characteristics and to where the hydrocarbons spread are discussed in Section 12 (see Tables 12-1, 12-3, 12-5, 12-7, 12-9, 12-11). Mass balances for condensate are not estimated for the CCAA, due to the high evaporation rate, but are estimated for the Kitimat Terminal.

Page 6-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

7 Incident Prevention and Response Prevention of accidents and injuries is of primary importance to Northern Gateway, whose policy is to operate in a manner designed to prevent incidents and implement emergency response to contain, remove and restore affected areas. In keeping with its policies on environmental protection, Northern Gateway has developed an OSRP concept founded on the vision of creating a model for response capability for the project-related activities; this includes: A combination of four key characteristics of a ―world class‖ OSR operation to be developed by Northern Gateway: a corporate commitment to ‗extended responsibility‘ to cover the marine approaches, as well as the pipeline and terminal operations an overarching strategy (the General Oil Spill Response Plan, or GOSRP) that provides an integrated management and operational approach for emergency response pre-staging of equipment and logistics support to enable a rapid response that is far better than regulatory requirements (i.e., an intended response time of 6 to 12 hours within the CCAA). Planning is in progress to determine the location of response centres which will govern the response time and therefore define the final response plan the OWA. Which will exceed the current requirements defined in the Canadian Shipping Act a robust risk reduction strategy (e.g., tanker vetting, tanker tracking, dedicated and tethered escort tugs, installation of improved aids to navigation, Use of ECDIS and Radar monitoring systems, use of local pilots on all project-related tankers in the CCAA, and the requirement for all tankers to be double-hulled) which is outlined in greater detail below. The marine strategy will include the following standards: all tankers using the Kitimat Terminal will be modern and double-hulled all vessels will be under the guidance of experienced and certified British Columbia-based marine pilots equipped with independent (PPU) navigation systems to provide guidance through coastal waterways only vessels meeting Northern Gateway‘s high safety and environmental standards will be permitted at the Kitimat Terminal all loaded tankers will be tethered to powerful escort tugs, the largest on Canada‘s West Coast, to facilitate safe passage through coastal waterways radar will be installed along the Northern and Southern Approaches to monitor all marine traffic and provide additional guidance to MCTS and to pilots vessel travel speed will be reduced to between 8 and 12 knots in the CCAA operational safety limits will be established to cover visibility, wind and sea conditions

April 2010 FINAL Page 7-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

first response stations and locally based personnel and equipment will be located at the Kitimat Terminal and in areas along the approaches to enable rapid response while docked at the Kitimat Terminal, tankers loading oil will be surrounded by a floating environmental protection system (i.e., a boom) Northern Gateway will require all tankers calling on the Kitimat Terminal to comply with an integrated approach to prevention and response for marine transportation of hydrocarbons, incorporating tanker design and prevention as key features. All applicable environmental, regulatory and statutory requirements will be addressed and compliance with the commitments will be monitored.

7.1 Canadian Coast Guard, Transport Canada and the Pilotage Authority Transport Canada is responsible for Port State Control, which is a ship inspection program whereby foreign vessels entering a sovereign state‘s waters are boarded and inspected to ensure compliance with various major international maritime conventions. Some of these include the International Convention for the Safety of Life at Sea (SOLAS), International Convention for the Prevention of Pollution from Ships (MARPOL), International Convention on Standards of Training Certification and Watchkeeping for Seafarers (STCW) and International Labour Organization Convention No. 147 (ILO 147) amongst others. The Safety and Security branch of Transport Canada is responsible for all Port State Control activities within Canada, and foreign ship inspections are carried out at all major ports by ship inspectors of the Marine Safety Branch. All vessels calling on the Kitimat Terminal will be required to meet CCG and Transport Canada requirements. Transport Canada has delegated the provision of Pilotage service to the Pilotage Authorities. The Pilotage Act created four pilotage regions with specific authorities, thereby replacing a large number of local pilotage districts. The four Pilotage Authorities - Atlantic, Laurentian, Great Lakes and Pacific are Crown corporations, responsible to Parliament through the Minister of Transport. Pacific Pilotage Authority was incorporated pursuant to The Pilotage Act on February, 1972 as a Crown corporation under Schedule III, Part I of the Financial Administration Act. The Pacific Pilotage Authority is in general governed by the following: Canada Shipping Act; Pilotage Act; General Pilotage Regulations; Pacific Pilotage Regulations; Authority by-laws. Tanker transit times in the CCAA will be in excess of 8 hours, and will therefore require at least two pilots to travel through Pilotage Waters. Objectives of the Pilotage Authority are to establish, operate, maintain and administer in the interests of safety an efficient pilotage service within a specific region. Pilots will board tankers from designated stations in the north: off Triple Island, near Prince Rupert Consideration will also be given to pilot boarding locations in Hecate Strait near Browning Entrance and Caamaño Sound

Page 7-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

7.2 Response Approaches and Capabilities Prior to commencement of shipping operations associated with the Project, Northern Gateway will complete project-specific emergency response plans for review Transport Canada. Contingency plans are prepared so that a response capability is in place to address accidents and malfunctions. The cornerstone of the contingency planning process for Northern Gateway is the GOSRP, which will be developed to describe measures and actions that would be implemented before or after an incident on land, at the Kitimat Terminal, along the Northern and Southern Approaches, or in the OWA. The GOSRP covers any oil spill with actual or potential consequences to personnel, the environment, property or the public at large. The GORSP outlines the actions and strategies that would be taken, in cooperation with government and other agencies, to manage a response. The Marine OSRP outlines emergency response measures that would be taken to mitigate the effects hydrocarbons released into the environment from and incident involving a project-related tanker. It is similar in concept to the GOSRP but will be more technical and provide support for the contingency plans, the tanker-specific SOPEPs and the tug SOPEPs (see Figure 7-1). The Marine OSRP will be a required component of all contingency plans. The SOPEPs will be required for each tanker and the support tugs of 300 gross tonnes or more. All tankers calling on the Kitimat Terminal will be required to provide a SOPEP as part of the tanker vetting process. Similarly, the Kitimat Terminal‘s Emergency Procedures Manual (Terminal OSRP) is an umbrella for the subplans that relate to oil spills and will include the TOM, the OPPP and the OPEP. The GOSRP and Marine OSRP will be completed and submitted to Transport Canada at least six months before the Kitimat Terminal begins handling bulk oil. All plans will be integrated with each other and with the British Columbia Provincial Marine OSRP and the CCG Pacific Oil Spill Contingency Plan (OSCP). The combined Marine OSRP, Project GOSRP and the Response Organization1 (RO) OSRP will secure compliance with the Canada Shipping Act and associated Response Organizations and Oil Handling Facilities Regulations (SOR/95-405). The detailed OSRPs will be followed to contain, remove or treat hydrocarbons. Key response actions are: safety notification (fire, police and emergency response) incident management source control assessment containment/recovery/protection strategies equipment and personnel deployment wildlife protection cleanup and decontamination demobilization and emergency response conclusion

1 Northern Gateway has proposed that participating coastal groups have direct ownership or involvement in a locally based RO in the CCAA. This could involve a partnership with an existing RO.

April 2010 FINAL Page 7-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

These actions are defined in the OSRPs; requirements for training and exercises also will be defined in the OSCPs. These requirements and policies will indicate the minimum standards expected and required of personnel to respond effectively. Terminal personnel, vessel operators, spill management and the RO will all be expected to participate in training, exercises and drills, so they are ready at any moment.

General Oil Spill Response Plan

Canada Coast Guard - Joint Contingency Plan, Canada-US Pacific Plan Dixon (Entrance)

BC Ministry of BC Geographic Environment MARINE Response Plans OSRP

Response Organization Area Plan

Marine Area Terminal Area

Marine Oil Spill Vessel Tug Ship Oil Terminal Oil Spill Response Plan Contingency Plan Pollution Response Plan Emergency Plan

Oil Pollution Prevention Vessel Ship Oil Pollution Plan Emergency Plan

Oil Pollution Emergency Plan

Terminal Operations Manual

Figure 7-1 Example of Contingency and Response Plan Integration

Page 7-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

The Marine and RO OSRPs will include sensitivity maps for the marine areas within the CCAA (Figure 7-2) and Geographic Response Plans (GRPs) that identify priority protection or response sites and strategies and tactics specific to geographic areas.

Roles and Responsibilities Although it is not a regulatory requirement, Northern Gateway is committed to ‗extended responsibility‘ for a response on marine approaches through ownership or direction of the certified RO. Response to a hydrocarbon spill in a marine environment involves the coordination of multiple agencies and groups with well-defined roles and responsibilities, which are described in the various OSRPs. Response actions are managed by implementing the Incident Command System (ICS). The GOSRP integrates the various emergency response documents to provide the framework for a successful response strategy. Table 7-1 lists some, but not all, key actions that might take place following an incident, including appropriate actions for Tier 1 (smaller spill) to Tier 3 (larger spill) responses. The tiers reflect maximum specific time requirements for response initiation, as well as increasing requirements for equipment and personnel. Some actions might be a single, one-off event, such as a notification, whereas others initiate a series of actions, some of which are concurrent and depend on other ongoing activities.

Table 7-1 Typical Sequence of Initial Response Actions

1. The vessel captain assumes role of the initial On-Scene Commander (OSC). 2. OSC notifies appropriate federal CCG, provincial, Northern Gateway and RO contractor contacts and designated coastal first response groups. 3. OSC initiates the first response actions prescribed in the OSRP. 4. OSC scales the problem and determines the most probable tier response level (based on a company philosophy of over-responding rather than under-responding). This assessment is reported to Northern Gateway contacts. TIER 1 RESPONSE LEVEL – FIRST RESPONSE 1. Onsite or local resources are deployed to contain, recover and clean up under direction of OSC and staff. 2. OSC informs the CCG Federal Monitoring Officer (FMO) of actions taken and intended actions. 3. OSC establishes a first response Command Post (CP) location. 4. The Crisis Management Team (CMT) mobilizes the Spill Management Team (SMT) to the level (depth of personnel) necessary to take over command and management of the response. TIER 2 and TIER 3 RESPONSE LEVEL – FIRST 24 HOURS 1. SMT conducts an assessment. 2. SMT recommends to the OSC which resources are appropriate to cascade to the response area. 3. SMT recommends response objectives, strategies and priorities to the OSC. 4. OSC informs CCG FMO of intended actions. 5. If appropriate, initial response OSC hands over to replacement OSC; CCG and the province are notified of the handover. 6. SMT establishes a finance section to provide funds, receive claims and document all costs and claims related to compensation. 7. If necessary, SMT establishes a new CP. 8. Participate in a Joint Information Center with the province and issues a press release on the incident and the response.

April 2010 FINAL Page 7-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Figure 7-2 Sample Coastal Sensitivity Map

Page 7-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

One of the keys to successful implementation of a response is that actions are agreed on and coordinated between the parties involved. To a large degree, the roles and responsibilities of emergency response are defined by regulations. The Responsible Party (RP) for a spill from a project-related tanker would be the tanker owner; however, through the planned implementation tanker‘s response plan, which is required to be integrated with Northern Gateway‘s GOSRP and Marine OSRP, Northern Gateway will oversee the response, management and implementation of the response operations and meet corporate commitments and objectives. The CCG, as the lead federal agency for all ship-source spills or pollution incidents in waters under Canadian jurisdiction, will advise the RP regarding spill response. In turn, Northern Gateway is advised of its responsibilities. Once satisfied with the RP‘s intentions and plans, the CCG will assume the role of Federal Monitoring Officer (FMO) on behalf of the Crown. The British Columbia Ministry of Environment (BC MoE) monitors on behalf of the province, participates in the response and coordinates local involvement. Environment Canada (EC) provides scientific and technical support and coordinates environmental input to the response as chair of the Regional Environmental Emergencies Team (REET).

7.3 Communications In the event of a spill from a project-related tanker, regardless of the origin or volume, the individual in command of the tanker would be required under Northern Gateway‘s GOSRP to report immediately to the Kitimat Terminal‘s Control Room; in turn, the Control Room Duty Operator would notify the : designated OSC response teams (Kitimat Terminal and RO) external contacts (CCG, province, public, community)

7.4 Response Objectives and Strategies Typical objectives and strategies for spills that would reach the marine environment are described in Table 7-2 (one or more of these might apply) and typical response activities are illustrated in Figure 7-3. Specific strategies for mass balance examples are provided in Section 12.

April 2010 FINAL Page 7-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Table 7-2 Examples of Response Objectives and Strategies

Objective Strategies Oversee the Safety of a. Identify hazards of spilled material (material safety data sheets [MSDS]) Citizens and Response b. Establish site control for hot zone (area with oil), warm zone Personnel (decontamination area), cold zone (non-oiled areas) and security c. Consider evacuations as needed d. Establish vessel and/or aircraft restrictions e. Monitor air in exposed and operational areas f. Develop site safety plan for response personnel g. Confirm that safety briefings are conducted Control the Source of a. Complete emergency shutdown procedures the Spill b. Conduct fire-fighting efforts c. Initiate temporary repairs d. Conduct salvage operations as necessary Manage Coordinated a. Complete or confirm notifications Response Effort b. Establish a unified command organization and facilities (e.g., CP) c. Confirm that Aboriginal and local (e.g., municipal) officials are included in RO d. Initiate Incident Action Plans (IAPs) e. Confirm that mobilization and tracking of response resources is carried out f. Account for personnel and equipment g. Complete documentation h. Evaluate planned response objectives vs. actual response (debrief) Maximize Protection of a. Implement pre-designated response strategies (e.g., booming of sensitive Environmentally areas) Sensitive Areas b. Identify resources at risk near the hydrocarbons c. Track hydrocarbon movement and predict their spread in the environment (spill trajectories) d. Conduct visual assessments (e.g., overflights) e. Develop/implement appropriate protection tactics Contain and Recover a. Deploy containment boom at the source Spilled Material b. Deploy containment boom at appropriate collection areas c. Conduct open-water skimming with vessels d. Evaluate time-sensitive response technologies (i.e., dispersants, in-situ burning) e. Establish waste recovery, transfer and temporary storage plan

Page 7-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Table 7-2 Examples of Response Objectives and Strategies (cont’d)

Objective Strategies Recover and a. Establish oiled wildlife reporting hotline Rehabilitate Injured b. Conduct injured wildlife search and rescue operations Wildlife c. Set-up primary care unit for injured wildlife d. Operate wildlife rehabilitation centre e. Initiate citizen volunteer effort for oiled bird rehabilitation Remove Stranded Oil a. Conduct shoreline cleanup assessment team (SCAT) surveys and develop cleanup priorities and plans b. Conduct appropriate shoreline cleanup efforts, including removal of sheens and mousse c. Clean oiled vessels, docks, facilities d. Inspect cleanup against endpoint criteria and sign-off segments as completed Reduce Economic a. Consider tourism, vessel movements and local economic effects Effects throughout response b. Protect public and private assets as resources permit c. Establish claims process Keep participating a. Provide forum to obtain stakeholder input and concerns Aboriginal groups and b. Establish a Joint Information Centre (JIC) Stakeholders Informed of Response Activities c. Conduct regular news briefings d. Manage news media access to response activities e. Conduct public meetings, as appropriate

April 2010 FINAL Page 7-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Spill occurs

Collect information on spill Collect information on environment

1. Evaluate oil characteristics and behaviour; 2. Monitor spill location and forecast oil movement and pathway

Is oil expected to reach key sensitive areas?

Yes No

Review Net Environmental Benefits Analysis

Protect key sensitive areas

Mechanical containment Yes Is mechanical containment No possible?

Mechanical collection

Is collection using Is manual Is washing Are dispersants or in- sorbents or collection possible? situ burning possible? skimmers possible? possible? No Yes No Yes No Yes No Yes

Perform collection using Perform manual Perform Apply dispersants skimmers or sorbents collection washing and/or perform in-situ burning

SCAT teams define oiling conditions Yes and initiate shoreline cleanup plan Has contact with key sensitive areas occurred? Clean shoreline segments to established endpoints No Transportation and storage of No waste, oil, and oiled materials Has the oil affected animals?

Yes Is it possible to end OSR efforts Rescue and rehabilitate wild animals

Yes No

Recycling of waste, documentation Planning steps for mitigation and restoration of measures and demobilization

Figure 7-3 Typical Response Activities for Hydrocarbons that Reach the Marine Environment

Page 7-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

7.5 Equipment and Personnel The Tier 1 (onsite) response capability for what is likely to be designated a Primary Area of Response (PAR) by the CGG is designed to exceed the Canadian Shipping Act (CSA) requirements by having trained, locally based personnel and equipment available to be mobilized immediately and deployed almost immediately (with Escort tugs) and with higher recovery capacity within the CSA the required six hours response time. Through its own resources, or those managed through the RO, Northern Gateway will have in place a model of response capability in the PAR that will greatly exceed CSA/CCG requirements. Onsite equipment includes resources at the Kitimat Terminal and at RO-designated response centres. Northern Gateway is exploring the potential for coastal groups to own, operate or directly participate in the RO. To date, only preliminary discussions have been held with organizations which may wish to collaborate.. Northern Gateway is committed to placement of response equipment in caches at four locations throughout the CCAA that will provide Tier 3 response times from 6 to 12 hours on water. As presently planned (see Table 7-1), Northern Gateway‘s response capability, as described in Level 1 terms, will be: more than 100 times the recovery capacity required by the CCG in the six-hour time frame for a PAR a Tier 3 – 2,500 t response time in the CCAA of 6 to 12 hours (Transport Canada guidelines indicate 18 hours) more than 2.5 times the CCG recovery capacity for a Tier 4 – 10,000 t situation, available in a maximum of 24 hours depending on distance from local response sites (Transport Canada guidelines indicate within 72 hours) part of a national and international mutual aid program to cascade additional resources, equipment and personnel designed to provide air and marine logistics support for personnel and equipment to initiate and sustain the response in a remote environment managed by a professional experienced spill management team capable of a sustained Tier 2 and 2+ response Northern Gateway envisions exceeding CCG requirements for: safety equipment (PPE [personal protective equipment], decontamination, gas monitoring, fire) response times boom (various sizes for distinct conditions) skimmers (for different oil types, water depths and operating platforms) vessels (tugs, oil spill response vessels [OSRVs], skiffs) pumps temporary storage (on water and shore) shoreline cleaning equipment wildlife hazing, capture and stabilization equipment

April 2010 FINAL Page 7-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Table 7-3 provides an overview of key OSRP equipment envisioned for the CCAA. Northern Gateway are reviewing the need and viability of extending response coverage in the open water area by installing additional satellite response centres to the north south and west of the confined channel area. Trained responders and logistical support will also be in place to mobilize and to maintain response assets.

Table 7-3 Overview of OSRP Equipment

Shuttle Location Boom Boats Skimmers Primary Storage Tanks Kitimat Terminal 2000 m 2 escort tugs 3 2 2 harbour tugs 2 terminal line boats Kitimat RO 3000 m 2 OSRVs 4 Minimum barge 6 3 2 tank shuttle boats capacity of 350 m 2 RO booming boats Kitimat Estuary 3000 m 4 booming boats 2 log mini-tugs PAR Pre-staged 2000 m 16 big local boats 2 Minimum barge 4 1 (four locations capacity of 350 m3 within the CCAA2) Escort Tug 1 400 m 1 escort tug 1 500 m3 internal (Tanker 1) Escort Tug 2 400 m 1 escort tug 1 500 m3 internal (Tanker 2) NOTE: a Local craft will be selected and incorporated under an arrangement such as the Fishermen’s Oil Spill Emergency Team (FOSET). b Northern Gateway are reviewing extension of satellite response centres to provide additional coverage of the open water area of the ship routes

Cascading of resources between other RO pre-staged sites and from mutual aid will expand capability considerably above these resources. Hydrocarbon recovery capability throughout the CCAA, based on caches of equipment (skimmers, barges and escort vessels) located within 6 to 12 hours of anywhere in the CCAA (based on a conservative speed for on-water delivery of 6 knots), is described in Table 7-4. Collectively, the recovery capacity envisioned for the ROs will provide a level of response that places it within the top terminal-port operations for oil preparedness worldwide.

2 The specific locations of the PAR pre-staged spill response equipment and personnel will be finalized in discussions with government agencies, participating coastal organizations and directly affected stakeholders.

Page 7-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Table 7-4 Proposed Hydrocarbon Recovery Capability

Effective Daily Location of Spill Response Recovery Capacity Equipment (EDRC) Storage Boom (m3/day) (m3) (m) Escort Tugs 960 1,000 800 Marine and Kitimat Terminal 2,400 10,750 7,000 South Douglas Channel 1,440 550 2,000 North CCAA 1,440 550 2,000 South CCAA 1,440 550 2,000 CCAA TOTAL 6,720 12,400 13,000

In addition to the specified OSRP equipment, personnel and logistical resources are required to operate and support response. This requires having dedicated trained personnel available 24 hours a day and the ability to mobilize additional contracted personnel. Response support activities will include air and on- water operations, in addition to booming and oil recovery. Air services will be pre-contracted and trained to provide overflights. A fleet of locally owned and operated support vessels will be identified to mobilize personnel and to support on-water activities. All support assets will be identified and inspected and arrangements made so that there is a robust fleet of transportation assets and crews. Further planning is in progress to determine the response recommendations for the OWA, which is anticipated to increase regional capacity and reduce response times within the region. Response times will be better than Transport Canada standards.

7.5.1 Escort Tugs It is envisioned that the escort tugs will provide a first line of prevention and emergency response. The project-specific, powerful, state-of-the-art tugs will accompany loaded tankers en route to and from the Kitimat Terminal. Emergency response equipment will be kept aboard the tugs to allow crews to deploy containment barriers should they be needed and as long as sea conditions allow safe deployment. High- capacity lightering pumps aboard the tugs will also be available for emergency transfers of oil.

7.6 Protection of Sensitive Areas The following types of sensitive areas have been identified for protection: environmentally sensitive and vulnerable areas (e.g., considering populations at risk, behaviour, location, habitat use, and seasonal-temporal aspects) traditional harvesting sites and locations of food, social and ceremonial (FSC) fisheries socioeconomic sensitivities (e.g., aquaculture, fisheries, tourism, shipping) cultural and archaeological sensitivities (e.g., traditional harvesting, historical sites)

April 2010 FINAL Page 7-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

Priorities for protection of sensitive areas will be assessed as part of the local, site-specific planning strategies that are typically developed as GRPs. Development of the GRPs will involve coastal organizations, local stakeholders and businesses that could be directly affected, and various levels of government. At the time of a spill, one of the initial objectives (see Table 7-2) is to assess oil movement and potential risk to sensitive areas. To that end, Northern Gateway has developed coastal sensitivity maps and oil pathway modelling that combines locally measured meteorological and oceanographic data to aid emergency response planning. Personnel in the Incident Management Team will request spill pathways so oil movement under conditions prevailing at the time of a spill (or during an exercise) can be identified. The pathway model provides timeframes considered for the deployment of protection countermeasures at key locations. On-water response strategies, combined with shoreline protection through booming, will be prioritized and implemented to limit potential oiling of sensitive areas.

7.6.1 Environmental Sensitivity Atlas A coastal environmental sensitivity atlas has been developed as part of background studies for project- related marine activities. The atlas identifies coastal sensitivities and types of shoreline, based mainly on information publicly available from the British Columbia government‘s Integrated Land Management Bureau (see example in Figure 7-2). The atlas is subject to field verification and refinement with input from local communities; however, it currently provides two key indices for shoreline oil sensitivity: coastline where long-term oil persistence is possible or likely due to combined low wave energy exposure and coarse, porous substrate (such as cobble or boulder) coastline where oil effects and persistence could be longer term due to low energy or coarse-grained beaches (Owens et al. 2008) Shoreline sensitivity information (e.g., areas of ecological or Aboriginal importance) and operational information (e.g., locations of airports, boat launches, and anchorages) are presented in a series of maps for the CCAA and OWA (see Coastal Sensitivity Atlas [Polaris 2010a, Polaris 2010b]). Northern Gateway has requested the assistance of coastal Aboriginal groups in ground truthing the existing coastal sensitivity maps (e.g., confirmation of shoreline types, locations of sensitive sites) and providing revisions on sensitive areas and other information. Failing this, additional ground truthing of the coastal sensitivity maps will be undertaken by Northern Gateway before beginning operations at Kitimat Terminal.

7.6.2 Key Sensitive Areas There are two general types of key sensitive areas: fixed locations (e.g., marshes) and mobile areas (e.g., shorebird concentration area or a whale feeding area). The mobile areas might vary geographically and temporally. An atlas helps to identify mostly the fixed locations, although areas frequented or commonly used by mobile species also can be illustrated on atlas maps (together with seasonal criteria). Key sensitive areas will encompass sites such as: sheltered fine-grained mud flats, coarse grained beaches, and marshes

Page 7-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

estuaries of rivers and streams of importance to salmon and other fish species Aboriginal traditional use and harvest areas Aboriginal coastal cultural sites (not provided in the Coastal Sensitivity and Operations Atlas) designated Marine Protected Areas key areas for migratory birds spawning areas for important fish species such as herring and eulachon marine mammal haulouts (the closest site to the proposed terminal is at the southern end of Ashdown Island, which is a known winter haulout for Steller sea lions and is more than 100 km south) Some of the preliminary priority protection sites, subject to field verification and consultation with local residents and agencies, are listed in Table 7-5. Before the tankers start calling at the Kitimat Terminal, Northern Gateway will complete planning processes with participating coastal Aboriginal groups and local stakeholders to identify sensitive areas and develop GRPs for specific areas.

Table 7-5 Examples of Sensitive Shoreline Areas

Location Sensitive Resources Timing Foch-Gilttoyees Bird concentration areas, including Marbled Murrelets, All year Marine Protected Area lagoon, park, wetlands, eulachon spawning, salmon- bearing river Gobeil Bay Protected area, beach and cabin All year Eagle Bay Park Coste Rocks Steller sea lion haulout, marine birds, park All year Emsley Cove Estuary, birds, marine wetlands, historic resources All year Miskatla Inlet Bird concentration area; wetlands Fall-winter (birds) Jesse Falls Marine Coastal ecosystem All year Protected Area Kitkiata Inlet Sheltered mudflats, marsh, Aboriginal fishery, bird Fall-winter (birds) concentration area Kiskosh Inlet Marsh, birds, Aboriginal fishery Hartley Bay Populated area, Aboriginal fishery, birds All year Dewdney and Glide Fens, bogs, and coastal ecosystem, birds All year Islands Ecological Reserve Ashdown Island Steller sea lion haulout All year

April 2010 FINAL Page 7-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

7.7 Shoreline Response At the outset of response operations, the spill management team will propose remediation endpoints for shoreline cleanup. These will be submitted to the CCG and reviewed by the REET, with input from regulatory agencies, participating Aboriginal groups and directly affected stakeholders. Endpoints are developed based on two key concepts: As Low As Reasonably Practical (ALARP) and Net Environmental Benefit. ALARP is an operational parameter that relates to practicality and safety. Net Environmental Benefit refers to the point at which the environmental benefits to be gained in continued removal of oil residues are outweighed by potential damage caused by the cleanup or treatment activity. For example, removal of relatively low levels of weathered hydrocarbons could require extensive disturbance of sediment or repeated washing of bedrock substrates. Once the endpoint is attained through cleanup and remediation measures, the oil residues are allowed to continue weathering through natural attenuation processes (biological degradation by microorganisms) that will further reduce their levels over time.

7.8 Financial Responsibility The response to a marine hydrocarbon spill would follow regulations and procedures and the costs will be reimbursed following established claims procedures. The RP for a spill from the marine terminal would generally be Northern Gateway, who will initiate, manage and implement the response operation. The tanker operator would be the RP if a spill results from a ship system failure. Northern Gateway and/or the tanker operator or their respective insurers will pay for the response and all reasonable claims in a timely manner. The tanker owner would be the RP for a spill from a tanker moving to or from the marine terminal or from an escort or harbour tug. The first response will be initiated, managed and implemented by the tanker owner‘s contracted representative (typically a certified RO) under the direction of Northern Gateway as prescribed in the Marine OSRP. The RP will pay for the response and claims, and then claim these costs from their insurers. The RP makes funds available, as necessary, for costs of the response and claims. For tankers, these are claimed through their insurance carrier, usually a Protection and Indemnity (P&I) Club, which is part of a larger non-profit mutual insurance system for marine transportation activities, including oil spills. Spill response financing and compensation amounts are addressed in Section 7.9. Agencies, authorities or private companies that participate in the response might have to bear the initial cost of their involvement or make arrangements with the RP to cover those costs. Costs associated with the response and damage or other claims are charged to the RP. These initial response costs can be supported, in some cases, by the Canadian Ship-source Oil Pollution Fund (SOPF) but later would be charged to the RP.

Page 7-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response

7.8.1 Response to Marine Spills Response Organisations (RO‘s) are certified by the federal government to respond to marine oil spills. ROs have plans, resources, training and equipment in place to effectively manage marine oil spill response operations. Oil handling facilities in Canada and marine vessels that transit Canadian waters must have agreements in place with certified ROs and an oil spill emergency plan. Oil handling facilities pay a bulk oil handling fee, and marine vessels pay an annual fee to ROs to maintain a level of preparedness to respond to an oil spill.

7.8.2 Ship Owner Liability Liability for marine oil spills in Canada is based on Canada‘s adoption of international conventions, which have been translated into federal legislation. Canada is a party to the International Convention on Civil Liability for Oil Pollution Damage (CLC). In Canada, the Marine Liability Act, S.C. 2001, c. 6 MLA) meets the requirements set out in the CLC. Under the CLC and MLA, ship owners are strictly liable for damage from oil spills. Whether or not the ship owner is at fault for the oil spill or negligence caused the oil spill, the ship owner would be responsible for oil spill costs unless the spill resulted from: an act of war, hostilities, civil war or insurrection Force Majeure (Act of God) an intentional act or omission of a third party or by government or other authority responsible for maintaining aids to navigation The MLA and CLC set limits on the liability of a ship owner, depending on the gross tonnage of the ship. Liability and compensation for marine oil spills is stated in Special Drawing Rights (SDR). The value of the SDR in Canadian dollars changes each day. Ships greater than 140,000 gross tonnes are liable for up to 89.8 million SDR (Can$152.2 million on October 1, 2009)3. So that the ship owner is able to cover its liability costs, they must establish a fund of the amount of its maximum liability, and must carry insurance. Insurance is ordinarily provided by international P&I Clubs, who provide coverage to ship owners against third party liabilities relating to the use and operation of ships.

7.9 Compensation In addition to ship owner liability, Canada has a tiered system of compensation for damage from oil spills not covered by a ship owner or its insurer. This system includes international funds available through Canada‘s participation in various international conventions and a national fund set up under the MLA.

7.9.1 International Oil Pollution Compensation Fund Under the first tier, compensation for oil spills is provided under the International Convention on the Establishment of the International Fund for Compensation for Oil Pollution Damage (IOPC Fund 1971, Internet site) as amended by the 1976 and 1992 protocols (IOPC). The fund established under this

3 On October 1, 2009 one SDR was equivalent to CAN$1.69936.

April 2010 FINAL Page 7-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 7: Incident Prevention and Response convention (IOPC Fund) provides compensation for pollution damage from oil spills in the following situations: where the damage exceeds the limit of the ship owner‘s liability under the CLC (in Canada the MLA) where the ship owner is exempt from liability under the CLC (MLA) because the damage occurred from a grave natural disaster, or wholly caused by an act or omission done with the intent to cause damage by a third party or by the negligence of public authorities in maintaining lights or other navigational aids where the ship owner is financially incapable of meeting obligations in full under the applicable CLC, and the insurance is insufficient to pay valid compensation claims The IOPC provides coverage such that the maximum compensation from the IOPC Fund and the ship owner or its insurer is 203 million SDR (Can$345 million).

7.9.2 Ship-Source Oil Pollution Fund The second tier in the system is the national fund, which in Canada is the Ship-Source Oil Pollution Fund (SOPF), established under the MLA. The maximum liability of the SOPF for one incident is $100 million SDR, adjusted annually ($154.4 million during the fiscal year commencing April 1, 2009). The total compensation available from the SOPF, IOPC Fund and the ship owner as of October 1, 2009 was $499.3 million SDR. The SOPF may be considered a fund of last resort because when all reasonable steps have been taken to recover payment of compensation from the owner of the ship and the IOPC and have been unsuccessful then compensation would be paid from the SOPF. It is also a fund of first resort because a person, other than a RO, may make a claim directly to the SOPF. If the SOPF pays the claim, the SOPF may then attempt to recover funds from the IOPC or the ship owner.

7.9.3 The International Supplementary Fund Canada recently amended the MLA to give effect to a third tier of compensation, the International Oil Pollution Compensation Supplementary Fund (Supplementary Fund). The amendments to the MLA will come into effect as soon as Canada ratifies the Supplementary Fund Convention. The Supplementary Fund makes additional compensation available so that the total amount payable for any one incident for pollution damage in Canada will be $750 million SDR (Can$1,274 million on October 1, 2009) including the amount payable from the ship owner or insurer and under the IOPC and SOPF. The Supplementary Fund is administered the same way as the IOPC Fund.

Page 7-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

8 Setting for the Open Water Area The OWA includes marine waters from the Alaskan border to the northern end of Vancouver Island, and from the 12 nautical mile limit of Canadian territorial waters (generally the continental shelf) landward to the northern fjords. The OWA is based on both ecological and administrative boundaries and is similar to the boundaries of the Pacific North Coast Integrated Management Area (PNCIMA) (Figure 8-1). The OWA is defined in terms of the British Columbia Marine Ecological Classification (BCMEC), which is a hierarchical classification that delineates marine areas in the province into Ecozones, Ecoprovinces, Ecoregions, Ecosections and Ecounits. Under the BCMEC system, benthic ecosections are classified based on wave exposure, depth, subsurface relief, seabed substrate, slope and current regimes. The Northern and Southern Approaches to be used by the project-related tankers and escort vessels fall within seven ecosections, including: Continental Slope, Dixon Entrance, Hecate Strait, North Coast Fjord, Queen Charlotte Sound, Queen Charlotte Straight and Vancouver Island Shelf, as described in Table 8-1 and shown on Figure 8-1. For example, the North Coast Fjord ecosection is characterized by a maze of waterways, inlets and glacial fjords. The fjords tend to be deep with steep sides and have relatively flat beds with thick sediments.

Table 8-1 Ecosections in the OWA

Marine Physiographic Ecosection Features Oceanographic Features Biological Features Continental Steep sloping shelf Strong across-slope and Upwelling zone, productive Slope down-slope turbidity currents coastal plankton communities and unique assemblages of benthic species Dixon Across-shelf trough Strong freshwater influence Mixture of inshore (neritic) and Entrance with depths mostly from mainland river runoff subpolar plankton species; <300 m; surrounded drives a north-westward migratory corridor for Pacific by low-lying coastal flowing coastal buoyancy salmon; some productive and plains (Hecate current and estuarine-like protective areas for juvenile Depression) circulation fish and invertebrate development Queen Predominantly shallow High current and high relief; Very important for marine Charlotte (<200 m), high relief very well mixed; moderate to mammals; migratory corridor Strait area with deeper fjord high salinity with some for anadromous fish; moderate areas freshwater inputs in the inlets shellfish habitat and fjords

April 2010 FINAL Page 8-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Table 8-1 Ecosections in the OWA (cont’d)

Marine Physiographic Ecosection Features Oceanographic Features Biological Features Hecate Very shallow strait Semi-protected water with Inshore plankton communities Strait dominated by coarse strong tidal current that with some oceanic intrusion; bottom sediments; promotes mixing; nursery area for salmon and surrounding coastal predominantly marine waters herring; abundant benthic lowlands invertebrates; feeding grounds for marine mammals and birds North Deep, narrow fjords Very protected waters with Low species diversity and Coast cutting into high restricted circulation, often productivity due to poor water Fjords coastal relief stratified exchange and nutrient depletion; unique species assemblages in benthic and plankton communities Queen Wide, deep shelf Ocean wave exposures with Mixture of inshore and oceanic Charlotte characterized by depths mostly >200 m and plankton communities; Sound several large banks dominated by oceanic water northern limit for many and inter-bank intrusion temperate fish species; lower channels benthic production Vancouver Narrow, gently sloping Open coast with oceanic wave Highly productive with inshore Island Shelf shelf exposures; northward, coast- plankton community; northern hugging buoyancy current due limit for hake, sardine, northern to freshwater influence; anchovy and pacific mackerel; seasonal upwelling at outer productive benthic community; margin rich fishing grounds for benthic fish and invertebrates

Modified from (Howes et al. 1997, Internet site)

Depths of the Northern and Southern Approaches range from abyssal to shallower areas, such as at Browning Entrance (128 m deep). Currents are typically less than 2 knots (i.e., currents within Queen Charlotte Strait exceed 3 knots about 8% of the time in some areas) (ILMB 2009a, Internet site). Channels tend to be wide; the narrowest channel sections (approximately 1.4 km) are in Douglas Channel, between Maitland and Emilia islands, and in Principe Channel adjacent to Dixon Island. Oceanographic and climatic conditions play a key factor in determining the species likely to be present in an area. A combination of global and local processes control productivity of specific areas. In general, solar heating, surface winds, freshwater inputs, local bathymetric features and tidal forces influence the circulation and mixing of ocean waters. Variability in climate and weather complicate the predictability of oceanic conditions.

Page 8-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Figure 8-1 Ecosections within the Open Water Area

April 2010 FINAL Page 8-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Currents within the OWA are driven by a combination of tidal and non-tidal water forces (Crawford et al. 2007). Non-tidal currents, including wind-driven, runoff/buoyancy-driven and bathymetric steering tend to dominate the tidal currents, whose oscillatory motion typically results in little long-term net movement. The subarctic current has the greatest large-scale influence within the OWA. This wind-driven, slow- moving (~5 to 10 cm/s), trans-Pacific current interacts with the continental slope and splits into two currents off the coast of British Columbia; the Alaska current flows north along the Alaska Gyre, and the California current flows around the East Pacific Gyre (Figure 8-2, Figure 8-3). The area where these two currents diverge is known as the transitional zone. This zone migrates north and south depending on atmospheric pressure systems. The transitional zone typically moves south during winter, meaning that most of the waters off coastal British Columbia experience a period of increased downwelling and lower productivity (Figure 8-2). During summer, downwelling is weakened and upwelling is noted in several areas where currents flow south (Figure 8-3). Figure 8-2 and Figure 8-3, derived from a figure produced by Thomson (1989) in Crawford et al., (2007), represent ocean circulation in the OWA during summer and winter. Figure 8-4 shows areas with oceanographic features (eddies, current systems, bottleneck areas and physically unique areas) that are considered ecologically and biologically important because they can be used easily as a measure for biological attributes (Clarke and Jamieson 2006). The areas identified in Figure 8-4 were originally identified as part of the Ecologically and Biologically Significant Areas (EBSA) identification process for the PNCIMA initiative.

8.1 Continental Slope The continental slope, or shelf break, has been identified as a proposed EBSA through the PNCIMA initiative (Clark and Jamieson 2006). The shelf break runs the length of the continental shelf and includes the canyons and troughs of Queen Charlotte Sound. Currents along the continental shelf generally flow north or south, depending on wind patterns and the forcing of cross-shelf sea surface slope due to runoff of fresh water from coastal areas (Crawford et al. 2007). Wind patterns and fresh water inputs are highly seasonal and create a northern flow during winter and southern flow or relaxation of northern flow during summer. Winter currents include the Davidson Current off Vancouver Island and the Haida Current off Haida Gwaii4 (Figure 8-2) (the Davidson Current might be disrupted by Queen Charlotte Sound). As a result of the Coriolis force, both of these northerly shelf break currents result in warmer low-nutrient surface waters building up along the coast and denser, high-nutrient waters being pushed deeper (Crawford et al. 2007). This downwelling is one of the primary reasons for the lower productivity in this area during winter. In summer, the Aleutian Low subsides and the shelf break currents weaken, often reversing to a southerly flow (Figure 8-3). This results in upwelling, where denser nutrient-rich waters are brought to the surface, which increases productivity. Upwelling during summer leads to accelerated primary production, which triggers increased biological activity. Given this, the shelf break has been designated an EBSA (Figure 8-4).

4 The name of Queen Charlotte Islands was changed to Haida Gwaii in December 2009. However, for consistency with source information used for mapping, Queen Charlotte Islands is used on all maps.

Page 8-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Figure 8-2 Winter Currents in the Open Water Area

April 2010 FINAL Page 8-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Figure 8-3 Summer Currents in the Open Water Area

Page 8-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Figure 8-4 Important Ecological and Biological Areas in the Open Water Area

April 2010 FINAL Page 8-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

8.2 Queen Charlotte Sound, Hecate Strait and Dixon Entrance Queen Charlotte Sound, Dixon Entrance and Hecate Strait are semi-contained areas characterized by relatively shallow waters and banks that promote strong tidal mixing. The tidal flood enters Queen Charlotte Sound, turns north and travels up Hecate Strait, where it is cut off by the flood tide travelling east in Dixon Entrance. This process reverses during ebb tides. Tidal currents are stronger at the surface and are likely influenced by density effects. Tides in this area might create baroclinic tides or internal waves due to tidal interaction with the shelf bathymetry. Spring tides over shallow areas and banks and around points of land result in strong tidal currents that remove local stratification and can bring dense nutrient rich waters to the surface. This action is noted over Goose Island Bank, Cook Bank, Rose Spit, western Dixon Entrance and, to a lesser extent, off Cape St. James (Jardine et al. 1993). Phytoplankton concentrations typically increase over these areas due to the increased nutrient availability. Tidal fronts associated with these features act to concentrate plankton, creating favourable feeding areas for many species. A large tidal front in Hecate Strait, present from spring through fall, has been identified as an EBSA due to the aggregation of zooplankton associated with this feature (Figure 8-4). Oceanographic characteristics of Queen Charlotte Sound and Hecate Strait are also affected by regional winds, internal waves and estuarine flow. During winter, southeasterly winds predominate, resulting in a general northward flow through Queen Charlotte Sound and Hecate Strait (Crawford et al. 2007). Summer winds are generally northerly, but are much weaker than winter winds, resulting in more variable currents and little net transport through Hecate Strait (Crawford et al. 1988). The northern currents during winter (October to April) are driven to the right by the Coriolis force, which results in a slightly higher sea level on the east side of Queen Charlotte Sound, Dixon Entrance and Hecate Strait. These northern surface currents displace deeper water in a process termed downwelling, which forces oxygen-rich water to move west across the bottom of these waterbodies during winter. Combined downwelling and wind-mixing in winter can create a well-mixed water column in many areas. This process increases the dissolved oxygen content of the bottom waters. During summer (June to August), the reverse process occurs and weak upwelling might occur in some areas, bringing nutrient-rich waters to the surface. Upwelling can be inhibited by the constant supply of fresh water that enters these coastal areas. Summer winds are also much weaker and likely to result in relaxation of downwelling rather than a true upwelling. This relaxation, combined with strong tidal mixing, can result in deep nutrient-rich waters being brought to the surface in the straits. Internal waves are formed at the interface of water masses of different densities such as the thermocline (Crawford et al. 2007) and typically result from the interactions of the tide with bathymetric features such as the shelf break. Due to the minor differences in density within the water column, compared with the water/air interface, internal waves can have large amplitude fluctuations. Internal waves are most pronounced near the shelf break in Queen Charlotte Sound and tend to be less in shallower waters throughout Hecate Strait (Crawford et al. 2007). Internal waves are capable of generating surface currents in excess of 0.5 m/s and might interact with plankton.

Page 8-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

Another major force driving coastal currents within the OWA is large-scale estuarine flow. Several of the fjords that connect with Queen Charlotte Sound, Dixon Entrance and Hecate Strait carry large amounts of fresh water into the region. Estuarine flow, where less buoyant water flows out of the fjords and estuaries, and more dense water flows towards the coast, is common throughout the OWA. Given the large number of fjords and freshwater inputs into the area, this flow is noted on a larger scale and can be inhibited or enhanced by local wind driven currents. The surface flow away from the coast is turned to the right by the Coriolis force and results in a northward current. This buoyancy-driven current contributes to the Alaskan Coastal Current that flows north over the continental shelf. As a result of these forces, currents within Queen Charlotte Sound, Dixon Entrance and Hecate Strait are subject to high variability. Summer residual currents within Queen Charlotte Sound are variable due to the weaker and inconsistent winds of summer. This results in little net movement through the area during the summer, which allows high concentrations of phytoplankton to develop. Strong outflow currents are noted near Cape St. James and Cape Scott, which are likely driven by summer northwest winds. In winter, currents tend to be more predictable and a general northward flow is observed through the majority of the area. A northerly flow is linked to downwelling during the winter.

8.3 Queen Charlotte Strait The southeastern areas of Queen Charlotte Strait experience extreme tidal mixing as tidal waters are forced through several narrows. These areas are so well mixed that little primary production occurs, as phytoplankton are unable to spend extended periods at the surface (Lucas et al. 2007). This allows nutrient-rich surface waters produced by tidal mixing to flow into the northwest areas of Queen Charlotte Strait and into Queen Charlotte Sound, where high levels of primary productivity occur. Queen Charlotte Strait also has several major inlets that bring fresh water into the area. The natural bottleneck of the narrows acts to concentrate salmon returning to spawn, which makes some areas in Queen Charlotte Strait preferred feeding habitat for several species of marine mammals. The area of highest primary productivity is the zone where water from Queen Charlotte Strait and Fitz Hugh Sound mix, providing maximum surface nutrients with thinnest surface layers. This is an ideal setting for phytoplankton growth (Mackas et al. 2007). Peak concentrations of chlorophyll have been observed in this area during May, as nutrients are rapidly depleted (Lucas et al. 2007). Overall flow from Queen Charlotte Strait is generally along the north coast of Vancouver Island, then south to Brooks Peninsula where waters are forced west offshore. Brooks Peninsula is often considered the dividing point on the West Coast, where little transport of plankton occurs between the north and south (Crawford et al. 2007).

8.4 North Coast Inlets and Fjords The north coast of British Columbia is characterized by glacier-carved fjords that tend to be deep, with steep sides, flat bottoms and sills or shallow ledges where they meet the ocean (Crawford et al. 2007). Freshwater input in these systems is at a maximum during late spring through summer (snowmelt) and is a major factor controlling fjord circulation and water properties (Crawford et al. 2007).

April 2010 FINAL Page 8-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area

The input of fresh water drives the estuarine circulation that is typical in these areas. During summer, when down-welling weakens over the continental shelf, deep-water renewal can occur in inlets and fjords (Crawford et al. 2007). Deep-water renewal during summer brings cooler, more saline, nutrient-rich and oxygen-poor water into the system. Basin renewals increase nutrient levels substantially in intermediate waters through entrainment, which results in the enrichment of outflow waters. These processes helps to boost the primary productivity of outflow areas (Crawford et al. 2007). Primary productivity is also higher in the outflow areas because particles have time to folliculate and settle out. This results in turbidity typically decreasing from the head towards the mouth of inlets, which allows for an increased euphotic zone and greater primary productivity near the mouth.

8.5 Vancouver Island Shelf The gently sloping shelf of Vancouver Island is generally considered to be the dividing point between the Alaska Coastal Current and the California Current domains. The California Current is characterised by moderate to strong upwelling winds in summer. The Alaska Coastal Current is characterised by relaxation of winter downwelling in spring and summer, rather than by upwelling winds (Lucas et al. 2007). During winter, the downwelling Davidson Current travels north along the Vancouver Island Shelf, close to shore (Crawford et al. 2007) (Figure 8-2). The southeast winds tend to push this current past Brooks Peninsula. In contrast, during summer, this current is often turned back at the Brooks Peninsula by the physical barrier of the peninsula and by wind-driven currents moving southward (Crawford et al. 2007). Brooks Peninsula often has an offshore flow of nearshore waters and is a north/south boundary for many organisms (Clarke and Jamieson 2006). Surface drifters deployed in the Queen Charlotte Sound drifted either southwest or grounded on Vancouver Island, but did not travel farther south of Brooks Peninsula, suggesting there was little transport of free-floating life (Crawford et al. 1999). Between the Brooks Peninsula and Scott Islands, plumes of cool coastal water often flow westward into the Pacific. It is expected that these come from northwest winds that pile up water along the shores of Vancouver Island (Lucas et al. 2007). Strong tidal flows and tidal mixing near the Scott Islands create an area of high productivity and nutrients (Clarke and Jamieson 2006; Lucas et al. 2007) in the surface waters, which supports seabird colonies on the Scott Islands (Lucas et al. 2007).

8.6 Uncommon Oceanographic Features Uncommon oceanographic features within the OWA include the transition zone, the Haida Eddies, El Niño events, bathometric parameters, current and tidal movements and the significant input of freshwater into the coastal marine system. The Haida eddies are believed to form along the west coast of Haida Gwaii (Yelland and Crawford 2005) when the winter outflow currents pass and interact with Cape St James to form one or two large eddies. The eddies are characterised by an anti-cyclonic rotation of currents more than 200 km wide and warmer than surrounding waters below the top 100 m. The eddies are also normally less saline and have a sea surface height anomaly of up to 0.5 m (Crawford et al. 2007). The necessary rotation of currents might result from buoyancy forces as the outflow current passes along the eastern side of the southern Haida Gwaii from the Hecate Strait (Miller et al. 2005). In the weeks and months following formation, the

Page 8-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 8: Setting for the Open Water Area eddies drift westward at typical speeds of 1 km/day (Yelland and Crawford 2005) into deeper waters of the Pacific ; as well, large amounts of warm and fresh coastal waters move out into the Gulf of Alaska (Crawford et al. 2007), along with the transport of plankton (Clarke and Jamieson 2006). Haida eddies are generally larger during winters of strong El Niño events that are associated with high sea levels along the central coast. In 1998, it was estimated that an eddy contained a volume of water approximately equivalent to Queen Charlotte Sound and Hecate Strait combined (Crawford et al. 2007). El Niño events occur every two to seven years and usually last six to eight months. Typically, the warm waters associated with El Niño begin to pass along the British Columbian coastline in late winter. The warm surface layer of the ocean thickens, preventing upwelling of nutrient-rich water and, resulting in less primary production and less food for marine animals, including commercial species, which in turn affects growth rates and behaviour (DFO 2003). The warmer waters enable unusual species (e.g., sunfish, tuna, Pacific mackerel) to move northward and migrate into British Columbian waters (DFO 2003). Mackerel are considered a problem species, as they feed and prey on juvenile salmon (DFO 2006). During El Niño events, spawning salmon returning to the Fraser River have been known to change their course of travel. For example, in El Niño years, salmon were displaced northward; over the years of observations, 10% to 90% moved around Vancouver Island via Johnstone Strait rather than the normal Juan de Fuca Strait route (DFO 2003, 2006). Coastal water zones of British Columbia are a maze of waterways, inlets and fjords that discharge fresh water into the marine environment. Fresh water, combined with offshore upwelling, produces a very nutrient-rich, highly productive marine environment. Freshwater input is a major controlling factor for fjord circulation and water properties, as the freshwater input drives estuarine circulation patterns (Crawford et al. 2007). Along the mainland, many of the freshwater inputs are driven by snowmelt and reach their maximum input in spring. The large volumes of freshwater inputs from the Columbia and Fraser rivers cause buoyant waters that initiate the coastal currents towards Alaska, subsequently feeding the Alaskan Coastal Current, which flows counter-clockwise around the Gulf of Alaska on the continental shelf from northern British Columbia through Alaska (Crawford et al. 2007). These coastal currents provide an important migratory corridor and habitat for salmon and other marine animals (Crawford et al. 2007). Much of the runoff that enters Queen Charlotte Sound comes from smaller rivers that ultimately feed in at Fitz Hugh Sound. Runoff to Hecate Strait and Dixon Entrance is primarily from the Skeena and Nass rivers (Crawford et al. 2007). The freshwater discharge causes a halocline, where the surface outflow of less dense water is accompanied by a landward flow of more saline ocean water at depth. This pattern of water movement is typical during summer and is enhanced through the deepwater troughs of Queen Charlotte Sound. During winter, the runoff is pushed back against the coast by wind, restraining the estuarine flow (Crawford et al. 2007). As described for the Vancouver Island ecosection, the tidal flows around southern Queen Charlotte Sound and Scott Island introduce nutrients to the surface waters and support seabird colonies on the Scott Islands (Lucas et al. 2007). Another important area in the OWA is Dogfish Bank; it is the largest shallow bank in the region and provides rearing habitat for a high diversity of invertebrate larvae (Clarke and Jamieson 2006). These uncommon oceanographic features, in combination, contribute to the concentrated phytoplankton biomass and high primary productivity found in the OWA (Clarke and Jamieson 2006).

April 2010 FINAL Page 8-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 9: Setting for the CCAA

9 Setting for the CCAA The CCAA includes Kitimat Arm, Douglas Channel and channels leading from Douglas Channel to open waters of Queen Charlotte Sound and Hecate Strait that will be used by vessels approaching the Kitimat Terminal (Figure 2-1). The network of waterways and channels connecting Kitimat with Queen Charlotte Sound and Hecate Strait are collectively known as the Kitimat fjord system. The approaches that will be used by tankers and escort tugs are part of the North Coast Fjord Ecosection of the Queen Charlotte Basin (QCB) Ecounit. This ecosection is characterized by a maze of waterways, inlets and glacial fjords. The fjords tend to be deep with steep sides, have relatively flat beds with thick sediments and contain glacial sills. Channels in the approaches have average depths greater than 365 m, with shallow areas at Browning Entrance (50 to 150 m) and between Banks and McCauley Islands (50 to 100 m). Channels are wide, with the narrowest at 1,400 m (Douglas Channel, between Maitland and Emilia islands). Along the coast of British Columbia, predominant winds are influenced by two major offshore pressure systems (Marine Physical Environment TDR (ASL 2010)). During winter, the Aleutian Low is associated primarily with southeasterly winds. In summer, the North Pacific High is associated with weaker northwesterly winds. Maximum sustained winds decrease by over 50% from the west coast of Hecate Strait to the inland waters of Kitimat Arm. Seasonal wind directions also vary between inland waters and coastal areas. In Kitimat Arm, southwesterly to southerly winds are most common between April and October and northerly winds are predominant from November to March, though stronger, southwesterly winds occasionally occur. These winds influence currents, water circulation and wave production in various portions of the approaches. The marine flora and fauna in the CCAA are typical of those in the highly seasonal, coastal marine ecosystem of the QCB. The QCB area provides valuable habitat for several commercial species, including five species of salmon (coho, chum, pink, sockeye and chinook), steelhead, many demersal and pelagic fish, and invertebrates such as crab and mussels. A number of ecologically sensitive species also occur within the QCB, including whales, seals, sea lions and some species of birds. Nearshore habitat along the approaches is characterized by a range of features including rocky shores, with some sandy beaches and estuaries. The predominant shoreline type in the CCAA is rock with gravel beach (29%), followed by rock, sand and gravel beach (25%) and rock cliff (21%), forming a combined 75% of the shoreline in the area. Species diversity of the rocky intertidal community is generally lower in inland waters (mostly rockweed) than along the Pacific Coast, where kelp and other species can also be present. The predominant intertidal fauna are barnacles, mussels, periwinkles and limpets. Species typical of the shallow subtidal community are sea urchins, moon snails, green sea anemones, sea stars and California sea cucumbers. The soft bottom estuaries are dominated by eelgrass, a marine vascular plant that provides important habitat for many juvenile fish and invertebrates. Commercially harvested bivalves such as butter clams and cockles inhabit sandy areas. Several fish species in the CCAA are important for commercial fisheries, recreational fishing, commercial recreational fishing and FSC fisheries. Kitimat Arm has several important salmon rivers, including the

April 2010 FINAL Page 9-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 9: Setting for the CCAA

Kitimat River, and there is a major pink salmon run in Bish Creek, just south of the proposed marine terminal location. Fish species commonly harvested include chum, coho, chinook and pink salmon, steelhead, eulachon and herring. Coves, estuaries and other nearshore areas offer habitat for juvenile salmon and other species, and serve as staging areas for adult salmon before their upstream spawning migrations in late summer and early fall. The marine mammals that occur most frequently in the area are killer whale, humpback whale, Dall‘s porpoise, harbour porpoise, Pacific white-sided dolphin, harbour seal and Steller sea lion. All but the harbour seal and Pacific white-sided dolphin are considered of special conservation concern by the federal Species at Risk Act (SARA), the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) or the British Columbia government. The CCAA is also an important waterfowl and seabird area. Large flocks of ducks and geese frequent the Kitimat estuary during fall and spring migrations. The many small estuaries, comprising slightly less than 3% of the shoreline within the CCAA, are also important areas (IAs) for wintering, migratory and breeding waterfowl. The Marbled Murrelet, a species listed as threatened by SARA and COSEWIC, is known to occur in the sheltered bays of the channels.

Page 9-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10 Effects of Hydrocarbons on the Biophysical Environment

10.1 Approach Regulatory agencies, Aboriginal organizations and communities have expressed concerns regarding the effects of hydrocarbons on the marine environment. Concerns include direct effects on ecosystem health, as well as effects on traditional harvesting, fishing and human health. The potential environmental effects of oil and condensate are assessed based on effects documented following past spills, particularly those in northern latitudes, Environmental monitoring following past spills provides an indication of the biological effects of hydrocarbon exposure, effectiveness of cleanup methods and recovery rates of affected populations and communities. Effects of hydrocarbons on marine organisms are influenced by a combination of biophysical factors, including: hydrocarbon physical and chemical properties (which determine behaviour in the environment and degree of toxicity) volume of hydrocarbons entering the environment oceanographic and physical conditions (e.g., currents, waves, weather and shore substrate) There is a range of sensitivities to hydrocarbons among the many marine species that could be exposed. When considering acute toxicity and the ability of a population to recover, the effect on a species will depend on the type of habitat used, sensitive times of year and life stage, and the ability to alter its behaviour in response to the presence of hydrocarbons (e.g., by avoidance). In general, species that depend on shoreline and intertidal habitat (some invertebrates and fish) or live on the water surface for a large portion of their life (some marine birds) are likely to show the most severe response to hydrocarbons. Table 10-1 provides a brief overview of documented effects of hydrocarbons on marine biota. Northern Gateway‘s proposed strategy is to integrate prevention and mitigation measures into standard operational practices. The actual probability of a hydrocarbon spill is low (see Sections 4 and 5), but should one occur, an effective response must be initiated quickly to limit potential effects. Sections 3 and 7 highlight key engineering and mitigation features and response plans designed to meet and exceed CCG-recommended response capabilities. The assessment addresses effects on the marine ecosystem by selecting and evaluating key components to characterize potential effects, highlight potentially vulnerable species (e.g., key life stages like breeding, feeding, spawning or migratory corridors) and identify sensitive habitat. Air, water and sediment, plankton, vegetation, invertebrates, fish, terrestrial mammals, marine birds and marine mammals are discussed. For some components, representative species are selected to represent responses of groups of organisms or species of conservation concern. The assessment

April 2010 FINAL Page 10-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

includes a discussion of the effects of hydrocarbon spills based on literature for past spills, mitigation specific to each component and recommended follow-up studies to document recovery.

Table 10-1 Overview of Effects of Hydrocarbons on Marine Biota and Terrestrial Wildlife

Group of Organisms Effects Plankton plankton might be at greater risk because of their proximity to the water surface and limited mobility. Effects on plankton could include decreased photosynthetic ability and changes in community structure (Kennish 1992; Varela et al. 2006; Gonzalez et al. 2009); however, they are capable of a fairly rapid recovery (NRC 1985; Kennish 1992) Algae (seaweed) adult rockweed might be relatively tolerant to oiling and resistant to smothering from hydrocarbons, but might be sensitive to high-intensity clean up techniques (van Tamelen and Stekoll 1996). sub-lethal effects on rockweed and kelp include leaf-loss, colour changes, accumulation of hydrocarbons and the slowing of reproduction (Fingas 2001) smothering can cause mortality or reduced growth rates to rockweed (Fingas 2001) hydrocarbons might cause loss of eel grass shoots, direct mortality, tissue contamination, inhabitation of shoot growth and increased physiological stresses (Zieman et al. 1984; Sloan 1999; Baca and Getter 1984) Invertebrates effects on invertebrates are physical smothering, altering metabolic and feeding rates, inhibiting reproduction and altering shell formation (Hoffman et al. 2003; USFWS 2004, Internet site) Fish the most vulnerable life stage of fish is the larval phase. During this time they are generally close to the water surface and have limited abilities to move away (Rice et al. 1976; Carls 1987; Paine et al. 1992) therefore hydrocarbons might cause mortality through ingestion of contaminated prey and oil droplets (Grahl-Nielsen et al. 1978; Norcross et al. 1996) adult fish are affected through the ingestion of residual hydrocarbons in the sediment or from contamination in the water. Sub-lethal effects include changes in heart and respiratory rates, gill structural damage and enlarged livers (Binderup et al. 2004) or mortality (Teal and Howarth 1984; Marty et al. 2003) Marine birds potential effects to marine birds include mortality from hypothermia (Szaro 1977; Stephenson 1997; Brown et al. 2003, Internet site), toxicity from ingestion of oil from preening or contaminated prey (Ainely et al. 1981) , reduced health due to increased energy expenditure in removing oil (Stephenson 1997; Brown et al. 2003, Internet site), reduced breeding success (Albers 1983; Fry et al. 1986; Valendo et al. 2005), asphyxiation (Leighton 1995) and loss or damage of habitat (Day et al. 1995, 1997) populations of localized species might be at risk, but evidence suggests no permanent damage occurs in regional seabird populations (e.g., Andres 1999; Wiens et al 2004)

Page 10-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-1 Overview of Effects of Hydrocarbons on Marine Biota and Terrestrial Wildlife (cont’d)

Group of Organisms Effects Marine mammals the full extent of effects on marine mammals is in large part difficult to quantify and predict (Geraci 1990; Loughlin 1994); however, some studies (e.g., Matkin et al. 2008) do show effects on individuals and populations sea otters are vulnerable due to fouling of insulating fur and ingestion of oil (Costa and Kooyman 1982; Ralls and Siniff 1990; Williams et al. 1988) Terrestrial wildlife potential to affect short-term health of individual animals that utilise the shoreline habitat and in extreme cases could result in mortality (Lipscomb et al. 1996; Hurst and Oritsland 1991; McEwan et al. 1974; Williams et al. 1988) toxicity or mortality can occur directly through oiling of plumage, skin or fur, ingestion of hydrocarbons and inhalation of vapour. Mortality can also occur indirectly through habitat loss or degradation, diminished food/prey resources and shifts to less productive habitats (Bowyer et al. 2003) hydrocarbon contact would be likely limited to feet and lower legs for most terrestrial based species The assessment of biota groups focuses on potential acute (i.e., short-term) effects of hydrocarbons on the biophysical environment in the CCAA and OWA, with some discussion of longer-term population-level effects. Longer-term ecosystem changes that could arise from bioaccumulation and biomagnification of contaminants in the marine environment are discussed in the Ecological Risk Assessment (Stephenson et al. 2010a). Past oil spills and the unmitigated hydrocarbon pathway examples described in Section 12 indicate the potential for substantial effects on some environmental components. For spills in northern cold water marine environments, such as the 1989 T/V Exxon Valdez spill of 41,600 m3 of crude oil, effects lasted a generation or more for some organisms; however, in the long term, these effects have not affected the structure, function and overall health of regional populations of most species (Harwell and Gentile 2006). Specific details for different biota are described below.

10.1.1 Geographic Extent of Study Area This assessment focuses on the CCAA and OWA and areas where hydrocarbons could travel (see Figure 2-1). The CCAA includes Kitimat Arm, Douglas Channel and approaches the vessels will use to enter Douglas Channel (Caamaño Sound and Squally Channel from the south; Principe Channel from the north). The OWA includes marine waters from the Alaskan border to Brooks Peninsula on Vancouver Island and from the limit of Territorial Waters landward into the northern fjords. The OWA is based on both ecological and administrative boundaries and is similar to the boundaries of the Pacific North Coast Integrated Management Area (PNCIMA). Section 12 provides examples developed to understand the potential fate of hydrocarbons at various times of year and assist in response planning. Depending on the time of year, prevailing wind and

April 2010 FINAL Page 10-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment oceanographic conditions, location and volume of the spill, and success of the response, hydrocarbons could move into various parts of the CCAA or OWA. The large scale of the OWA makes the use of Fisheries Management Areas (FMAs) of limited use for descriptive purposes. Fishing grounds and biological characteristics of target species are based more reliably on the biophysical ocean environment, such as provided under the British Columbia Marine Ecological Classification (BCMEC) (Section 8). The geographical distribution of marine species and fishing efforts is discussed in terms of the seven Ecosections identified in the OWA and depicted in Figure 8-1.

10.1.2 Assessment Data The use of appropriate scale data was an important consideration when examining data sources. Where possible, the assessment focussed on regional databases and information from broad-scale planning activities. Although information from site-specific studies was considered when developing regional databases, the specific results were not included due to the scale of the information versus the resolution of the assessment, and the tendency for detailed studies to create a bias towards areas that have been intensively studied, rather than areas that are regionally ecologically important. Further, many of these site-specific studies would not be visible on maps at the scale used in this analysis. Specific characteristics considered when identifying areas of importance included the high mobility, widespread distribution, annual variation and large information gaps that exist for marine species. Background documents produced to support integrated management in the PNCIMA were used to prepare information about the OWA, including an Ecosystem Overview Report (Lucas et al. 2007), a Marine Use Analysis report (MacConnachie et al. 2007) and identification of EBSAs (Clarke and Jamieson 2006) . DFO identified EBSAs through a two-phase process as part of the PNCIMA initiative. In Phase One, almost the entire PNCIMA was identified as being an important area (IA) for at least one species (Clarke and Jamieson 2006). Phase Two limited the areas by restricting the EBSAs to physical features, including oceanographic features, bottleneck areas and the sponge bioherms or reefs (Clarke and Jamieson 2006). The rationale for limiting EBSAs to physical features is that species are typically associated with unique ocean features. This process identified 15 EBSAs covering 45,182 km2 (44%) of the PNCIMA (Clarke and Jamieson 2006), which are shown in Figure 8-4. Clarke and Jamieson (2006) noted that the scale of assessment for identification of the LOMA EBSAs was not useful for identifying shoreline and fjord EBSAs. The Coastal Resources Information Management System (CRIMS) was also used as a data source in this assessment. Since 1979, British Columbia has been collecting resource information in a systematic and synoptic manner using peer-reviewed provincial Resource Information Standards Committee (RIC) methodologies. Both biophysical and human use data are collected. The CRIMS provides data and analysis for coastal resource management, conservation, protection and planning. For this review, CRIMS data were viewed online and relevant information was requested from the Integrated Land Management Bureau (ILMB 2009b, Internet site). Fish and fish habitat data from the internet-based GIS application MAPSTER were also incorporated into the OWA assessment. Many of the layers included in the OWA were obtained from DFO after viewing them on MAPSTER. Datasets of use to the OWA represent specific themes such as fish species presence and

Page 10-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment distribution, marine habitat and administrative boundaries. MAPSTER provided the viewing tool to locate useful data, which were then requested from DFO.

10.2 Air Evaporation of volatile constituents is an important weathering process. Evaporation can account for 13 to 75% of the volume of diluted bitumen, synthetic oil or condensate within 24 hours of hydrocarbons entering the environment (see Table 6-1). Some factors that influence the rate of evaporation are wind speed, solar radiation, formation of emulsions (which limits evaporation in water), air and water temperature, vapour pressure of the hydrocarbon and wave activity. The oil constituents that affect air quality are volatile organic carbons (VOCs), which are defined as any compound of carbon that is involved in atmospheric photochemical reactions. Heavy oils contain the smallest volatile fraction, whereas light oils and condensates contain the most. On evaporation, VOCs leave the oil surface and their atmospheric concentration is governed by meteorological conditions. They disperse in the atmosphere rapidly by simple diffusion and mixing, even in calm conditions, at rates influenced by wind speed, wind direction, atmospheric stability, air temperature and roughness of the water surface. VOCs are reactive and they are dispersed and chemically broken down by photo-oxidation. Most VOCs have relatively low toxicity, so concentrations near the initial location of the hydrocarbons might not be of concern. VOCs such as BTEX might be present in sufficiently high concentrations to be of concern, particularly for responder and public safety, but concerns can be mitigated by evacuation to a safe distance. Farther afield, VOCs are well dispersed and of less concern. Dispersion and photo- oxidation lessen their potential effect on human health and the environment. The effects are most pronounced in the immediate hours after a spill.

10.3 Water and Sediment Some of the hydrocarbons would dissolve into the water column and eventually disperse; however, the concentration of dissolved hydrocarbons in seawater under surface oil rarely exceeds 1 mg/L. Compounds that take a long time to biodegrade, such as PAHs, will settle in the sediment and contribute to chronic toxicity. Rapid containment and removal of hydrocarbons, in conjunction with emergency response, will reduce the effects on water and sediment. No specific mitigation measures are used for water and sediment beyond those proposed for the general response. Monitoring will include determination of amounts of hydrocarbons and specific hydrocarbon constituents in various substrates (e.g., muds, silts, sand, and cobbles) and will continue until hydrocarbon concentrations return to pre-spill levels. The use of dispersants and their effects on water and sediment would be investigated upon approval of the type and use of dispersants as a response measure.

10.4 Plankton Plankton form the base of the marine food web and are the foundation for marine ecosystems. They are typically most abundant in surface waters, but some species occur at a range of depths. Plankton includes plants (phytoplankton) and animals (zooplankton such as copepods, invertebrate larvae and euphausiids)

April 2010 FINAL Page 10-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment that provide food for many larger species such as larval fish, juvenile salmonids and filter-feeding organisms ranging in size from barnacles to humpback whales. Planktonic organisms have limited motility and their horizontal movement is controlled largely by ocean currents. There are limited data on plankton assemblages in Douglas Channel, but communities are most likely typical of mid-latitude subarctic marine systems. Information about phytoplankton and zooplankton in the open waters of Hecate Strait–Queen Charlotte Sound is provided by Ware and McQueen (2006), Lucas et al. (2007) and Clarke and Jamieson (2006). Within the CCAA, DFO identified Caamaño Sound as an ‗Important Area‘ because of its particularly high productivity as a result of tidal mixing (Clarke and Jamieson 2006). NOAA (2009a, Internet site) maintains a database on plankton for the North Pacific Ocean. Many marine organisms migrate to British Columbia‘s productive waters to feed. The coastal areas and upwelling zones of the OWA are rich in concentrated phytoplankton biomass and high in primary productivity. In general, regions of high productivity provide important feeding habitat for many marine organisms. Many species also depend on the specific physiographic nature of local areas to aid in concentrating their prey. Much of the OWA is considered an important feeding area to at least one species of marine organism at some point during the year. Abundance and concentrations of marine organisms feeding on plankton are generally highest during spring, summer and fall, but little is known about abundance, distribution or habitat usage by many species during the winter months. Mesoscale factors such as eddies and currents structure the plankton assemblages in open portions of the OWA. In nearshore areas, currents and nutrient availability are the main factors influencing plankton. Additionally, topographic complexity of the seabed and shoreline can account for smaller-scale variations in plankton abundance, such as through retention of nutrient-rich water, maintenance of water column stratification and depth profiles that maintain phytoplankton within the photic zone. As a result, the composition of both phytoplankton and zooplankton in the OWA is patchy over time and space; however, several areas have been observed to retain higher plankton concentrations on a seasonal basis. The west side of Haida Gwaii has high concentrations of macroscopic zooplankton, largely as a result of periodic summer upwelling along the Queen Charlotte Shelf. Around southern Queen Charlotte Sound and Scott Island, higher nutrient levels can support higher plankton levels. Also, Dogfish Bank serves as a rearing area for a large diversity of invertebrate larvae (Clarke and Jamieson 2006). Increased tidal mixing over Goose Island Bank, Cook Bank, Rose Spit and western Dixon Entrance results in nutrient pulses that lead to increased phytoplankton concentrations. Tidal fronts further concentrate plankton, leading to aggregation of zooplankton, and supported by high concentrations of phytoplankton (i.e, this occurs in spring through to fall in Hecate Strait). Reduced summer current activity in areas such as Queen Charlotte Sound results in low net water movement through the area, allowing increased concentrations of phytoplankton to develop.

10.4.1 Potential Effects of Hydrocarbons on Plankton Potential effects of a condensate are considered to be greater than those associated with diluted bitumen or synthetic oil, given the greater amount of condensate that can dissolve and disperse in the water (Section 12), and are discussed here. Plankton might be at greater risk from hydrocarbons than other biotic groups due to their presence near the water surface and limited ability to move away. The ERA describes the risk of acute lethality for plankton communities from exposure to condensate, related to

Page 10-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment levels of BTEX and TPH, which would be short-lived (Marine ERA for Kitimat Terminal Operations TDR, Stephenson et al. 2010b). However, while laboratory studies have shown adverse effects, such as decreased photosynthetic ability and changes in community structure (Kennish 1992; Varela et al. 2006; Gonzalez et al. 2009), no mass toxicity to phytoplankton has been reported from the field, either from a spill or from chronic input conditions (National Research Council [NRC] 1985). Effects of spills on plankton were also not mentioned as a concern in the updated review on the effects of oil by the NRC (2003). Fluctuations in plankton populations after oil spills have often been attributed to natural variables such as climate and oceanographic conditions. Copepods are the most researched species of zooplankton and have been shown to accumulate hydrocarbons via ingestion, which might lead to effects on feeding and reproduction. Both phytoplankton and zooplankton recover from an oil spill fairly rapidly, mainly due to recruitment from other areas and to intrinsic biological characteristics such as wide distribution, large numbers, short generation times and high fecundity (NRC 1985; Kennish 1992).

10.4.2 Mitigation Measures No specific mitigation measures are used to protect plankton communities, other than those described in general for prevention, response and environmental protection (e.g., booming) (Section 7). Although monitoring of plankton communities could be conducted, and would provide useful information about the marine food web, their expected rapid recovery suggests that monitoring programs would be more informative if efforts were focused on water and sediment quality and comparisons of these data to guidelines for protection of aquatic life.

10.5 Marine Vegetation Marine vegetation includes macrophytes (seaweeds) and aquatic angiosperms (flowering plants), which provide habitat for a wide range of animals. Any change in health of marine vegetation because of exposure to hydrocarbons would be considered a change in habitat, and of concern under the federal Fisheries Act. Fish habitats are defined as ―spawning grounds and nursery, rearing, food supply and migration areas on which fish depend directly or indirectly to carry out their life processes‖, and fish are defined as all the life stages of all ―shellfish, crustaceans, marine animals and any parts of shellfish, crustaceans or marine animals.‖ Adverse effects on marine vegetation documented after a hydrocarbon spill include direct damage to plants through oiling and indirect effects related to cleanup efforts (removal of oiled vegetation). Intertidal areas denuded of algae after a spill usually re-populate within several growing seasons once the oil has been mostly removed; however, full recovery of associated communities can take up to a decade (Peterson et al. 2003).

10.5.1 Baseline Conditions Baseline information about marine vegetation in the CCAA is provided in Bates (2008) and is summarized here. Species of brown, green and red macroalgae occur in the intertidal and subtidal zones in various preferred habitats. Brown algae make up the bulk of the algal biomass. Red algae are the most diverse group, stretching from the intertidal zone to the deepest parts of the sunlit waters. Common species include the green alga Ulva spp. (sea lettuce), brown algae such as the kelps Nereocystis

April 2010 FINAL Page 10-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment luetkeana (bull kelp) and Macrocystis integrifolia (giant kelp), as well as Fucus distichus (rockweed), and red alga Chondracanthus exasperates or Turkish towel (Bates 2008). The eelgrass Zostera marina is the most common marine angiosperm in the CCAA, forming perennial beds along protected stretches of shoreline (Pojar and MacKinnon 1994). Kelp and eelgrass beds provide many important ecosystem functions including residence, refuge, foraging, breeding and nursery habitats for many marine invertebrates, fishes, seabirds and some mammals (Hall 2008). They also contribute to sediment stability, which is important for nutrient cycling (Fargo et al. 2007). Floating kelp rafts might also serve as habitat for larval and juvenile fishes, effectively transporting them among spatially isolated local populations of adults (Hobday 2000; Kingsford 1992; Kokita and Omori 1998). Eelgrass and kelp beds have been identified as major habitats vital to the overall health of the marine ecosystem in British Columbia (CIT 2003, Internet site). Commercially important fish species, such as Pacific herring, spawn on intertidal and subtidal plants (Hall 2008). Kelp beds in the CCAA are composed primarily of bull kelp, with some giant kelp beds in the northwest around Browning Entrance. Eelgrass beds are composed mainly of Zostera marina (Fargo et al. 2007). Commercially important fish species, such as Pacific herring, spawn on intertidal and subtidal plants (Hall 2008). Eelgrass and kelp beds have been identified as major habitats vital to the overall health of the marine ecosystem in British Columbia (CIT 2003, Internet site). Marine macrophytes are restricted to coastal areas of the OWA, given that most of the seabed is below the photic zone. Baseline vegetation surveys of coastlines within the OWA have not been completed, but it is assumed these areas are typical of outer coast Northeast Pacific seaweed assemblages, retaining areas of high species richness, biomass and percent cover where rocky substrate is available. These assemblages are exposed to greater wave action, nutrient availability and supply of dispersing reproductive propagules than those within the CCAA.

10.5.1.1 Rockweed Rockweed is ubiquitous and has the largest macrophyte biomass throughout the CCAA; it is considered the predominant alga in the mid to high intertidal regions on the Pacific Coast (Druehl 2000). It is perennial, persisting through winter, when sexual reproduction is at a peak, and growing mainly during summer (Ang 1991).

10.5.1.2 Eelgrass and Surfgrass Eelgrass prefers muddy or sandy substrates and its occurrence and seasonal distribution along the coast is influenced by desiccation, temperature, salinity and water motion. Shoots are susceptible to scour and wave action so are typically found in more sheltered waters. Eelgrass is perennial and, in this region, reproduces mainly asexually by lateral shoots, which develop from late March to June. Some flowering might occur in May and June, but few seeds mature into plants (Phillips 1984). Spatial distribution of eelgrass in the CCAA has not been completely defined (Clarke and Jamieson 2006; Hall 2008); however, much of the shoreline is not ideal for eelgrass growth as it is steep and rocky. Small patches of eelgrass have been identified around the proposed terminal location, with the largest bed found in the Bish Creek estuary to the south of the terminal. Other patches were found in Minette Bay and the Kitimat River

Page 10-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment estuary (Marine Fish and Fish Habitat TDR). When available, eelgrass beds have been included in the coastal sensitivity atlas prepared for the Project. In exposed and semi-exposed areas along shorelines within the OWA, eelgrass is replaced by the vascular plant surfgrass (Phyllospadix scouleri), which is similar to eelgrass in many developmental and ecological characteristics but is restricted to rocky substrates and areas with moderate to substantial water motion.

10.5.1.3 Kelp Bull kelp forms extensive beds on bedrock, reefs and boulder fields 3 to 20 m deep (Nicholson 1970; Vadas 1972; Springer et al. 2007). It is essentially an annual species, although in some populations, individuals produced late in the season might successfully overwinter and survive a second year (Chenelot et al. 2001). Maximum growth occurs in summer and early fall; mortality of bull kelp sporophytes peaks in the winter, primarily due to being washed out by winter storms (Springer et al. 2007). Like eelgrass, spatial distribution of kelp beds is not completely known for the CCAA and presence of IAs has not been documented (Clarke and Jamieson 2006; Hall 2008). However, according to the Coastal Resource Information Management System (CRIMS) (ILMB 2009b, Internet site), there are several kelp beds within the CCAA around the Estevan Group, the western edge of Campania Island, the northern tip of Banks Island and the southwestern edge of Porcher and Goschen islands. Along OWA shorelines, there is a much higher diversity of species in the kelp assemblages than in the CCAA. The northeast Pacific has the highest kelp diversity in the world, and OWA coastlines represent the southern extent of the ‗northern‘ northeast Pacific kelp flora. Several canopy-forming kelp species have inherent flotation, allowing them to remain buoyant and viable after detaching from the rocky substrate. These floating kelp mats might be found throughout the OWA, provide habitat and refuge for invertebrates, seabirds and fish and can allow long-distance dispersal of kelp to occur.

10.5.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Vegetation

10.5.2.1 Rockweed Observations of adult rockweed plants at heavily oiled sites suggest it might be relatively tolerant to oiling and resistant to smothering but the species is sensitive to high-intensity cleanup techniques such as power washing and hot water washing (van Tamelen and Stekoll 1996). Rockweed is likely to be most vulnerable to hydrocarbon exposure during summer, when the growth rate is greatest (Ang 1991). Absorption of hydrocarbons from the water column can lead to sub-lethal effects, including frond loss, colour change, accumulation of hydrocarbons and slowing of reproductive and growth rates (Fingas 2001). Mortality or reduced growth can also occur through smothering (by limiting access to light and nutrients) or toxicity of the hydrocarbons. Following EVOS, rockweed disappeared from the mid and upper intertidal zones. Other factors might be attributed to the loss, including trampling in frequently accessed areas, cleanup activities, competition from other seaweeds, and severe weather. Hot water and high pressure washing cleanup techniques might have contributed to loss of plants (van Tamelen and Stekoll 1996). The dramatic loss of rockweed triggered indirect effects such as algal blooms, declines in associated invertebrate populations (limpets,

April 2010 FINAL Page 10-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment periwinkles, whelks) and inhibited recovery of the rockweed, whose recruits avoid desiccation under the canopy of adult plants (Peterson et al. 2003). Following the initial decline in abundance, recovery of biomass was noted within three years, although recovery of reproductive capacity took longer, particularly in the upper intertidal zone. Few differences between oiled and unoiled areas were observed two to three years after the spill, although fluctuations in abundance of rockweed in areas treated with hot-water washing were noted for several years (Houghton et al. 1997). Delayed indirect effects extended over a decade, as opportunistic species of flora and fauna were gradually replaced by single-aged stands of Fucus, which died in cycles, starting the whole process again (Ott et al. 2001, Internet site; Peterson et al. 2003).

10.5.2.2 Eelgrass and Surfgrass Observations of eelgrass after exposure to hydrocarbons suggest vegetative growth might recover rapidly after initial loss of shoots, but that effects on associated invertebrate communities last longer. Eelgrass is probably most sensitive to hydrocarbons during spring and summer, as those are critical times for seed production and shoot growth (Wright 2002). Additionally, intertidal and high subtidal eelgrass beds have a greater risk of exposure to hydrocarbons during the low summer tides. Potential effects of oiling include direct mortality due to smothering or toxicity, indirect mortality of associated species due to loss of food or habitat alteration, uptake of hydrocarbon contaminants into tissues and incorporation into food webs and increased physiological stresses (Zieman et al. 1984; Baca and Getter 1984; Sloan 1999). Studies following the 1978 Amoco Cadiz and 1989 Exxon Valdez incidents provide information about effects on eelgrass communities. The Amoco Cadiz spill resulted in heavy oiling of eelgrass beds, with blackening and loss of many shoots in the first weeks post-spill; however tissue production soon returned to normal (Jacobs 1980). Dean et al (1998) reported that the modest damage to plants did not persist for more than a year following EVOS; noted effects included a slight reduction in eelgrass density and production of flowering shoots. Recovery of benthic and infaunal communities associated with eelgrass habitat was much slower than recovery of the plants themselves. Six years after EVOS, many affected species had not recovered, which was partially attributed to concentration and persistence of oil in the soft sediment (Dean and Jewett 2001). In sheltered habitats, such as the Kitimat River estuary and Minette Bay, hydrocarbons left untreated could persist for several years in the fine-grained sediment of the tidal flats and marshes. The same would be true for sheltered estuaries throughout the OWA.

10.5.2.3 Kelp Because kelp inhabit deeper water, they are less vulnerable to effects from hydrocarbons than intertidal species such as rockweed, although rafts of bull kelp floating at the surface, which are common throughout the CCAA and OWA, particularly after storms, might be susceptible to oiling. Kelp are likely most vulnerable during summer and early fall, when growth rates are highest. In the CCAA and coastal areas of the OWA, kelp beds are typically found along more exposed areas of coast, such as Porcher and Goschen islands in Browning Entrance, where wave action would hinder settlement (and subsequent concentration) of oiled sediments.

Page 10-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Kelp will respond similarly to rockweed, with frond loss, colour change, accumulation of hydrocarbons and slowing of reproductive and growth rates (Fingas 2001). Following EVOS, Dean and Jewett (2001) noted lesser effects of oiling and cleanup activities for kelp than for eelgrass, and no differences in density or biomass of kelps in oiled and reference areas. Dean et al. (1996) observed a higher proportion of bull kelp at oiled sites in more exposed habitats, suggesting recent recruitment in response to the disturbance caused by the spill. Recovery of invertebrate communities associated with vegetation was faster for kelp beds than for eelgrass, with most of the affected groups recovering within two to four years after the spill (Dean and Jewett 2001).

10.5.3 Potential Effects of Condensate on Marine Vegetation Condensate would have fewer effects than synthetic oil or diluted bitumen, as it would evaporate quickly from the water surface, spread more slowly and cover less shoreline habitat. However, it is lighter than diluted bitumen and more easily entrained in the water column, making it likely to affect subtidal vegetation (French-McCay et al. 2009) such as kelp and eelgrass.

10.5.4 Mitigation Measures Cleanup with skimmers and sorbents will reduce the amount of residual hydrocarbons in the marine environment. Specific mitigation for rockweed is not suggested, because it is ubiquitous on rocky substrates throughout the CCAA and it is not possible to boom off all these areas. Based on historical cleanup programs, aggressive cleanup activities (i.e., hot water and high-pressure washing) would not be recommended as they result in greater damage than leaving the oil in place (van Tamelen and Stekoll 1996). Protection of kelp and eelgrass beds will be a high priority due to the more specialized ecosystem functions they provide (food and habitat for many species). Exclusion booms would be placed around sensitive eelgrass habitats in intertidal areas (e.g., Bish Cove, Kitimat River estuary, Minette Bay) to restrict movement of hydrocarbon and around kelp beds. Oil should be removed from around marine vegetation using absorbent pads rather than dispersants, hot water or high pressure washing, given the detrimental effects reported (Baca and Getter 1984).

10.5.5 Follow-up Monitoring Northern Gateway would like to engage coastal Aboriginal communities in the establishment and monitoring of permanent environmental monitoring transects and sites. Information from these monitoring programs would be used to obtain information on baseline environmental conditions, including the quantity and quality of important harvested species such as marine vegetation. Offers to undertake a monitoring program have been extended to several coastal Aboriginal communities (Gitga‘at, Kitkatla and Lax Kw'alaams) as part of ongoing engagement activities, and similar offers will be extended to other coastal communities as engagement progresses. A sampling program would be implemented to assess the extent of effects on marine biota, including rockweed, eelgrass and other species where appropriate. For example, the sampling program might target affected areas and continue until vegetation has recovered. Analyses of affected marine vegetation might also be undertaken for comparison to baseline measurements to determine the success of the response measures and the recovery of marine vegetation.

April 2010 FINAL Page 10-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.6 Marine Invertebrates Marine invertebrates live on the surface (epifauna) or within sediment (infauna) and include deposit feeders, filter feeders, grazers and predators. Marine invertebrates provide food for many marine species and are an important link for transferring energy and nutrients between trophic levels. Marine invertebrate communities are often used as indicators of ecosystem health because many species are sensitive to pollution and sudden changes in their environment. Invertebrates might be useful indicators of both short- term and long-term effects from exposure to hydrocarbons because many species are relatively sessile and can provide a localized time-series of effects and rates of recovery. Species that inhabit shorelines, such as mussels, are generally more vulnerable to acute effects that occur when oil reaches the shoreline, whereas organisms that inhabit the ocean bottom might be exposed to long-term effects of hydrocarbons that have dispersed and remain in the bottom sediments.

10.6.1 Baseline Conditions Marine invertebrates are ubiquitous throughout the OWA. The ecozones in the OWA provide habitats for 3,800 to 6,000 marine invertebrate species (DFO 2002b, Internet site). Widespread distribution of benthic invertebrate species throughout the OWA results in many IAs based on uniqueness, aggregations and fitness (Clarke and Jamieson 2006). Most of the important habitats for benthic organisms occur along the coast (intertidal and subtidal areas), although offshore areas are used by crabs, sponges and corals. Important Areas for benthic invertebrates are shown in Figure 10-1 and the invertebrates associated with these areas are listed in Table 10-2, which includes mainly commercially important species and some ecologically important groups. The marine invertebrate communities in the CCAA are typical of those in the highly seasonal, mid- latitude, coastal marine ecosystem of the QCB (Hall 2008). Species diversity of the rocky intertidal community is generally low, mainly consisting of barnacles, mussels, periwinkles and limpets. Sea urchins, moon snails, green sea anemones, sea stars and California sea cucumbers are common in shallow subtidal habitat. Northern abalone was once prevalent, but is currently listed as Threatened under the SARA due to drastic population declines. Glass sponges are present in the CCAA, including around the proposed Kitimat Terminal, where subtidal videos taken in 2007 showed a low density of loose aggregations associated with steep cliffs. Glass sponges are known to form large reef complexes in Hecate Strait and Queen Charlotte Sound (Leys et al. 2004). The reefs in these two areas occur at depths of 165 to 240 m (Conway et al. 2001). They are being considered for protective status (Jamison and Chew 2002), but do not currently receive any formal provincial or federal protection. Damage induced by trawling fish boats is the main risk identified for the glass sponge reefs in Hecate Strait and Queen Charlotte Sound. The soft bottom estuaries are dominated by eelgrass, a marine vascular plant that provides important habitat for invertebrates such as Dungeness crab. Commercially harvested bivalves such as butter clams and cockles also inhabit sandy areas.

Page 10-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Figure 10-1 Summary of Biologically Important Areas for Marine Benthos in the OWA

April 2010 FINAL Page 10-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-2 Marine Invertebrates Associated with Ecosections in the OWA

Marine Ecosection Benthic Species of Importance Continental Slope Coastal areas provide homes to uncommon assemblages of benthic species, including northern abalone (rocky substrates in waters up to 10 m deep), many species of corals and sponges (near Haida Gwaii and northern parts of Vancouver Island), Tanner crabs (little data available), and sea urchins, as well as typical species (barnacles, blue mussels, etc.). Dixon Entrance There is high primary productivity due to nutrient availability and oceanographic features, with numerous benthic species, including Dungeness crabs, sponges, corals, razor clams, Northern abalone, shrimps, sea cucumbers and sea urchins. Some areas have higher productivity of benthic invertebrates. Hecate Strait There is high primary productivity due to nutrient availability and oceanographic features, with numerous benthic species and high benthic productivity. Species include Dungeness crabs (Dogbank, Prince Rupert harbour, with larval aggregations off McIntyre Bay in summer), hexactinellid sponges (in deep water troughs), corals, razor clams, Northern abalone, prawns, sea cucumbers and sea urchins. Queen Charlotte There is relatively low benthic production, with a variety of species, including Sound hexactinellid sponges (in deep water troughs), corals, shrimps, Olympia oyster (southern Sound), manila clam, razor clam, geoduck clam (Bella Bella), Northern Abalone, sea cucumbers and sea urchins Queen Charlotte Strait The areas of the Strait are mainly shallow, with maximum depths of 200 m, some areas of high relief and deeper fjords. This area supports many species of invertebrates and is considered moderate shellfish habitat. Drury Inlet is the most productive area for prawns, with large commercial harvests and a humpback shrimp bycatch; Olympia oysters occur in isolated small populations in a few places; giant red sea-cucumber and Northern abalone occur sparsely in suitable habitats; and there are scattered populations of coral North Coast Fjords The fjords provide sheltered areas and often have restricted circulation, causing stratification with fresh water inputs. They tend to have poor water exchange and nutrient depletion, with low productivity and species diversity, which creates an uncommon assemblage of benthic organisms. IAs have been identified for Dungeness crabs and green urchins (adjacent to Prince Rupert), tanner crabs (Observatory Inlet, Douglas Channel and Kitimat Arm), sea cucumbers (fjords between Prince Rupert and Bella Bella), manila clams (Bella Bella) and green sea urchins Vancouver Island A northward, coast-hugging buoyancy current associated with freshwater Shelf influences runs along the west coast of the island, bringing high nutrient waters in the summer and seasonal upwelling towards the outer margin, generating productive benthic communities. Klaskino Inlet, at the north end of the shelf, has a population of the Olympia oyster (protected species). Northern abalone giant red sea cucumber, green sea urchins and prawns occur along the coastline.

SOURCES: Howes et al. 1997, Internet site; Clarke and Jamieson 2006; Lucas et al. 2007

Page 10-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

The Marine Fish and Fish Habitat TDR developed for the Kitimat Terminal provides a complete discussion of baseline marine invertebrate data at the Kitimat Terminal. Dungeness crab (Cancer magister) and blue mussel (Mytilus trossulus) were identified as key indicators (KI) for routine effects of project-related vessel traffic at the marine terminal. These are two of the species that could be affected by hydrocarbons and reflect different types of habitat use, feeding strategies behaviour and sensitivity to hydrocarbons.

10.6.1.1 Dungeness Crab Dungeness crabs occur from Alaska to California, from the intertidal zone to a depth of about 180 m (DFO 2002b, Internet site). They live in bays and inlets, around estuaries and on the continental shelf. Although sometimes found on mud and gravel, they are most abundant on sandy bottoms and in shallow waters around eelgrass beds. Adults are important predators, feeding on clams, other crustaceans and small fish (DFO 2002b, Internet site). Aggregations of adult Dungeness crab are found in the shallow waters of Dogbank in Hecate Strait and nearby McIntyre Bay in Dixon Entrance. The eddy associated with McIntyre Bay is important for its retention of crab larvae (Clarke and Jamieson 2006). The Prince Rupert area is also recognized as an important area breeding area for Dungeness crab (Clarke and Jamieson 2006). Dungeness crab recruitment is affected by variations in oceanic conditions and, as a result, abundance (total catch) fluctuates greatly from year to year (DFO 2007, Internet site). Dungeness crabs have seasonal migration patterns based on reproductive status. Mating typically occurs outside estuaries and in nearshore habitats during summer, from May to August. During winter, they tend to migrate to deeper waters, where females bury themselves in soft sediments and remain inactive, from the time of fertilization to release of larvae in spring (Alaska Department of Fish and Game 1985). Crabs release very large numbers of larvae, which provide a valuable food source for Pacific herring, Pacific sardine, rockfish, salmon and many deposit and suspension feeders. Larvae remain in the water column for three to four months and are dispersed by ocean currents before settling on suitable substrates. Juveniles reside in shallow coastal waters, tidal flats and estuaries, living in beds of eelgrass and other aquatic vegetation for several months (DFO 2002b, Internet site).

10.6.1.2 Blue Mussel The blue mussel (Mytilus edulis) is native to the Pacific Northwest and is predominant on the hard shoreline of sheltered coasts of British Columbia (Gosling 1992), including the CCAA and OWA. Blue mussels are typically abundant in rock shelf, estuary and boulder beach habitats, as was confirmed in intertidal surveys conducted for the Project in 2005, 2006, 2008 and 2009. Blue mussels are harvested recreationally and by Aboriginal groups. As the predominant intertidal species in the region, they are considered a useful indicator of environmental change. Mussels form densely aggregated beds from the upper intertidal to subtidal zone on hard rocky shores, gravel/cobble substrates, and soft sediment shores in protected habitats. Adult mussels are attached to the substrate by byssal threads and are essentially immobile. The vertical distribution of mussels in the intertidal zone is determined by abiotic factors (temperature and desiccation set the upper limit of mussel

April 2010 FINAL Page 10-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment distribution) and biotic factors (predation and competition set the lower limit) (Connell 1972; Paine 1974). Major predators include sea stars, gastropods, crabs, fishes, shorebirds, sea ducks and sea otters.

10.6.1.3 Other Species Information about other invertebrate species targeted in commercial, First Nations and recreational fisheries in the OWA and CCAA is provided below. There are many biologically important areas for invertebrates within these areas, as summarized in Table 10-3 and Figure 10-1.

Shrimp and Prawn Shrimp commonly occur in soft-bottom habitats at 50 to 200 m depths, whereas prawns occur on rocky bottom habitats, mostly between 50 and 70 m. They occur along the entire British Columbia coast (Pellegrin et al. 2007), with the majority of commercial shrimp catches from the Vancouver Island Shelf (ILMB 2009b, Internet site) and prawn catches from southern Haida Gwaii and, to a lesser extent in fjords associated with Queen Charlotte Strait (DFO 2008b). Shrimp spawn in late autumn to early winter, and larvae are pelagic for approximately three months before they settle (Pellegrin et al. 2007). Nearshore juvenile populations are found predominantly in channels and estuaries and are particularly sensitive to habitat alteration (Lucas et al.2007).

Tanner Crab The Tanner crab is a large, deep water spider crab noted for its scarlet/orange colouration (DFO 2008d, Internet site). Two species of Tanner crabs are commercially exploited in the OWA (Chionoecetes tanneri and C. Bairdi) as an exploratory fishery, given the limited biological information on population levels at this time (Clarke and Jamieson 2006).

Sea Cucumber (Parastichopus californicus) The giant red sea cucumber lives on a variety of substrates and is most abundant in areas of moderate current on cobble, boulder or crevassed rock (DFO 1999). They have limited mobility but are known to migrate to deeper waters, which might protect them from commercial harvesting pressure and provide protected areas for spawning (DFO 1999).

Sea Urchin (Strongylocentrotus spp.) Sea urchins often have a distinct upper limit to their distribution and tend to aggregate in preferred habitats (Pellegrin et al. 2007). Areas of commercial and recreational fisheries of green and red sea urchins are concentrated in Queen Charlotte Sound, Hecate Strait and around the Haida Gwaii (ILMB 2009b, Internet site, referenced as Queen Charlotte Islands). The green urchin population is recovering from previous fishing pressure (Pellegrin et al. 2007). Distribution is patchy, and high numbers are found near Queen Charlotte and Johnstone Straits, and Prince Rupert. Green urchins are relatively mobile and migrate seasonally to deeper waters from their preferred depths of less than 140 m (Clarke and Jamieson 2006). Red sea urchin stocks are currently considered stable (Pellegrin et al. 2007). Populations occur in rocky habitats with moderate to strong currents (Clarke and Jamieson 2006) at depths ranging from

Page 10-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment intertidal to 90 m (DFO 2009, Internet site). Sea urchins avoid areas exposed to extreme wave action and sandy/muddy substrates (DFO 2009, Internet site).

Olympia Oyster (Ostrea conchaphila) The Olympia oyster is British Columbia‘s only native oyster and is listed under the SARA as a species of ―Special Concern.‖ It has limited distribution due to specialized habitat requirements and vulnerability to temperature extremes (Gillespie 1999). Populations were historically much larger, but a period of heavy fishing pressure caused drastic declines; currently, Olympia oyster populations are considered stable but at low densities (SARA 2007, Internet site). Olympia oysters are found along the west coast of Vancouver Island and in several places in Queen Charlotte Sound and Haida Gwaii (Gillespie 1999; Clarke and Jamieson 2006).

Northern Abalone (Haliotis kamtschatkana) The Northern abalone is listed under the SARA as a ―Threatened‖ species. It commonly occurs in subtidal areas less than 10 m deep (Williams 1989) and prefers rocky substrates that provide refuge in cracks and crevices on exposed or semi-exposed coastlines (DFO and the Abalone Recovery Team 2004). Abalone is currently sparsely distributed on the coast with no identified areas of high concentrations. The abalone fishery has been closed since 1990, and surveys show no signs of significant recovery, primarily due to the illegal exploitation of the resource and the reintroduction of sea otters (DFO and the Abalone Recovery Team 2004; Pellegrin et al. 2007).

Clams Five clam species comprise the majority of commercial, recreational and First Nations fisheries: Razor, Manila, Geoduck and, to a lesser extent, Butter and Littleneck clams (Pellegrin et al. 2007). Clam beds occur in coastal areas throughout the OWA (ILMB 2009b, Internet site) and typically spawn from May to August (Lucas et al. 2007). Razor clams (Siliqua patula) burrow just below the surface and are found on sandy beaches with high wave action from mid-tidal to 20 m deep (Lucas et al. 2007). The largest known stock occurs from Masset to Rose Spit in Haida Gwaii. Manila clams (Tapes philippinarum) commonly occur in the upper and mid-intertidal zones, creating permanent burrows in mixed substrates of sand, mud and gravel (Clarke and Jamieson 2006; Lucas et al. 2007). Commercially important habitats have been identified near Bella Bella (Clarke and Jamieson 2006). Geoduck clams (Panopea abrupta) can live for more than 100 years. They occur from the intertidal zone to depths of at least 110 m, buried up to 1 m in sand, silt and gravel (Pellegrin et al. 2007). Cockles inhabit the lower half of the intertidal zone in soft substrates (Lucas et al. 2007; Pellegrin et al. 2007). They have shallow burrows and large adults are often exposed. Cockles are hermaphroditic and do not have a designated spawning times (Lucas et al. 2007). They are considered ubiquitous throughout British Columbia but are not abundant in any one location (Lucas et al. 2007).

April 2010 FINAL Page 10-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Octopus One of the more mobile molluscs is the octopus. The most common species in British Columbia is the giant Pacific octopus (Enteroctopus dofleini). Its occupies rocky subtidal areas to depths of over 100 m, establishing dens in caves or among rocks. Smaller individuals might dig dens in sand-shell substrate 13 to 30 m deep (Williams 1989; Sport Fishing Institute of British Columbia 2008, Internet site). Giant Pacific octopi inhabit deeper waters from February to April and August to October (Williams 1989; Sport Fishing Institute of British Columbia 2008, Internet site). Females brood their eggs on the roof of their dens; larvae are planktonic until they are about 50 mm long, at which point they become benthic. Adult octopi are nocturnal predators, feeding on crabs, bivalves, gastropods and small fish (Williams 1989).

Sponges and Corals Sponges are sessile organisms that can form three-dimensional, reef-like structures that provide complex habitat for diverse communities (Clarke and Jamieson 2006). Sponges can be found on a variety of substrates from soft mud/sand to submerged timbers and vertical rock walls at depths from intertidal to very deep sea (Pellegrin et al. 2007). There are five known Hexactinellid sponge reef complexes stretched over 1,000 km2 in Hecate Strait and Queen Charlotte Sound (Clarke and Jamieson 2006; Guilbault et al. 2006, Internet site). These structures are globally unique and estimated to be between 8,500 to 9,000 years old. They can reach 18 m high, and usually occur at depths from 165 m to 230 m (DFO 2000). The sponge reefs provide relatively high habitat complexity and likely support diverse communities not found elsewhere. They are on the British Columbia Conservation Data Centre ―red list‖, but are not yet protected under SARA or given Marine Protected Area status (Pellegrin et al. 2007). However, since 2002, DFO has closed the four Hecate Strait sponge reefs to trawling. Nine areas have been identified in the OWA as IAs for corals based on trawl surveys, ranging from the northern end of Vancouver Island through Queen Charlotte Sound, Hecate Strait, Dixon Entrance and the Continental Shelf. Further research is required to map areas of importance for corals (Pellegrin et al. 2007).

10.6.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Invertebrates Oil can be directly toxic to marine invertebrates or can affect them through physical smothering, altering metabolic and feeding rates, inhibiting reproduction and altering shell formation (Hoffman et al. 2003; USFWS 2004, Internet site). Many invertebrates are filter feeders or scavengers, so they can be exposed to hydrocarbon contaminants through ingestion or uptake from the water column. Effects can be acute or chronic. Intertidal invertebrates are especially vulnerable when oil becomes highly concentrated along the shoreline, as the sediments can become reservoirs for the hydrocarbons. Some invertebrates can survive exposure, but might accumulate elevated levels of contaminants in their bodies that can be passed on to predators. Subtidal invertebrates, including glass sponges, are not expected to be vulnerable to direct effects of oiling, given the depth at which they occur; however, depending on their concentrations, hydrocarbon contaminants in bottom water or sediment could result in chronic toxicity. Details are described below for potential effects on Dungeness crab and blue mussels, species which can be

Page 10-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment considered representative of organisms than inhabit the types of habitat that would most likely be affected by exposure to hydrocarbons.

10.6.2.1 Dungeness Crab Hydrocarbons can adversely affect health and habitat of Dungeness crabs, given that crustaceans tend to be more sensitive to toxic effects than are other taxa (Peterson et al. 1996). Dungeness crabs are likely most sensitive in summer, when they inhabit shallower water and females are moulting and reproducing; larvae are most sensitive in the surface water in spring and summer. In a controlled study, female Dungeness crabs held on oiled sediments through a reproductive cycle produced much fewer larvae than did control crabs (Suchanek 1993). Crabs are exposed through direct contact with hydrocarbons mixed in the water column and ingestion of contaminated sediments. Physical smothering by oil (exclusive of toxicity) prevents respiration, inhibits or prevents movement and creates excess weight and/or shearing forces on affected mobile organisms (Suchanek 1993). Planktonic larvae are vulnerable to oil at the surface (Cucci and Epifanio 1979). After EVOS, densities of subtidal crabs and seastars were much lower at oiled sites (Dean and Jewett 2001); however, numbers were similar to those in non-oiled areas within two years (Armstrong et al. 1995). Studies of long-term effects of hydrocarbons on salt-marsh fiddler crabs, Uca pugnax, showed reduced density, ratios of males to females, and juvenile settlement, heavy overwinter mortality and behavioural anomalies (Krebs and Burns 1978). Culbertson et al. (2007) reported that fiddler crabs collected in a salt marsh with oiled sediments made shallower burrows, had slower escape responses and had short-term lowering of feeding rates compared to crabs from a non-oiled marsh.

10.6.2.2 Blue Mussel Mussel health and habitat are likely to be affected by hydrocarbons. Organisms directly coated with oil can suffer high mortalities (Boyd et al. 2001) due to fouling of gill filaments, which affects gas exchange and feeding rates, ultimately resulting in less energy for growth and reproduction (Suchanek 1993). Hydrocarbons can also interfere with byssal thread activity, resulting in loosening of mussels from their substrates and washing into unsuitable habitats (Suchanek 1993). Hydrocarbons in the water column can affect the larval life stages of mussels (Wu and Zhou 1992). Mussels filter particles from large volumes of water and can concentrate PAH and other lipophilic contaminants; this makes them better indicators of the presence of biologically active contaminants, compared to levels in water (Peterson 2000). Analysis of hydrocarbon contamination of mussels and four species of infaunal clams after EVOS demonstrated widespread and locally long-lasting effects in the intertidal system. Although contaminant levels in mussels and clams declined over time, pools of partially weathered oil remained at least until 1996 in the sediments below and among the mats of mussel byssus, cobbles and fine sediments (Peterson 2000). Because mussels are a prominent source of food for marine mammals, coastal birds, macro-invertebrates and fishes, contaminants can be transferred into important intertidal food chains (Peterson 2000).

April 2010 FINAL Page 10-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.6.3 Potential Effects of Condensate on Marine Invertebrates Condensate would have a limited rate of spread and affect less total area compared to diluted bitumen or synthetic oil. Condensate is a lighter hydrocarbon, and less persistent; however, there would be higher levels entrained in the water column, which would likely have greater effects on floating larval stages and subtidal invertebrates, as it is unlikely that condensate would reach the shoreline and be taken up by mussels and other filter feeders. The ERA conducted for a hypothetical condensate spill at the terminal indicated that PAH levels in subtidal sediment would be well below thresholds that indicate a potential risk of chronic adverse effects on subtidal benthic invertebrates; the ERA also indicated that levels of other hydrocarbon compounds (shorter aliphatic hydrocarbons, BTEX) in condensate would slightly exceed those thresholds, although these compounds would not be expected to persist in sediment (Stephenson et al. 2010b). In general, it is unlikely that exposure to condensate would result in chronic toxicity to benthic invertebrates.

10.6.4 Mitigation Measures Clean-up using skimmers and sorbents would reduce the amount of residual hydrocarbon in the environment. Aggressive cleanup activities (i.e., hot water and high-pressure washing) would not be recommended because greater damage can result than leaving the oil in place. Sensitive shorelines would be protected by exclusion booming because of the tendency of oil and bitumen to accumulate and persist in the intertidal zone. As part of the response planning, the use of dispersants and their effects on marine invertebrates will be investigated. Dispersants would only be used upon approval by DFO and Environment Canada for the type and use of dispersants as a response measure.

10.6.5 Follow-up Monitoring As noted earlier, Northern Gateway has invited several coastal First Nations to participate in the establishment and monitoring of environmental transects that would include monitoring of marine invertebrates. The monitoring program might include studies of the distribution, abundance and quality of selected marine invertebrates. For example, sampling could be undertaken to assess the levels of contaminants such as PAHs in tissue of selected species and sediments. Such monitoring would typically continue until hydrocarbon concentrations have returned to baseline levels. A monitoring program is essential to identify whether harvesting closures are needed and to assess effectiveness of cleanup measures.

10.7 Fish, Fish Habitat and Marine Fisheries Management Fish and fish habitat are protected under the Fisheries Act, which defines fish habitat as areas on which fish depend directly or indirectly to carry out their life processes. These areas include: spawning grounds nursery, rearing, food supply and migration areas

Page 10-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

This includes freshwater, estuarine and marine environments that directly or indirectly support fish stocks or fish populations that sustain, or have the potential to sustain, subsistence, commercial or recreational fishing activities. In the event of a spill, the most relevant sections of the Act include Section 35, which prohibits any harmful alteration, disruption or destruction of fish habitat (HADD) without authorization, and Section 36, which prohibits the deposition of a deleterious substance into waters used by fish. The OWA is home to a number of fish species with important ecological, social and economic value. These fish provide food for other fish, birds and mammals, and provide food for humans through FSC, commercial, recreational and commercial-recreational fisheries. This section addresses ecological considerations of fish and fish habitat as well as fisheries management concerns for FSC, commercial, recreational and commercial-recreational fisheries (closures, tainting of fish resources); Section 11.4 addresses fishing from a non-traditional resource use perspective and Section 11.5 discusses socio- economic aspects. Hydrocarbons could directly affect marine fish and lead to acute effects such as fouling of gills or ingestion of residual hydrocarbon, or indirect effects through contamination of prey and habitat. FSC, commercial, recreational or commercial–recreational fishing could also be affected by closure of fisheries to allow for recovery of populations or from tainting or elevated hydrocarbon levels, and also through fouling of fishing vessels and gear. Egg and larval stages are generally the most vulnerable to exposure to hydrocarbons because they inhabit the upper water column and cannot swim away to avoid contaminated areas. Adult fish in deeper water are less vulnerable, but those at the surface (e.g., migrating salmon) might be exposed. Most of the effects are expected to be associated with hydrocarbons mixed into the water, especially in nearshore areas, and, in some cases, contaminated sediment. Given the wide range of fish species in the OWA, and their diverse habitat requirements, species representative of a range of habitat uses, differing vulnerabilities and different socio-economic as well as ecological importance are discussed. Based on these criteria, seven species are used to assess the potential effect of hydrocarbons on fish and fish habitat: Eulachon: Eulachon is a culturally and ecologically important species in the region. It supports a traditional Aboriginal harvest and is an important prey item for larger fish and other marine predators. Annual migrations to spawning grounds increase overall biodiversity in the area by supporting a vast array of predators that follow them to their spawning grounds. Pacific herring: Pacific herring is an important ecological and commercial species in the area. It represents a food source for many larger fish and marine mammals and is important to fishers for herring roe, spawn on kelp, food and bait. Rockfish: Rockfish species are representative of demersal fish closely associated with bottom habitat. They are long-lived, late maturing species that typically have a defined home range for most of their lives. This group includes the boccacio, a threatened species currently under review for addition to Schedule 1 and protection under SARA. Pacific salmon: All species of salmon and their lifecycles (pink, chum, coho, chinook and sockeye) are considered. These are targeted commercially, recreationally and by FSC fishers.

April 2010 FINAL Page 10-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

FSC fisheries: FSC fishing in the CCAA and OWA is very important for local Aboriginal residents as it forms a primary component of the Aboriginal diet. Commercial fisheries: There are extensive commercial fisheries throughout the OWA and CCAA that contribute to the local and regional economy. Recreational and commercial-recreational fisheries: Recreational and commercial-recreational fisheries operate throughout the coastal areas of the OWA and the CCAA and are important to the local and regional economy for enjoyment and as a food source. Commercial and recreational fishing landing statistics indicate the region produces notable quantities of fish and shellfish. There are also two aquaculture sites in Principe Channel near McCauley Island. Any changes to these resources could adversely affect the livelihood of local Aboriginal communities, commercial fishers, commercial-recreational lodges and charter businesses.

10.7.1 Baseline Conditions Baseline conditions can be described in terms of: (1) Ecoregions (which organize information based on ecological function that can be used to describe fish presence/absence and uses in the regions); (2) EBSAs (to identify areas of heightened management concern due to their uniqueness, aggregation, and/or fitness consequence characteristics); or IAs (which have been identified for a variety of fish species based on spawning and rearing habitat). There is an approximate 60% overlap of fish IAs with EBSAs in the OWA (Clarke and Jamieson, 2006b), creating a system that allows for interpretation of fish distributions and quantification of fish presence on an ecological backdrop. IAs, EBSAs and key habitat use for many species are shown in Figure 10-2 and summarized in Table 10-3. Additional socio-economic information about FSC, commercial and recreational fishing is discussed in Section 11. Information includes descriptions of eco-tourism, fishing lodges and other recreational- commercial activities, marina locations and recognized prime fishing spots.

Table 10-3 Important Areas for Fish in the OWA and CCAA

Marine Ecosection Species (Importance) Important Area Key Habitat Use Continental Green Sturgeona (high) Shelf Break Slope Sablefish (high) Shelf Break, Cape St. James, Brooks Peninsula, Scott Islands Pacific Halibut (high) Shelf Break, Cape St. Spawning James Sole/Flounder (moderate) Shelf Break Pacific Hake (moderate) Shelf Break, Cape St. James

Page 10-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-3 Important Areas for Fish in the OWA and CCAA (cont’d)

Marine Ecosection Species (Importance) Important Area Key Habitat Use Dixon Eulachon (moderate) McIntyre Bay Summer Grounds Entrance Walleye Pollock N/A (moderate) Pacific Halibut (high) Learmouth Bank, Shelf Rearing Break, McIntyre Bay Pacific Hake (low) Shelf Break Herring (low/moderate) McIntyre Bay Adult Feeding Hecate Strait Pacific Cod (high) Chatham Sound, Sponge Shallow Rearing Reef Bioherm, Dogfish Bank Walleye Pollock Hecate Strait Front, Spawning and Rearing (moderate) Chatham Sound Sablefish (high) Dogfish Bank, Hecate Strait Front Sole/Flounder (high) Sponge Reef Bioherm, Shallow Rearing Chatham Sound, Dogfish Bank Herring (moderate/high) Hecate Strait Front, Cape Spawning and Rearing St. James, Chatham Sound Queen Eulachon (moderate) Hecate Strait Front, Shelf Adult Feeding Charlotte Break Sound Pacific Cod (high) Scott Islands, Sponge Reef Spawning and Rearing Bioherm Lingcod (high) Shelf Break, Scott Islands Spawning and Rearing Sablefish (high) Cape St. James, Scott Spawning and Rearing Islands, Shelf Break Pacific Halibut (high) Shelf Break, Cape St. Spawning James Sole/Flounder (high) Cape St. James Shelf Break Spawning Scott Islands Spawning and Rearing Rockfishb (high) Scott Islands, Sponge Reef Rearing and Feeding Bioherm, Shelf Break Pacific Hake (moderate) Cape St. James, Scott Feeding Islands, Shelf Break Herring (moderate/high) Scott Islands, Cape St. Feeding James, Hecate Strait Front

April 2010 FINAL Page 10-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-3 Important Areas for Fish in the OWA and CCAA (cont’d)

Marine Ecosection Species (Importance) Important Area Key Habitat Use Queen Pacific Cod (high) North Island Straits Charlotte Lingcod (high) North Island Straits Strait Sablefish (high) North Island Straits Sole/Flounder (high) North Island Straits Pacific Hake (moderate) North Island Straits Herring (high) North Island Straits Migration Salmon (moderate) North Island Straits Rearing, Feeding, Staging Vancouver Green Sturgeon (high) Shelf Break, Brooks Staging/aggregation Island Shelf Peninsula Lingcod (high) Scott Islands, Brooks Spawning and Rearing Peninsula Sole/Flounder (high) Scott Islands Rockfishb (high) Scott Islands, Brooks Peninsula Pacific Hake (moderate) Scott Islands, Brooks Peninsula Herring (moderate/high) Scott Islands, Brooks Peninsula NOTES: Key habitat use is described where information is available. a Green sturgeon are listed under the SARA (at risk) and by COSEWIC (special concern). b Rockfish include the following listed species: Canary rockfish (COSEWIC under review), Rougheye rockfish (COSEWIC under review). SOURCE: Adapted from Clarke and Jamieson 2006

Page 10-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Figure 10-2 Summary of Biologically Important Areas for Ground, Pelagic and Anadromous Fishes in the OWA

April 2010 FINAL Page 10-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.7.1.1 Biologically Important Areas for Ground, Pelagic and Anadromous Fishes by Ecosection In general, little is known about functional habitat use for most life history stages for marine species in British Columbia (Lucas et al. 2007). Further, the wide range of habitats used by fish across various life history stages makes identifying universally IAs difficult. Most of the OWA is considered important habitat for at least one species. Areas of overlap are considered particularly important. Some of these overlapping areas are concentrated around the northern end of Hecate Strait, the western end of Dixon Entrance near Langara Island, the central and western Queen Charlotte Sound around Goose Island Bank and North Bank and the associated troughs, and Cook Bank around the Scott Islands. For pelagic fishes, their highly migratory nature results in the majority of the OWA being considered habitat. However, it can be argued that the nearshore environment is of increased importance to overall population health due to its importance to the early life stages of most of the key species for rearing. With the exception of only a few species (e.g., English sole), the majority of adult flatfish are found at considerable depths (Lucas et al. 2007). For many species, however, planktonic, larval juveniles settle in shallower water and some species spawn in shallower water, thus making the embayments and sheltered waters nearer to shore important for groundfish. The important habitat areas identified in Figure 10-2 and Table 10-3 generally comprise key spawning and rearing areas for a number of pelagic and benthic species. Other habitat functions provided by these IAs include migration and feeding. Rockfish Conservation Areas are also presented in Figure 10-2, as these have been identified as areas of key importance for rockfish species. Each individual data layer represents a single species of fish, with the exception of sole and flounders, which are generally grouped together for management purposes. Areas of darker orange in the figure indicate areas of overlap between important habitat areas. This overlap of IAs indicates regions of particular importance to these species.

10.7.1.2 Information for Individual Fish Species Many fish species are transient in space and time, and through life stages, across the OWA and CCAA, so the IAs are useful to limit the scope of this analysis to spawning and juvenile rearing habitat, which are the two sensitive and limiting life stages for most fish species (Clarke and Jamieson 2006). Fisheries data available for adult fish are generally separated by fishery management area, which is not an ecologically based system, making comparisons across life stages more difficult. For these reasons, fish habitat use is described below for ecoregions and also for individual species.

10.7.1.3 Eulachon Eulachon is an anadromous species that spawns in rivers in the CCAA and some mainland coastal rivers. Migration to spawning grounds occurs between January and April and results in formation of dense schools throughout Douglas Channel. Rivers in the area that consistently support Eulachon spawning include Kildala River, Kitimat River and possibly other small channels off Gardner Channel (Hay and McCarter 2000). Gilttoyees Inlet and Foch Lagoon are used on an irregular basis (Hay and McCarter 2000). Eggs hatch three to five weeks after being laid; larvae disperse immediately and reside in nearshore marine environments and estuaries for several weeks (British Columbia Forest Service 1998),

Page 10-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment feeding on plankton until they are large enough to move out to sea. Adults likely spend much of their sea life in Hecate Strait and Queen Charlotte Sound (British Columbia Forest Service 1998; Hay and McCarter 2000). Within the CCAA, DFO identified Douglas Channel and Gardner Canal as IAs for eulachon spawning aggregations (Clarke and Jamieson 2006). Additionally, feeding areas include the offshore banks of Calvert Island and Pearl Rock Ground, the Cape Scott ground, the southeast edge and northeast corner of the Goose Island bank, the southeast edge of Middle Bank (North Bank), the entire 50 to 100 fathom edge from Rose Spit, Two Peaks, White Rocks, Butterworth edge, Bonilla Island down to the Horseshoe ground and southeast to Cape St. James, and the entire Chatham Sound (Lucas et al. 2007).

10.7.1.4 Pacific Herring Pacific herring is a schooling pelagic species that inhabits nearshore and continental shelf environments. Herring are part of a complex food web as spawn, eggs, juveniles and adults, forming a large part of the diet of a wide variety of piscivorous fishes, mammals and marine birds. Spawning migrations move from offshore feeding grounds (west coast of Vancouver Island and Hecate Strait) to inshore spawning grounds during October to December (Clarke and Jamieson 2006). In British Columbia, herring spawn in late winter, from February to as late as July, peaking in March from the high tide line down to 20 m depth. Herring are deposit spawners, and their sticky eggs coat spawning substrates. Therefore, spawning habitat requirements for Pacific herring include intertidal and subtidal vegetation A small resident population of Pacific herring resides in the upper reaches of Kitimat Arm. This population is distinct from the larger, migratory stocks of Hecate Strait and coastal British Columbia. In general the resident population does not migrate, although there is evidence of post-spawning migrations to the mouth of Kitimat Arm (Triton 1993). Herring that migrate into the CCAA to spawn are assumed to belong to the Central Coast stock that extends from the southern tip of Banks Island south to Johnstone Strait (DFO 2001). Spawning locations vary from year to year, but generally include Kitimat Arm, the southwest side of Hawkesbury Island and Hartley Bay. Within the Kitimat fjord complex, there are spawning beds on both sides of Douglas Channel, on the west side of Ursula Channel and on the south side of Coste Island. Adult Pacific herring are also found in Kitkatla Inlet, just north of Browning Entrance and in Kitasu and Weeteean bays south of Caamaño Sound. Near the proposed Kitimat Terminal, spawning occurs along the foreshore between Kitamaat Village and Minette Bay, in Clio Bay, Kildala Arm and on Coste Island north of the terminal to Minette Bay. The Kitimat Arm Pacific herring population spawns in Minette Bay, south of Kitamaat Village (Triton 1993). Of the five major spawning areas for herring stocks in British Columbia, three are within the boundaries of the OWA (Clarke and Jamieson 2006). Additionally, the bottleneck of Queen Charlotte Strait and Johnstone Strait serves as a major migration route for herring and is characterized as a high-value IA. Herring generally spawn over a four-day period, typically in late March (Hay et al. 1989; Triton 1993). They can spawn on a range of intertidal and subtidal vegetation, including rockweed, at depths between the high tide mark and -11 m (chart datum). Juveniles are known to rear in the upper end of Kitimat Arm, including Minette Bay.

April 2010 FINAL Page 10-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.7.1.5 Rockfish There are approximately 40 species of rockfish in British Columbia, which are generally slow-growing, long-lived with low fecundity. Considering the many rockfish species and their life stage habitat preferences, the entire coast from subtidal to depths of at least 2000 m could be considered important habitat (Lucas et al. 2007). However, the identification of IAs considers only the species associated with the ‗outside‘ rockfish fishery (i.e., Pacific Ocean perch, Yellowtail and Yellowmouth rockfish); these do not include the inshore rockfish populations that are also of conservation concern, but limited more to the nearshore environment. Rockfish generally do not disperse far from natal waters; thus, spawning and rearing habitats might be extrapolated to be important across all life stages. Rockfish prefer boulder habitat and rocky outcrops that provide complex substrate and cover (crevices, caves, rock ledges, areas of vertical structure such as kelp forests). The IAs for rockfish spawning and rearing extend along the length of the continental slope and into the deeper areas of Queen Charlotte Sound and Dixon Entrance (Clarke and Jamieson 2006). Several rockfish species are present in the CCAA, mostly in areas of boulder habitat and rocky outcrops that provide complex substrate and cover. Most species are long lived and reproduce annually throughout their life. Mature rockfish generally release live larvae when ocean productivity is high, which in British Columbia is between April and June (Love et al. 2002). Larvae are pelagic for up to one year, and tend to aggregate before they settle to the bottom. Locations of larval aggregations in the CCAA are unknown. Adult rockfish prefer rocky outcrops that provide complex substrate and cover in crevices, caves and rock ledges, in addition to vertical structure such as kelp forests.

10.7.1.6 Pacific Salmon Salmon species are anadromous, meaning eggs are laid in freshwater. Juveniles spend some time in freshwater before swimming to sea to grow and mature. Between March and November, depending on the species, adult salmon move into Kitimat Arm on their way to spawn in the Kitimat River and other area streams. Chum and pink salmon juveniles spend very little time instream: immediately after emerging from the gravel, they migrate downstream to the estuary and to Kitimat Arm and Douglas Channel, where they tend to aggregate close to shore in discrete schools during the first few weeks (DFO 2002a, Internet site). Fry of sockeye, coho and chinook spend longer, up to a year, in the stream before migrating to the ocean. Salmon are most vulnerable to effects from hydrocarbons when adults are migrating toward the Kitimat River and when juveniles are migrating into the estuary and Kitimat Arm. There are five species of salmon (pink, chum, sockeye, coho and chinook), all of which are anadromous. Populations of salmon are generally classified by watershed and river; their strong homing instincts result in distinct genetic populations. There are 383 major populations of salmon that return to the North Coast Fjords and Queen Charlotte Strait ecoregions (Hyatt et al. 2007). Adults and juveniles undergo extensive migrations to and from these areas. Pink salmon are the most abundant species in the area and are widely distributed throughout the coastal watersheds and Haida Gwaii. Chum have a similar distribution. Sockeye are also widely distributed and mostly spawn in a few rivers on the Central Coast. Coho and chinook salmon are most likely to be

Page 10-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment residents within the OWA throughout their lifespan. Because they are piscivorous, they are closely linked to the health of many of the forage fish populations discussed above. DFO considers Johnstone and Queen Charlotte straits an IA because they form a highly important migration route for sockeye and steelhead salmon. Coho salmon, in contrast, seem to remain in marine areas closer to the entrance of their natal streams. DFO has identified the Broughton Archipelago- Johnstone Strait area as important for coho stocks (Clarke and Jamieson 2006).

10.7.1.7 Other Species

Pacific Cod Pacific cod (Gadus macrocephalus) are a relatively short-lived, fecund species. They migrate from shallow water in spring and summer to deeper waters in fall/winter and spawn in February and March. Larvae are planktonic and settle near the bottom in shallow waters (5 to 11 m), making them sensitive to environmental changes or disturbance (Lucas et al. 2007). They are commercially abundant at depths of 18 to 130 m. Pacific cod densities are highest in deep trough habitat within central Queen Charlotte Sound and southeast Dixon Entrance (Lucas et al. 2007). A large shallow water rearing area in Hecate Strait and two smaller spawning and rearing areas around Goose Island Bank and Cook Bank have been identified as IAs for Pacific cod (Clarke and Jamieson 2006; Fargo et al. 2007).

Walleye Pollock Walleye Pollock (Theragra chalcogramma) are considered the most abundant fish in the North Pacific and are ubiquitous along the West Coast (Lucas et al. 2007). While pollock are most abundant along the continental shelf and slope at depths between 100 and 400 m, during various stages of their lifecycle they might also be found nearshore, in large estuaries, coastal embayments and open ocean basins (Lucas et al. 2007). Spawning occurs in March and April, and juveniles might occur in a variety of habitats, including eelgrass beds (over sand and mud substrates), and gravel and cobble substrates (Lucas et al. 2007). There are six important spawning and rearing areas identified for Walleye Pollock in the OWA and CCAA, three in Hecate Strait and Queen Charlotte Sound and three in the Northern Fjords. These include the east side of Dixon Entrance, Finlayson Channel, Portland Inlet, Whale Chanel, the middle of Hecate Strait (west of the northern tip of Banks Island) and Browning Entrance (Clarke and Jamieson 2006). Additional spawning habitat has been identified in inlets on east Moresby Island as well as in mid-water inlets along Hecate Strait and Queen Charlotte Sound (Lucas et al. 2007).

Pacific Hake The migratory stock of Pacific hake (Merluccius productus) offshore of the west coast of Vancouver Island sometimes extends into the OWA (Clarke and Jamieson 2006). Offshore hake migrate annually into British Columbia waters, from late spring to early fall. Hake abundance fluctuates greatly from year to year and appears to be greatly influenced by climate. This, combined with their migratory nature, makes it difficult to define IAs; however, several IAs were identified in Queen Charlotte Sound and Dixon Entrance, including the Scott Islands and surrounding areas along the coast of Vancouver Island and the deep troughs of Queen Charlotte Sound (Clarke and Jamieson 2006).

April 2010 FINAL Page 10-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Lingcod Lingcod (Ophiodon elongatus) are generally non-migratory and can be found at depths of 3 to 400 m in habitats associated with high invertebrate and algal cover and strong currents. Males move to nearshore areas in October to prepare for spawning. Spawning occurs from December to March, when females migrate nearshore to deposit egg masses in nesting sites previously established by the males (Clarke and Jamieson 2006). Juveniles, unlike adults, are often found on flat substrates in shallower waters and later move to areas of high relief and rockiness as they mature. In the OWA, densities of lingcod appear to be highest in northern Moresby Trough, the southern edge of Dixon Entrance and around Goose Island Bank (Lucas et al. 2007). Lingcod spawning is known to occur off the northeast coast of Vancouver Island, spanning the Vancouver Island shelf and Queen Charlotte Sound ecoregions; thus DFO has identified this as an area of particular importance (Clarke and Jamieson 2006).

Sablefish Sablefish (Anoplopoma fimbria) spawn in very deep water (1000 m) from January to March. Juveniles rear in nearshore waters and shelf habitats until two to five years of age, at which point they migrate back to offshore waters and into the fishing grounds (Lucas et al. 2007). However, juvenile sablefish are also highly mobile and can move between nursery areas in Hecate Strait to the Gulf of Alaska and the Bering Sea. Sablefish spawning and rearing habitats span from Hecate Strait, through Queen Charlotte Sound, onto the Vancouver Island Shelf and the Continental Slope. Various life stages of sablefish can be found in most areas of the OWA west of, and including, the Continental Slope (Lucas et al. 2007).

Pacific Halibut Pacific halibut (Hippoglossus stenolepis) are commonly found along the continental shelf and slope, from relatively shallow waters to at least 1000 m. While mature fish might move extensively, most tend to stay on the same grounds and make only seasonal migrations from shallow feeding areas in spring to deeper spawning grounds in winter (Lucas et al. 2007). Halibut spawn in waters along the continental shelf at depths between 180 and 450 m from November to March, after which eggs rise to the surface, and larvae become planktonic and drift into shallow, nearshore environments (Lucas et al. 2007). Locally important spawning grounds include around the Goose Group, around Cape St. James and areas in northern Hecate Strait and Dixon Entrance (Clarke and Jamieson 2006).

Sole and Flounders Sole and flounder are generally found on sand, gravel, or mud substrates across a wide range of depths. Adults of certain species might show a preference for certain substrate types but generally prefer coarser substrate over finer sand/silt (Perry et al. 1994). Most species spawn in winter. For the purposes of identifying IAs, the following species were considered in the sole/flounder group: Petrale, Butter, Rock, Dover, English soles, and Arrowtooth flounder. Sole nursery and spawning areas were identified across Hecate Strait, Queen Charlotte Sound, Dixon Entrance and the Continental Slope.

Page 10-30 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Pacific Sardine Pacific sardine (Sardinops sagax) is a migratory fish that breeds in California and migrates to British Columbia waters in the summer to feed (Clarke and Jamieson 2006). Similar to hake, the northern limit of their distribution depends on water temperature and can extend into Dixon Entrance in warm El Nĩno years. There were no areas within the sardine summer distribution identified as IAs by DFO (Clarke and Jamieson 2006).

Albacore Tuna Albacore tuna (Thunnus alalunga) is a highly migratory pelagic species. It is mainly piscivorous, feeding on many of the forage fish species present throughout the OWA. Tuna feed on pelagic (sardines, herring, anchovy) and demersal (rockfish) fish species, as well as some invertebrates (squid, euphausiids). Tuna migrate with changing sea surface temperature, preferring warmer waters where prey species aggregate. This migratory behaviour places them in British Columbia‘s offshore waters in July and August (Schwiegert et al. 2007).

10.7.1.8 FSC Fisheries FSC fisheries are important to coastal Aboriginal communities and can constitute a large proportion of their daily food requirements. Targeted species include the five species of Pacific salmon, halibut, herring, octopus, shellfish, crabs, groundfish, cockles and clams. Although FSC fishing occurs throughout the CCAA and OWA, exact locations, seasons of harvest and relative importance of the locations are not known. However, it is assumed that activities are greatest around coastal settlements (e.g., Kitamaat Village, Kitkatla, Hartley Bay, Lax Kw'alaams, Kitsela and Kitsumkalum) due to accessibility of the resource.

10.7.1.9 Commercial Fisheries Commercial fishing occurs throughout the OWA. Fishing effort occurs throughout the year but is most intense during the salmon openings in July and August. While the local economic returns are greatest for the five salmon species, there are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. Groundfish species are landed year round. Detailed commercial fisheries information is available in TERMPOL 3.3.

10.7.1.10 Recreational and Commercial-Recreational Fisheries Recreational and commercial-recreational fisheries (fishing lodges, charters) in the CCAA and OWA are most prevalent during the summer (June to September), when the conditions are more desirable. The marina at Kitamaat Village is a base for many recreational fishers, who access the surrounding areas with their own boat or a charter. Local and out of town fishers target the same species as the commercial fishers. Socio-economic aspects of the recreational and commercial-recreational fishing are discussed in Section 11.

April 2010 FINAL Page 10-31

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.7.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Fish and Fish Habitat The direct effects of hydrocarbons tend to be similar for all selected fish species, given their similarity in physiology. However, the magnitude of effects can vary depending on life stage and environmental conditions. Life stages with an obligatory nearshore or surface water period, such as for rearing or spawning, are highly vulnerable to the effects of hydrocarbon exposure (Birtwell and McAllister 2002). Spring is generally the most sensitive time, when spawning adults, eggs, larvae and juveniles are present in nearshore areas. Larval fish are the most vulnerable life stage because they generally live close to the water surface where surface oil would be and have limited capacity to move away (Rice et al. 1976; Carls 1987; Paine et al. 1992). Larvae might also ingest oil droplets and oil-contaminated prey, which can result in altered feeding patterns and mortality (Grahl-Nielsen et al. 1978; Norcross et al. 1996). Larvae that hatch in uncontaminated waters can be affected if they drift into the area of a spill. Adult fish might be exposed to contaminants through ingestion of residual hydrocarbon in sediment or uptake from water, which can lead to sub-lethal effects, including changes in heart and respiratory rates, gill structural damage, enlarged livers (Binderup et al. 2004), reduced growth and fin erosion (Moles and Norcross 1998). Following an oil spill on the Lithuanian coast, commercial catches were down approximately 90%, suggesting fish avoided the oil-contaminated waters, but catches returned to levels similar to pre-spill catches by the following spring (Binderup et al. 2004). Avoidance would probably provide only a short-term benefit, as it could also lead to disruption of schooling behaviour or premature occupation of a different water depth or higher-salinity waters. Adult mortality following acute exposure to hydrocarbons has been reported; however, it can be difficult to attribute the cause conclusively and exclusively to the spill (Teal and Howarth 1984; Marty et al. 2003). Sand lance have been shown to avoid sand contaminated with crude oil (Pinto et al. 1984). In addition to the effects described below for individual species and groups, interactions among the species might lead to broader ecosystem effects, although these can be difficult to confirm, given the influence of unrelated natural events that can confound interpretation. A spill during spring would have the greatest effect on overall fish populations through reductions in plankton levels and mortality of herring eggs and juveniles, which would lead to reduced amounts of prey for juvenile salmonids and other species, as well as any direct effects of oil on these species. Following EVOS, there was a 75% reduction in 1993 herring numbers, although this could not be attributed completely to the spill, and a decline in salmon number due to direct and indirect (habitat and prey abundance) factors (Peterson 2000). There is ongoing debate about links between EVOS, the reduction in the 1989 year class and the collapse of the herring population in 1993 because of confounding environmental factors, insufficient knowledge about natural processes affecting recruitment and varying methods of assessing population size (Pearson et al. 1995; Thorne and Thomas 2008). There would also be effects on fish at other times of year, such as when salmon are returning to spawn, but to a lesser extent. Any biological effects on fish resources would have implications for other users, including birds, mammals and humans. The socio-economic implications of a spill are discussed in Section 11.5.

Page 10-32 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.7.2.1 Eulachon Depending on the time of year, location and prevailing wind and currents, bitumen or oil could spread to spawning rivers used by eulachon; however, the Kitimat River, which is the primary spawning area, has a low probability of being affected by a spill. Between January and April, large numbers of eulachon migrate to and congregate in spawning grounds in rivers adjacent to Nanakwa Shoals, and in associated rivers and streams of the Kitimat River, Foch Lagoon, Gilttoyees Inlet, Bish Creek, Kildala River and Gardner Channel. The larval stage is generally the most vulnerable (Carls 1987), so a spill during spring would likely have the greatest effect on the overall population, through toxicity or ingestion of contaminated prey, and indirectly through effects on plankton food sources. The extent of effects would depend on how long the larvae are in contact with the hydrocarbon and the concentration of hydrocarbons mixed in the water (Law and Hellou 1999).

10.7.2.2 Pacific Herring All life stages of herring are vulnerable, although eggs are likely the most sensitive (Carls et al. 1996b). Eggs, larvae and juveniles are present in the surface and nearshore, shallow areas, where they will be highly vulnerable (Birtwell and McAllister 2002). Adult herring migrated through an oil spill zone in the Baltic Sea. Their subsequent consumption of crude oil was thought to have led to impairment of egg deposition and larval survival (Birtwell and McAllister 2002). Studies of adult herring indicate they come to the surface to release gas bubbles and to escape predation, and might come into contact with oil on the water surface (Thorne and Thomas 2008). Some studies following EVOS suggested large effects on all life stages of Pacific herring (Brown et al. 1996; Norcross et al. 1996; Thorne and Thomas 2008) and others suggested minor population effects (Pearson et al. 1995). Reported effects included severe genetic defects in larvae, persisting even after oil was not detectable (Norcross et al. 1996) and sub-lethal effects (premature hatching, low weights, reduced growth, increased morphologic and genetic abnormalities) in newly hatched larvae of the 1989 year class (Brown et al. 1996). Brown et al. (1996) measured egg biomass and reported that approximately 53% of herring eggs deposited in 1989 were within the oil trajectory and that over 40% were affected by oil at toxic levels, resulting in large decreases in larval production. There is ongoing debate about links between the 1989 oil spill, the reduction in the 1989 year class and the collapse of the herring population in 1993 (75% drop in returning adults) because of confounding environmental factors, insufficient knowledge about natural processes affecting recruitment and varying methods of assessing population size (Pearson et al. 1995; Thorne and Thomas 2008). Indirect effects can occur when herring larvae ingest oil-contaminated prey. Carls (1987) noted that larvae exposed to hydrocarbons dissolved in water had a reduced feeding rate. Normal feeding rates resumed when surviving larvae were returned to clean water, but the mechanism of reduced feeding was not determined.

10.7.2.3 Rockfish The larval stage of rockfish, present in spring, is the most vulnerable stage that could be affected by hydrocarbons. The strategy of long life with annual reproduction creates natural between-year variation in larval survival (Love et al. 2002) and lowers the risk that the population will be reduced due to one

April 2010 FINAL Page 10-33

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment unsuccessful reproductive season. Small numbers of adults in the spill area might die, but fish from surrounding areas would likely move into the vacant habitat, resulting in no long-term effect on the population. Rockfish was the only fish found dead in large numbers following EVOS. While oil was present in their stomachs, the link between exposure and mortality was not confirmed (Marty et al. 2003). Several species of rockfish from oiled and reference sites were analyzed for microscopic lesions after the spill; pigmented macrophage aggregates and hepatic megalocytosis, fibrosis and lipid accumulation were found, but were correlated more with age and species rather than to previous oil exposure (Marty et al. 2003). Conservation areas have been established to help protect rockfish populations identified by DFO as in need of protection and conservation (Figure 10-2). These protected areas within the OWA can be considered the most sensitive to hydrocarbons.

10.7.2.4 Pacific Salmon Hydrocarbons can affect all life stages of salmon. They have an obligatory residence period in shallow estuarine and nearshore coastal waters, and it is not known whether, in the presence of hydrocarbons, they would remain there or avoid the area. Salmon would not be exposed as eggs or emergent fry in freshwater, but the juveniles are highly susceptible to environmental disturbances when moving into the estuary and saltwater, making this the time of greatest vulnerability (Birtwell and McAllister 2002). Juvenile salmon in Kitimat Arm are particularly vulnerable in winter through early summer when they are present in nearshore areas and in estuaries (Birtwell and McAllister 2002). Adults will be vulnerable in the fall as they move up the Douglas Channel to spawn, particularly if river flows are low and the adults stage in estuaries. Adult salmon have shown behaviours ranging from no avoidance of oil (in a test situation) to avoidance (by migrating adults) (Birtwell and McAllister 2002). Following EVOS, there were reduced growth rates of juvenile salmon in the year of the spill and lower return numbers in subsequent years (Carls et al. 1996a; Geiger et al. 1996; Wertheimer and Celewycz 1996; Willette 1996). Effects were greatest in the year of the spill and decreased in subsequent years. Pink salmon eggs had elevated embryo mortality in oil-affected streams during the fall seasons of 1989 to 1992, and in the lower intertidal zone up to 1996 (pink salmon also spawn in intertidal areas of Prince William Sound). In intertidal areas, elevated embryo mortalities were observed in intertidal areas in 1990 through 1992, decreasing in severity over time; however, no difference in survival of embryo to pre- emergent fry was detected for the 1989 to 1991 brood years (Bue et al. 1996). Juvenile pink salmon from oiled nearshore marine habitats had elevated levels of PAHs and cytochrome P4501A in 1989, compared to juveniles in reference areas; PAH levels had reduced by 1990 (Carls et al. 1996b). Based on the types of tissues most affected, Carls et al. (1996b) concluded that ingestion of whole oil was an important route of contamination for juvenile salmon and contributed to the reduced growth observed in oiled areas in 1989. Wertheimer et al. (1996) reported similar findings for juvenile pink salmon during initial residence in nearshore waters. EVOS resulted in contamination of juvenile pink salmon habitat, including approximately 31% of spawning streams and much nearshore rearing habitat in southwest Prince William Sound (Geiger et al. 1996).

Page 10-34 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.7.2.5 Commercial Fisheries During and following an incident that would result in hydrocarbons entering the marine environment, commercial fisheries in the affected area would likely be closed to harvest for some time to allow fish populations time to recover. DFO would make decisions about how long to maintain the closure, based on the extent and duration of the spill and contamination of commercially targeted fish and shellfish. Fisheries management issues are discussed below and socio-economic aspects are discussed in Section 11. Yender et al. (2002) provided a summary of oil spills in the United States and Europe where seafood monitoring was conducted and closures implemented. The length of time the fishery was closed depended more on site-specific characteristics of the physical surroundings and prevailing weather conditions than on the volume spilled. Examples provided by Yender et al. (2002) include: entire season closure for herring and salmon and advisories for shellfish harvest in some areas following EVOS closures of two months for wild finfish and 12 months for farmed fish following the 95,000 m3 spill of light crude oil from the T/V Braer in the Shetland Islands in January 1993 closures of three months for marine finfish, six months for whelks and crustaceans, four months for cockles, and eight to 19 months for mussels following the 80,000 m3 spill of light crude and heavy fuel oil from the T/V Sea Empress in Wales in February 1996 a 21-day closure for bivalves following a 265 m3 spill of bunker oil and marine diesel from the M/V New Carissa near Coos Bay Oregon in February 1999 (Gilroy 2000; Michel 2000; Mauseth and Challenger 2001) a 49-day closure for mariculture oysters and rock crab following a 17 m3 spill of fuel oil from the M/V Kure in Humboldt Bay California in November 1997 (Challenger and Mauseth 1998) a 15-day closure for lobster, scallop, clam and mussel following a 680 m3 spill of fuel oil from the T/V Julie N in Portland Maine in September 1996 (Mauseth et al. 1997) Hydrocarbons in the marine environment could result in tainting of local seafood resources, even if contaminant levels are low and there are no food safety concerns. Altered smell, taste or appearance of seafood would affect its palatability, marketability and economic value. Tainted seafood is defined as containing abnormal odour or flavour not typical of the seafood itself (International Standards Organization 1992). Both actual and potential contamination can affect commercial use of the seafood. Loss of confidence in seafood safety and quality can adversely affect markets, even after any actual risk has subsided, resulting in adverse economic consequences. Commercial fisheries generally target adult fish, which would probably be unlikely to be contaminated or tainted by a spill because their exposure might be brief while moving through an affected area. Also, any exposed finfish can rapidly remove hydrocarbons (Yender et al. 2002) and would not tend to store hydrocarbons in muscle tissue (Binderup et al. 2004). The exception would be if a large amount of fresh, light oil is mixed into the water column or if bottom sediments (particularly nearshore sediment used for spawning) become contaminated.

April 2010 FINAL Page 10-35

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Yender et al. (2002) summarized contaminant and tainting conditions for seafood following oil spills in the United States and Europe: total PAH values of 12 to 186 μg/kg in wild salmon, which rapidly declined following a February 1996 spill of 80,000 m3 of light crude and heavy fuel oil from the T/V Sea Empress in Milford Haven, Wales; there were no reports of tainting of salmon total PAH values of up to 19,000 μg/kg (and no tainting) reported for whelk, mussel and cockle (no tainting of whelks, no tainting data for mussels) following the T/V Sea Empress spill total PAH values of 1 to up to 2,650 μg/kg in finfish, depending on species, which declined to background in one to two months, following the January 1993 spill of 95,000 m3 of light crude oil from the T/V Braer in the Shetland Islands; no tainting for many fish species, but tainting was noted for one month for haddock and dab, two months for plaice and seven months for caged salmon up to 1,400 μg/kg in velvet crab and lobster, with lobster tainted for one month and no tainting reported for velvet crabs; also up to 3,580 μg/kg for whelk and scallop for 12 to 18 months (suspected tainting of scallop for two months) following the T/V Braer spill total PAH values of less than 15 μg/kg in Dungeness crab, with no tainting tests conducted, and up to 1,200 μg/kg for up to three weeks (no tainting reported) following a 265 m3 spill of bunker oil and marine diesel from the M/V New Carissa near Coos Bay Oregon in February 1999 total PAH values of 5 to 360 μg/kg in rock crab for one-half month, with no tainting reported, and up to 4,600 μg/kg in oysters, with no tainting reported following a 17 m3 spill of fuel oil from the M/V Kure in Humboldt Bay, California in November 1997 These case studies indicate that wild finfish are seldom tainted, and if they are, it is for a short time (one to two months). Laboratory experiments have reported tainting in salmon after eight-hour exposure to 0.4 mg/L of the water-soluble fraction of crude oil (Ackerman and Heras 1992) and in Arctic char after four hour exposure to 50 mg/L of a crude oil (Lockhart et al. 1996). Shellfish are more likely than finfish to become contaminated because they are more vulnerable to exposure in the intertidal areas, less efficient at metabolizing hydrocarbons and less mobile, with more contact with sediments, which can become contaminated and serve as a long-term source of exposure. Yender et al. (2002) indicated PAH levels were reduced in crabs and lobster relatively quickly (one to two months), with little or no tainting reported. In contrast, PAH levels in clams, oysters, mussels and other shellfish were elevated for longer (4 to 17 months), depending on species, with few reports of tainting. Potential biological effects on mussels, other shellfish and crabs are discussed in Section 10.6.

10.7.2.6 FSC Fisheries As detailed above for commercial fisheries, both actual and potential contamination of seafood can affect FSC harvest of seafood. It is assumed that loss of confidence in seafood safety and quality will affect traditional FSC harvesting activities, resulting in economic and social consequences for communities that depend on the resource. During various engagement events, Aboriginal organizations expressed concerns regarding contamination of traditionally harvested marine foods and the need to people to have to travel

Page 10-36 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment to other areas to harvest uncontaminated foods. Closures for conservation reasons would also reduce access to fish for FSC purposes. Areas of high fishing effort are identified in TERMPOL 3.3.

10.7.2.7 Recreational and Commercial-Recreational Fishing Effects of hydrocarbons on recreational and commercial-recreational fishing are likely to be similar to those described for commercial fisheries. If contamination of seafood is detected, the species will be closed to harvesting; this could have adverse economic consequences for fishing lodges and charters, which depend on recreational fishing customers as their source of livelihood. Loss of confidence in seafood safety and quality could adversely affect commercial recreational and recreational fishing, even after any actual risk has subsided, resulting in adverse economic consequences. Closures for conservation reasons would also affect recreational and commercial-recreational fishing.

10.7.3 Potential Effects of Condensate on Fish and Fish Habitat Fish respond rapidly to toxic volatile compounds present in condensate, which can penetrate (especially via the gills) and disturb the respiratory, nervous and circulatory systems and disrupt enzyme activity. Behaviour changes such as excitement, increased activity or scattering in the water are the first signs of exposure to condensate. Longer exposure can lead to chronic poisoning, which involves cumulative effects at the biochemical and physiological level. A typical effect is gas embolisms, which includes rupture of tissues (especially in fins and eyes), enlarging of the swim bladder, disturbances of the circulatory system and other pathological changes. As with crude oil contamination, the most vulnerable periods for exposure to condensate are the early life stages (Patin 1999).

10.7.4 Mitigation Measures To limit the effects of hydrocarbon exposure on fish and fish habitat, the following types of mitigation would be implemented, as appropriate: exclusion booming around the entrance to sensitive spawning habitat (e.g., estuaries and stream outlets) to protect salmon fry exclusion booming to protect rearing areas around estuaries where juvenile eulachon are present (January to April) exclusion booming to protect known herring spawning and rearing areas issuing of public advisories about commercial fisheries and areas that might be affected, including information about the geographic extent over which the hydrocarbons spread and guidelines for seafood consumption issuing of a general advisory to recreational and traditional users not to consume oiled or tainted fish or shellfish; with the advisory in place for a fixed period defined by the regulatory agencies or until an assessment and species monitoring program indicate a return to levels considered acceptable by regulatory agencies

April 2010 FINAL Page 10-37

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

development of a compensation plan with commercial fishers based on fish landings in previous years, with FSC users based on landed harvests from previous years, and with commercial- recreational lodges and charters for lost business based on previous years activity Fisheries and Oceans Canada and certified responders will be engaged in the development of Geographic Specific Plans for the protection of fish habitat and response. Aboriginal organizations and public stakeholders (e.g., fisheries) who could be potentially affected by an accidental release will also be consulted in the development of the plans. Regulators and local interests would help to priotitize the priority areas for protection of fish habitat and fish and the types of response measures that may be employed.

10.7.5 Follow-up Monitoring As noted earlier, Northern Gateway invited coastal Aboriginal groups to participate in the establishment and monitoring of environmental transects and sites that would include monitoring of marine fish. Information from these transects and sites would be used to characterize baseline conditions. given the importance of fishing to local communities, representatives of the FSC, commercial, recreational and commercial-recreational fishers would be involved in design of the monitoring study and review of results. The following types of monitoring programs could be implemented, as applicable, to address concerns related to the health of finfish and shellfish: salmon returns to major river systems in the CCAA over the long term, with comparison to annual escapement data collected by DFO contaminant levels in water and sediment within an appropriate distance from the release contaminant levels and tainting in fish and invertebrate tissue loss of income for commercial fishers and commercial-recreational businesses loss of harvesting opportunities by the FSC fishery Monitoring would typically continue until specific end points are achieved and residual hydrocarbons reach acceptable background levels.

10.8 Marine Birds Marine birds is a broad term applied to avian species that are dependent on aquatic habitats for at least part of their life cycle. In addition to the diversity and abundance of avian life they represent, marine birds are important components of the freshwater and marine environments in which they are found (Milko et al. 2003). Marine birds make extensive use of coastal wetlands, nearshore and offshore habitats, including islands, islets and cliffs. The large populations of marine birds common to the British Columbia coast are an integral part of the marine ecosystem. The Pacific coast is also an important migratory corridor for millions of birds, especially shorebirds and waterfowl (Slattery et al. 2000). Many of the colonial breeding marine birds found in British Columbia do not breed elsewhere in Canada (Campbell et al. 1990).

Page 10-38 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Marine birds can be divided into four main groups based on seasonal abundance, habitat use and breeding distribution: breeding resident, winter resident, summer visitor and spring and fall migrant. They can also be loosely grouped based on life history characteristics and behavioural traits that include: pelagic seabirds (e.g., alcids and tubenoses), which spend most of their time on open water waterfowl (e.g., loons and ducks), which generally breed inland, but migrate, moult, or overwinter in marine habitats shorebirds (e.g., oystercatchers and sandpipers), which breed inland but forage coastally during spring and fall migrations coastal raptors (e.g., eagles and ospreys), which live close to the coast and make widespread use of marine resources Birds that spend much of their life on the water (e.g., alcids) are more vulnerable to the effects of hydrocarbons than those that spend much of their life on land or in flight (Piatt et al. 1990; Kauffmann et al. 2009, Internet site). Marine birds are protected nationally, as non-game birds, under the Migratory Birds Convention Act and provincially under the British Columbia Wildlife Act. The MBCA regulates hunting seasons and prohibits commercial exploitation and wilful destruction of migratory birds, taking of their eggs and disturbance of their nests. The Canada Wildlife Act provides for the coordination of wildlife programs and policies for birds not protected under the MBCA. In British Columbia, all species are classified as one of Red-, Blue- or Yellow-listed. Red- and Blue-listed species are considered for formal designation as Endangered or Threatened, either provincially under the British Columbia Wildlife Act or nationally by COSEWIC. Species at risk are protected under SARA, which applies to federal lands and protects all wildlife species listed as at risk and their critical habitat.

10.8.1 Baseline Conditions Large numbers of breeding, migrant and wintering marine birds, including seabirds, waterfowl, shorebirds, gulls and coastal raptors, occur in the near and offshore waters of British Columbia‘s north coast. The CCAA provides habitat for more than 100 marine bird species during some part of their life cycle. The Marine Birds TDR provides a thorough discussion of the abundance, distribution and seasonal use of the CCAA by marine birds. There are large numbers and concentrations of wintering and migratory birds in the CCAA and also locally breeding populations. Biologically important areas for birds in the OWA are shown in Figure 10-3 and summarized by ecosection in Table 10-4. Many of the areas shown were identified by consolidating distributional and sightings data from a variety of sources, including: BirdLife International‘s Important Bird Area (IBA) maps by species (BirdLife International 2008, Internet site), BCCDC‘s Species and Ecosystems Explorer (ILMB 2009b, Internet site), Table K.3 from McFarlane Tranquilla et al. (2007), and the Coastal Resource Information System (ILMB 2009b, Internet site). All layers can be linked directly to references, with the exception of the IBA of Significance layer at Aristazabal Island, which is a known IBA to BirdLife International; this area was based on expert knowledge.

April 2010 FINAL Page 10-39

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Figure 10-3 Summary of Biologically Important Areas for Marine Birds in the OWA

Page 10-40 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-4 Important Bird Areas Associated with Ecosections in the OWA and CCAA

Marine Ecosection Important Bird Areas (IBA) Haida Gwaii (northern Haida Gwaii provides essential year-round habitat for many species, including Continental Slope, Blue-listed Ancient Murrelet (most of the known nesting areas in British Dixon Entrance, Columbia and Canada are located on Haida Gwaii, 74% of global breeding western Hecate Strait, population). Also globally significant IBAs for Leach’s Storm Petrel, Fork-tailed northwestern Queen Storm Petrel, Brant, Pigeon Guillemot, and Cassin’s Auklet. Nationally Charlotte Sound) significant IBAs for Peale’s Peregrine Falcon and Marbled Murrelet. Known breeding habitat for Pelagic Cormorant, Red-breasted Merganser, Sand Hill Crane, Short-billed Dowitcher. Scott Islands (mid The Scott Islands and surrounding waters are an area of global IBA for many Continental Slope, species. Roughly 90% of British Columbia’s Blue-listed Tufted Puffins occur northern Vancouver here (Global IBA). The entire British Columbia population of Red-listed Thick- Island Shelf, southern billed Murre breed exclusively on Triangle Island. The entire population of Queen Charlotte Common Murre and Horned Puffin (Red-listed) breed only here and on the Sound) Kerouard Islands (QCI). Hecate Strait and Shallow waters of Hecate Strait are an area of particularly high marine bird Queen Charlotte Sound densities. Black Oystercatchers have been known to nest along 80% of British Columbia’s shoreline, and 30 to 35% of the global population breeds on the British Columbia coast; areas around Princess Royal Island are a globally important IBA. The northwest has also been identified as a globally significant IBA for Pacific Loon. Coastal Inlets (North Queen Charlotte Strait is a globally significant IBA for Leach’s Storm-Petrel. Coast Fjords, Queen Northern inlet waters (particularly around Prince Rupert) are global IBAs for Charlotte Strait waterfowl and Rhinoceros Auklets. Both ecosections are national IBAs for Black Turnstone. Vancouver Island Shelf Important habitat and breeding grounds for many species, including Red-listed Marbled Murrelet (globally significant IBA). The northern part of the shelf is globally significant IBA for Leach’s Storm Petrel. Southern part is globally significant IBA for Western Grebe, Surf Scoter, Brandt’s Cormorant and many species of shorebird and gull, and is a nationally significant IBA for Northern Fulmar.

The IBA of Significance (orange layer) is a composite of IAs for key marine bird species in British Columbia (Clarke and Jamieson 2006; Canadian Wildlife Service 2007; BirdLife International 2008, Internet site). Sea bird habitat (green layer) is a compilation of important habitat for bald eagles, black oystercatchers, cormorants, dabbling ducks, diving ducks, fulmar, shearwaters, petrels, geese, great blue heron, gulls, loons, grebes, shorebirds and other species (ILMB 2009b, Internet site). Each species layer was overlaid so that areas of importance to multiple species appear darker. Important bird breeding colonies, as well as habitat for alcids and marbled murrelets were highlighted separately. Globally significant populations of some species of colonially breeding marine seabirds, such as Cassin‘s Auklets, Ancient Murrelets and Rhinoceros Auklets, occur in the OWA. Marine birds make extensive use of coastal wetlands, as well as nearshore and offshore habitats, and are an integral part of the coastal marine ecosystem. As of 1991, an estimated 5.6 million colonial seabirds were nesting at 503 sites in the OWA (reported by Rodway in McFarlane Tranquilla et al. 2007), and many of these do not breed

April 2010 FINAL Page 10-41

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment anywhere else in Canada. Shelf waters, especially near inshore banks, support the highest abundance and diversity of birds (Morgan et al. 1991). The British Columbia coast is also a highly important corridor for millions of migrating birds, particularly waterfowl and shorebirds. A list of the marine bird species most likely to be found in the OWA is located in Appendix B, Table B-1. Marbled Murrelet, Surf Scoter, Bald Eagle and Black Oystercatcher are considered representative bird species because they occur in large numbers, are sensitive to changes in the environment and provide important functions in the marine ecosystem. Moreover, they contribute to local and global biodiversity, and hold social, cultural and aesthetic values. In addition, several other species of wintering shorebirds forage along exposed mussel beds and shoreline habitats at low tide. For example, Black Turnstones, Surfbirds and Rock Sandpipers seek small invertebrates among scattered shells and seaweed. Crows and ravens commonly prey on shellfish and scavenge at low tide.

10.8.1.1 Marbled Murrelet The Marbled Murrelet is a species of national (Threatened) and provincial (Red-listed) conservation concern. Marbled Murrelet numbers along the north coast of British Columbia are estimated to range from 10,000 to 14,600 birds, approximately 12% of the North American population and 18% of the British Columbia population (Piatt et al. 2007; Marine Birds TDR). Marbled Murrelets that occur in the CCAA are part of a genetically distinct population that ranges from the eastern Aleutian Islands to northern California (Friesen et al. 2005, 2007). Marbled Murrelets reside year-round in the CCAA. They are solitary breeders, but when observed at sea, they are typically in groups of two to 10, although local concentrations of up to 400 have been observed (e.g., Gilttoyees Inlet Foch Lagoon, the waters around McCauley Island and Principe Channel; Steventon and Holmes 2002; Marine Birds TDR). They typically forage and find shelter in shallow protected bays and inlets in the CCAA.

10.8.1.2 Surf Scoter The Surf Scoter is a provincially Blue-listed species. In British Columbia, the Surf Scoter is the most numerous wintering sea duck, frequently outnumbering all other marine birds (Savard et al. 1998). However, the North American population is declining, most notably in northwestern Canada and Alaska (Sea Duck Joint Venture 2004, Internet site). Surf Scoters breed inland, on shallow lakes in the boreal forest and tundra, but migrate and winter along the coast of British Columbia, where as much as 50% of the world population might occur. However, population estimates for this species in general are quite poor (Campbell et al. 1990). Surf Scoters forage in intertidal areas on molluscs, crustaceans and insects. In spring, they congregate in large groups of up to 50,000 individuals in areas where Pacific herring are spawning. Surf scoters are present in all waterways of the CCAA, with notable concentrations during winter (e.g., off Hawkesbury and Farrant Islands; Marine Birds TDR) and late summer when males moult in areas within the CCAA, such as Coste and Hawkesbury Islands (Boyd et al. 2001). In addition, they mass locally at herring spawn locations. Some non-breeding birds remain in the CCAA throughout the year.

Page 10-42 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.8.1.3 Bald Eagle The Bald Eagle is the only coastal raptor that uses the CCAA throughout the entire year. Individuals breed in the CCAA and several nest sites have been identified during field surveys (Marine Birds TDR). Bald Eagles are considered a top predator and feed on a variety of terrestrial and aquatic species, including salmon and intertidal organisms. As such, biomagnification of contaminants has affected the North American population and makes eagles vulnerable to hydrocarbons in marine waters (e.g., Nisbet 1989). The Bald Eagle population has been steadily increasing since the 1960s. In coastal British Columbia, breeding and wintering populations have increased at a higher rate than elsewhere in Canada (Environment Canada 2000, Internet site). In addition, the coastal wintering population is larger than the breeding population on the coast each year, suggesting that coastal British Columbia serves as a wintering area for eagles from elsewhere in western North American (Environment Canada 2000, Internet site).

10.8.1.4 Black Oystercatcher The Black Oystercatcher (Haematopus bachmani) is a resident shorebird that occurs along the Pacific Coast of North America from mainland Alaska and the Aleutian Islands south to Baja California, Mexico. Most breed between south-coastal Alaska and coastal British Columbia (Andres and Falxa 1995, Internet site; Morrison et al. 2001). Throughout its range, it is uncommon and patchily distributed; however, the global population is ranked as secure (BC MoE 2009a, Internet site). The oystercatcher is yellow-listed in British Columbia. The oystercatcher occurs throughout the CCAA and likely breeds in Principe Channel and other areas of suitable habitat (Marine Birds TDR). Oystercatchers are year-round obligate users of the intertidal zone feeding on marine invertebrates, particularly bivalves and other molluscs (limpets, whelks and chitons (Andres and Falxa 1995, Internet site). Access to foraging habitat is strongly dependent on tides and surf action, and most feeding is done at low tide (Andres and Falxa 1995, Internet site). Oystercatchers are long lived (e.g., greater than10 years) and form permanent pairs (i.e. mate for life; Andres and Falxa 1995, Internet site). Breeding habitat is exclusively associated with the high tide margin of the intertidal zone, and includes mixed sand and gravel beaches, cobble and gravel beaches, exposed rocky headlands, rocky islets, and tidewater glacial moraines (Jehl 1985). Nests less than one metre from waterline are often flooded during extreme high tides (Andres and Falxa 1995, Internet site). Breeding territories are usually in close proximity to dense mussel beds (Groves 1984). The availability of suitable shoreline along the west coast likely limits global population growth (Andres and Falxa 1995, Internet site). The period from hatching until first flight is the most critical life history stage (Groves 1984). In winter, flocks concentrate on protected, ice-free tidal flats with dense mussel beds (Hartwick and Blaylock 1979). Winter flocks seldom range more than 50 km from nesting sites (Terres 1980).

April 2010 FINAL Page 10-43

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.8.1.5 Other Species

Loons and Grebes Loons and grebes breed on inland lakes during summer and migrate to coastal wintering grounds in the fall. They are divers and feed just offshore or in large inshore bays and inlets throughout the year, although their abundance in the marine environment is highest in winter and during the spring and fall migrations. Red-throated loons also collect food for their young from marine waters and return to freshwater during the breeding season. There are five species of loon, four of which (Common, Pacific, Yellow-Billed, and Red-Throated) occur in coastal British Columbia. Six of North America‘s seven grebe species have been observed in British Columbia. Horned, Red-necked and Western grebes are common throughout the winter and during spring (April to May) and fall (September to October) migration. Pied- billed Grebes are rare visitors to British Columbia waters.

Tubenoses Tubenoses include albatrosses, fulmars, shearwaters, petrels and storm petrels. These species generally breed in the Southern Hemisphere, after which they gather in large numbers off the coast of British Columbia during the non-breeding season. Black-footed Albatross are common in Pacific coastal waters of the continental slope, transitional Pacific and subarctic Pacific during late summer. Laysan and Short- tailed Albatross are rarer. Northern Fulmar nest in Alaska in summer and are found in large numbers in northern British Columbia waters during winter. Murphy‘s Petrels are usually seen offshore, especially in spring. Sooty and Short- tailed Shearwaters are common to Hecate Strait while Pink-footed, Flesh-footed and Buller‘s shearwaters prefer offshore waters. Sooty Shearwaters are the most abundant shearwater in the OWA, particularly around Haida Gwaii, where they gather over deep waters in large feeding flocks that can number hundreds of thousands. Storm petrels are most common on the coast, but highest densities are seen over deep pelagic waters.

Cormorants Three species of cormorant are found in British Columbia (Double-Crested, Brandt‘s and Pelagic) and all are year-round residents. Pelagic and Brandt‘s Cormorants are provincially Red-listed, and the double- crested cormorant is provincially Blue-listed. All three species breed in small colonies along the coast. They often gather in large coastal flocks and roost and nest in prominent places. Brandt‘s Cormorants are associated with the open ocean more often than the other two species and often roost on offshore rocky outcroppings. All three species are piscivorous.

Waders There are only two species of waders regularly seen British Columbia waters: the American Bittern and the Great Blue Heron. The American Bittern occurs mostly in southern parts of the OWA in winter. The Great Blue Heron is a resident along the coast, and is regularly seen foraging along the water‘s edge. Waders are commonly found in marshes and mudflats or in shallow waters of the intertidal zone.

Page 10-44 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Geese and Swans Geese and swans are present in protected bays and estuaries of the coast year-round; however, there is no recorded nesting in the OWA. Abundance is highest from August to April. The Greater White Fronted Goose migrates in large flocks in spring and fall, offshore and along the coast, with the Snow Goose. Almost the entire global population of Brant geese migrates along the coast every spring.

Diving Ducks Diving ducks spend the majority of their life on the water and are widespread and common throughout coastal British Columbia. Although abundance is generally highest during migration and through winter, many species are year-round residents.

Dabbling Ducks Dabbling ducks frequent nearshore waters along coastal British Columbia, particularly in shallow protected bays, inlets and estuaries. These species might feed in marine environments but do not generally nest on the coast. Some species are present year-round, with the Mallard being the most abundant. Others, such as the American Wigeon, are present in greatest numbers during winter. During spring and fall migrations, Northern Pintails and American Wigeon can also be seen resting in large flocks offshore.

Coastal Raptors Coastally dependent and commonly occurring birds of prey include Bald Eagles (discussed above), Osprey and Peale‘s Peregrine Falcon. These three species are found on open, saltwater and estuarine habitats. Peale‘s Peregrine Falcon preys primarily on colonial-nesting seabirds. Ospreys consume almost exclusively fish and occur along the entire coast.

Rails, Coots and Cranes The American Coot is found in marine and brackish waters, including marshes and tidal mudflats in fall and winter. The Sandhill Crane is present year-round. Although it is not commonly seen in saltwater, it uses low-lying patches of shoreline and breeds on Haida Gwaii and coastal mainland islands.

Shorebirds The Black Oystercatcher (discussed above) is the only shorebird found year-round in coastal British Columbia. Other shorebirds, although widely distributed, use the area mostly during spring and fall migrations (when shorebird abundance peaks) or as winter visitors. A few species, such as the Semi- palmated Plover, Killdeer, Spotted Sandpiper and Short-billed Dowitcher also breed in coastal British Columbia.

Gulls, Jaegers, and Terns There are 30 species of gulls, jaegers, and terns found in British Columbia. They are abundant and distributed widely year-round. The Glaucous-winged Gull is the only colonially breeding gull on the coast, though Mew Gulls also occasionally breed on the coast. Although many species are constrained to

April 2010 FINAL Page 10-45

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment the nearshore environment some, such as the Black-legged Kittiwake and Arctic Tern, are predominantly pelagic and can be found offshore in winter (kittiwake) and during migrations (tern).

Alcids Thirteen species of alcids (puffins, auks, auklets, murres, murrelets) occur along coastal British Columbia. Alcids generally spend most of their lives in the marine environment, inhabiting nearshore and offshore habitats year-round. Most species feed on the open ocean and breed in dense colonies on sea cliffs and offshore islands. This high density renders colonies vulnerable to environmental or anthropogenic disturbance.

Kingfishers The Belted Kingisher occurs in British Columbia and is present year-round in low densities. Kingfishers are strictly piscivorous, and perch on coastal snags and branches, hunting for prey in water immediately adjacent to the shore. They require clear, ice-free water for diving and vertical earthen banks for nesting.

10.8.2 Potential Effects of Diluter Bitumen or Synthetic Oil on Marine Birds The potential effects of diluted bitumen on marine birds can be acute or chronic and include: mortality from hypothermia or toxicity of hydrocarbons ingested during preening or consumption of contaminated prey reduced health from increased energy expenditure for preening to remove oil from their feathers or switching to a non-contaminated, but inferior prey-base (i.e., less plentiful foraging grounds or foraging trips are longer and require more energy) reduced breeding success due to loss of mate or transfer of oil from adult to eggs or nestlings asphyxiation from toxins evaporating into the air above the surface of the water loss or damage to habitat Although most studies have focused on marine birds, similar effect mechanisms are predicted for birds that use intertidal habitat (e.g., waterbirds, waterfowl) and land (e.g., passerines, raptors) as a component of their life cycle requirements. Lesser effects are predicted for shorebirds and wading birds (e.g., Great Blue Heron, Rock Sandpiper) than for waterfowl (e.g., dabbling ducks, geese), as these species typically spend little time in the open water, where the likelihood of mass feather contact with hydrocarbons is greatest. Many waterbirds and shorebirds are most vulnerable during spring and fall migrations, when there is potential for large flocks of staging birds (i.e., Black Turnstones, sandpipers) to use marine shorelines. Shorebirds and wading birds also might be exposed to hydrocarbons if they forage in intertidal areas that are fouled with oil. Some species also nest close to the waterline (e.g., oystercatchers). Effects on marine birds are noted at an oil thickness of 0.01 mm or more (French-McCay et al. 2004). Direct contact with oil results in clogging of feathers, which might lead to loss of buoyancy, drowning and loss of thermal insulation. Hypothermia and death are common, as the loss of thermal insulation often

Page 10-46 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment coincides with increased metabolic costs and stress as birds attempt to maintain their core body temperatures (Stephenson 1997; Pribilof Islands Wildlife Protection Subgroup 2001, Internet site; Mazet et al. 2002). Hypothermia effects are most severe during sensitive periods such as moulting. Behavioural changes such as excessive preening and difficulty flying might also deplete energy stores quickly and put birds at higher risk of predation (Stephenson 1997). Hydrocarbon exposure might have immediate effects on reproductive success of waterbirds, as the oiled birds spend excess time preening, so do not invest as much time and energy in breeding and nesting. This is thought to affect lifetime reproductive success (Albers 1983; Albers 1990; Fry 1995). Ingestion of hydrocarbons also can affect reproductive ability (Jenssen 1994). Eggs can be affected if oil is transferred from adults to the egg surface; this has been associated with reduced embryo survival and hatching success (Szaro and Albers 1977), possibly due to a reduction in gas exchange. Oil ingestion during preening of oiled feathers can lead to toxic effects, including organ damage and, in sufficiently high doses, organ failure. Oil ingestion can also reduce reproductive rates if it occurs during breeding season. Adult birds are more resistant to toxic effects of hydrocarbons, and must ingest large amounts to cause direct mortality (Jenssen 1994). Adults that have ingested or are exposed to oil have been shown to exhibit reduced egg-laying, hatchability and breeding success (Fry et al. 1986; Velando et al. 2005; Ainley et al. 1981) and PAH accumulation can lead to altered immunological defences (Rocke et al. 1984) that can reduce survival of individuals over time. Ingestion of contaminated prey from the intertidal zone can lead to mortality for raptors such as Bald Eagle and scavengers such as American Crow and Raven, which are often attracted to the shoreline to feed on compromised and dead birds. Given the tendency for tidal action to deposit floating carcasses on shorelines, this cycle of deposition and subsequent scavenging could persist in the weeks immediately following a spill.

10.8.2.1 Marbled Murrelet Birds such as Marbled Murrelets and other alcids have a low reproductive rate, spend most of their time on the water, dive to feed and have a flightless moulting stage; thus, they are vulnerable to hydrocarbon exposure (Speich et al. 1991; Berger 1993; Ford et al. 1990; Ford et al. 2001). Over 80% of birds recovered after the Nestucca and Exxon Valdez incidents were alcids (Berger 1993; Piatt et al. 1990). Marbled Murrelets would be at greatest risk during fall, when juveniles and adults are found on the water and adults are moulting and flightless, making it difficult to navigate around surface oil. During spring and summer, the risk of effects would be less because the young are in the nest and the adults can fly away from surface oil. However, effects would occur when oiled adults abandon their nest or transfer oil to their eggs or chicks. The exposure of eggs to as little as 5 μL of oil can result in embryo mortality (Albers 1978; Hoffman 1978), and the transfer of oil to nestling feathers might result in reduced insulation and hypothermia (Szaro 1977). Murrelets are least vulnerable during winter when nesting and moulting are finished. Hydrocarbons in the marine environment would affect Marbled Murrelet foraging habitat, and any reduction in abundance or availability of prey (small fish such as herring and sand lance, which feed and spawn intertidally) could force them to forage in less productive areas. Because oil tends to accumulate in intertidal areas for extended periods, this habitat is often strongly affected.

April 2010 FINAL Page 10-47

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Recovery of Marbled Murrelet populations could take four to five years, based on reproduction and survival rates (Beissinger 1995; Cam et al. 2003; McShane et al. 2004). Recovery could take longer if chronic effects related to oil or environmental conditions (e.g., reduced food availability) have reduced bird health. The loss of birds from the nesting population in the CCAA could have serious implications for the local population, but would likely have very little effect on the coastal population unit, which consists of birds from Alaska to California (Friesen et al. 2005, 2007). Additionally, effects would be limited if the murrelets remain in fjords or inlets such Gilttoyees Inlet and Foch Lagoon, which can be boomed.

10.8.2.2 Surf Scoter Exposure to diluted bitumen has the potential to affect Surf Scoter populations, due to the large numbers that use coastal waters during migratory and wintering periods. Further, Surf Scoters feed in shallow intertidal areas, where oil tends to accumulate. Male Surf Scoters would be most vulnerable during preening, in mid-to-late summer, when they, along with sub-adults and non-breeding females, are moulting at sea (Savard 1988). During summer, Surf Scoters tend to occur in fjords and shallow waters, such as around Coste Island and Hawkesbury Island (Boyd et al. 2001); these areas could be boomed. During summer, breeding females and newly fledged chicks would not be present, as they do not return to the coast until early October. Relatively small numbers of birds (less than 100 individuals) have been recorded in fall in shallow bays and estuaries with rocky substrates within the CCAA. The Surf Scoter is sensitive to habitat changes resulting from hydrocarbons, which would cause changes in prey abundance or availability. They are particularly vulnerable in spring when they forage at herring spawn sites. The Kitkatla Channel and Goschen Island North to Porcher Island IBA (IBA BC119) support large numbers of spring migrating surf scoters (IBA Canada 2009, Internet site). Surf Scoter population recovery from hydrocarbon expsosure would be expected to be relatively rapid (two years) based on recovery rates reported for Prince William Sound after EVOS (Day et al. 1997).

10.8.2.3 Bald Eagle Bald Eagles would be vulnerable to a spill in the coastal OWA and CCAA because they inhabit the area year-round. Effects would be largest during the nesting period, herring spawning season (spring) and spring and summer salmon migrations. They would be particularly affected (King and Sanger 1979), if oil moves into intertidal areas where they spend considerable time foraging. While the effects of oiled plumage on survivorship are less certain than in other birds, mortality can occur. Adult eagles with oiled plumage could transfer oil onto eggs or fledglings; this could affect nesting success, as small quantities of oil on eggs can lead to embryo mortality or cause deformities, especially during the early incubation phase (Albers 1980; Albers and Szaro 1978). Ingestion of oil from contaminated prey and feather preening can lead to reproductive effects such as reduced hatchability of eggs and breeding success (Fry and Lowenstein 1985; Velando et al. 2005; Ainley et al. 1981), possibly because of altered egg laying and incubation behaviour (Holmes et al. 1978). Immediately following EVOS, traces of oil were documented in 24 of 66 Bald Eagle eggs, but this decreased to 1 in 54 eggs the following year (Bence and Burns 1995). This suggests short-term effects on nest productivity (Bowman et al. 1993; Albers and Szaro 1978). In general, nest productivity and population of Bald Eagles recovered

Page 10-48 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment one to two years after the spill, and no long-term effects on densities or reproduction were documented (White et al. 1995). Effects of oil ingestion appear to be less serious for Bald Eagles than for other species. Blood samples taken from Bald Eagles within an oiled area appeared relatively normal within two to three months after a spill (Gibson and White 1990). In addition, Bald Eagles have been observed to avoid oiled areas (Gibson and White 1990 ) and might avoid eating contaminated prey if non- contaminated prey is available (Pattee and Franson 1982), thereby reducing the probability of adverse health effects. Indirect effects include reduction in water clarity, which would impede the birds‘ ability to forage. It is expected that Bald Eagle mortality would be low and limited to the immediate spill area.

10.8.2.4 Black Oystercatcher Black Oystercatchers are highly vulnerable to the effects of hydrocarbons that drift to shore because of their dependency on intertidal habitat (Andres 1998). Oystercatchers would be susceptible to effects such as hypothermia and ingestion of hydrocarbons during preening and foraging. Depending on the location of the spill, variable numbers could be affected as individual and small numbers of oystercatchers are distributed in the CCAA throughout the year. Oystercatchers would be most vulnerable during the breeding season when hydrocarbon exposure could cause the direct mortality of adults, eggs and chicks. Indirect mortality of eggs and chicks could also occur if mortality of the provisioning adults occurred. The oystercatcher prey base is comprised almost exclusively of filter-feeding invertebrates (e.g., mussels), such that contaminant biomagnification would quickly occur in oystercatchers foraging in contaminated intertidal zones. Contaminated mussel beds could provide a chronic source of exposure (Andres 1994). Data collected following EVOS, showed oystercatchers experienced direct mortality, disruption of breeding activity and reduced survival of chicks. In addition, cleanup activities disrupted breeding birds into 1990 (Andres 1998). Productivity of pairs after 1990 suggests that any ingestion of oil that occurred in 1989 or 1990 had little effect on the subsequent reproductive performance of oystercatchers in some instances (Andres 1998). However, the presence of elevated hydrocarbon concentrations in the feces of chicks provided evidence that oystercatchers were exposed to persistent shoreline oil until 1993 (Andres 1999). Depending on the location, the geographic extent of the spread of hydrocarbons and the time of year, variable local oystercatcher mortality and disruption of breeding would occur. Effects would be limited to intertidal zones affected by the hydrocarbons. Recognizing that Black Oystercatchers are long-lived and monogamous, it could take approximately two to five years for a locally affected breeding population to recover from a large number of mortalities (Agler et al. 1994; Andres 1998; Day et al 1997). Although ingestion of hydrocarbons might occur in subsequent years, research from the Exxon Valdez incident suggests that productivity of breeding oystercatchers is not reduced in following years. In non-breeding seasons, buffer zones restricting boat and human traffic could be established around feeding and roosting concentrations. In areas at risk to large-scale environmental perturbations (e.g., oil spills), baseline information on breeding density, nonbreeding population size, reproductive success, prey abundance and prey quality should be collected.

April 2010 FINAL Page 10-49

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.8.3 Potential Effects of Condensate on Marine Birds Condensate would have fewer effects than diluted bitumen or synthetic oil because condensate does not spread as far or cover as much water. Protective measures and mitigation would be similar for all hydrocarbons transported to and from the Kitimat Terminal. Condensate would disperse rapidly from the water surface, but the evaporating toxins could cause asphyxiation of marine birds in the immediate area; however, the evaporates would disperse quickly and any effect would be very short-term.

10.8.4 Mitigation Measures General emergency responses are described in Section 7. Geographical, seasonal and species–specific plans will be developed, when appropriate, before operations to identify areas of important marine bird use that could be at risk, based on information presented in this assessment for the CCAA, traditional and local knowledge and future studies. Northern Gateway, in cooperation with the Canadian Wildlife Service (CWS) and certified responders, will develop a Marine Bird Protection Plan before operations, which will specify and guide all mitigation related measures. Aboriginal organizations and affected stakeholders (e.g., bird watchers, ecotourism operators) will also be consulted in the development of these plans. Plans will include habitat protection measures (i.e., booming of sensitive habitat areas), a Marine Bird Hazing Plan, a Marine Bird Cleaning Plan and a Marine Bird Follow-up Plan. Exclusion booming of sensitive habitat will depend on the location and movement of the hydrocarbons and time of the year. Sensitive areas that would benefit from booming include protected fjords and inlets branching off the main waterways where birds forage and find refuge, like Bish and Emsley coves. These and other sensitive habitats will be noted on the coastal sensitivity maps that are being developed. The CWS provides direction on spill responses to manage affected birds, including hazing of non-oiled animals and cleaning of oiled animals. The hazing plan will be comprehensive and include information such as type, quantity, location, maintenance and use of necessary equipment, contact information for experts trained in hazing techniques, coordination of hazing measures and responsibilities and roles of relevant stakeholders and experts. Hazing strategies would focus on flushing and keeping birds away from the hydrocarbons. The most effective deterrent device is usually sound (e.g., rockets, mortars, gas cannons, Breco buoys) (Koski et al. 1993), but small vessels can also be used to move birds away from hydrocarbons. Sound might not be effective for Bald Eagles in the long term, as they are known to habituate to sound. In open waters, where booming will be logistically difficult, scare tactics designed to flush birds will be used. Capture and salvage (e.g., by dip netting) might be successful for removing small numbers of moulting birds (Lougheed 1999). Pyrotechnics and bioacoustics are the most effective hazing techniques for some species. Pyrotechnics work well for hunted bird species, but might cause diving birds to dive rather than fly away. Use of pyrotechnics will depend on the season, as some birds cannot fly during some stages of their moult (California Department of Fish and Game 2005, Internet site). Gas exploders and lasers vary in effectiveness, and success depends on the species being hazed. Techniques or combinations of techniques that deter the most species should be employed (Gilsdorf et al. 2002). Flammable gases in the hydrocarbons might prevent the use of pyrotechnics, propane cannons, flares, or other items that use a battery or produce a spark.

Page 10-50 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Oiled birds should be removed if for no other reason than to remove a continuing source of contamination to other wildlife and the environment (Jessup 1998). CWS (1999, Internet site) uses the criterion of ‗time to recovery‘ to assess rehabilitation priority and assigns a higher priority to rare species with low reproductive capacity than to species that are more likely to re-establish quickly. Oiled birds are not usually cleaned unless they are designated as species at risk. CWS (1999, Internet site) requires that an oiled bird rehabilitation program be established where there is reasonable hope of rehabilitation success or if species at risk are affected. Success of a rehabilitation program depends on a fast response time and the state of the individual birds, so response resources should be stationed at intervals along the CCAA. Oiled birds are usually transported to a rehabilitation centre, where oil is manually cleaned off the plumage. The goal of the bird rehabilitation program is to re-introduce rehabilitated birds to their natural habitats and breeding populations (CWS 1999, Internet site). Rehabilitation has had variable success. Jessup (1998) estimated that 50 to 60% of birds that go through rehabilitation successfully re-enter their breeding population. In cases where birds cannot be saved or are low priority for rehabilitation (i.e., non-species at risk), CWS (1999, Internet site) advocates the most humane action, which is often euthanasia, done under the direction of CWS by persons with appropriate permits (CWS 1999, Internet site).

10.8.5 Follow-up Monitoring As noted earlier, Northern Gateway has invited coastal First Nations to participate in the establishment and monitoring of environmental transects and sites that would include monitoring of marine birds, particularly in coastal areas. Information from these transects and sites would be used to characterize baseline conditions. In the event of a spill, the health and recovery of affected bird populations would be monitored. A follow- up program for marine birds may be implemented, as appropriate, to: identify the number of birds affected by direct oiling and the number of associated mortalities assess effects of hydrocarbons on feeding habitat and prey (a follow-up program for marine fish, including important prey species such as Pacific herring, will also be carried out; see Section 10.7) determine contaminant levels (e.g., PAHs) in preferred prey species (e.g., non-lethal sampling of liver tissue from captured birds) assess environmental health risks to marine birds monitor the recovery of the local marine bird population through recruitment The follow-up program should continue until recovery of the populations has been documented and contaminant levels in prey and key habitats are reduced to baseline levels.

April 2010 FINAL Page 10-51

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.9 Marine Mammals and Turtles Marine mammals are a valued social and cultural resource in British Columbia. Though evidence regarding the short-term vulnerability of marine mammals to hydrocarbons is not conclusive (as reviewed below), Northern Gateway is committed to integrating knowledge about marine mammals and their response to hydrocarbon exposure appropriately into protection and mitigation planning. DFO will be the lead agency responsible for mandating mitigation of effects and treatment of affected marine mammals. Marine mammals fall under the same category as fish in the federal Fisheries Act, which provides for the protection of fish and fish habitat. The Fisheries Act stipulates, ―no person may carry out any work or undertaking that results in the harmful alteration, disruption or destruction of fish habitat‖ unless authorized by the Minister of Fisheries and Oceans. Species at risk are protected under the SARA, which applies to federal lands and protects all wildlife species listed as at risk and their critical habitat. Once a species is listed on Schedule 1, protection and recovery measures are developed and implemented. If added to Schedule 1, a species‘ populations and ‗critical habitats‘ are afforded legal protection. A list of species of marine mammals in the CCAA that are listed on Schedule 1 of SARA is found in Appendix B. The marine mammal regulations under the Fisheries Act apply to management and control of fishing for marine mammals in Canadian waters. The regulations state, ―no person shall disturb a marine mammal except when fishing for marine mammals under the authority of these regulations‖. The Marine Mammal Regulations are under review and a draft of proposed amendments was released in March 2005 that largely address management of tourism-related activities (DFO 2005, 2008c, Internet sites). Sea turtles (Order Testudines, Suborder Cryptodira) are also discussed here. The leatherback is the only marine turtle listed under SARA.

10.9.1 Baseline Conditions Marine mammals and sea turtles are found throughout the OWA and CCAA, from far up coastal inlets to well offshore. There are large knowledge gaps concerning abundance, distribution and critical habitat of many species, particularly for those that are rare or that inhabit remote, seldom visited areas. Ten marine mammal species found in the assessment area are currently protected under SARA, including blue whales, fin whales, grey whales, harbour porpoises, humpback whales, killer whales, North Pacific right whales, sei whales, sea otters and Steller sea lions. Marine mammals found within the assessment area belong to two orders. The first, Cetacea, encompasses whales, dolphins, and porpoises, and can be further broken down into two suborders: Mysticeti (baleen whales) and Odontoceti (toothed whales). The second, Carnivora, includes the Family Mustelidae, which includes sea otters, and the Suborder Pinnipedia, of which members of the families Phocidae (true seals) and Otariidae (sea lions) occur in the OWA and CCAA. Several species of marine mammals range within the CCAA. Because of the large distances that marine mammals travel, baseline information for the entire CCAA (Kitimat Arm, Douglas Channel and the north and south approaches to Douglas Channel from the open Pacific), is relevant to the assessment for the OWA.

Page 10-52 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Northern resident killer whales, humpback whales and Steller sea lions are assessed for effects from exposure to hydrocarbons. It is assumed that species of the same taxonomic group experience similar effects associated with hydrocarbons (i.e., that information provided for killer whales would apply to white-sided porpoise, as they are both in the Suborder Odontoceti) (Evans and Raga 2001). In addition to these three selected species, potential effects from hydrocarbons on other marine mammals present in the CCAA (i.e., transient killer whales and harbour seals) are also discussed. There is limited knowledge of habitat use along British Columbia‘s north-central coast for most cetaceans. Some species use the area year-round, whereas others use it as part of annual migration routes or summer feeding grounds. Identification of IAs for cetaceans is complicated by a lack of data about calving grounds and social understanding for many species (Hall 2008), as well as the wide annual ranges and large uncertainties about abundance of some populations. Some information about habitat use comes from sources for which the accuracy and level of effort are not known (Lucas et al. 2007), for example, those based on historic commercial whaling statistics or general sightings. A number of marine mammal and turtle species range within the OWA (Figure 10-4; Table 10-5). Because of the large distances that marine mammals can travel, baseline information for the entire OWA, including the CCAA, is relevant to the assessment. Figure 10-4 is a compilation of IAs for certain key marine mammals and sea turtles in British Columbia identified by DFO (Lucas et al. 2007) and CRIMS (ILMB 2009b, Internet site). The Important Marine Mammal Habitat on this figure (orange layer) is a composite of IAs (breeding, feeding, or migrating) for blue whale, sperm whale, grey whale, sei whale, fin whale, northern fur seal, leatherback turtle (Lucas et al. 2007), California sea lion and northern fur seal (ILMB 2009b, Internet site). Each species layer was overlaid so that areas of importance to multiple species show up darker on the figure. Killer whale critical habitat and potential critical habitat, known Steller sea lion rookeries, major humpback whale concentrations and sea otter distribution were highlighted separately. Inlets and coastal areas are under-represented in PNCIMA source layers. Only 10 of the 37 identified marine mammals and turtles have mapped distributions that were available, thus there is a large knowledge gap for this species group. A general list of marine mammal species that might be present in the OWA is located in Appendix B, Table B-2.

April 2010 FINAL Page 10-53

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Figure 10-4 Summary of Biologically Important Areas for Marine Mammals in the OWA

Page 10-54 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-5 Marine Mammals and Turtles Associated with Ecosections in the OWA and CCAA

Marine Ecosection Areas and Species of Importance Continental During summer, strong upwelling makes this an area of high primary productivity and a Slope very important feeding area for marine mammals, particularly baleen and toothed whales. Historical whaling data and current sighting records suggest this is key habitat, especially for endangered species such as the blue whale, sei whale, and leatherback turtle. The continental slope is also an important migratory corridor for many species such as northern fur seal and grey whale, and is one of the most likely habitats for beaked whales. Dixon Dixon Entrance, particularly the southern half during summer and early fall, is a known Entrance area of concentration of humpback whales and killer whales. For northern resident killer whales, portions of Dixon Entrance might be potential critical habitat (DFO 2008c). Fin and sei whale might also use the western waters; grey whales are thought to pass though during their annual migration, and large numbers of porpoise are commonly seen in this area. Hecate Strait The southern portion is an important feeding area for marine mammals, particularly humpback whales from late spring through early fall. The northeastern section has also been identified as feeding habitat for northern fur seals (Lucas et al. 2007). One of only three Steller sea lion rookeries in British Columbia is in eastern Hecate Strait. Rare sightings of Cuvier’s beaked whales, Risso’s dolphins, and leatherback turtles have also been made in the waters of Hecate Strait, and grey whales might use these waters as a migratory corridor. Queen The waters of Queen Charlotte Sound that border the continental slope provide Charlotte important breeding habitat for two of only three Steller sea lion rookeries in British Sound Columbia waters. The eastern waters near Goose Island are also established sea otter habitat, and small numbers of fin and humpback whales forage in these waters. Queen The southeastern portion of Queen Charlotte Strait has been identified as critical habitat Charlotte for northern resident killer whales, which are present from July to November (DFO Strait 2008c, Internet site). Summer resident grey whales and humpback whales also feed in this area. North Coast The semi-protected waters and inlets of the north coast fjords are potentially critical Fjords habitat for killer whales and humpback whales. This area also provides haulout sites and foraging areas for a large proportion of British Columbia’s harbour seal population. Vancouver This highly productive area provides important feeding habitat to many species of Island Shelf marine mammals, including killer whales, Steller sea lions (with the largest breeding rookery in British Columbia in the northern portion), and large concentrations of humpback whales. A small portion of the north Pacific grey whale population is resident in British Columbia throughout the summer to feed in these waters. A large proportion of the leatherback and green sea turtle sightings are recorded here and it is also one of only two areas in British Columbia where sea otters have established.

April 2010 FINAL Page 10-55

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.9.1.1 Marine Mammal Vulnerability by Key Life Stages

Feeding areas The coastal areas and upwelling zones of the OWA are rich in concentrated phytoplankton biomass, and are high in primary productivity. In general, regions of high productivity provide important feeding habitat for most species of marine mammal and sea turtle. Many species also depend on the specific physiographic nature of local areas to aid in concentrating their prey. The vast majority of the OWA is considered an important feeding area to at least one species of marine mammal or turtle at some point during the year. Abundance and concentrations of marine mammals are generally highest during spring, summer and fall, but for many species there is very little known about abundance, distribution, or habitat usage during the winter months.

Breeding areas Many species of marine mammal and sea turtle breed in tropical and sub-tropical waters and migrate to British Columbia‘s productive waters to feed. Among species that breed in the OWA, some require particular breeding sites and others breed throughout the region. For example, Steller sea lions breed in only three areas in British Columbia, while harbour seals breed throughout the coastal areas. Breeding behaviour is rarely observed in many toothed or baleen whales and it is possible that some animals, such as the sperm whale, might breed within the OWA.

Migratory corridors Little is known about exact migratory corridors for many marine mammal and turtle species that transit the OWA. However, the continental slope and Vancouver Island shelf likely provide important migratory habitat, as might the subarctic and transitional Pacific. The waters of Queen Charlotte Sound, Hecate Strait, the North Coast fjords and Dixon Entrance might also provide an ‗inside route‘ for animals migrating north and south along the coast. A more detailed understanding of migratory corridors in this region is needed.

Haulout sites and rookeries Pinnipeds (seals and sea lions) require both terrestrial and marine habitat. Although they forage for food in the marine environment, pinnipeds require access to haulout sites for rest and thermal regulation. They also go ashore at places known as rookeries, where they mate, give birth and nurse their young. The ability to access marine resources close to terrestrial haulout sites is very important to these species. Breeding habitats for pinnipeds are also particularly vulnerable to disturbance.

10.9.1.2 Killer Whales The northern resident (NR) killer whale is relatively well studied, of conservation concern (COSEWIC 2008, Internet site; BC MoE 2009b, Internet site) and known to frequent the CCAA (Ford 2005, pers. comm.; Marine Mammals TDR). As top predators, NR killer whales rely on lower trophic levels and can be considered an indicator of ecosystem health. Their communication methods, hearing and physiology

Page 10-56 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment are generally similar to those of other toothed whales, although transient whales do show some marked differences. The northern resident killer whale is listed as threatened under COSEWIC and on Schedule 1 of SARA because of its small population, low potential growth rate and presence of anthropogenic threats that might limit its potential recovery or lead to further population decline (COSEWIC 2008, Internet site). Detailed understanding of critical habitat for NR killer whales is recognized by DFO as a data gap, and thus a national recovery strategy now exists under Schedule 1. In 2006, the northern resident killer whale population numbered 244 individuals (Ford 2008, pers. comm.). In 2007, the R1 pod had four matrilines (R2, R5, R17 and R7) made up of 34 individuals (Ellis et al. 2007). The W1 pod had only one matriline (W3) and consisted of three individuals. Thus, as of 2007, the R clan consisted of 37 individuals, 11 of which are known to be females that have previously given birth, as well as two calves born in 2006 (Ellis et al. 2007). In 2007, the other two clans, A and G, had 120 and 81 individuals, respectively (Ellis et al. 2007). As not all members of the northern resident population are seen each year, population data are considered less precise than for the better-studied southern resident population (DFO 2008c, Internet site). Studies conducted for Northern Gateway and by Ford (2006) suggest the outer regions of the CCAA (Caamaño Sound, Whale and Squally Channel, and possibly Wright Sound) are important northern resident killer whale habitat, especially from April to June (Clarke and Jamieson 2006; DFO 2008c, Internet site). Information on abundance (e.g., peak numbers or proportion of the population) of northern resident killer whales in this region was not located. Available data suggest whales from pods A01, A04 and A05 are much more commonly observed than other pods (Ford 2006). The general spatial and temporal distribution of the northern resident population is largely driven by prey populations (Clarke and Jamieson 2006), primarily chinook salmon from May to September, with some individuals switching to chum salmon in October. From December to April, the population's range is extended but is not believed to leave the north-central coast region. It might be possible that the population moves north to southeast Alaska. Field surveys done by Northern Gateway identified northern residents in the outer region of the CCAA during summer only (Camaaño Sound; I31 pod [G clan], C1 pod [A clan], June and July of 2006 and 2009, respectively) and in less confined regions (e.g., north of Wright Sound). Opportunistic data collected by the BCCSN (which is not corrected for effort) in the CCAA (1985 to 2009) demonstrates killer whale use of this region, but does not differentiate between residents and transient killer whales (Marine Mammals TDR).

10.9.1.3 Humpback Whales The north Pacific humpback whale is a baleen whale that occurs in the CCAA and has conservation status. Humpback whales in the North Pacific are listed as ‗of special concern‘ on British Columbia‘s Blue List (BC MoE 2009c, Internet site) and as threatened on Schedule 1 of SARA (COSEWIC 2003a, Internet site). As of September 2009, a formal recovery strategy under SARA is being developed. As such, critical habitat (as defined by SARA) of this species has not yet been formally designated. However, ‗IAs‘ for humpback whales, ranked as high, include many of the inland waterways (Douglas Channel, Lewis Passage, Campania Sound) of the CCAA (Clarke and Jamieson 2006).

April 2010 FINAL Page 10-57

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Humpback whales are found in North Pacific waters throughout the year. They are most abundant from May and October, when they are believed mainly to be feeding (Chittleborough 1965), although recent observations suggest the period might range from June to November (Cetacealab 2009, Internet site). Seasonal field surveys carried out by Northern Gateway from 2005 to 2009 within the CCAA suggest humpback whales might be absent in winter (February) and present as early as April, with peak abundance in fall (Marine Mammals TDR). Humpback whales were primarily observed in the outer regions of the CCAA (Wright, Squally, Campania, Caamaño, Napean Sounds, Lewis Passage and Whale Channel) and less so in Douglas Channel and Kitimat Arm. Data obtained from the BCCSN (not corrected for effort; Marine Mammals TDR) support the predominant use of outer regions of the CCAA and also indicates sightings near Coste Island and Kitamaat Village. Opportunistic sightings in Kitimat Arm in December 2005 (Wheeler 2006, pers. comm.) also indicate year-round presence within the CCAA. Humpback whales in British Columbia display extremely strong site fidelity to very localized feeding grounds (Rambeau 2008), as reported elsewhere in the North Pacific (Darling and McSweeney 1985; Baker et al. 1986; Craig and Herman 1997; Mizroch et al. 2004) and North Atlantic (Clapham et al. 1993). Site fidelity is thought to be a learned, maternally directed behaviour, with whales returning to the feeding areas they first visited as calves with their mothers (Martin et al. 1984; Baker et al. 1987; Clapham and Mayo 1987). Abundance estimates of north Pacific humpback whales in the CCAA are not available at this time, so it is not possible to quantify the importance of this habitat to the overall population. Given the high site fidelity shown by feeding humpbacks, the CCAA is likely important to the individuals that commonly feed there.

10.9.1.4 Steller Sea Lions The Steller sea lion is a pinniped (seals and sea lions) and is a species of special concern on Schedule 1 of SARA (COSEWIC 2003b, Internet site). Steller sea lions occur year-round within Kitimat Arm (Marine Mammals TDR) and are reasonably well studied. Steller sea lions are high trophic level consumers and can be considered indicators of ecosystem health. They are species of special concern primarily because of the small number of breeding sites and their susceptibility to human disturbance and oil spills. A formal management plan has not been developed to date. Steller sea lions are year-round residents in the OWA. When they are not at haul-out sites along the coast, they are generally distributed along the continental shelf and shelf break, within 60 km of land in summer, and 200 km in winter (Heise et al. 2007). Breeding occurs from May to August on only three breeding areas within the OWA (off the northeastern tip of Vancouver Island, the southern tip of Haida Gwaii and the northern mainland coast, west of Banks Island). Steller sea lions are widespread throughout coastal waters of British Columbia and were observed within the CCAA during all marine mammal field studies carried out by Northern Gateway. They are present year round (Marine Mammals TDR), although sightings in the Kitimat Arm region were documented in spring only; opportunistic sightings also suggest they are present in winter (sighting of an individual at Bish Point in December 2005, Anderson 2005 pers. comm.). Animals were consistently observed at Ashdown Island (a known haulout site) and at Isnor Rocks (outside the CCAA).

Page 10-58 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Rookeries are used from May through August for birthing, breeding and nursing pups. At this time, the majority of the population are at the rookeries (and possibly not in the CCAA). Danger Rocks, west of Banks Island, in the OWA, is the closest rookery to the Kitimat Terminal (COSEWIC 2003b, Internet site). Steller sea lions show a high site fidelity to their birthing and mating sites (Raum-Suryan et al. 2002). Year-round haulouts are used continuously, and winter haulouts are used primarily during the non- breeding season (Bigg 1985). Three winter haulout sites are located close to, but outside of, the CCAA (Warrior Rocks, Bonilla Island and Isnor Rocks; COSEWIC 2003b, Internet site). Ashdown Island, 75 km from Kitimat Arm off the southern tip of Gil Island (in the CCAA), is categorized as a major winter haulout site (COSEWIC 2003b, Internet site). Steller sea lions are known to travel long distances from haulouts in winter.

10.9.1.5 Other Species

Transient Killer Whales The West Coast transient killer whale population is listed as threatened under COSEWIC, based on the November 2008 assessment. This population is listed on Schedule 1 of SARA due to the very small number of mature individuals (approximately 122) and threats from high levels of anthropogenic contaminants and disturbance (COSEWIC 2008, Internet site). Prey depletion is also a threat to the population (COSEWIC 2008, Internet site), although the prey base of pinnipeds and cetaceans is considered stable or increasing. This population of killer whales occurs throughout coastal waters of British Columbia (COSEWIC 2008, Internet site). Field studies in the CCAA suggest transient killer whales predate on porpoises, and possibly seals, in winter and early spring. Transients were identified in Kitimat Arm (February 2009), Douglas Channel (April 2009) and Principe Channel (February 2009) and averaged three individuals per group (Marine Mammals TDR). Transient killer whales prey on a variety of marine mammals, including harbour seals, harbour porpoises, Dall‘s porpoises and Steller sea lions. They are also known to consume California sea lions, Pacific white-sided dolphins, grey and minke whales, and less commonly, river otters, northern elephant seals and seabirds (COSEWIC 2008, Internet site; Baird and Dill 1996; Ford et al. 1998, 2005).

Offshore Killer Whales Offshore killer whales are the least known of the British Columbia killer whale ecotypes. Sightings of offshore killer whales are fairly infrequent, although over 250 individuals have been identified within the OWA. They are suspected of spending the majority of their time offshore.

Sperm Whale Male sperm whales are found along the continental slope and offshore waters of the OWA. Although not common, they can be seen regularly off the west side of Haida Gwaii. It has also been suggested that they might calve in offshore British Columbia waters (Gregr and Trites 2001).

April 2010 FINAL Page 10-59

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Pacific White-Sided Dolphin Pacific white-sided dolphins are abundant, permanent residents of the pelagic waters of the OWA and regular visitors inshore. These dolphins are inquisitive and active, and will frequently approach vessels and interact with other species. They feed on fish and squid, often in small groups when inshore. They frequently travel offshore in much larger aggregations of a few hundred to a few thousand individuals.

Dall’s Porpoise Dall‘s porpoises are common offshore and in deep inshore waters close to land, throughout the OWA. Little is known about the migration of this species, but they are found in small groups within the OWA in all months of the year, and are preyed upon by transient killer whales.

Harbour Porpoise Harbour porpoises are particularly sensitive to human activities and are displaced by underwater noise. They are found in coastal shallow waters (less than 200 m) throughout the OWA, particularly in regions where physiographic features aid in concentrating prey (small fish and squid). Calving takes place between May and September.

Risso’s Dolphin and Northern Rightwhale Dolphin Mainly found in deep waters over the continental shelf, the OWA is at the northern limit of the normal distribution for these species. However, recently there have been fairly regular sightings of these dolphins along the continental slope and in western Hecate Strait.

Beaked Whales - Baird’s, Stejneger’s, Hubbs’, and Cuvier’s Beaked whales are pelagic species that inhabit deep, offshore waters and are found only rarely in the OWA. Sightings are thought to be uncommon, primarily because these are elusive, deep-diving animals found far from land and human presence. Distribution is known mostly from stranded specimens, although in recent years there have been sightings of Cuvier‘s beaked whales in Hecate Strait and Baird‘s beaked whales along the continental slope.

Short-Finned Pilot Whale, Pygmy Sperm Whale, Dwarf Sperm Whale, False Killer Whale, Striped Dolphin, and Common Dolphin (Long-Beaked and Short-Beaked) These species are mainly tropical, sub-tropical or warm temperate in their distribution. They occasionally follow warm waters beyond normal range and have rarely been seen in the OWA. Little is known about their abundance or distribution in the area due to their tendency to use deep water areas.

Grey Whale The eastern North Pacific population of grey whale passes through the OWA once in spring (February to June), as the group travels north to its Arctic feeding grounds, and again in winter (December to January) as individuals return south to breed. There is also a small sub-population of summer resident whales that spends eight to nine months feeding off Vancouver Island and the central coast.

Page 10-60 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Minke Whale The smallest of the North Pacific baleen whales, minke whales are shy, inconspicuous, mostly solitary animals. They prey mainly on small schooling fish in coastal inlets and nearshore areas; however, very little is known about the distribution or movements of North Pacific minke whales.

Fin Whale Fin whales were hunted extensively in the OWA and, although no longer common or widespread, they are regularly seen in Dixon Entrance and along the continental slope, and have been seen in Hecate Strait, Queen Charlotte Sound and inlets of the North Coast fjords. They are present in the OWA year round, although historically, commercial catches were highest in July and August. Fin whales feed primarily on euphausiids, as well as on copepods, squid, and fish.

Blue Whale, Sei Whale and North Pacific Right Whale Although historically common to the OWA, commercial hunting dramatically reduced the population size of these whales, leaving all three species endangered in British Columbia waters. Sightings of blue whales are relatively rare, although there have been several recent sightings along the continental slope, particularly in August. Hydrophones have detected blue whale calls far offshore between August and March. Sei whales are a pelagic species that were historically abundant in continental slope waters of the OWA; however, in recent years, there have only been two sightings of sei whales, one off southeast Moresby Island and one in western Dixon Entrance. The North Pacific right whale is the most endangered of the world‘s whales, and there have been no sightings within the OWA in the last 60 years.

Harbour Seal Harbour seals are not considered a species of conservation concern by COSEWIC, the province or under SARA. They are common in coastal areas, inlets and estuaries throughout British Columbia, demonstrate site fidelity and move locally during feeding, breeding and moulting (Olesiuk et al. 1990). They are mobile, opportunistic predators that consume seasonally or locally abundant prey, which include commercial and non-commercial fish species (Olesiuk 1993; NMFS 1997, Internet site; Browne et al. 2002). Harbour seals are considered non-migratory (Bigg 1981). Aerial censuses indicate approximately 100,000 harbour seals in British Columbia waters (Olesiuk 1999). Abundance estimates specific to the CCAA region (eastern Queen Charlotte Sound, central coast) were not located. Harbour seals were found in all seasons throughout the CCAA (Marine Mammals TDR), more commonly observed along coastlines (during bird surveys) than in open water (middle of channels). Vessel-based surveys of Gilttoyees Inlet and Miskatla Inlet and the mouth of Foch Lagoon (western Douglas Channel) regularly reported harbour seals, often in greater numbers than elsewhere in the CCAA (Marine Mammals TDR).

April 2010 FINAL Page 10-61

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Northern Fur Seal and Northern Elephant Seal Most of the northern fur seals in British Columbia are believed to be from populations that breed in the eastern Bering Sea (Pribilof Islands and Bogoslof Island) in summer, although some likely come from California and Asia. Individuals migrate through British Columbia in the early winter and again in late spring. Because fur seals are predominantly pelagic, offshore waters along the continental slope and shelf break provide important migratory habitat. However, northern fur seals are also known to feed in Queen Charlotte Sound, and there is a large overwintering population that feeds feed on small schooling fishes and squid in Hecate Strait and the North Coast fjords from January through April (Heise et al. 2007). Northern elephant seals breed in California and Mexico from December to March, after which they disperse from the rookeries to forage on the continental slope. Adult males and pups tend to move north along the coast as far as Alaska, though they are relatively uncommon in British Columbia waters.

Sea Otter Sea otters spend their life on the water surface. They are listed as of special concern under COSEWIC and on Schedule 1 of SARA. Since the species was extirpated and subsequently re-introduced to British Columbia, it has repopulated 25 to 33% of its historical range; however, the numbers of animals are not yet stable (less than 3,500) and still require careful monitoring (COSEWIC 2007, Internet site). Oils spills are considered the greatest threat to sea otters (Government of Canada 2009, Internet site), and their proximity to existing oil tanker approaches makes them particularly vulnerable (COSEWIC 2007, Internet site). Sea otters are not known to be present in the CCAA, as reported in field studies (Marine Mammals TDR) but could recolonize the area.

Leatherback Sea Turtle The leatherback is the largest and most wide-ranging extant reptile and is critically endangered. It is the most commonly seen sea turtle in British Columbia. Leatherbacks breed in tropical and subtropical waters and are most likely to be found in the OWA between June and November. They feed primarily on jellyfish and tend to be more locally abundant where jellyfish densities are high. Sightings are infrequent but span the entire British Columbia coast all the way offshore to the boundary of the Canadian Exclusive Economic Zone (EEZ) (Pacific Leatherback Turtle Recovery Team 2006). The Vancouver Island shelf and continental slope appear to be IAs for leatherbacks. Key knowledge gaps include distribution and seasonal movements as well as habitat requirements and feeding behaviour.

Green Sea Turtle Green sea turtles are normally a tropical species, but occasionally follow warm ocean currents northwards. They and leatherbacks are the only sea turtles regularly seen in British Columbia waters. However, as water temperatures in British Columbia are close to the minimum thermal tolerance of this species, occurrence in the OWA might be mostly accidental. Live sightings and fresh carcasses have been recorded in the OWA mainly between September and December on the west coast of Vancouver Island or the east coast of Graham Island (McAlpine et al. 2007).

Page 10-62 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Loggerhead Sea Turtle and Olive Ridley Sea Turtle Although loggerhead and Olive Ridley sea turtles have not previously been sighted in the OWA, there have been sightings in Washington and Alaska, suggesting they might be occasional visitors to British Columbia waters (McAlpine et al. 2007).

10.9.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Marine Mammals This discussion focuses on short-term acute effects for marine mammals. Acute effects (i.e., under two years) of hydrocarbon exposure on marine mammals as a group are difficult to quantify and predict, largely owing to the many variables involved (e.g., location and volume of the spill, degree of product weathering, chemical constituents, time of year, marine mammal behaviour, ecology and distribution). Marine mammals tend to be long lived and range widely. Many feed at the top of the food chain, which, over a lifetime, exposes them to many influences on health and poses a challenge in making direct links with individual events. Effects could be short or long term and result in population- or community-level changes, depending on some or all of the factors listed above. The few studies of potential effects of hydrocarbons on marine mammals are either based on captive animals (i.e., so might not adequately represent field conditions and behaviours), or on sightings made opportunistically during a spill (i.e., so might lack access to robust data on abundance and distribution for many years prior to, and after the spill). Wursig (1990) provides an overview of potential effects of hydrocarbons on marine mammals. Species that occur in restricted areas (such as the CCAA) for at least part of their lives are more likely to encounter oil than those that range widely (e.g., feeding humpback whales, porpoises, dolphins). Cetaceans that feed at the surface or bottom are more likely to contact oil than those that generally feed in the water column. As a group, baleen whales appear to be more vulnerable in view of their small population size, general feeding strategies and abundance in selected or restricted habitats for feeding and reproduction. Known acute effects of hydrocarbons on marine mammals fall into four general categories: changes in distribution physical oiling immediate ingestion of oil inhalation of oil vapour These are discussed below for toothed whales, baleen whales and sea lions. Cetaceans would be exposed to the hydrocarbons mainly while the oil is present at the water surface. Seals, sea lions and otters might be exposed to oil on shorelines. As discussed below for toothed whales, there is inconclusive information about the species‘ ability to detect and avoid hydrocarbons. French-McCay et al. (2004) undertook an oil spill modelling exercise to predict potential effects to wildlife and marine mammals. As a basis for this model, it is assumed the probability for a marine mammal to encounter surface oil is related to the percentage of time an animal spends on the water or shoreline. In validating their model, French-McCay et al. (2004) found reasonable predictive accuracy for sea otters and seals affected by EVOS and underestimation of wildlife mortality where an unusually high abundance of species occurred or where abundance data were not available.

April 2010 FINAL Page 10-63

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Long-term effects were recognized for one pod of killer whales in Prince William Sound, but not for the entire population of the Sound 17 years after EVOS (Harwell and Gentile 2006). Based on the precautionary principle and identified effects on individuals, mitigation to protect marine mammals are a high priority for Northern Gateway.

10.9.2.1 Toothed Whales (Killer Whales) Studies examining a whale‘s ability to detect and avoid hydrocarbons are inconclusive. Smith et al. (1983) found that captive bottlenose dolphins initially avoided oiled areas in the tanks but eventually contacted the oil. Geraci and St. Aubin (1985) suggest that captive bottlenose dolphins rely on vision to detect thick oil and on tactile response for avoidance. Geraci (1990) examined historical interactions and effects on whales and dolphins from oil and noted several occasions where animals swam through a spill without any apparent effects. They noted that adverse effects on cetaceans, such as sickness, stranding or mortality, tended to be associated with crude or bunker C oil. Scholz et al. (1992) suggested that cetaceans are able to detect and avoid oil and other hydrocarbons. Observations on dolphins suggested they detected surface oil but not sheen oil and did not avoid traveling through it (Smultea and Wursig 1995). Killer whales were observed swimming through oiled areas following EVOS, with no indication of avoidance or erratic behaviour (Loughlin 1994). Dahlheim and Matkin‘s (1994) initial analysis of the effects on killer whales suggested no definitive links to mortality. However, their later studies indicated that most of the AB resident killer whale pod (AB pod) was documented swimming through heavy sheen in Prince William Sound six days after the spill and that thirteen individuals (eight juveniles and five reproductive-age females), or 33% of the AB pod, died by the following summer. Furthermore, 41% of the AT1 group of transient killer whales had also died (Matkin et al. 2008). Matkin et al. (2008) postulated that seven missing individuals of the AB pod died as a result of inhaling oil or oil vapour and cited Griffiths et al. (1987; not reviewed here) as an example of respiratory stress in cetaceans (seven dolphins associated with a spill in the Arabian Sea). Cetaceans must breathe at the surface, where large quantities of mono-aromatic hydrocarbons evaporate off (Matkin et al. 2008). Depending on amount of hydrocarbon vapour at the surface and duration of exposure, effects of vapour inhalation range from mild irritation to sudden death (Geraci and St. Aubin 1988). No studies demonstrating effects of physical oiling (e.g., eye irritation) and direct ingestion of oil by toothed whales were located. This might be partly because cetaceans typically feed underwater (not in contact with hydrocarbons on the water surface) and because carcasses typically sink and are not available immediately following a lethal exposure. After extensive studies, Geraci (1990) concluded that direct surface fouling poses little, if any, risk to the thermoregulatory capabilities of whales. A diet of contaminated prey would not account for acute mortalities immediately following a spill (Matkin et al. 2008). Both resident and transient killer whales are susceptible to direct ingestion of hydrocarbons, despite their differing prey base (residents consume primarily fish and transients consume primarily mammals). While the transient killer whales might ingest pinnipeds or other small cetaceans, they are highly mobile and might move away from a contaminated area in search of prey. Matkin et al. (2008) noted that oiled seals were unusually lethargic and approachable, so might have been easier prey for transient killer whales. While resident killer whales are also highly mobile, they might be more affected

Page 10-64 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment by the loss of opportunity to hunt for prey during important times of the year (such as the spawning chinook salmon). Following EVOS, five harbour porpoises were found dead in Prince William Sound. The three animals that were autopsied showed elevated levels of hydrocarbons in blubber and liver tissues. However, levels of assimilated oil were not high enough to determine with certainty that the animals died from exposure to crude oil (Dalheim and Matkin 1994). The deaths might have been the result of a combination of factors, including acute toxicity of crude oil, increased energy expenditure from epidermal fouling, reduced prey abundance and increased susceptibility to parasitism or disease (Albers and Loughlin 2003).

10.9.2.2 Baleen Whales (Humpback Whales) Short-term acute effects (physical oiling, ingestion of oil, inhalation of vapours) are presumably similar for humpback whales and killer whales. However, humpback whales feed using baleen to filter prey, so their ability to feed might be greatly affected through fouling of baleen plates, which might decrease filtration capability and water flow (Haebler 1994). In laboratory studies using humpback baleen, exposure to heavy crude oils doubled the resistance to water flow (required for filtering seawater). Other studies determined that over 70% of oil that adhered to baleen would be lost within 30 minutes and 95% would be lost within 24 hours of rinsing with salt water (Geraci and St. Aubin 1988). Combined evidence of baleen fouling leads to the conclusion that heavy oil could interfere with feeding efficiency of baleen whales for at least several days and whales that remained in a heavily oiled area, where the rate of baleen fouling exceeds the rate of cleaning, would be most vulnerable (Geraci and St. Aubin 1988). It is not known whether humpbacks avoid oil spills (NMFS 1991, Internet site). Photo-identification studies on humpback whales following EVOS suggest these whales were not severely affected (Loughlin et al. 1996). The studies also showed that short-term changes were likely restricted to temporary avoidance of areas within Prince Williams Sound, although interpretation of results was complicated by factors such as the wide distribution of the North Pacific humpback whales and unequal survey effort in Prince Williams Sound between pre-spill and post-spill years (von Ziegesar et al. 1994). The number of dead grey whales (20) found in the Gulf of Alaska during 1989 (year of EVOS) exceeded the total number of stranded whales reported for the entire 1975 to 1987 period, although lack of pre-spill data precludes identification of definitive links with the oil spill (Loughlin 1994).

10.9.2.3 Pinnipeds (Steller Sea Lions) The main effect of hydrocarbons on Steller sea lions is likely to be from direct contact with oil when the source of the spill is near their haulout sites; lesser effects might include absorption through skin, incidental ingestion of oil directly or from prey, exposure to vapours and fouling of fur (COSEWIC 2003b, Internet site). Sea lions primarily use blubber to thermoregulate, rather than relying solely on their fur, so oiled fur would not interfere with thermoregulation (Geraci and St. Aubin 1988; COSEWIC 2003b, Internet site). Steller sea lions near oil spills have been observed with oil deposits in their throats or around their lips, jaw and neck (Calkins and Pitcher 1982). Studies conducted following EVOS (e.g., Calkins et al. 1994)

April 2010 FINAL Page 10-65

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment indicate that sea lions were seen swimming in or near surface oil. Those that were near or contacted oil could have become contaminated through inhalation, absorption through the skin, or ingestion either directly or by consuming contaminated prey; however, none of the data presented and analyzed regarding sea lions provided conclusive evidence of an effect on this species from EVOS (Calkins et al. 1994). There was no evidence of harbour seals avoiding oiled water or shorelines after EVOS. Some animals were oiled, although few carcasses were found, perhaps due to the tendency for carcasses to sink. In the early weeks after the spill, seals surfaced in and swam through floating oil while feeding and moving to and from haulout sites (e.g., Lowry et al. 1994). In heavily oiled areas, many seal pups became oiled and newborn pups were sometimes seen with oil around their nose. Oil fouling might affect seal locomotion, with heavy oiling causing flippers to stick to the body; contact with oil also reduces the insulative value of hair, but in healthy seals this is not likely to be a major problem as they rely primarily on blubber for insulation. Contact with oil can irritate or damage sensitive tissues, especially mucous membranes; some heavily oiled seals were observed having difficulty keeping their eyes open. The seals became cleaner over time if they were not repeatedly exposed to oil. Non-pups moulted their pelage in August and no further evidence of external oiling was seen within one year. Various types of skin lesions in harbour seals were probably caused by crude oil. Examination of dead oiled seals suggested the observed brain lesions were related to inhalation of toxic fumes and mortality could have resulted from behavioural disorientation, lethargy and stress response (Ott et al. 2001, Internet site). The lesions were found to affect the part of the brain responsible for sensing the environment and might have interfered with thermoregulatory functions.

10.9.2.4 Mustelids (Sea Otters) The effects of hydrocarbon exposure on sea otters are well documented. EVOS resulted in mortality of 2,650 (Garrott et al. 1993) to 3,905 sea otters (DeGange et al. 1994), and the population is not yet considered completely recovered (Monson et al. 2000). Otters rely on a layer of air trapped inside dense fur (other marine mammals have a layer of blubber) and an extremely high metabolic rate for warmth. Maintaining fur integrity and function requires constant and intensive grooming so, in the event of a spill, the most serious effect of oil will be fouling of the insulative fur. Contamination of as little as 30% of the body surface can lead to death from hypothermia or pneumonia (Costa and Kooyman 1982; Williams et al. 1988). Grooming of oiled fur can lead to ingestion of oil and inhalation of fumes, resulting in injury of lungs and other internal organs (Ralls and Siniff 1990). Sea otters aggregate in sexually segregated groups of up to 200 animals, so a spill has the potential to oil many otters at one time. Otters typically congregate near kelp beds, where oil tends to accumulate (Ralls and Siniff 1990).

10.9.3 Potential Effects of Condensate on Marine Mammals Acute effects of condensate on marine mammals are more difficult to quantify and predict than for synthetic oil and diluted bitumen exposure. Literature about condensate spills and their effects on marine mammals was not found; however, it is assumed the highly volatile condensate will evaporate quickly, with some dissolution into the water column and subsequent weathering (Section 6). Exposure would be possible during the first day or so after the spill. Hence, effects associated with physical fouling (e.g., filter feeding baleen or hair fouling) are expected to be less than for the more viscous hydrocarbons.

Page 10-66 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

However, inhalation of vapours could result in acute affects (narcosis, lung tissue damage, disorientation leading to drowning), as stated by Geraci and St. Aubin (1988) and as implicated in the reduction of populations of two groups of killer whales following EVOS (Matkin et al. 2008).

10.9.4 Mitigation Measures Vessel operations described in Section 3 and response measures described in Section 7 will be important in reducing the probability of an incident and mitigating any effects on marine mammals. Areas of important marine mammal use that could be at risk from a spill will be identified prior to operations and used to develop geographically, seasonally and species applicable plans to protect marine mammals. Northern Gateway, in cooperation with DFO and certified responders, will develop a Marine Mammal Protection Plan before operations, which will specify and guide mitigation related measures. Aboriginal organizations and affected stakeholders (e.g., research specialists, ecotourism operators) would also be consulted in the development of response plans. The plans will include a Marine Mammal Hazing Plan, a Marine Mammal Cleaning Plan, and a Marine Mammal Follow-up Plan. The hazing plans will be comprehensive and will include information such as type, quantity, location(s), maintenance and use of necessary equipment, as well as contact information for experts trained in hazing techniques, coordination of hazing measures and responsibilities and roles of relevant stakeholders and experts. General marine mammal hazing techniques include: acoustic deterrence (e.g., artificial sounds that would induce an animal to move away from an affected area, underwater playbacks of recordings from marine mammals that might either attract or repel marine mammals away from the affected area) boat traffic that generates noise and water movement helicopters that generate considerable noise and wave movement at close range fire hoses that can direct streams of water at whales (NOAA 2009b, Internet site) Seals, river otters and seas otters (if present) might be cleaned and re-introduced into the environment. A Marine Mammal Cleaning Plan will specify all aspects associated with safe cleaning of affected animals. The plan will specify cleaning products, techniques, location of cleaning activities, necessary veterinarian expertise, animal care facilities and infrastructure, and other details.

10.9.5 Follow-up Monitoring Northern Gateway has invited interested coastal Aboriginal communities (i.e., Gitga‘at, Kitkatla) to participate in or undertake independent research on marine species, including marine mammals. Similar offers will be extended to other coastal communities as engagement progresses. Information from these studies would be used to better understand the seasonal abundance and distribution of marine mammals, and characterize baseline conditions. Follow-up and monitoring will be undertaken, as appropriate, to evaluate acute effects on marine mammals (follow-up and monitoring relating to longer-term, chronic effects, is discussed in Section 13.4, Follow up and Monitoring). Details associated with the required follow-up and monitoring programns

April 2010 FINAL Page 10-67

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment would be specific to different areas of the CCAA and OWA, and will be included in the Marine Mammal Follow-up Plan. The plan will include the appropriate duration, spatial coverage, frequency or effort, survey design, roles and responsibilities of monitoring staff and required expertise and other applicable details.

10.10 Terrestrial Mammals Hydrocarbons have the potential to affect the short-term health of terrestrial animals that use shoreline habitat. With a lengthy shoreline in the CCAA and OWA, there are many areas where terrestrial animals could interact with hydrocarbons would have drifted ashore. Effects on wildlife are typically documented through toxicity or mortality (number of individuals in the species groups affected) and areal extent of habitat loss. Diluted bitumen or oil will have a greater effect than condensate, as condensate evaporates quickly and is not expected to reach intertidal or shoreline areas.

10.10.1 Baseline Shoreline habitats present are varied and range from gently sloping sand and gravel beaches and mud flats to high vertical rock cliffs. Between these extremes are rugged, rocky shorelines, which are predominant in the CCAA. The intertidal zone is highly productive, with many invertebrates that provide prey for terrestrial birds and mammals. Carcasses (e.g., fish, birds, seals) from the marine environment are often found in the intertidal zone, attracting bird and mammal scavengers. Black-tailed deer, black bear (including the Kermode subspecies [Ursus americanus kermodei]), grizzly bear and grey wolf are the most common large mammals found in the area. Grizzly bear are typically confined to the mainland, but the other species occur on the islands. Kermode bears are known to occur on the mainland and islands of the CCAA. All these species potentially forage on accessible shoreline habitats. Black-tailed deer might forage on seaweed exposed at low tides. Black bear (including the Kermode subspecies) and grizzly bear are omnivores and opportunistic feeders, and are drawn to shoreline habitat, particularly at low tides, where invertebrates such as crabs and large worms are found under rocks. Bears and wolves often scavenge on shorelines when carcasses are exposed. North American river otters are widely distributed throughout their range in North America, and conservation initiatives have re-established some populations (Polechla 1990). River otters use freshwater and marine habitat, as well as associated shoreline habitat. They are nocturnal and consume mainly fish that are abundant, midsized and close to shore (Sefass and Polechla 2008, Internet site), amphibians (mostly frogs) and crustaceans (mainly crayfish) and might also consume small mammals, molluscs, reptiles, birds and fruits. Transient killer whales are their main natural predators in the water. Although the marine mammal surveys carried out by Northern Gateway were not designed specifically to detect otters, there were a few sightings along coastlines in Whale Channel in February and at the mouth of Foch Lagoon in April 2006. They are likely distributed throughout the CCAA, very close to coastlines.

Page 10-68 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

10.10.2 Potential Effects of Diluted Bitumen or Synthetic Oil on Terrestrial Wildlife There are two types of effects of hydrocarbons on wildlife: toxicity, potentially leading to mortality, and reduced habitat availability. Toxicity or mortality can occur directly through oiling of plumage, skin or fur, ingestion of hydrocarbons and inhalation of vapour. Mortality can also occur indirectly through habitat loss or degradation, diminished food or prey resources and shifts to less productive habitats. Both terrestrial and semi-aquatic wildlife species might be adversely affected. Species that use or regularly access the intertidal zone, such as semi-aquatic mammals (river otter) are the most vulnerable. Ungulates (e.g., deer) and other terrestrial mammals such as bears, wolves and coyote that contact, inhale or ingest the hydrocarbons could also be affected. In the unlikely event that SARA-listed species or species of management concern are affected by a spill in the CCAA or OWA, Northern Gateway will consult with provincial and federal regulators and participating Aboriginal groups, to identify possible options and management strategies. Lethal and sublethal effects from hydrocarbons have been documented for terrestrial and semi-aquatic mammals. Semi-aquatic species will be the most vulnerable due to their consistent habits of hauling out on suitable shorelines to forage and rest; juveniles are at greatest risk during spring and summer. River otters are known to swim in the sea, potentially for long distances, and to den on or near banks. Animals can be coated with hydrocarbons in the water or stranded along the shoreline, reducing the thermoregulatory capacity of the fur and leading to death from hypothermia (McEwan et al. 1974; Williams et al. 1988; Hurst et al. 1991). Exposed animals might succumb rapidly to hypothermia (Lipscomb et al. 1996) and might attempt to compensate by increasing metabolic rates (Hurst et al. 1991). They can also inhale vapour or ingest or absorb hydrocarbons through the skin. Contact can lead to fatal internal damage to lungs, livers and other organs. After cleanup, aquatic furbearers might be chronically exposed, via ingestion, to hydrocarbons that remain in the environment, and might experience habitat loss and a decline in food availability along affected shorelines. While some semi-aquatic mammals can avoid heavily oiled areas, the contamination effectively reduces the amount of habitat available over the short term. Decreased prey abundance due to contamination, avoidance of contaminated prey or feeding areas, changes in prey vulnerability, or reduced predator success could lead to changes in food availability. Effects on river otters were studied following EVOS (Bowyer et al. 2003). In 1989, river otters inhabiting heavily oiled areas had reduced body mass, elevated biomarkers in their blood, ate a less diverse diet, had larger home ranges and selected habitat differently than otters living in areas that were not heavily oiled (Bowyer et al. 2003). Data on habitat selection clearly demonstrated that otters avoided oiled shorelines in 1990, which would have helped reduce effects on individuals. Indirect and direct effects could have contributed to mortality that occurred after 1990. Based on presence of biomarkers in blood of otters from oil and non-oiled areas, recovery of the local population was considered to have occurred by 1999 (Bowyer et al. 2003). For terrestrial-based mammals such as grizzly and black bear, the effect mechanism is the same as for semi-aquatic furbearers: contamination of fur, ingestion of hydrocarbons during preening and longer-term effects from ingestion of contaminated food. Species such as grizzly and black bear—including Kermode bear (Ursus americanus kermodei)—could ingest scavenged prey along contaminated shorelines; ingestion would, at most, lead to mortality for a limited number of individuals.

April 2010 FINAL Page 10-69

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Larger terrestrial-based mammals might encounter oil or bitumen incidentally along affected shorelines and are at a low risk of exposure as they are highly mobile and adept at avoiding contaminated areas (e.g., using scent) within their relatively large home ranges. Hydrocarbon contact would be likely limited to feet and lower legs. Ungulates such as black-tailed deer feeding or moving along marine shorelines are not likely to encounter hydrocarbons in amounts that result in mortality.

10.10.3 Mitigation Measures Emergency response measures described in Section 7 will be important in reducing the probability of incidents and mitigating any effects on terrestrial mammals. Based on information presented in this assessment, traditional and local knowledge and future studies, shoreline areas that could be at risk and are accessible to, and potentially used by, terrestrial mammals will be identified prior to operations. This will focus the development of geographically, seasonally and species applicable plans to protect terrestrial mammals. Northern Gateway will engage the Province of British Columbia, the CWS and certified responders in the development of a Terrestrial Mammal Protection Plan prior to the start of operations. Aboriginal organizations and affected stakeholders will also be consulted in the development of the plans. The plans will specify mitigation measures and emergency response, and include a Terrestrial Mammal Hazing Plan, a Terrestrial Mammal Cleaning Plan and a Terrestrial Mammal Follow-up Plan. The hazing plan will be comprehensive and will include information such as type, quantity, location(s), maintenance and use of necessary equipment; contact information for experts trained in hazing techniques, coordination of hazing measures and responsibilities and roles of relevant stakeholders and experts. Oiled wildlife should be removed from the environment if for no other reason than to remove them as a continuing source of oil to other wildlife and the environment (Jessup 1998). A rehabilitation program must be established soon after the spill to be effective, so response resources should be stationed at intervals along the CCAA to facilitate rapid response. Affected wildlife are usually transported to a rehabilitation centre, where oil is manually cleaned off, commonly with liquid dishwashing detergents. Rehabilitation time largely depends on the state of individuals at the time of arrival. The British Columbia Ministry of Environment would be consulted to determine the most humane treatment for the oiled animal. Management strategies for non-oiled mammals include active or passive hazing to keep them away from oiled areas, and trapping and relocating them. Passive hazing refers to physical exclusion of individuals from an affected area. Wildlife fences or snow fences can be built around affected areas to exclude larger species (BP Exploration [Alaska] Inc. 2006, Internet site) from shoreline areas; the bottoms of the fences can be buried to prevent burrowing into the affected area by smaller mammals. For passive hazing to be successful, wildlife attractants within the affected area should be limited through proper waste management and disposal practices, including regular removal of animal carcasses deposited on shorelines (BP Exploration [Alaska] Inc. 2006, Internet site).

Page 10-70 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Active hazing refers to the use of devices to deter wildlife from entering an affected area. It is most effective in a small area. A common approach for birds, hazing has also been effective for terrestrial mammals (Gilsdorf et al. 2002; Ronconi et al. 2004; Sellers and Miller 1999, Internet site). There must be suitable habitat outside the affected area for the hazed animals to move into and it might be preferable to limit human access and activity in these areas (California Department of Fish and Game 2005, Internet site). The presence of cleanup crews might be effective in discouraging wildlife, as suggested by Sellers and Miller (1999, Internet site) for bears following EVOS. Gas exploders, lights, sirens and guard animals are the most effective techniques for mammals (Gilsdorf et al. 2002). Hazing should begin as soon as possible to avoid attracting wildlife to the area due to attractants (e.g., new scents) and, should extended hazing be necessary, techniques can be varied to reduce habituation (Gilsdorf et al. 2002). Animal- activated hazing techniques are the least likely to lead to habituation, as there is a stronger association between the animal behaviour and the deterrent (Ronconi et al. 2004). Precautions include avoiding the use of pyrotechnics, propane cannons, flares or other items that use a battery or produce a spark if there are flammable gases in the area and not using loud noises if there are sensitive non-target animals in the area. Trapping and relocating wildlife away from an affected area can be effective in preventing further oiling, especially for larger animals, including those that do not respond well to hazing (California Department of Fish and Game 2005, Internet site). Immediate relocation after capture might be important to increase survival. Ben-David et al. (2001) found that captive river otters released into the wild had a lower survival probability than non-captive river otters; survival probability was the same for captive unoiled otters and captive oiled otters. The relocation area should contain suitable habitat for the relocated species. After various oil spills, regional repopulation in the affected area was faster if the refuge was close to, and had a similar habitat type to, the original habitat (Cairns and Elliot 1987). Similar habitat is important, as the relocated animals require less time and resources for adaptation. Animals relocated to sub-optimal habitats might be more likely to return to the affected area. Territoriality and habitat carrying capacity of the species being relocated should be assessed before relocation. To increase habitat quality, human access to the relocation areas can be restricted, for example by purchasing private land considered to be beneficial for affected species and resources, as was done following EVOS (Weiner et al. 1997).

10.10.4 Follow Up and Monitoring Northern Gateway would like to engage coastal Aboriginal communities in the establishment and monitoring of permanent environmental monitoring transects and sites, including the collection of information on terrestrial wildlife that frequent shoreline and intertidal areas. Information from these monitoring programs would be used characterize baseline environmental conditions, including the quantity and quality of important harvested species. Offers to undertake a monitoring program have been extended to several coastal Aboriginal communities (Gitga‘at, Kitkatla and Lax Kw'alaams) as part of ongoing engagement activities, and similar offers will be extended to other coastal communities as engagement progresses. A follow-up program would be conducted, as appropriate, to assess the effectiveness of mitigation measures and cleanup techniques. Of particular importance would be implementation of a carcass monitoring and recovery program along affected shoreline areas to limit hydrocarbon exposure, via

April 2010 FINAL Page 10-71

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment ingestion, by species that typically scavenge remains. Local wildlife rehabilitation centres, which will be established as part of the response plan by Northern Gateway, will be asked to assess the extent of wildlife that might have been affected and cleanup and recovery success attained for the various species groups.

10.11 Summary of Effects on the Biophysical Environment Diluted bitumen or synthetic oil would likely have more extensive effects than would condensate because condensate would evaporate quickly and be dispersed in the water column, degrade, and be less likely to reach shoreline habitat. Bitumen or oil would spread over the water surface and could reach the shore within several hours, depending on location. Surface oil would be detectable for days, depending on response implementation and meteorological and oceanographic conditions. Shoreline remediation of sensitive habitats in affected areas could result in temporary (one- to two-year) loss of habitat complexity. A spill in the CCAA or OWA could affect shoreline vegetation (e.g., rockweed), intertidal invertebrates, fish that use nearshore habitat (e.g., herring spawning), waterbirds (e.g., Marbled Murrelet, Surf Scoter) during a sensitive phase such as moulting, and marine mammals (e.g., whales, seals and sea lions). However, waterbirds and species such as sea otter would be most vulnerable. Environmental effects could occur through both short-term mechanisms (acute toxicity from ingestion of oil and oiled prey, oiling of animals, inhalation of vapour) and long-term mechanisms (chronic effects from loss of habitat or uptake of contaminants). Lesser effects could occur for species that use shoreline habitat less frequently, such as scavenging bears, deer, crows or ravens that might feed on contaminated food sources. Changes in habitat quality and mortality of prey organisms could result in further effects. For example, loss of herring eggs and fry following a spring spill would lead to loss of prey and longer-term effects on salmon populations; these changes could affect human use of the fisheries resources, as well as bird and mammal predators. The extent of effects on biota would be determined by their vulnerability (i.e., how likely they are to be present in the area of a spill, exposure pathway, sensitivity of the life stage at that time of year) and the time needed for recovery of a population. For example, fish species that inhabit subsurface water would not be directly affected; however, Marbled Murrelets, which have a small population and are of conservation concern, could suffer mortality or long-term effects if individuals are present in the area of the spill. Vulnerable areas within the OWA are summarized by ecosection in Table 10-6 for invertebrates, fish and fisheries, marine birds, marine mammals and turtles and traditional uses. Emergency response plans, booms around sensitive habitat and cleanup procedures would be implemented immediately to mitigate adverse effects. Monitoring would be necessary so that recovery would continue post-cleanup.

Page 10-72 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-6 Summary of Vulnerable Areas within the OWA

Commercial and Marine Mammals Marine and Subsistence Ecosection Oceanography and Sea Turtles Marine Birds Anadromous Fish Fisheries Invertebrates Dixon Across-shelf trough Potential critical Langara Island, Area of overlapping Area of overlapping An important area for Entrance with depths mostly habitat for northern northwest of Graham important habitat use high effort in the aggregation of <300 m; surrounded resident killer whales Island, supports both for many species of commercial fisheries: Dungeness crab by low-lying coastal and humpback globally and nationally fish: eulachon, groundfish trawl, larvae and adults plains; strong whales. Important important breeding walleye, pollock, schedule II near McIntyre Bay. freshwater influence migratory corridor and areas for Ancient Pacific halibut, Pacific groundfish, sablefish The largest stock of from mainland river location for species Murrelet, Pelagic hake, herring. trap, sablefish razor clam beds is runoff drives north- such as fin, sei and Cormorant, Pigeon longline, crab trap, found from Masset to westward flowing grey whales as well Guillemot and several rockfish. Rose Spit. coastal buoyancy as porpoise. pairs of Peregrine current and estuarine- Falcon. This area is like circulation. also an important feeding ground for the threatened Marbled Murrelet. Queen High current and high Identified as critical An identified IBA of Area of overlapping Area of overlapping Supports a diversity of Charlotte relief area; very well habitat for northern global significance for important habitat use high effort in benthic organisms, Strait mixed; moderate to resident killer whales, Leach's Storm-petrel, for many species of commercial fisheries: including prawns, high salinity with and provides summer and nationally fish: Pacific cod, groundfish trawl, shrimps, sea urchins, some freshwater feeding grounds for important for Black lingcod, sablefish, shrimp trawl, Northern abalone, inputs in the inlets grey and humpback Turnstone. sole/flounder, Pacific schedule II Giant red sea and fjords. whales. hake, herring, salmon. groundfish, red sea cucumber and small urchin dive, prawn population of Olympia trap, green sea oyster. urchin, geoduck dive, crab trap, and rockfish.

April 2010 FINAL Page 10-73

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-6 Summary of Vulnerable Areas within the OWA (cont’d)

Commercial and Marine Mammals Marine and Subsistence Ecosection Oceanography and Sea Turtles Marine Birds Anadromous Fish Fisheries Invertebrates Hecate Shallow strait An identified feeding Particularly high Area of overlapping Area of overlapping The unique Strait dominated by coarse ground for humpback marine bird densities important habitat use high effort in Hexactinellid sponges bottom sediments; whales and northern in shallow waters of for many species of commercial fisheries: are found associated semi-protected water fur seals and an strait. The fish: Pacific cod, groundfish trawl, with deeper troughs; with strong tidal important area for northwestern area walleye pollock, schedule II aggregations of adult current that promotes Stellar sea lion has been identified as sablefish, groundfish, crab trap, Dungeness crab mixing; predominantly rookies; serves as a an important area for sole/flounder, herring. rockfish. associated with marine waters. migratory corridor for Pacific Loon. Known Dogbank. grey whales. Rare breeding areas for sightings of Cuvier's Red-breasted beaked whale, Merganser, Sandhill Risso's dolphin and Crane and Short- leatherback turtles billed Dowitcher have have been made. been identified adjacent to Haida Gwaii. North Coast Deep, narrow fjords Critical habitats for Fjords, particularly Area of overlapping Area of overlapping As identified for Fjords cutting into high northern resident killer around Prince Rupert, important habitat use high effort in Dungeness crab and coastal relief, very whales and are IAs for waterfowl for many species of commercial fisheries: green sea urchin protected with humpback whales, and Rhinoceros fish: inside rockfish, shrimp trawl, aggregations. Tanner restricted circulation foraging areas for Auklets. A nationally salmon, eulachon, schedule II crab, giant red sea and often stratified. harbour seals. important area for herring. groundfish, red sea cucumber and Manila Black Turnstone. urchin, prawn trap, clams also occur in geoduck dive, crab protected areas. trap, rockfish.

Page 10-74 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-6 Summary of Vulnerable Areas within the OWA (cont’d)

Commercial and Marine Mammals Marine and Subsistence Ecosection Oceanography and Sea Turtles Marine Birds Anadromous Fish Fisheries Invertebrates Vancouver Open coast with The highly productive Provides important Area of overlapping Area of overlapping High nutrient levels Island Shelf oceanic wave waters support habitat and breeding important habitat use high effort in and primary exposures; northward, feeding grounds for grounds for Marbled for many species of commercial fisheries: productivity support coast-hugging many marine Murrelet. The fish: green sturgeon, groundfish trawl, benthic communities buoyancy current due mammals (humpback southern section is lingcod, sole/flounder, shrimp trawl, and rich fishing to freshwater whales, killer whales, identified as an IBA rockfish, Pacific hake, groundfish schedule grounds. A population influence; seasonal Stellar sea lions, for Western Grebe, herring. II, red sea urchin dive, of -protected Olympia upwelling at outer small portion of the Surf Scoter, Brandt's prawn trap, geoduck oyster are found in margin. north Pacific grey Cormorant, Northern dive, crab trap, the northern part of whale) in summer. A Fulmar and many rockfish. the shelf. Other large proportion of the species of shorebirds species such as leatherback and and gulls. The northern abalone, green sea turtle northern section is an giant red sea sightings are recorded IBA for Leach's Storm cucumber and sea here and it is also one Petrel. The Scott urchins occur in of only two areas in Islands and cracks and crevices British Columbia surrounding waters and in intertidal areas. where sea otters have are globally important re-established. for the provincially Blue-listed Tufted Puffins, and roughly 90% of British Columbia's population occurs here. The entire British Columbia population of the provincially

April 2010 FINAL Page 10-75

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 10: Effects of Hydrocarbons on the Biophysical Environment

Table 10-6 Summary of Vulnerable Areas within the OWA (cont’d)

Commercial and Marine Mammals Marine and Subsistence Ecosection Oceanography and Sea Turtles Marine Birds Anadromous Fish Fisheries Invertebrates (cont’d) Red-listed Thick-billed Murre breed exclusively on Triangle Island, and the entire population of both the Common Murre and the Horned Puffin (provincially Red-listed) breed only here and on the Kerouard Islands (Haida Gwaii). Queen Wide, deep shelf Important breeding Important breeding Area of overlapping Area of overlapping low benthic diversity Charlotte characterized by habitat for Stellar sea areas for Black important habitat use high effort in with low densities of Sound several large banks lion rookeries; Oystercatcher, for many species of commercial fisheries: some unique and inter-bank established sea otter particularly around fish: eulachon, Pacific groundfish trawl, assemblages, channels, ocean wave habitats near Goose Princess Royal Island. cod, lingcod, shrimp trawl, including exposures with Island; small There are also Global sablefish, Pacific groundfish schedule Hexactinellid sponges depths mostly >200m populations of fin and IBA nesting colonies halibut, sole/flounder, II, red sea urchin dive, associated with deep and dominated by humpback whales of Leach's and Fork- rockfish, Pacific Hake, prawn trap, geoduck troughs. Areas near oceanic water forage in these tailed Storm Petrels, herring. dive, rockfish. Bella Bella are intrusion. waters. and for Cassin's important for Geoduck Auklet along eastern clams. Queen Charlotte Sound.

Page 10-76 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11 Effects of Hydrocarbons on the Human Environment

11.1 Approach There is considerable literature about the effects of hydrocarbon spills on the human environment, mostly related to major open-water spills. According to information from the International Tanker Owners Pollution Federation Limited (2009, Internet site), general effects of spills include adverse effects on traditional harvesting, adverse effects on industries that use seawater, damage to fishing boats and gear, interruption of fishing activities (harvesting and sales) and contamination of coastal amenity areas (recreation and tourism). However, there have been very few attempts to quantify these effects, and most of the studies have occurred after the spill. Quantification of adverse effects of marine hydrocarbon releases on the human environment is a difficult task. Analysis of effects following a spill event is typically impeded by absence of background information regarding pre-event conditions. Prediction of potential effects is difficult, as the extent of effects is dependent on the volume of hydrocarbon released, the location of the event, the nature of marine resources affected, the extent of traditional and non-traditional marine activities in the affected locality, and the duration of clean-up and recovery. Given these variables, and the geographic scope of the OWA and CCAA, the approach taken in this assessment is to provide a general description of baseline conditions relating to heritage resources, traditional and non-traditional marine uses and the socio-economic setting.

11.2 Heritage Resources

11.2.1 Baseline Conditions Heritage resources include precontact period archaeological sites, historic sites and paleontological resources. Systematic field surveys of heritage resources have not been conducted for the entire shoreline in the CCAA, so baseline conditions are unknown; however, the CCAA is considered to be an area of high archaeological sensitivity, particularly in sheltered coastlines, which are rich in a variety of important archaeological sites. A review of site records maintained by the British Columbia Archaeology Branch indicates a large number of previously recorded archaeological sites in the CCAA. Within the OWA, Parks Canada also maintains a database of archaeological sites known in the Gwaii Haanas National Park Reserve and Haida Heritage Site. Precontact resources that could potentially be affected by a hydrocarbon spill include habitation sites (e.g., house depressions, villages, plank house remains, shell middens), ceremonial or religious sites (e.g., rock art, human burials, poles), culturally modified tree sites, transportation sites (e.g., trails and canoe skids), subsistence features (e.g., fish traps), artifact scatters and isolated finds. Historic resources include residential sites, ceremonial sites (human remains and burials) and shipwrecks (Rogers 1992; Northern Maritime Research 2002). Some individual sites may be classified as both precontact period and historic period, such as a historic cemetery placed directly over a

April 2010 FINAL Page 11-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment precontact shell midden. Within the tidal zone there is high potential for a number of important sites older than 5,000 years, due to the complex glacial history of the region (Fedje et al. 2005). No paleontological sites have been identified within the CCAA to date, but sedimentary strata of the Alexander Terrane and Quaternary deposits in marine terraces might contain fossils. Northern Gateway has offered most coastal Aboriginal groups the opportunity to undertake their own Traditional Knowledge study. These studies, combined with archaeological surveys, would be used to define important sites that might require protection. The studies would be completed prior to the start of operations of the terminal, if possible. Additionally, Northern Gateway has offered four Aboriginal groups the opportunity to participate in verification of a draft coastal sensitivity atlas that includes identification of culturally and environmentally sensitive sites and areas. The participating nations are the Gitga‘at Nation (Hartley Bay), the Gitxaala Nation (Kitkatla), the Kitselas First Nation, and the Kitsumkalum First Nation.

11.2.2 Potential Effects of a Hydrocarbon Release on Heritage Resources Degradation, contamination and physical loss of heritage material and interpretive context are the key issues related to hydrocarbon spills. The time of year the spill occurs and recovery time are not relevant, except as they relate to the physical extent and containment. Containment and recovery would be very important in reducing the amount of hydrocarbons reaching the shoreline and intertidal zone. Hydrocarbons can affect heritage resource sites in two ways. First, if hydrocarbons reach the intertidal zone or shoreline, chemical interactions with the physical cultural remains can limit the ability to interpret the sites. Many types of specialized analyses conducted during archaeological studies (e.g., radiocarbon dating, blood protein residue analysis, stable isotope analysis and ancient DNA analysis) are important for enhancing site interpretations. If the hydrocarbons contaminate organic samples, some forms of testing might not be possible or reliable. Radiocarbon dating is particularly difficult in this region due to the marine reservoir effect where marine samples appear ‗older‘ than terrestrial samples of equivalent age (Dumond and Griffin 2002); however, it is difficult to predict specific effects on dating due to variability in hydrocarbons and degree of impregnation at different depths with a site (Reger et al. 1992). It is also possible that rock art pigments could be affected by hydrocarbons, but there is insufficient data to determine effects of hydrocarbon contamination on the pigments. Palaeontological sites are less susceptible to degradation because of hydrocarbon contamination due to the effects of fossilization. Secondly, sites can sometimes be more affected by spill containment and cleanup activities than by the initial contamination. Boom placement, substrate removal, bioremediation and other cleanup activities such as establishment of staging areas and access trails might mechanically disturb heritage resources. Any activities that result in ground disturbance can affect heritage resource sites by disturbing site contents and site contexts, which are critical for archaeological interpretation. Shipwrecks deep beneath the ocean surface will probably not be affected, but shallow shipwrecks could be affected. The introduction of clean up and monitoring personnel into a previously undisturbed area can also have unanticipated secondary effects, such as illegal artifact or fossil collection and looting. An increase in illegal artifact collection (including deliberate excavations in productive midden sites) during cleanup activities was a major effect after EVOS (Bittner and Reger 1995).

Page 11-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.2.3 Mitigation Measures Mitigation requirements for heritage resources would be issued under the British Columbia Heritage Conservation Act. The Ministry maintains a database of all known sites and this information would be used, in combination with the coastal sensitivity atlas and information from the traditional knowledge studies to determine the location of known sites relative to where the hydrocarbons move. GRPs, developed in cooperation with regulators, participating Aboriginal groups and directly affected public stakeholders would identify priority sites for protection and response in the CCAA. Containment is the primary method of limiting effects. The British Columbia Archaeology Branch would immediately release a series of mitigation measures, depending on the volume hydrocarbons and the success of containment activities. Where appropriate, mitigation could include Archaeological Impact Assessments of lands affected by hydrocarbons and cleanup activities to document and record previously unrecorded sites and evaluate known sites. It would be important to consult with Aboriginal groups, as they will have concerns about potential ancestral sites (particularly those of a religious or ceremonial nature, such as human burial sites). All responders would need to be educated to prevent illegal collection of artifacts or fossils. Northern Gateway has also prepared a preliminary coastal sensitivity atlas for the assessment area that includes some information on culturally important sites, including archaeological sites. Northern Gateway has offered participating Aboriginal groups (the Gitga‘at Nation [Hartley Bay], the Gitxaala Nation [Kitkatla], the Kitselas First Nation, and the Kitsumkalum First Nation) the opportunity to undertake truthing of the sensitivity atlas, including the verification of culturally important sites. If this work is not undertaken by coastal Aboriginal communities, Northern Gateway would retain professional archaeologists to provide input to the updating and finalization of the atlas. The British Columbia Archaeology Branch and Parks Canada would be invited to comment on the atlas and provide input on response measures and priorities. If it is determined that sites with heritage value have been affected, additional mitigation requirements, such as monitoring of response activities, detailed mapping and documentation of sites, shovel testing, controlled surface collection, scientific excavation and specialized analysis such as stem round sampling of CMTs could be required. If a portion of a site is damaged by hydrocarbons or by cleanup activities, compensatory studies outside the affected area might be required. Compensatory excavations are in place so that information lost due to contamination or ground disturbance is partially recovered through controlled scientific study of related site areas. After EVOS, local Aboriginal groups requested the establishment of regional repositories and display facilities for artifacts recovered during cleanup activities. Site monitoring programs and public education programs were also conducted.

11.2.4 Follow-up Monitoring Follow-up studies will be required after cleanup to document the effects to heritage resources. Northern Gateway will follow all requirements issued under the British Columbia Heritage Conservation Act to mitigate effects.

April 2010 FINAL Page 11-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.3 Traditional Marine Resource Use Aboriginal groups in the assessment area have considerable knowledge of the area, based on their traditional uses and knowledge of the environment. Aboriginal peoples have particular concerns about effects of hydrocarbons because of their long association with and dependence on the sea for food, transportation and other purposes. Fisheries resources used for food, social and ceremonial purposes, and potential effects of hydrocarbons on these resources, are discussed in Section 10.7. Socio-economic characteristics at the main population centres in the CCAA (Kitamaat Village, Hartley Bay and Dolphin Island) are discussed in Section 11.5. Northern Gateway has offered to meet with Aboriginal groups in the CCAA and OWA. The engagement area includes Coastal First Nations within the CCAA, and those who have interests in the OWA that are in proximity to marine shipping routes, comprised of the Turning Point members. To date, engagement has varied depending on the level of interest and preference of each individual community. Some communities declined to participate, some have been interested in simply receiving information from Northern Gateway about the Project, while others have engaged in a more comprehensive way and talked about a variety of issues, including environmental study methodologies, shipping and emergency response issues. The Haisla, Gitga‘at and Gitxaala are the three main Aboriginal groups within the CCAA or nearby areas that would potentially be affected by a hydrocarbon release. Other Aboriginal communities, while not in the CCAA, have territories that extend into the CCAA, including Lax Kw'alaams, Kitselas, Kitsukalum and Metlakatla and traditional areas important to these Aboriginal groups could, therefore, potentially be affected by a hydrocarbon release. Northern Gateway continues to engage with participating Aboriginal groups in the CCAA and has offered to fund and/or assist in the completion of ATK community reports. Two types of ATK include traditional use information and traditional environmental knowledge. Together, they represent an understanding of specific natural and cultural environments that has been acquired over generations and may be used to enhance analysis of a project‘s effects. The information shared is typically qualitative and based on cultural norms and values, sensory perceptions and oral traditions. Traditional use information focuses on human activities and associated sites or areas of cultural importance within a landscape. Examples can include travel (e.g., waterways, important landmarks), harvesting (e.g., resource use and harvesting areas, special use sites such as fish camps), habitation areas (e.g., occupation areas, meeting areas, gathering places, campsites) and spiritual sites and sacred landscapes (e.g., burial sites, sacred sites, spiritual sites, sacred geography). Traditional environmental knowledge is broad-based and represents acquired wisdom about cultural and ecosystem variability within a landscape and across time. Traditional environmental knowledge, when shared with practitioners of technical scientific disciplines, permits a consideration of Aboriginal perspectives when analyzing the effects of a development, including accidents and malfunctions, on a landscape. Information about the quality, quantity, trends and changes in elements related to a number of disciplines will be recorded when provided by Aboriginal communities.

Page 11-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Together, traditional use and traditional environmental knowledge represent an understanding of specific natural and cultural environments acquired over generations and may be used to enhance analysis of the potential effects of a hydrocarbon spill. They will also inform future emergency response planning.

11.3.1 Baseline Conditions The opportunity to participate in ATK community reports has been extended to Aboriginal groups. To date, ATK community reports on the coast have been limited. As such, first-hand information from communities in the assessment area is not available at this time. The Haisla, Gitga‘at and Gitxaala are the three Aboriginal groups within the CCAA or nearby areas that could be affected by a hydrocarbon release. Traditional territories and locations of reserves are discussed below. Other Aboriginal communities, while not in the CCAA, do have territories that extend into the CCAA, including Lax Kw'alaams, Kitselas, Kitsukalum and Metlakatla. Socio-economic characteristics of Aboriginal communities in the CCAA are discussed in Section 11.5.

11.3.1.1 Haisla Nation The Haisla Nation (Kitamaat Village Council) traditional territory is located in Kitimat Arm, Kidala Arm and Gardner Canal in Douglas Channel, as well as the Kitimat, Kitlope and Kidala River valleys (INAC 2009, Internet site). A total of 648 of the 1,632 registered members of Kitamaat Village Council reside on reserve (INAC 2009, Internet site), which include Kitamaat No. 1, Kitamaat No. 2, Walth No. 3, Tahla (Kildala) No. 4, Jugwees (Minette Bay) No. 5, Bees No. 6, Kitasa No. 7. Kuaste Indian Reserve (Mud Bay, Kildala Arm) No. 8, Kildala River (Thala) No. 10, Henderson‘s Ranch No. 11, Tosehka (Eagle Bay) No. 12, Gilttoyees No. 13, Misgatlee No. 14, Crab River (Crab Harbour) No. 18 and Ja We Yah‘s No. 99.

11.3.1.2 Gitga’at Nation The Gitga‘at Nation (Hartley Bay) territory is near the mouth of Douglas Channel. Currently, Hartley Bay has 670 registered members, of which 151 reside on reserve (INAC 2009, Internet site). The reserves include Kitkahta No. 1, Gill Island No. 2, Quaal No. 3, Quaal No. 3A, Kulkayu (Hartley Bay) No. 4, Kulkayu (Hartley Bay) No. 4A, Lachkul-Jeets No. 6, Gibble Island No. 10, Turtle Point No. 12 and Kunhunoan No. 13.

11.3.1.3 Gitxaala Nation The Gitxaala Nation (Kitkatla) has 1.766 registered members, of which 284 reside on reserve (INAC 2009, Internet site). The main reserve is Dolphin Island No. 1, on an island in Browning Entrance of Hecate Strait (INAC 2009, Internet site). Other reserves are Tsimtack No. 7, Toowartz No. 8, Citeyats No. 9 and Kumowdah No. 3. There is currently very limited information on Aboriginal use of fish, wildlife and vegetation resources for communities in the region; however, data from the 2001 Aboriginal Peoples Survey (Statistics Canada 2002, Internet site) showed that: 32% of adults on the Kitamaat 2 Reserve reported having fished in the previous 12 months

April 2010 FINAL Page 11-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

12% of adults hunted (all cases were for fish or food) 20% of adults reported gathering wild plants (in 75% of cases wild plants were being gathered for food) Thus, there is clear evidence that Aboriginal people in the area harvest fish, wildlife and plants for food as well as social and ceremonial use, although the exact nature of this use and key harvesting locations are not documented. Northern Gateway will continue to explore opportunities with the coastal Aboriginal communities to complete ATK community reports to better document existing use (i.e., important species, locations and seasons of harvesting). Coastal Aboriginal groups with marine interests in the OWA are detailed below.

11.3.1.4 Lax-Kw’alaams First Nation Lax-Kw‘alaams First Nation is located around Port Simpson, approximately 30 km northwest of Prince Rupert. Lax Kw‘alaams First Nation has 78 reserves, which cover a combined area of 11,898.7 ha. The main reserve is Lax Kw‘alaams No. 1, located in Port Simpson, British Columbia, approximately 30 km northwest of Prince Rupert (INAC 2009, Internet site). Lax Kw‘alaams First Nation has 3,228 registered members, of which 760 reside on reserve (INAC 2009, Internet site).

11.3.1.5 Kitasoo/Xaixais Nation The Kitasoo/Xaixais Nation is located on British Columbia‘s central coast at Klemtu, on the east shore of Swindle Island. The Nation consists of 15 reserves, which cover a combined area of 718.4 ha. Kitasoo/Xaixais Nation has 504 band members (INAC 2009, Internet site), of which approximately 294 reside on reserve, which include Canoona No. 2, Dil-ma-sow No. 5, Gander Island No. 14, Goo-ewe No. 8, Kdad-eesh No. 4, Kinmakanksk No. 6, Kitasoo No. 1, Lattkaloup No. 9, Mary Cove No. 12, Oatswish No. 13, Quckwa No. 7, Saint Joe No. 10, Skilak No. 14, Ulthakoush No. 11, and Weeteeam No. 3. The most populated reserve is Kitasoo No. 1, located at Klemtu on the Central Coast.

11.3.1.6 Council of the Haida (Old Masset Village Council, Skidegate Village Council) The Council of the Haida (Old Masset Village Council, Skidegate Village Council) was formed in 1974 as a political entity representing First Nations on Haida Gwaii. The Constitution of the Council of the Haida (Old Masset Village Council, Skidegate Village Council) was formally adopted in 2003 and mandates the Council of the Haida to settle the issue of title and rights on Haida Gwaii. Regional representatives are elected by Haida citizens in four regions: Skidegate, Masset, Prince Rupert and Vancouver. There are also seats for representatives from the Skidegate and Old Masset Band Councils. The Skidegate Band consists of 11 reserves covering 841.8 ha. There are 1,467 band members, of which 699 reside on reserve. Reserves include: Black Slate No. 11, Cumshewas No. 7, Deena No.3, Kate No. 6, Khrana No. 4, Lagins No. 5, New Clew No. 10, Skaigha No. 2, Skedance No. 8, Skidegate No. 1, and Tanoo No. 9. The most populated reserve is Skidegate No. 1, located on the southeast corner of Graham Island, Haida Gwaii. (INAC 2009, Internet site).

Page 11-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

The Old Masset Village Council consists of 26 reserves covering 907.7 ha. Of 2,687 registered band members, 683 reside on reserve, which include Ain No. 6, Cohoe Point No. 20, Daningay No. 12, Egeria Bay No. 19, Guoyskun No. 22, Hiellen No. 2, Jalun No. 14, Kioosta No. 15, Kose No. 9, Kung No. 11, Lanas No. 4, Mammin River No. 25, Masset No. 1, Meagwan No. 8, Naden No. 10, Naden No. 23, Owun No. 24, Saouchten No. 18, Satunquin No. 5, Ssuk No. 17, Tatense No. 16, Tiahn No. 27, Tlaa Gaa Aawtlaas No. 28, Yagan No. 3, Yan No. 7, Yasitkun No. 21, and Yatze No. 13. The most populated reserve is Masset No. 1, located approximately 5 km northwest of the village of Masset, on the east shore of Masset Sound. There are 2,687 registered band members, of which 683 reside on reserve (INAC 2009, Internet site).

11.3.1.7 Heiltsuk Tribal Council The traditional territory of the Heiltsuk spans British Columbia‘s central coast, with the First Nation located on Campbell Island, across from Bella Bella (British Columbia Treaty Commission 2010, Internet site). Heiltsuk consists of 23 reserves, including Belle Bella No. 1, Clatse No. 5, Elcho No. 6, Grief Island No. 2, Hoonees No. 2, Howeet No. 8, Island No. 14A, Kajustus No. 10, Kisameet No. 7, Kluemt No.15, Kokyet No. 1, Koqui No. 6, Kunsoot No. 9, Kyarti No. 3, Neekas No. 4, Noota No. 4, Pole Island No. 14, Quartcha No. 3, Tankeah No. 5, Tcimotf No. 1A, Werkinellek No. 11, Yellertlee No.2, and Yeo Island No. 13 (Aboriginal Canada Portal 2010a, Internet site).

11.3.1.8 Metlakatla First Nation The Metlakatla First Nation is located 5 km west of Prince Rupert on the Tsimshian (Tsimpsean) Peninsula, on British Columbia‘s north coast. Metlakatla First Nation consists of 16 reserves, which cover a combined area of 3,464.4 ha. Metlakatla has 794 band members, of which approximately 109 live on reserve, which include Avery Island No. 92, Dashken No. 22, Edye No. 93, Khtahda No. 10, Khyex No. 8, Kshaoom No. 23, Lakelse No. 25, Meanlaw No. 24, Rushton Island No. 90, S1/2 Tsimpsean No. 2, Scuttsap No. 11, Shoowahtlans (Shawtlans) No. 4, Squaderee No. 91, Tuck Inlet No. 89, Tugwell Island No. 21, and Wilnaskancaud No. 3. The most populated reserve is the Tsimpsean Reserve No. 2, located about 5 km west of Prince Rupert on the Tsimpsean Peninsula (INAC 2009, Internet site).

11.3.1.9 Kitselas First Nation Kitselas traditional territory extends in the Skeena Valley from the Kitselas Canyon westward to the Terrace area and eastward to Lorne Creek, a distance of about 55 km (Kitselas 2010, Internet site). The territory includes the mountains down to the creeks on both sides of the Skeena River. Kitselas First Nation has 10 reserves, which cover a combined area of 1,069.4 ha. Kitselas has 517 band members, of which approximately 186 live on reserve, which include Chimdimash 2, Chimdimash 2A, Ikshenigwolk 3, Ketoneda 7, Kitselas 1, Kshish 4, Kshish 4B, Kulspai 6, Port Essington, and Zaimoetz 5. The most populated reserve is the Kulspai 6, located in Terrace, British Columbia (INAC 2009, Internet site).

April 2010 FINAL Page 11-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.3.1.10 Kitsumkalum First Nation The Kitsumkalum traditional territory is the land around Kalum Mountain and the entire Kitsukalum watershed. The Kitsumkalum also claim heritage sites and resources down the Skeena River and in coastal areas (Kitsumkalum 2010, Internet site). Kitsumkalum First Nation has four reserves, which cover a combined area of 597 ha. Of the 657 band members, approximately 201 live on reserve, which include Dalk-Ka-Gila-Quoeux 2, Kitsumkaylum 1, Port Essington, and Zimagord 3. The most populated reserve is Kitsumkaylum 1, located in Terrace, British Columbia (INAC 2009, Internet site).

11.3.2 Other Communities of Interest In addition to the First Nations discussed above, there are a number of communities including First Nation communities, on the north coast of Vancouver Island, and as such within the geographic boundary of the Open Water Assessment area even though they are more remove from project-related shipping activities. First Nations include the Kwakiutl Nation, the Quatsino First Nation, the Tsatlasikwala Nation, and the Gwa‘Sala-Nakwaxda‘xw First Nation.

11.3.3 Potential Effects of a Hydrocarbon Release on Traditional Marine Use Aboriginal organizations have raised concerns during various engagement activities with Northern Gateway that an accidental spill would affect their traditional lands, culture, practices and lifeways. The concern is that hydrocarbons from a spill would have systemic effects on the food chain, human health, future well being of local communities and, ultimately, on the ability of Aboriginal groups to exercise their rights to resources in the area. The potential for incidents associated with project-related marine transportation is the greatest environmental concern expressed to date in the Aboriginal engagement. Aboriginal resource management strategies are strongly rooted in precautionary principles and adaptive management (Berkes 1999). Cultural heritage often determines how one perceives the environment and the resources contained within it (Haggett 1983; Jordan and Rowntree 1986). While each Aboriginal group is unique, many share certain fundamental environmental values. This type of cultural knowledge and experience often leads to very different interpretations of risk and environmental sustainability than that of mainstream society (British Columbia First Nations Environmental Assessment Working Group 2000, Internet site; Paci et al. 2002). Effects of a hydrocarbon release on traditional marine use would be dependent on the volume and location of the spill, and the nature and extent of traditional marine use in the area as well as the duration of clean-up and recovery (see Section 12 for a description of the effects of selected hypothetical spill sizes and locations).

11.3.4 Mitigation Measures In the unlikely event of a hydrocarbon release into the marine environment, immediate steps would be taken to respond to the event. This would include the numerous measures described elsewhere in this application, such as containment booming, exclusion booming for protection of sensitive sites or communities, execution of clean-up and remediation plans, and compensation.

Page 11-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

For traditional marine uses in particular, Northern Gateway would work to ensure that the protection of culturally and environmentally sensitive sites was given high priority. Pre-identification of such sites through Geographic Response Plans prepared in conjunction with participating Aboriginal groups would be important for achieving that objective. Direct involvement of Aboriginal communities in the Response Organization structure for the area would also assist in ensuring that appropriate priority is given to protection of traditional marine uses. Such capability would be available not only for the Project, but for any other marine activities that could encounter emergency situations. For portions of the OWA more remote from project-related shipping, Northern Gateway would contribute toward regional marine safety planning and also respond to concerns and suggestions from those communities as they are raised.

11.3.5 Follow-up Monitoring The biophysical studies described in Section 10 would provide valuable information to Aboriginal groups about the state of the environment, especially the fisheries. As noted earlier, Aboriginal organizations would be invited to participate in the environmental monitoring studies to characterize existing environmental conditions. In the event of a spill, it would be appropriate for Elders to conduct ceremonies, and to monitor effects on the environment.

11.4 Non-traditional Marine Use The waters and land adjacent to the OWA and CCAA are used for a variety of recreational, residential and commercial purposes. Many of these were identified in the Non-traditional Land Use TDR and include: communities and settlements foreshore development, including commercial leases and marinas marine fishing (traditional, recreational and commercial) eco-tourism (lodges and boat excursions) non-consumptive recreational activities (e.g., marine wildlife viewing, bird watching, hiking, boating, camping), especially in parks and protected areas forestry-related operations involving log dumping, sorting, booming and scaling and boom towing within and beyond the area marine transportation Section 10.7 considers the potential effects on marine fishing, commercial, recreational and traditional use (FSC fishing).

April 2010 FINAL Page 11-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.4.1 Baseline Conditions

11.4.1.1 Settlements and Communities There are numerous communities and settlements located throughout the OWA. The largest of these are the City of Prince Rupert and the District Municipality of Kitimat. Other communities and settlements on the northern coast include the District Municipality of Port Edward, the unincorporated settlement of Dodge Cove and the Port Simpson 1, Lax Kw'alaams 1, S1/2 Tsimpsean 2, Dolphin Island 1, and Kulkayu (Hartley Bay) 4 Indian Reserves. Haida Gwaii is included in the OWA. The following communities and settlements are located in this part of the OWA: the Village of Masset, the Village of Port Clements, and the Village of Queen Charlotte; the unincorporated settlements of Tlell and Sandpit, and the Masset 1 and Skidegate 1 Indian Reserves. The OWA includes communities in the Central Coast area, including the unincorporated settlement of Bella Coola and Hagensborg, and the Bella Bella 1, Bella Coola 1 and Katit 1 Indian Reserves. Parts of the north end of Vancouver Island have also been incorporated into the OWA, including the following communities and settlements: the District Municipality of Port Hardy, the Town of Port McNeill, the Villages of Port Alice and Alert Bay, three unincorporated communities (Sointula, Hyde Creek and Cal Harbour), and nine populated Indian Reserves (Tsulquate 4, Alert Bay 1A, Kippase 2, Quatsino Subdivision 18, Alert Bay 1, Quaee 7, Gwayasdums 1, Hope Island 1, and Fort Rupert 1). Communities in the CCAA include Kitamaat Village (located on the east side of Kitimat Arm), Hartley Bay (located on the west side of Douglas Channel north of Wright Sound), and Kitkatla (located on Dolphin Island, at the north end of Browning Entrance). Section 11.5.1 provides more information on socio-economic characteristics of these settlements.

11.4.1.2 Commercial Leases, Marinas, Moorages and Aquaculture Sites There are commercial leases and marinas, moorages and aquaculture sites located throughout the OWA, and these are too numerous to mention individually. However, as shown in Figure 11-1, these facilities are clustered around Prince Rupert, at the northern end of Kitimat Arm, in the south part of the central coast near Rivers Inlet, at the north end of Vancouver Island, and throughout Haida Gwaii. Within the CCAA there are three commercial leases in Kitimat Arm (a wharf on Minnette Bay and two marinas at the northern end of Kitimat Arm). There are two other marinas, not identified as leases, at the northern end of Kitimat Arm. Additional commercial recreation leases and licences are located throughout the CCAA (Figure 11-2). There are two aquaculture sites in Principe Channel near McCauley Island.

Page 11-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Figure 11-1 Summary of Human Use in the Open Water Area

April 2010 FINAL Page 11-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Figure 11-2 Recreational Areas in the CCAA and on Nearby Land and Water

Page 11-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

The marinas at Minnette Bay and at north end of Kitimat Arm are used by residents from Kitimat, Kitamaat Village, Terrace and visitors from outside the region. A private moorage residential licence is also located on Minnette Bay. Eight ocean charters and pleasure boats operating from the marinas constitute some of the marine traffic in Kitimat Arm (Kitimat Chamber of Commerce 2009, Internet site; Thompson 2009, pers. comm.). July to October is the season of highest boat traffic in Kitimat Arm. Mooring capacity in the MK Bay Marina at Kitamaat Village is for a maximum of 122 boats, and the marina is booked to capacity during the peak season. Boats, yachts and sea vessels are mostly privately owned and used for pleasure, fishing and sightseeing (Grant 2009, pers. comm.).

11.4.1.3 Commercial Recreational Fishing, Eco-Tourism and Lodges Numerous eco-tourism businesses and lodges operate in and adjacent to the OWA. According to BC Stats (2008b), there are 15 fishing lodges in the north coast region of British Columbia and they generated $5.6 million in 2007. Many of theses lodges are on Haida Gwaii, but others are scattered along the mainland coast and are accessible only by air or water. Some information on the location of fishing lodges in the OWA is provided in Figure 11-1. There are two floating lodges operating in the vicinity of the CCAA. One is the North King Lodge, located on the northwest end of Aristazabal Island. It can accommodate 32 people and offers fishing adventures, but will host corporate groups. Visitors fly in from Vancouver. The other is King Pacific Lodge, located on Barnard Harbour on Princess Royal Island. The lodge is operated by Rosewood Hotels and Resorts and is one of their 9 facilities in North America and 21 facilities worldwide. Guests travel to the lodge in a privately chartered plane from Vancouver to Bella Bella and then fly via floatplane to the lodge. It was selected as the Best Resort in Canada by the 2008 Condé Nast Readers' Choice Awards. King Pacific Lodge has also developed a working relationship with the Gitga'at Nation, including a student mentoring program, an elders' breakfast program, joint educational initiatives and hospitality training. Various ecotourism-based companies operate in the OWA and many of them use the inside passage and parts of the CCAA. A list of ecotourism operations that have commercial recreational licences or leases in the CCAA includes: Ocean Lights II Adventures Ltd, based in Vancouver, offers eight-day adventures aboard the 71-ft ketch, Ocean Lights II, in September and October (Ocean Light II Adventures 2009, Internet site). Visitors fly in and out of Prince Rupert or Bella Bella by float plane. Maple Leaf Adventures Corp., based in Victoria, offers eight-day trips from Kitimat to Bella Bella in September and April aboard the 92-ft schooner, the Maple Leaf. Pacific Yellowfin Charters, based in Vancouver, offers eight-day trips from Bella Bella to Kitimat, Hartley Bay in September aboard the 114-ft yacht, the Pacific Yellowfin. Duen Sailing Adventures, based in Brentwood Bay, offers trips to the Great Bear Rainforest aboard the 75-ft ketch, the S.V. Duen.

April 2010 FINAL Page 11-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Due West Charters, based in Richmond, offers mid-coast cruises on the 85-ft trawler, the Curve of Time. Northwest Angling Adventures of North Vancouver also has a number of moorage sites in the CCAA.

11.4.1.4 Commercial and Recreational Fishing Parts of the OWA are intensively used for subsistence, recreational and commercial activities. The original settlement of the region occurred at locations with a secure abundant food supply and proximal to other natural resources required for human survival and settlement. An assessment of the use of marine resources in PNCIMA, which includes the OWA, was completed in 2007 (MacConnachie, et. al. 2007) and provides a good overview of current and potential future human activities. The OWA is a key area for commercial fishing in British Columbia. Available data suggest that the PNCIMA accounted for: 75 to 85% of the groundfish trawl catch (excluding hake) between 1996 and 2005 65 to 90% of the groundfish hook and line catch between 1995 and 2004 60 to 90% of the sablefish long line and trap fishery between 1990 and 2004 75 to 90% of the salmon fishery between 1996 and 2005 (all species; all gear) Coastal areas in the PNCIMA are also commercially fished for a number of invertebrate species including crab, shrimp, prawn, geoduck clams and urchins. In recent years, there has been extensive aquaculture development in the PNCIMA. Most finfish (including salmon) and shellfish farms in British Columbia are located in the PNCIMA, including 78 of 123 finfish sites. Salmon aquaculture has created more than 3,500 direct and indirect jobs in British Columbia, while shellfish aquaculture provides another 800 direct jobs. Many of these aquaculture sites are located on the northwest side of Vancouver Island and in the Queen Charlotte Strait around the Broughton archipelago. Considerable recreational fishing occurs in the tidal areas around Haida Gwaii and numerous locations along the mainland. Recreational catches of coho and chinook salmon and halibut have been increasing steadily since 1995. In terms of total salmon harvests (including commercial), recreational fishermen accounted for 33% of the total harvest of chinook salmon and 15% of the coho salmon harvest. Within British Columbia, recreational fishing is a $0.5 billion dollar industry that generated 4,300 jobs in 2003. First Nations living along the West Coast have a long history of using fish, invertebrates and marine plants for food, medicine, and ceremonial purposes. There are currently no quantitative estimates of the extent of such activities in the PNCIMA, but key fish species are known to include salmon, eulachon and surf smelt. The opportunity to prepare or participate in ATK community reports has been offered to Aboriginal communities along the coast to obtain a better understanding of their use of aquatic and biological resources in the OWA.

Page 11-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Within the CCAA, recreational and commercial fishing in Kitimat Arm and Douglas Channel are important sources of recreation, business and food for many local residents, business operators, guides and tourists. The area attracts anglers from the local area, other parts of British Columbia, Canada, the United States and abroad. There are no current estimates of total recreational angling activity in the CCAA, but a 1982 study of people using four boat launch sites in the Kitimat area determined that 44% of fishing activities occurred in Kitimat Arm (to Lorette Island and including Kildala Arm), 37% occurred in the inner portions of Douglas Channel (Hilton Point to Helen Point), 16% occurred in Devastation Channel, 2% occurred in Garner Canal, and the remaining 1% occurred in outer Douglas Channel and Verney Passage (Canadian Resourcecon Ltd. 1982). Surveys of five local charter and lodge operators identified a number of prime fishing spots near the proposed terminal. These include Moon Bay Marine Area, Half Moon Bay, Bish Creek, shoreline areas in and around rivers and creeks, the eastside channel of Kitimat Arm, and a popular area described as ‗The Wall‘ on the western shore of Kitimat Arm. Caamaño Sound, Principe Channel, Campania Sound, Estevan Sound, Squally Channel and Wright Sound are also popular marine locations, although they are not accessed by all businesses. Giltoyees Inlet, Kildala Inlet, waters in and around Hecate Strait, Nepean Sound and Browning Entrance are less frequently accessed. June to August are the most popular months for recreational fishing and wildlife viewing. Specialized fishing for halibut, rock cod and other bottom-dwelling fish usually occurs in February, March and April. During the peak season, 40 to 70 boats are on Douglas Channel, consisting of 20% local fishers and 80% people from outside Kitimat. Two of the local charter operators interviewed suggested that 40% to 50% of non-local anglers were on guided trips. There are no comprehensive data on commercial fishing in the area. The CCAA is used for a variety of commercial fishing activities. Areas with the highest commercial fishing activity area are located north of the CCAA at the mouth of the Skeena River, although there are pockets of high activity within the CCAA, especially at the south end of Douglas Channel and at Browning Entrance near Kitkatla.

11.4.1.5 Non-Consumptive Recreational Activities Non-consumptive recreation (including SCUBA diving, kayaking, and sailing) occurs throughout the OWA. There are numerous high use underwater diving sites (see Figure 11-1) concentrated in Kitimat Arm and Camaaño Sound, with other sites located along the central coast. There are three trails and one recreation area near the CCAA: Douglas Channel Trail, North Cove Trail, Kitamaat Village Shoreline Trail and North Cove Recreation Area. There are several recognized SCUBA dive sites in Douglas Channel near Maitland Island and around Campania Island (Squally Channel and Estevan Sound). Based on input during public consultation meetings, residents of Kitimat also use a number of areas off the Bish Cove Road (including Emsley Cove, Bish Cove and Half Moon Bay) for a variety of recreational activities.

April 2010 FINAL Page 11-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.4.1.6 Parks and Protected Areas – Provincial There are numerous parks and protected areas in the OWA. Naikoon Provincial Park is a large park in the northeast corner of Haida Gwaii, and offers almost 100 km of beaches. Tow Hill and Rose Spit ecological reserves are also located on the northeast coast. There are several marine parks near Prince Rupert; these are located at Kitson Island, Klewnuggit Inlet, Lowe Inlet, and Union Passage. In 2006, the British Columbia Park Act was amended to include a new designation called a ‗conservancy‘ to protect special areas announced as part of the Central Coast and North Coast Land and Resource Management Plans. Although conservancies provide levels of protection similar to Class ―A‖ parks, they recognize traditional Aboriginal uses but also allow opportunities for low impact, compatible economic activities. Some of these conservancies are shown in Figure 11-3, including five conservancies in the immediate area around Prince Rupert. Several designated areas in and around the CCAA are protected under the British Columbia Parks Act and the British Columbia Protected Areas Act. Several marine parks are located in the central coast area, including Green Inlet Marine Park, Oliver Cover Marine Park, Jackson Narrows Marine Park, Sir Alexander Mackenzie Marine Park, Codville Lagoon Marine Park and Penrose Island Marine Park. The central coast also includes the Hakai Luxvbalis Conservancy, 115 km southwest of Bella Coola. This area is managed under an agreement between the Heiltsuk Nation and the Province of British Columbia and is the largest provincial marine protected area on the British Columbia coast. Land and marine parks are located at the north end of Vancouver Island. The most important of these include Lanz and Cox Islands Provincial Park, which protects some of the most important seabird nesting colonies in the world. Cape Scott Provincial Park, at the tip of Vancouver Island, has more than 115 km of ocean frontage (including 30 km of remote beaches), and the Broughton Archipelago Marine Park, British Columbia's largest marine park, consists of dozens of undeveloped islands and islets at the mouth of Knight Inlet. Within the CCAA there are six parks and two protected areas adjacent to Douglas Arm, three marine parks in Grenville Channel, and three ecological reserves surrounding the entrance to Caamaño Sound. Most of these include a small portion of associated protected marine habitat. Nearshore habitat is sensitive to degradation and pollution by hydrocarbons. Eelgrass beds and estuaries provide important habitat for waterfowl and juvenile fish. This habitat is protected in Dala-Kildala Rivers Estuaries Park, Eagle Bay Park, Foch-Gilttoyees Park and Protected Area and Jesse Falls Protected Area. Coste Rocks Park, directly adjacent to an approach, is an exposed section of intertidal habitat used as a haulout by harbour seals and an undersea garden that holds high value as fish habitat and a recreational SCUBA diving site. Grenville Channel is a common route for boaters travelling the inside passage. Three marine parks provide protection of ecological values, safe anchorage and scenic opportunities to visitors. Union Passage Marine Park, located at the south end of Grenville Channel between Pitt Island and Farrant Island, conserves 395 ha of the North Coast Fjords Marine Ecosection (NCF) and 1,000 ha of land. As part of the Central Coast and North Coast Land and Resource Management Plans, the Parks Act was amended in 2006 to include a new designation called a ‗conservancy‘ to protect special areas.

Page 11-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Conservancies provide a level of protection similar to Class ‗A‘ parks, but traditional Aboriginal uses and low effect, compatible economic activities are also allowed. As shown in Figure 11-3, there are about 10 conservancies located near the CCAA, with another 12 nearby. Kitasoo Spirit Bear Conservancy on Princess Royal Island (103,000 ha) is one of the largest of the largest and is part of efforts to conserve more than 200,000 ha of the Spirit Bear‘s habitat.

11.4.1.7 Parks and Protected Areas – Federal The largest federal protected area in the OWA consists of the Gwaii Haanas National Park Reserve and Haida Heritage Site, a remote collection of 138 islands in the southern part of Haida Gwaii. There are also numerous DFO Rockfish Conservation Areas throughout the OWA. Shown in Figure 11-3, these include three on the west and south coasts of Haida Gwaii, eight between Prince Rupert and Banks Island, eight along the central coast south of Banks Island, two large conservation areas at the north end of Vancouver Island and numerous small conservation areas in the north end of Queen Charlotte Strait. There are three DFO Rockfish Conservation Areas within the CCAA: Otter Passage, Goschen Island and West Aristazabal Island (see Figure 11-3). Rockfish conservation areas are set aside to protect important rearing habitat and recover reduced stocks. Parks Canada is currently selecting candidate sites for National Marine Conservation Areas (NMCA) in the Queen Charlotte Sound marine region. Four candidate sites have been identified for consideration, including the area around Banks Island and Aristazabal Island (including Douglas Channel). After candidate sites have been chosen, Parks Canada will assess the feasibility of establishing an NMCA in the chosen locations.

11.4.1.8 Forestry-Related Operations Forestry operations occur throughout the OWA. Harvested logs are typically assembled into booms at various coastal locations and eventually moved by water to processing facilities (i.e., lumber mills or pulp and paper mills). While there are some processing facilities in the OWA, on Haida Gwaii, near Bella Coola and at Port Hardy, most of the processing facilities are located on central or southern Vancouver Island or in the lower Mainland. Thus, there is extensive movement of log booms from the OWA south to log processing facilities. Within the CCAA, North Coast Log Handling Ltd. owns four water leases for forestry operations within Kitimat Arm, including two on Minnette Bay and two on Cleo Bay. Minette Bay is used for log dumping, sorting, booming and scaling, and log booms are towed to the Cleo Bay barge sites two to three times per month (i.e., 50 to 100 trips per year during a slow year and 150 to 200 trips per year during a busy year). According to Seaspan International, a general estimate of barge traffic related to forestry operations includes 100 barge transits to and from Haida Gwaii, 20 transits to and from Prince Rupert and 20 transits from Cleo Bay in the Kitimat Arm area. About 95% of barge traffic from Cleo Bay is related to North Coast Log Handling Ltd. operations. North-south barge traffic in the CCAA transect Caamaño Sound, Principe Channel and Laredo Channel. Other routes transect Grenville Channel, McKay Channel, Princess Royal Channel, Squally Channel and Whale Channel between Grenville and Laredo Channel, and between Douglas Channel and Laredo Channel. Seaspan International tows barges for 37 forestry- related businesses within the CCAA.

April 2010 FINAL Page 11-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Figure 11-3 Conservancies, Marine Parks, Parks and Protected Areas in and Near the CCAA

Page 11-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

11.4.1.9 Marine Transportation There is extensive marine traffic throughout the OWA, including commercial coastal and inland shipping, as well as ferry traffic. Prince Rupert is the largest port in the OWA and is the terminus for much of the marine traffic. However, Skidegate, Port Hardy and Bella Coola are used by BC Ferries, and the Port of Kitimat is also used for commercial shipping. Cruise vessels are another key component of the tourism industry in the OWA. As noted in MacConnachie et.al. (2007), 16% of all cruise vessels using the Inside Passage Marine Highway stopped in Prince Rupert; it is estimated that the associated tourist spending amounted to about $4 million. Other key ports of call in the OWA for tour ships include Alert Bay and Bella Coola. This component of the tourism industry is expected to increase over time. A more detailed assessment of marine transportation in the OWA is provided in the TERMPOL Study 3.2 Origin, Destination and Marine Traffic Volume Survey. Commercial ships move goods and raw materials in and out of the Port of Kitimat for the Rio Tinto Alcan aluminum smelter, the Eurocan pulp and paper mill (scheduled to cease operations in 2010) and the Methanex terminal. Shipping servicing these operations peaked at approximately 280 ship calls (600 transits) per annum in 1992/93, and had reduced to 100 calls per annum in 2007 (Port of Kitimat Vessel Traffic 1976 to 2007). Traffic is expected to increase once the Sandhill and Kitimat LNG projects begin operations. The Sandhill Project will have a marine terminal north of the proposed Kitimat Terminal, and the Kitimat LNG Project will have a marine terminal at Bish Cove. BC Ferries and some cruises also operate in the CCAA. Ferries cross Wright Sound between Grenville Channel and Whale Channel as part of the inside passage route to and from Prince Rupert. Cruise ships travelling to and from Alaska might also use the outer passage (Principe and Laredo Channels).

11.4.2 Potential Effects of a Hydrocarbon Release on Non-Traditional Marine Use The potential effects of hydrocarbons in the OWA or the CCAA on non-traditional marine use would depend on the volume and location of the spill, the extent to which it spreads, the amount of time required for containment and clean-up, and the nature and extent of the non-traditional marine use in the locally affected area. Effects could include: temporary restrictions on marine access to various communities, primarily those only accessible by boat (e.g., by BC Ferries, other ferries or private vessel). contact of oil with wharves, docks, boats and related equipment for marinas and other operations with foreshore leases. restrictions on access to marinas in Kitimat Arm and other communities for a short time short-term disruption of ecotourism businesses, including possible stranding of cruise-based operations, loss of clients and fouling of equipment, as well as long-term damage to the reputation of the region as pristine aquatic wilderness and resulting loss of business for operators. disruption to and temporary restrictions on access to marine fishing (recreational and commercial) areas and activities, as well as fouling of vessels and gear.

April 2010 FINAL Page 11-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

disruption (fouling of logs) and temporary restriction of access to and transportation of log booms for forestry operation could be a problem throughout the OWA as much of the timber harvested in coastal areas is moved by tug to southern locations for processing. A spill in the CCAA could affect a number of log booming and towing operations in Kitimat Arm. disruption of, or temporary restrictions on, commercial ships that would normally move through the OWA or CCAA. This includes ships moving in or out of the major ports, primarily Prince Rupert, the Port of Kitimat, Bella Bella and Port Hardy. From the perspective of non-traditional marine use, any hydrocarbons that would spread to the shorelines near any of the coastal areas that are either intensively used for marine activities or are heavily populated would be of consequence. A spill in the open water areas where the oil does not reach the shores is considered less problematic because commercial shipping, eco-tourism operators, commercial and recreational fishermen and other marine users would be able to adjust their travel plans to avoid the affected area. The duration of non-traditional marine use restrictions would depend on how long it takes for the spill to be contained so that shipping and other marine activities can resume. Based on available information it appears that a spill during the summer months would represent an event with the greatest potential for effects from the perspective of recreational use (i.e., highest number of users), but a spill in September would represent the greatest potential for effects on ecotourism (i.e., peak number of customers). While most potential adverse effects associated with a spill would be relatively short-term (weeks to months), there might be some longer term effects on the commercial and recreational fishing. Hydrocarbons can taint local seafood resources, thereby affecting their safety, marketability and palatability for recreational and commercial fishermen. It is expected that DFO would make the decision about the length of closure of commercial and recreational fisheries in affected areas. There might also be temporary declines in fish populations that affect fishing success rates and the quality of resident and non- resident fishing. Closures for any length of time or decreased harvesting rates would likely result in commercial and recreational fishermen traveling to other, less favourable sites to engage in similar activities. The severity of the effects of a spill in the OWA or the CCAA will ultimately depend on how long it takes for the spill to be contained so that shipping and other marine activities can resume. Based on available information it appears that a spill during the summer months would represent the worst case event from the perspective of recreational use, but a spill in September would represent the worst situation for ecotourism. While most potential adverse effects associated with a spill would be relatively short-term, depending on the amount of time individual areas are less accessible to recreational or commercial uses, there may be some longer term effects on the commercial and recreational fishing. As noted in Section 10.7.2, hydrocarbon releases can result in tainting of local seafood resources, thereby affecting their safety, marketability and palatability for recreational and commercial fishermen. DFO will ultimately make the decision about the length of closure of commercial and recreational fisheries in affected areas. There may also be temporary declines in fish populations that affect fishing success rates and the quality of resident and non-resident fishing. It is expected that closures for any length of time or decreased harvesting rates

Page 11-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment would result in commercial and recreational fishermen traveling to other, less favourable sites to engage in similar activities.

11.4.3 Mitigation Measures Timely containment and clean-up is the most important mitigation measure in respect of disruption of non-traditional marine use activities. Containment would allow most marine activities to continue, although during this process there might be minor inconveniences for boaters, recreational and commercial fishermen and ecotourism operators who have to travel farther to avoid affected waters. Prompt containment and clean-up would also limit the duration of adverse effects such as fouling and aesthetics for boaters, fishermen, foreshore users and local residents. The effects on non-traditional marine use would also be mitigated through immediate notification of relevant local and provincial authorities, fisheries resource user groups and recreation groups; where required, advisory signs will be posted relating to fish consumption and recreational use at strategic locations and on selected websites (Kalum Forest District, District of Kitimat and British Columbia fishing websites). Where the spill would result in fouling of boats, ships, and marine and foreshore infrastructure, effects would be mitigated through the provision of compensation for clean-up, repair and/or replacement, as per standard practices. Compensation is described in Section 7.9. Northern Gateway proposes to establish a Fisheries Liaison Committee that will include Aboriginal, commercial and local fisheries representatives who will provide advice on means to reduce the routine effects of the terminal operations and vessel movements on marine fisheries and other marine users. The committee will also provide a forum for discussion of measures to be taken to mitigate effects of hydrocarbon releases on other marine users.

11.4.4 Follow-up Monitoring There should be minimal long-term effects on non-traditional marine use. However, in the near term (i.e., one or more fishing seasons), effects could occur for recreational and commercial fish stocks and associated recreational and commercial fishing. As noted in Section 10.7.5, a monitoring program will be employed to determine when fish stocks and other marine life have sufficiently recovered so that affected areas can be fully reopened to recreational and commercial fishing and other affected activities.

11.5 Socio-Economic A hydrocarbon spill in the OWA or CCAA would have the potentially affect the well-being of people who live, work or spend time in areas affected by the spill. Depending on the size and location of the spill, economic well-being could be adversely affected as a result of changes in traditional marine uses (Section 11.3) and non-traditional land and resource use activities (Section 11.4). Social well-being of regional residents could also be affected.

April 2010 FINAL Page 11-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

A key socio-economic issue for community well-being is the extent to which any resulting property damage, prohibitions or restrictions on access, transportation, recreation, commercial activities or resource use adversely affects land or resource owners, resource users and the general public. Where individuals, companies or Aboriginal communities can demonstrate quantifiable loss resulting from a spill, compensation for such damages (e.g., lost harvesting opportunities), increased costs of operations, increased travel costs, lost opportunity costs, and cleanup or replacement of equipment) would be provided according to standard industry practices (see Section 7). Details of financial responsibility and compensation are provided in Section 7.9. A second potential socio-economic effect relates to adverse effects on the social well-being of residents of nearby communities. Studies on the effects of EVOS led Picou and Gill (1996, Internet site) to conclude that ―renewable resource communities,‖ consisting of individuals ―whose primary cultural, social and economic existences are based on the harvest and use of renewable natural resources,‖ are particularly vulnerable to the effects of an oil spill. The study states that a spill might sever the traditional relationships between these communities and seasonal harvest activities. From the residents‘ perspective, a spill can alter their perception of personal safety and security within their immediate biophysical environment and this can result in social disruption, psychological stress and loss of institutional trust. A third issue relates to cleanup and remediation activities and the effects these might have on local communities and infrastructure. While clean-up activities can potentially provide employment and business opportunities for local residents, the introduction of non-resident clean-up and remediation workers into local communities can create community conflict and segmentation.

11.5.1 Baseline Conditions The socio-economic region (SER) for the assessment, shown in Figure 2-1, includes the portion of the Kitimat-Stikine Regional District (RD) around and including the District Municipality (DM) of Kitimat, the Skeena-Queen Charlotte RD, the Central Coast RD, and the Mount Waddington RD, which includes the northern end of Vancouver Island. As shown in Table 11-2, this area includes 10 villages, towns and cities, 16 Indian reserves with populations of 100 or more, and 13 rural areas comprising regional district electoral areas (RDEAs). In 2006, the SER had a combined population of 44,096 people (Statistics Canada 2007, Internet site; BC Stats 2009a, Internet site), with 45% living in the Skeena-Charlotte RD. Prince Rupert alone accounted for 29% of the population of the SER. The Mount Waddington RD accounted for 26% of the population, the DM of Kitimat (including Kitamaat Village) accounted for 21%, while the Central Coast RD accounted for the remaining 7%. Overall, 74% of the population lived in the 10 communities, 11% lived in the rural areas and 15% lived on Indian reserves.

Page 11-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Table 11-1 Communities, Indian Reserves and Rural Areas included in the OWA

Skeena- Queen Charlotte Regional District

Kitimat-Stikine Central Coast Regional Mount Waddington Regional Regional District Haida Gwaii Prince Rupert Region District District

Communities Kitimat Masset Prince Rupert Port Hardy Port Clements Port Edward Port McNeill Queen Charlotte Port Alice Alert Bay

Kitamaat 2 (Kitamaat Masset 1 Lax Kw'alaams 1 Bella Bella 1 Tsulquate 4

Indian Reserves Indian Village) Skidegate 1 S1/2 Tsimpsean 2 Bella Coola 1 Alert Bay 1A Dolphin Island 1 (Kitkatla) Katit 1 Kippase 2 Kulkayu (Hartley Bay) 4 Quatsino Subdivision 18 Alert Bay 1 Quaee 7

Kitimat-Stikine D RDEA Skeena-Queen Skeena-Queen Charlotte A Central Coast A RDEA Mount Waddington A

Rural Areas Rural Charlotte D RDEA RDEA Central Coast C RDEA RDEA Skeena-Queen Skeena-Queen Charlotte C Central Coast D RDEA Mount Waddington C Charlotte E RDEA RDEA Central Coast E RDEA RDEA Mount Waddington B RDEA

Mount Waddington D RDEA

April 2010 FINAL Page 11-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

The population of the SER has been dropping steadily since at least 1996. Between 1996 and 2001, the population decreased 10% from 55,111 to 49,496. An additional decline of 10.9% was reported for the period between 2001 and 2006. On average, this represents a loss of nearly 1,100 people per year, resulting in a 20% decrease between 1996 and 2006. Despite the decline in total population, the Aboriginal population in the SER has increased slightly; in 2001 there were 13,566 Aboriginal people in the SER (Statistics Canada 2002, Internet site), and this number increased 3% to 13,974 in 2006 (Statistics Canada 2007, Internet site). In 2006, Aboriginal people accounted for 30.9% of the SER population, although this ranged from 13.3% in the Kitimat-Stikine RD to 62.6% in the Central Coast RD. Only 48% of Aboriginal people in the SER lived on reserves, and the on-reserve population decreased 5.2% between 2001 and 2006, suggesting that most of the growth in the Aboriginal population has occurred in the major communities. All of the regions in the SER rely on marine resources to some extent and could be adversely affected by a spill should it occur in their particular area, and be of sufficient size to create measurable effects. The extent of commercial dependence on marine resources is evident from the percentages of total income (after tax) directly attributed to fisheries and tourism. This information is available for some of the larger communities and regions (see Table 11-2) and indicates that regional residents rely on employment for between 64% and 86% of their annual incomes, with the balance coming from government transfers and other sources. Table 11-22 also shows that income from public sector employment, which is reliant on provincial government funding, accounts for a large amount of employment income, especially in the Central Coast region. This means that goods producing industries (industrial employment), which generate wealth, account for between 29% and 71% of incomes in the local communities. The relative importance of various industries as a source of employment varies considerably among the communities and regions. Table 11-2 shows that, although fishing provides a relatively small source of income from industrial employment in most communities (except Kitimat), it is of particular importance in Prince Rupert Region (48% of income) and the Central Coast (36%). In most large communities, forestry accounts for more income than fishing (especially in Port Alice [67% of income] and Port McNeil [77%]). While tourism employment contributes relatively small amounts of income in most communities, this sector accounts for 31% of industrial employment income in the Queen Charlotte Islands, 32% in the Central Coast Region and 24% in the Prince Rupert region. Within the SER, Kitimat is unusual because of the very large component of regional income generated by the mining and mineral processing industries (72%). Overall, the data show that the parts of the SER that are most reliant on marine resources include the Prince Rupert region, the Queen Charlotte Islands and the Central Coast region.

Page 11-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Table 11-2 Income Dependencies (After Tax) of Communities and Regions in the OWA, 2006

Kitimat- Skeena-Queen Stikine RD Charlotte RD Mount Waddington RD

Prince Queen Central DM of Rupert Charlotte Coast Alert Port Port Port Source of Income Kitimat Region Islands RD Bay Alice Hardy McNeill Employment Income (%) 76 73 70 76 70 64 75 86 Public Sector (%) 34 44 44 66 46 25 33 26 Industrial Employment (%) 61 45 51 29 44 67 59 71 Forestry (%) 9 15 39 18 42 67 41 77 Mining & Processing (%) 72 3 0 0 0 0 5 2 Fishing (%) 0 48 19 36 29 12 27 10 Tourism (%) 9 24 31 32 16 9 16 8 Construction (%) 11 9 11 14 13 12 11 3 Other (%) 5 11 4 5 10 8 8 3 Transfer Payments (%) 14 18 18 16 19 22 16 8 Non-employment Income 10 9 12 8 11 14 9 6 (%) TOTAL (%) 100 100 100 100 100 100 100 100

NOTE: Socio-economic information in this report is based on areas from the Alaskan boarder to the Northern Vancouver Island and out to the Territorial Sea; therefore, community information, study areas and assessment numbers cannot be compared directly with the rest of the Application.

SOURCE: BC Stats 2009b, Internet site

11.5.2 Potential Effects of a Hydrocarbon Release on Socio-economic Conditions The magnitude of the effects would depend on the nature of the spill: its location time of year geographic extent length of time the spill interfers other activities effort required for clean-up manner in which claims for compensation are considered and reimbursed As noted in Section 11.4.2, diluted bitumen or synthetic oil in the SER could affect various marine-based activities, including fishing, tourism, commercial shipping, in addition to the well-being of any communities affected by the spill. A major spill near one or more of the communities could result in temporary loss of employment and income and temporary restrictions on boat access to communities, which could have adverse effects, especially if isolated communities are affected.

April 2010 FINAL Page 11-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

While some communities might be more vulnerable than others, it is expected that, in most cases, a spill would have minimal net effects on employment and income. Where individuals or businesses or Aboriginal groups can demonstrate a quantifiable loss as a result of restrictions imposed on marine or foreshore access or activities, compensation would be available to offset these losses. This would mean that businesses could be compensated for loss of revenue, damage to boats and/or equipment, and/or any operating costs of having to travel to alternate sites to pursue commercial fishing or other activities. Similarly, Aboriginal groups would be eligible for compensation for quantifiable losses. A spill in the CCAA could result in temporary restrictions on or closure of areas to boating or fishing and this would affect the well-being of local residents as well as commercial operators who transport tourists to locations in Kitimat Arm, Douglas Channel, Squally Channel or Principe Channel by boat. For tourism operators this could result in an immediate loss of revenues, if tourists choose to go elsewhere, and/or increased operating costs as commercial operators take their clients to secondary sites further away. A spill during the spring or summer months would have the greatest effects on commercial tourism. The financial effects for the tourism industry could be of longer duration, depending on how long it takes for sport fish populations and environmental conditions to return to pre-spill conditions. Serious effects on recreational activities would occur if access across the entire width of Kitimat Arm, Douglas Channel, or Grenville Channel was restricted for any length of time during late spring or summer, whereas the high magnitude effects on ecotourism would be in September. The full economic and social costs of a spill would only be known after the cleanup has been completed, compensation has been paid, and environmental and social conditions return to near normal. Studies have shown that the socio-economic costs of spills, including tourism and recreation income loss, effects on port operations, and effects on commercial fishing, can be as large as the costs of clean-up (Etkin et. al. 2003, Internet site). Clean up activities would be undertaken by trained responders located at various sites within the SER; opportunities exist to train emergency responders from First Nations communities and there might be requirements for additional workers, including members of Aboriginal groups and other regional residents.

11.5.3 Mitigation Measures Timely containment and clean-up is the most important mitigation measure for limiting adverse effects on socio-economic conditions and community well-being. Containing the hydrocarbons so that most marine and foreshore activities in either the OWA or the CCAA are unaffected would limit effects on boaters, fishermen, tourism operations and other commercial activities. Prompt containment and clean-up would also limit the duration of fouling and compromised aesthetic conditions (visual, odour and noise) for regional residents, tourists and commercial operators. Fair and prompt reimbursement of compensation claims would substantially mitigate both economic and social costs associated with a spill. The strict liability and compensation scheme created under the Marine Liability Act and associated international conventions are designed to accomplish that objective by avoiding litigation, and providing access to an internationally backed pool of compensation funds for use in such circumstances. Northern Gateway would also work with Aboriginal and non-native communities in advance to develop geographic response plans that identify areas of key importance to specific communities and resource users, as well as priorities and strategies for protection of these areas. The plans could also describe

Page 11-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment mitigation strategies that include consideration of the additional stress on the community that might result from bringing large numbers of non-residents into the community for any length of time.

11.5.4 Follow-up and Monitoring The majority of effects on community well-being would be short term and reversible once hydrocarbons in the OWA or CCAA have been contained and cleaned-up. No long-term effects on socio-economic conditions would be expected as a result of timely compensation for economic losses. Nevertheless, Northern Gateway would work with the local community to identify and resolve any other longer-term socio-economic issues related to a spill and clean up activities.

11.6 Summary of Effects on the Human Environment A large spill of diluted bitumen or synthetic oil from a tanker in the OWA or in the CCAA could adversely affect the human environment by interfering with use of marine resources (fish, shellfish, crustaceans) by Aboriginal, recreational and commercial fishers and by people involved in the tourism industry (e.g., fishing guides, lodge and marina operators, ecotourism operations, and accommodation and service providers). Log booming and related forestry activities could also be affected. There would be short-term effects while the emergency response is underway; however, effects could last longer if there are tainting and contamination concerns about fisheries resources, if there is a fishery closure for consumption or for recovery of a population, or if hydrocarbons damage the reputation of the area for eco-tourism. In situations where individuals or commercial operations experience a clear economic loss, compensation will be paid and Northern Gateway will identify opportunities to remunerate adversely affected parties as promptly as possible. Members of the local communities could be employed to assist with cleanup and remediation. The extent of socio-economic effects would likely only become clear once the costs of cleanup have been documented. Health and safety considerations for responders and the general public would be dealt with through the OSRP. Hydrocarbons that move onshore could also damage heritage resources such as pre-Contact and post- Contact physical remains or palaeontological resources (fossils) through contamination or during the cleanup. Although there are some records of heritage resources along the coast, more detailed study and mitigation plans would need to be developed at the time of a spill, under direction of the provincial regulatory agency, to avoid damage. Effects of condensate would be much less than for viscous hydrocarbon, given the high volatility of condensate, leading to less area and fewer numbers of organisms being potentially affected. Emergency response plans and cleanup procedures would be implemented immediately to mitigate adverse effects to the human environment. Monitoring would be necessary so that recovery would continue post-cleanup.

April 2010 FINAL Page 11-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Table 11-3 Summary of Areas and their Uses within the Open Water Area

Commercial and Ecosection Subsistence Fisheries Human Use Traditional Use Continental Area of overlapping high Used by trade vessels and Black cod has been utilized Slope effort in commercial seasonally by passenger in the Continental Slope fisheries: groundfish trawl, vessels. The continental area as a staple food sablefish trap, sablefish slope region has limited source in the absence of longline, rockfish utilization by humans. abundant salmon.

Dixon Area of overlapping high Sponges and corals are An important harvest area. Entrance effort in commercial located at the northwestern Intertidal species such as fisheries: groundfish trawl, tip of Graham Island, which razor clam and other more schedule II groundfish, is also an important area pelagic species such as sablefish trap, sablefish for recreational fishing. eulachon are harvested. longline, crab trap, rockfish

Queen Area of overlapping high Supports a large number of First Nations communities Charlotte effort in commercial industrial and aquaculture use fish and intertidal Strait fisheries: groundfish trawl, sites. RCAs offer protection species for food and for shrimp trawl, schedule II to areas that were showing cultural practices. groundfish, red sea urchin evident rockfish population dive, prawn trap, green sea declines. Numerous urchin, geoduck dive, crab marinas and small craft trap, and rockfish harbours located throughout the Strait, indicating heavy use by humans throughout this area. Hecate Area of overlapping high Potential for wind energy. The Haida village of Strait effort in commercial BC Ferries services a route Skidegate borders Hecate fisheries: groundfish trawl, across Hecate Straight. Strait on Haida Gwaii. The schedule II groundfish, RCAs protect rockfish from 800+ residents of crab trap, rockfish further decline in the Skidegate continue to historically heavily fished engage in harvesting and areas. The Gwaii Haanas cultural practices in their National Marine traditional lands. Conservation Area and National Park provides protection to the southern portion of Haida Gwaii, within the Queen Charlotte Sound.

Page 11-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 11: Effects of Hydrocarbons on the Human Environment

Table 11-3 Summary of Areas and their Uses within the Open Water Area (cont’d)

North Coast Area of overlapping high A number of industrial sites First Nations communities Fjords effort in commercial are scattered through the utilize on the return of fisheries: shrimp trawl, ecosection. There are salmon to rivers such as schedule II groundfish, red some popular sites for Nass River, Skeena River, sea urchin, prawn trap, recreational SCUBA diving. Kitimat River, Bella Coola, geoduck dive, crab trap, BC Ferries services some Smith and River inlet. rockfish routes that pass through Gitga'at First Nation people the fjords, providing access utilize resources in Hartley to some otherwise remote Bay, including cockles. locations. Much of this area is protected by provincial park legislation, some of which extends into the fjord waters. Vancouver Area of overlapping high This area supports a large First Nations on western Island Shelf effort in commercial portion of British Vancouver Island use the fisheries: groundfish trawl, Columbia's aquaculture resources available within shrimp trawl, groundfish sites. Contains RCA the Vancouver Island shelf schedule II, red sea urchin designated areas. region, dive, prawn trap, geoduck dive, crab trap, rockfish

Queen Area of overlapping high Potential for wind energy Queen Charlotte Sound is Charlotte effort in commercial generation. Sensitive accessed by First Nations Sound fisheries: groundfish trawl, sponges, corals and communities mostly by shrimp trawl, groundfish sponge reefs are located in boat to fish. schedule II, red sea urchin the center of Queen dive, prawn trap, geoduck Charlotte Sound, which are dive, rockfish protected from trawl fisheries. Rockfish conservation areas and provincial parks protect a large portion of the nearshore habitat. Recreational fishing activities are popular in nearshore areas. The inside channels are popular cruise ship routes during the summer.

April 2010 FINAL Page 11-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12 Mass Balance Examples for Response Planning

12.1 Introduction Most accidents and malfunctions are improbable, preventable and readily addressed through planning, design, tanker vetting, hazard analysis and corrective action, emergency response planning, and mitigation measures. Northern Gateway considers accident prevention a priority. With proper planning and application of mitigation measures, the probability of a hydrocarbon spill occurring within the CCAA or OWA, or at the Kitimat Terminal, is low (see Sections 4 and 5). However, to allow for even the lowest risk, provisions will be made for a response to any potential spill within the CCAA or at Kitimat Terminal This section provides examples of response activities (containment, clean up and restoration) that would occur in the event of a large spill (a minimum of approximately 10,000 m3) in the CCAA or OWA, or for a medium spill of 250 m3 at the terminal. Because it is impractical to assess the potential for and effects of the full range of spills that might occur, hypothetical mass balance examples are used to represent spill conditions. By focusing on spill examples likely to have the greatest consequences, and where there could be environmental effects for a range of components of the marine and human environments, the assessment encompasses any effects of lower consequence situations.

12.2 General Spill Response The GOSRP will include an outline of the responsible organization and the procedures established, in cooperation with government and other agencies, for managing a spill response. The general mitigation and cleanup measures outlined in Sections 3 and 7 would be in place. Although the response actions and times might differ depending on the spill circumstances, a response as described in the mass balance examples might involve the following: upon detection, communicate with authorities specified in the OSRP make notifications per the response plan protocol and activate the ICS upon notification, deploy emergency response teams from local and regional pre-staged locations in Kitimat and several other locations within the CCAA within a maximum 6 to 12 hours in the CCAA, the first response team (in addition to the escort tug response capability) is on site with emergency response spill equipment the duration of cleanup and recovery operations will be defined by cleanup endpoints (see Section 7.7) as agreed upon by the response team, government agencies, coastal Aboriginal organizations and directly affected stakeholders coastal remediation would likely entail longer-term operations

April 2010 FINAL Page 12-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Response efforts would consider: vulnerability of the various organisms to contact with oil (e.g., marine birds that spend their life on the water would likely have more contact with hydrocarbons than would demersal fish) the type and magnitude of effect (e.g., irritation of mucus membranes or mortality)

12.3 General Mitigation The mass balance examples discussed below include specific mitigation for the locations and conditions modelled. These are in addition to the mitigation that would be in place for any spill, as discussed in Sections 3 and 7. The key mitigation for any spill would be deployment of containment booms around the tanker, initially using equipment on the escort tugs. Other measures include: immediate notification of relevant local and provincial authorities, coastal Aboriginal organizations and resource users staging of response equipment at several locations throughout the CCAA and in Kitimat to reduce response time5 tracking of the movement of hydrocarbons and deployment of exclusion booming to protect sensitive areas in the likely pathway of a movement. Priorities for protection of sensitive areas would be developed in the GRPs based on input from government agencies, Aboriginal organizations and directly affected stakeholders use of passive or active hazing techniques and other methods to deter or exclude wildlife from the area of the spills and areas in the likely pathway of moving hydrocarbons manual recovery and cleanup of surface hydrocarbons using skimmers, sorbents, vacuum systems, washing, and removal of stranded hydrocarbons, oiled vegetation and oily debris ongoing shoreline cleanup operations posting of a fish-consumption advisory on British Columbia fishing websites posting of a recreational use advisory on websites additional mitigation measures for marine birds and marine mammals are described in Sections 10.8 and 10.9, such as rehabilitation of oiled animals and birds. Compensation would be provided for damages to marine infrastructure and marine resource users.

5 Northern Gateway has offered coastal organizations the opportunity to participate in the RO for the Project. This could involve locally based emergency response personnel, vessels and equipment in a number of coastal communities.

Page 12-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.4 Selection of Mass Balance Examples for the CCAA and OWA Hypothetical mass balance examples provide a specific context to discuss hydrocarbon behaviour, environmental effects and cleanup effectiveness. This information is important in: further refining tanker specification and operational plans to reduce the risk of incidents developing and implementing an emergency response plan to contain, control and recover a spill developing and implementing a habitat protection plan to protect sensitive environments and sites Examples are used for diluted bitumen and synthetic oil for different geographic and environmental conditions, navigational hazards, and inflow and outflow conditions (winds and currents). Although the likelihood of these hypothetical spill examples is low, each example provides an opportunity to determine: key design features to avoid the potential for a spill appropriate mitigation (or contingency) measures to limit spills the behaviour of liquid hydrocarbons in the environment the probability of a spill (discussed in Sections 4 and 5) effective response plans and equipment and personnel needs potential social, economic and biophysical effects of spills Information from these examples provide information that is pertinent to how GRPs might be developed. Northern Gateway will work with Aboriginal organizations, potentially affected stakeholders and various levels of government (i.e., municipal, provincial and federal) to develop GRPs for key sites in the CCAA and OWA. These plans will identify and prioritize response activities and requirements (e.g., personnel, vessels, aircraft, equipment) with respect to: protection of culturally- and environmentally-sensitive areas containment and removal of hydrocarbons shoreline cleanup shoreline rehabilitation The six pathway examples developed are based on proximity to navigational hazards or to unique or sensitive environmental features, or both. The locations are Emilia Island, Wright Sound, and Principe Channel, Caamaño Sound, Browning Entrance and the Kitimat Terminal (see Figure 12-1). A hypothetical spill of 10,000 m3 from tankers is discussed for: Emilia Island (synthetic oil under outflow conditions during winter) Principe Channel (diluted bitumen under inflow conditions during summer) Ness Rock in Caamaño Sound (diluted bitumen under typical winter conditions) Butterworth Rocks in north Hecate Strait (synthetic oil under typical summer conditions).

April 2010 FINAL Page 12-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Figure 12-1 Mass Balance Example Locations

Page 12-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

This volume approximates a worst case consequence of a laden Suezmax tanker grounding. The volume was derived through the QRA, which provides a theoretical calculation of the consequence from a breach of a double hull and two internal tanks of the most common ship size, Suezmax, due to grounding. The estimated chance of a spill exceeding this volume due to a grounding incident which results in a spill, involving any size of the proposed tankers, is 5.5 percent. The Wright Sound example, although highly unlikely, is for 36,000 m3, which represents the potential volume of oil that might be spilled as a result of a collision between a VLCC and another vessel that results in a breach of the double hull and almost complete drainage of two internal tanks. The location of the breach in the VLCC for this example is such that the collision breach would result in a greater loss of oil than with a grounding breach, due to the relative difference between the oil level in the tanks and sea level along the hull of the VLCC at the breach point. In addition to the vessel based pathway examples, two examples are included for accidental releases during loading or offloading operations at the terminal. These examples were developed based on a 250 m3 spill of diluted bitumen from a tanker loading and a 250 m3 spill of condensate from a tanker unloading at the terminal (under inflow conditions during summer). This is approximately the amount that would be released before automatic shutdown occurs (47 seconds for oil or bitumen, 60 seconds for condensate). The volume estimated in the quantitative risk assessment (QRA) (DNV 2010). These examples are credible representations of accidents and malfunctions that might occur during cargo loading and unloading operations because of equipment failures or other issues. As it is not reasonable to consider all the possible accidents and malfunctions that might occur, the examples represent the hydrocarbons that will be transferred at the marine terminal under assumed meteorological conditions. These examples were also selected to represent higher consequence events (i.e., maximum volumes of hydrocarbons, potentially significant environmental outcomes) that would tend to address the effects of less likely or lower consequence scenarios. Synthetic oil spill was not used as an example at the terminal due to its similarity in properties to that of diluted bitumen. Key design features to avoid a spill are discussed in Section 3. Important mitigation measures to reduce the potential for spills include: using a tethered escort tug and close escort tug for laden oil tankers, and close escort tugs for those in ballast using enhanced navigational aids as proposed by British Columbia Coastal Pilots, and installing radar systems along important sections of the ship routes using an Electronic Chart Display and Information System in conjunction with standalone personal pilot units, which allow navigation capability independent of the ships navigations systems boarding and control of all inbound and outbound vessels by licensed, experienced Canadian marine pilots who are very familiar with the local waterways and weather conditions requiring that all tankers have a double hull vetting of all tankers and assured compliance with IMO standards (e.g., requiring electronic chart display and information systems on oil tankers, watch-keeping certification, inspections etc.)

April 2010 FINAL Page 12-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

reducing speed within the CCAA to 10 to 12 knots and, in certain areas, to 8 to 10 knots observation of weather restrictions for vessels entering the CCAA and for berthing at the Kitimat Terminal Implementing such mitigation measures, and in particular mandating use of tethered escort tugs and close escort tugs in the CCAA, reduces the probability of an incident considerably, as discussed in Section 4.

12.4.1 Description of a Mass Balance Example The mass balance for each example is based on physical properties of the liquid hydrocarbons and prevailing oceanographic and meteorological conditions at the selected site and is based on the conservative assumption that there is no spill mitigation (containment or removal of any of any potential oil spill). Although the fate and behaviour of the hydrocarbon differs for each example, the following generalizations apply: condensate would evaporate and disperse within 24 hours of the spill (generally within approximately 12 hours) synthetic oil would remain on the water surface for 48 to 168 hours before it evaporates, disperses naturally or is stranded along shorelines. diluted bitumen would be more persistent than condensate or synthetic oil because of the lower proportion of volatile compounds and lower rate of evaporation Estimates of the mass balance and time for hydrocarbons to reach shore or sensitive areas is used to identify and prioritize areas that can be protected with booms. Mass balance numbers show the distribution of spilled synthetic oil or diluted bitumen in four environmental compartments (evaporated, on the water surface, dispersed in the water column and on the shoreline) over time (3 to 15 days depending on the example). The mass balance quantifies the hydrocarbon chemistry and weathering processes discussed in Section 6. GRPs will be developed before the start of operations of the marine terminal. These plans provide opportunities for coastal Aboriginal organizations, government agencies, communities and directly affected stakeholders to prioritize response activities to protect sensitive biophysical and human areas. GRPs would include information similar to what is provided below for each mass balance example, but at a greater level of detail than what is shown here.

Page 12-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.5 Example 1: Wintertime Spill at Emilia Island This hypothetical example simulates a spill of 10,000 m3 of synthetic oil in Douglas Channel just south of Emilia Island as a result of tanker grounding. The spill is assumed to last 13 hours, with a higher percentage of the total volume spilled immediately, and slower release over the subsequent 12 hours (i.e., likely mass balance of the synthetic oil without a boom in place). The example illustrates the unmitigated fate of the hydrocarbon over three days. And is premised on the ship remaining at the spill site (as opposed to being taken to a safe haven) over the total release time. A February spill results in interactions with the greatest range of marine organisms (see Section 10) and represents wind and currents characteristic of winter outflow conditions in Douglas Channel. Typical winds at that time of year are northeast at 0 to 12 m/s, and would generate waves that could mix some of light synthetic into the upper water column. The model assumes current speeds in the range of 5 to 34 cm/s. This segment of Douglas Channel is approximately 3-km wide, with water depths greater than 100 m. Emilia Island is located east of Drumlummon Bay, west of Maitland Island and just south of the coastal boundary of the Foch-Gilttoyees Park and Protected Area. It is northeast of the Stair Creek Conservancy.

12.5.1 Spill Characteristics The hypothetical examples are based on an unmitigated spill (in other words, there is no allowance for standard mitigation measures, such a containment boom or oil removal). As a result, these examples overestimate the quantity and distribution of hydrocarbons likely to interact with the environment. As part of a precautionary approach, the assessment of potential effects is based on examples that are worse than are likely to occur (as mitigation measures would be applied). In this mass balance example, over the first eight hours winds and currents would transport synthetic oil south and west down Douglas Channel. Oil is predicted to reach the northern shore of Douglas Channel approximately 8 hours after the spill and, after 24 hours, would be stranded from the initial point of contact across from Grant Point on the southwest end of Maitland Island down to Kiskosh Inlet, with a total affected shoreline of 12 km. After 40 hours, estuarine currents from Kitkiata and Kiskosh inlets would transport some synthetic oil across Douglas Channel, affecting approximately 3 km on the southwest shore of Hawkesbury Island. No synthetic oil is predicted to remain on the water surface after 57 hours. The mass balance developed for synthetic oil and its fate over 72 hours is shown in Table 12-1 and Figure 12-2. Initially, all the oil is predicted to be on the water surface, but evaporation would begin immediately. The oil would move southwesterly and not be expected to contact nearby park and protected areas. After 72 hours, no oil is predicted to be on the surface; approximately 38% would be on the shoreline, 40% naturally dispersed into the water column and 22% evaporated. With no mitigation, the synthetic oil would move southwest reaching just beyond Hartley Bay.

April 2010 FINAL Page 12-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-1 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill off Emilia Island during Winter

Environmental Compartment Hours After Evaporated into Spill Water Surface Air Ashore Water Column (%) (%) (%) (%) 4 80 11 0 8 6 75 13 0 12 12 57 15 5 22 18 22 19 27 33 24 7 20 36 37 48 0 22 38 40 72 0 22 38 40

100

75 Water Surface Water Column Ashore Evaporated % 50

25

0 0 12 24 36 48 60 72 Hours After Release

Figure 12-2 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill off Emilia Island during Winter

Page 12-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.5.2 Mitigation Key responses for a spill at Emilia Island are listed in Table 12-2.

Table 12-2 Response for Emilia Island Example

Activity Description Assumptions Safe to commence initial response operations The cause of the spill has no effect on the size or duration of the spill event No injuries are associated with the incident Safety All personnel accounted for and no injuries. Emergency responders assure safety, proper PPE and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and spill response actions Secure spill source Contain and recover oil on water Protect priority sensitive areas Coordinate response actions with Unified Command Procedures to The vessel Captain identifies the compromised tanks and begins tank-to-tank stop a discharge transfers to remove oil from damaged tanks. Emergency lightering pumps on escort tugs are used to begin the offload of cargo from damaged tanks to internal tanks on tugs or barges and are transferred to the Kitimat Terminal Notifications The Captain notifies the RO Dispatcher, CCG and Northern Gateway, as dictated in the Marine OSRP. The RO Dispatch notifies the RO Duty Officer, Spill Response Team and Terminal Supervisor. Northern Gateway, OSC requests notification be made (or confirmed) to CCG, Environment Canada, and British Columbia Ministries. Fire prevention No ignition source or open lights are present, other marine craft are kept clear of and control the area. In the event of fire, the Captain and escort tugs activate the Marine Fire Response Teams and notify the Kitimat Fire Department. Discharge Visual assessment is maintained by crew on vessel, responders and shore tracking personnel. The OSC dispatches a helicopter or aircraft for overflight documentation, tracking, and initial guidance to on-water oil spill containment and recovery operations. In addition, oil pathways begin using real-time weather and oceanographic conditions, and oil weathering properties. Pathway examples are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground-truth for forecasting. Protection of Responders refer to the Marine OSRP and to pertinent GRPs to identify priority environmentally sensitive areas near or in the pathway of the moving synthetic oil. Targeted sensitive areas priority protection sites, subject to Unified Command approval, are: estuary of the Foch-Gilttoyees Protected Area Kitkiata and Kiskosh inlets herring spawning areas across from Kiskosh Inlet Aboriginal communities at Hartley and Malsey bays shorelines of Stair Creek Conservancy boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command

April 2010 FINAL Page 12-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-2 Response for Emilia Island Example (cont’d)

Activity Description Containment and A decision is made by the IC whether the ship remains at the incident location or control strategies is relocated to an alternate location or safe haven. Initial containment is provided around the vessel through deployment of boom aboard the escort tugs. Response teams from RO bases are deployed to add additional secondary containment boom. Floating storage barges and/or other tankers are dispatched to site to provide capacity for emergency lightering operations. Oil recovery Skimmers and pumps on the escort tugs and from RO bases are used to recover strategies oil from inside containment boom. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is limited from reaching shorelines by directing sweep and recovery operations from overflights. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline. Shoreline cleanup SCAT teams deployed for assessment and cleanup recommendations. strategies Cleanup endpoints defined and submitted to CCG for oiled shore types. Primary oiled shoreline types, based on trajectories, will consist of rock cliff and rock with gravel beach. Oiled segment cleanup undertaken as defined in the IAP based on SCAT team recommendations. Recommended cleanup techniques for the two primary shoreline types are: Rock cliff (spot washing, natural cleaning) Rock with gravel beach (spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas) Segments are inspected for meeting endpoints and signed-off, if so. Recovered oil All wastes are segregated into oiled versus non-oiled and liquid versus solid. transfer and Recovered oily water mixture is pumped to available tanks on vessels, barges, Storage bladders and to shore. Oily liquids are transferred from floating storage to available tanks at the Kitimat Terminal. Oily solids are handled in specified drums. Wildlife protection Wildlife at risk in the immediate spill area is assessed. Wildlife hazing kit is readied in case of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate capture and hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area.

Page 12-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Meteorological conditions will strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures will be based on hydrocarbon behaviour and distribution, degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation is immediate deployment of containment booms around the tanker, using equipment on the escort tugs. In this winter mass balance example for Emilia Island, wind during the three-day period blows from the northeast at speeds up to 12 m/s, which would generate waves that mix some oil into the upper water column and generate surface currents that would move floating oil. In this region, currents are driven by a combination of tides, fresh water input and winds, and for this example are modelled at 5 to 34 cm/s. Given these conditions, responders would have eight hours to place exclusion booms around sensitive habitats and areas such as: an embayment at Kitkiata Inlet to reduce effects on Aboriginal communities at Kitkiata Creek and Quaal River and on shoreline bird habitat herring spawning areas on the eastern shore of Douglas Channel across from Kiskosh Inlet a 1.5-km-wide opening from Halsey Point to Dawson Point to reduce effects on the estuary and Aboriginal communities at Malsey Bay near the Gabion River shores of Stair Creek Conservancy estuary of the Foch-Gilttoyees Protected Area Aboriginal communities at Hartley and Malsey bays

12.5.3 Potential Effects on Key Resources at Risk Unmitigated effects described below illustrate conservative estimates of potential situations for the biophysical and human environment. Booming around the tanker would contain some of the oil, skimmers and booms would be used to remove additional oil and exclusion or redirection booming at sensitive areas would protect many areas described below.

12.5.3.1 Unmitigated Effects on the Biophysical Environment The aquatic organisms most vulnerable to a spill at Emilia Island during winter are those that use shoreline habitat, as discussed generally in Section 7. In addition, attention would be focused on species of conservation concern (e.g., Marbled Murrelet, Surf Scoter, killer whale and humpback whale). The affected shore would be predominately rock cliff and rock and coarse-grained beaches. The latter shorelines have the potential for penetration and remobilization of oil. The synthetic oil could potentially reach Kitkiata Inlet, a shoreline bird area near Kitkiata Creek. It is unlikely to reach prime Marbled Murrelet areas because these birds tend to forage in shallow protected inlets and bays away from the predicted movement of the synthetic oil. Herring could be affected because they could be returning to spawn along shorelines across from Kiskosh Inlet in February, although the bulk of the fish would return in March. Surf scoters congregate where herring spawn and they winter off nearby Hawkesbury Island. Synthetic oil could also reach Gabion River, where there is a salmon hatchery. The southern extent of the

April 2010 FINAL Page 12-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning spread of synthetic oil would reach beyond Hartley Bay to Dawson Point, where herring spawn and shoreline birds congregate. Synthetic oil on the water surface, dispersed in water and coating the shoreline would result in short term effects on water quality and potentially longer-term effects on sediment quality. Eulachon migrating in February might be exposed to, and affected by, hydrocarbon fractions dissolved in the water column. Organisms in contact with shoreline oil could include rockweed, kelp and other algae, intertidal marine invertebrates and shorebirds. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Diving birds and marine mammals could come in contact, inhale or ingest oil, from cleaning their fur or feathers, and might be at risk of hypothermia. These organisms would be most vulnerable while oil is on the surface in the first 48 hours, particularly as the channel is relatively confined in the area affected by the spill. Terrestrial mammals such as river otters, black-tailed deer, bears and wolves, or birds such as ravens or crows, which feed and scavenge along the shoreline could come into contact with stranded oil.

12.5.3.2 Unmitigated Effects on the Human Environment Hydrocarbons could affect heritage resources and traditional marine uses in the intertidal and shoreline regions. A spill at Emilia Island during winter would likely affect marine food resources; in turn, these changes could affect current and future well-being, and ability of Aboriginal groups to exercise their rights. In addition to mortality of some fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting (for example of herring or eulachon moving into the area to spawn). Groundfish species are landed year round. Aboriginal groups would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. In this example, Aboriginal groups at Kitkiata Creek and Quaal River (in the Kitkiata Inlet) and at Malsey Bay near the Gabion River would be particularly vulnerable to the effects from synthetic oil. It is difficult to generalize the effects because no consultations have occurred regarding traditional uses in these areas, which in addition to harvesting of food, could include areas of cultural and sacred importance or periodic habitation areas. A spill might affect heritage resource sites through contamination with oil or through physical damage associated with cleanup, although there is little documentation of heritage resources in the area. A spill at Emilia Island would have minor effects on non-traditional marine uses at Hartley Bay and Kitkatla Inlet because winter is not typically a peak recreational season. The likely effects would be aesthetic disturbances and restricted access to shorelines and marinas during the cleanup. The channel width is wide enough to allow vessel traffic to avoid the spill; however, the marina at Malsey Bay and vessels in contact with oil would be fouled. Given that hydrocarbons could move down-channel 20 km as far as Hartley Bay, that community could experience temporary disruption of vessel traffic, and loss of local fish and shellfish resources over one season or more (e.g., from herring mortality or contamination of shellfish). As noted earlier, Northern Gateway has offered coastal Aboriginal organizations the option of directly participating in the RO for emergency events. Assuming this is the case, trained responders, vessels and equipment from coastal communities would be engaged immediately in the response.

Page 12-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.6 Example 2: Summertime Spill at Principe Channel The second hypothetical example simulates a spill of 10,000 m3 of diluted bitumen midway along Principe Channel, off Anger Island, in July, as a result of tanker grounding. It is assumed the bitumen is spilled over a period of 13 hours, with approximately 20% of it released immediately and the remainder being slowly released over the next 12 hours. A July spill would interact with the greatest range of aquatic organisms (see Section 10) and represents wind and currents characteristic of summer inflow conditions in Principe Channel. Typical winds at that time of year are southeast from 0 to 9 m/s and would generate waves that might mix some oil into the upper water column. Modelled current speeds are in the range of 5 to 27 cm/s for the assumed conditions. This segment of Principe Channel is a relatively narrow stretch ranging from 2- to 3-km wide, with chartered depths in excess of 100 m. Anger Island contains numerous inlets. Its northwestern shores form a boundary for Aboriginal and commercial fisheries.

12.6.1 Spill Characteristics The hypothetical examples are based on unmitigated spills (no allowance for standard mitigation measures, such a containment boom or oil removal). As a result, these examples overestimate the quantity and distribution of hydrocarbons likely to interact with the environment. As part of a precautionary approach, the assessment of potential effects is based on examples that are worse than are likely to occur (as mitigation measures would be applied). In this mass balance example, winds and currents would transport oil north and east up Principe Channel over the first eight hours. Bitumen would first reach Anger Island approximately 10 hours after the spill. By the end of 24 hours, diluted bitumen is predicted to be stranded along 8 km of Anger Island, 17.5 km of Banks Island, and 2 to 3 km of shoreline on the southwest coast of McCauley Island. Over the next two days, the bulk of surface oil is predicted to move farther northwest up Principe Channel, stranding along the southwest coast of McCauley Island and a total of 25 km of Banks Island. At the end of the fourth day a shift in winds would transport the remaining surface oil to the southeast, further affecting Banks Island. The mass balance analysis shows the predicted distribution of diluted bitumen over an eight-day period (see Table 12-3 and Figure 12-3). Initially, all it would be on the water surface, but evaporation would begin immediately. The diluted bitumen is predicted to flow northwest, skirting the edge of Aboriginal and commercial fisheries areas. The first shoreline contact would occur at hour 10. By the end of day eight, none of the bitumen is predicted to remain on the water surface, approximately 83% of the bitumen would be ashore, 15% would be lost through evaporation and 2% would be dispersed into the water column. With no mitigation, the bitumen would move as far north as Whalen Point.

April 2010 FINAL Page 12-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-3 Mass Balance Estimates for a Hypothetical 10,000 m3 Diluted Bitumen Spill in Principe Channel during Summer

Environmental Compartment Evaporated Days After Spill Water Surface into Air Ashore Water Column (%) (%) (%) (%) 0.25 98 2 0 0 0.5 88 3 9 1 1 41 7 52 1 3 15 11 72 2 5 4 13 81 2 6 0 14 84 2 7 0 14 84 2 8 0 15 83 2

100

75

% 50 Water Surface Water Column Ashore Evaporated 25

0 0 1 2 3 4 5 6 7 8 9 Days After Release

Figure 12-3 Mass Balance Estimates for a Hypothetical 10,000 m3 Diluted Bitumen Spill in Principe Channel During Summer

Page 12-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.6.2 Mitigation Key mitigations, assumptions, safety, initial response objectives and activities, protection of environmentally sensitive areas and species, containment and control strategies and cleanup responses are described in Table 12-4.

Table 12-4 Response for Principe Channel Example

Activity Description Assumptions Safe to commence initial response operations The cause of the spill has no effect on the size or duration of the spill event No injuries are associated with the incident Safety All personnel accounted for and no injuries. Emergency responders assure safety, proper PPE and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and response actions Secure spill source Contain and recover oil on water Protect priority sensitive areas Coordinate response actions with Unified Command Procedures to stop a The vessel Captain identifies the compromised tanks and begins tank-to- discharge tank transfers to remove oil from damaged tanks. Emergency lightering pumps on escort tugs are used to begin offload of cargo from damaged tanks to internal tanks on tugs or barges and are transferred to the Kitimat Terminal. Notifications The Captain notifies the RO Dispatcher, CCG and Northern Gateway, as dictated in the Marine OSRPs. The RO Dispatch notifies the RO Duty Officer, Spill Response Team and Terminal Supervisor. Northern Gateway, OSC requests notification be made (or confirmed) to the CCG, Environment Canada and British Columbia Ministries. Fire prevention and No ignition source or open lights are present, other marine craft are kept control clear of the area. In the event of fire, the Captain and escort tugs activate the Marine Fire Response Teams and notify the Kitimat Fire Department. Discharge tracking Visual assessment is maintained by crew on vessel, responders and shore personnel. The OSC dispatches a helicopter or aircraft for overflight documentation, tracking and initial guidance to on-water containment and recovery operations. In addition, oil pathways begin using real-time weather and oceanographic conditions, and oil weathering properties. Oil pathway examples are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground- truth for forecasting. Protection of Responders refer to the Marine OSRP and to pertinent GRPs to identify environmentally priority sensitive areas near in the pathway of the bitumen. Targeted priority sensitive areas protection sites, subject to Unified Command approval, are: Anger Anchorage area and fishery areas Dory Passage, Alexander Shoal and Canaveral Passage Boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command.

April 2010 FINAL Page 12-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-4 Response for Principe Channel Example (cont’d)

Activity Description Containment and Initial containment is provided around the vessel through deployment of control strategies boom aboard the escort tugs. Response teams from RO bases are deployed to add additional secondary containment boom Floating storage barges and/or other tankers are dispatched to site to provide capacity for emergency lightering operations. Oil recovery strategies Skimmers and pumps on the escort tugs and from RO bases are used to recover oil from inside containment boom. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is limited from reaching shorelines by directing sweep and recovery operations from overflights. Re-direction boom may be used to intercept and divert oil to the shoreline for recovery along relatively non-permeable shorelines. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline (central portion of channel). Shoreline cleanup SCAT teams deployed for assessment and cleanup recommendations. strategies Cleanup endpoints defined and submitted to CCG for oiled shore types. Primary oiled shoreline types, based on trajectories, consist of rock cliff and rock with gravel beach. Oiled segment cleanup undertaken as defined in the IAP based on SCAT team recommendations. Recommended cleanup techniques for the three primary shoreline types are: Rock cliff (spot washing, natural cleaning) Rock with gravel beach (spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas) Rock, sand, and gravel beach (spot washing and vacuum, flushing and recovery, tilling, berm relocation, manual cleaning for small areas) Segments inspected for meeting endpoints and signed-off, if so. Recovered oil transfer All spill response wastes are segregated into oiled versus non-oiled and and storage liquid versus solid. Recovered oily water mixture is pumped to available tanks on vessels, barges, bladders and to shore. Oily liquids are transferred from floating storage to the 10,000-m3 oil tank at the Kitimat Terminal. Oily solids are handled in specified drums. Wildlife protection Wildlife at risk in the immediate spill area are assessed. Wildlife hazing kit is readied in case of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate capture and hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area.

Page 12-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Meteorological conditions will strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures would be based on hydrocarbon behaviour and distribution, degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation for any spill is immediate deployment of containment booms around the tanker, using equipment on the escort tugs. In this summer mass balance example for Principe Channel, wind during the first four days is predicted to blow from the southeast at speeds of 0 to 9 m/s, then switch to northwest for the next four days, with maximum speed of 5.7 m/s. Wind blowing over the water surface would generate waves that mix some diluted bitumen into the upper water column and surface currents that would transport floating bitumen. The currents are driven by a combination of tides, fresh water input and winds. Given these conditions, responders would have 10 hours to place exclusion or redirection booms at sensitive areas such as: the western point of Anger Island and along the boundaries of Aboriginal and commercial fisheries areas Canaveral Passage, Alexander Shoal and the southern shores of the island adjacent to Meet Point, which are shorebird areas

12.6.3 Potential Effects on Key Resources at Risk Unmitigated effects described below illustrate conservative estimates of potential situations for the biophysical and human environment. Booming around the tanker would contain some of the oil, skimmers and booms would be used to remove additional oil and exclusion or redirection booming at sensitive areas would protect many areas described below.

12.6.3.1 Unmitigated Effects on the Biophysical Environment The aquatic organisms most vulnerable to a spill in Principe Channel during summer would be those that use shoreline habitat. In addition, attention would be focused on species of conservation concern (e.g., Marbled Murrelet, Surf Scoter, killer whale and humpback whale). The affected shore would be predominately rock cliff, rock and coarse-grained (gravel) beaches, and rock, sand, and gravel mixed beaches. The latter two have the potential for penetration and remobilization of oil where oil might persist. In this example, the diluted bitumen would reach the western banks of Anger Island and flow northwest, reaching Whalen Point midway up McCauley Island. The southern part of McCauley Island is prime shorebird habitat (e.g., Marbled Murrelet). Anger Island is characterized by numerous inlets and sustains Aboriginal and commercial fisheries. Diluted bitumen on the water surface, dispersed in water and coating the shoreline, would result in short- term effects on water quality, and potentially longer-term effects on sediment quality. Organisms in contact with shoreline oil would include rockweed, kelp and other algae and intertidal marine invertebrates. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Marine mammals and birds tend to follow prey. During this time of year, salmon would be migrating into Douglas Channel and other areas, which could increase the potential for presence of predators. Animals

April 2010 FINAL Page 12-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning could inhale or ingest oil from cleaning their fur or feathers. Oiled fur or feathers pose the risk of hypothermia. A spill could have large effects on the Marbled Murrelet population because the birds moult during summer and cannot fly to escape the oil. Marine mammals would be most vulnerable when bitumen is on the surface, mostly in the first 48 hours, because the confined channel limits oil-free areas as the whales breach for air. Terrestrial mammals such as river otters, black-tailed deer, bears and wolves, or birds such as ravens or crows, which feed and scavenge along the shoreline could come into contact with stranded oil.

12.6.3.2 Unmitigated Effects on the Human Environment Diluted bitumen moving into intertidal and shoreline regions could affect heritage resources and traditional marine uses. A spill in Principe Channel during summer would likely affect marine food resources; in turn, these changes could present a temporary limitation to the ability of Aboriginal groups to exercise their fishing and harvesting rights. In addition to mortality of some fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting (for example salmon moving into the area to spawn). Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook); there are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. Aboriginal groups would be particularly sensitive to the effects of a spill because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. In this example, the southern border of an Aboriginal fishery near the western shores of Anger Island would be particularly vulnerable to the effects from a spill. It is difficult to generalize the effects because no information has been provided regarding traditional uses in these areas, which in addition to harvesting of food, could include areas of cultural and sacred importance or periodic habitation areas. A spill might affect heritage resource sites through oil contact or through physical damage associated with spill cleanup, although there is little documentation of heritage resources in the area. The possible effects would be aesthetic disturbances and restricted access to shorelines and marinas during the cleanup. The channel is wide enough to allow vessel traffic to avoid the spill; however, vessels and marine infrastructure in contact with oil would be fouled and require cleanup and decontamination. Given that the diluted bitumen could move up-channel more than 20 km, as far as Whalen Point, communities in the area could experience temporary disruption of vessel traffic and loss of local fish and shellfish resources over one season or more (e.g., from salmon mortality or contamination of shellfish). As noted earlier, Northern Gateway has offered coastal Aboriginal organizations the option of owning and directly participating in the RO for emergency events. Assuming this is the case, trained responders, vessels and equipment from coastal communities would be engaged immediately in the response.

12.7 Example 3: Summertime at Wright Sound This hypothetical mass balance example simulates a spill of 36,000 m3 of diluted bitumen in Wright Sound, midway between Promise Island and Gil Island, in July, as a result of a tanker collision. It is assumed that two compartments on a VLCC are ruptured, despite the double hull configuration, and that all the diluted bitumen in the two compartments eventually drains. The mass balance model assumes the

Page 12-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning diluted bitumen is spilled over a period of 13 hours, with most of it released immediately, and the rest more slowly over the next 12 hours. A July spill would interact with the greatest range of marine organisms and the conditions modelled describe conditions typical of Wright Sound in the summer, with inflow winds opposing the seaward-directed estuarine flow at the surface. Winds would blow mostly from the south-southwest at speeds varying from 0.5 to 10.5 m/s, with short periods of northerly winds. Wright Sound has a width of about 3.7 km at the western end where it meets Grenville Channel and up to 5.5 km at the eastern end where the Sound opens up to Douglas Channel and Verney Passage. Wright Sound has water depths in excess of 360 m.

12.7.1 Spill Characteristics The hypothetical examples are based on unmitigated spills (no allowance for standard mitigation measures, such a containment boom or oil removal). As a result, these examples overestimate the quantity and distribution of hydrocarbons likely to interact with the environment. As part of a precautionary approach, the assessment of potential effects is based on examples that are worse than are likely to occur (as mitigation measures would be applied). In this mass balance example, over the first 15 hours, persistent estuarine outflow, aided by ebbing tidal currents, is predicted to transport oil south and west toward Fin Island. Diluted bitumen would first reach Fin Island approximately eight hours after the spill. As tidal currents reverse and winds from the south- southwest continue, it is predicted that bitumen will be stranded on southern Farrant Island, northwestern Gil Island, and in the region of Hartley Bay north of the spill site between 13 and 24 hours after the spill. Ebbing tidal currents would transport oil through Cridge and Lewis Passages but opposing winds would keep most of the surface oil in Squally Channel between 24 and 48 hours after the spill. Bitumen would also enter Hartley Bay and southern Grenville Channel approximately 35 hours after the spill. By the end of Day 3, oil is predicted to have spread farther up Grenville Channel and enter Whale Channel. By Day 5, oil would be stranded on Campania and Gil Islands and 200 km of shoreline would be affected. Starting on Day 5 the wind is modelled to begin to reduce in velocity and change to northerly and the tide would carry some of the bitumen out to open waters through Caamaño Sound. By Day 15, less than 1% of the bitumen is predicted to remain on the water surface, 6% would be in the water column, 76% would be on the shore, and about 17% would have evaporated. The mass balance developed for diluted bitumen over 15 days is shown in Table 12-5 and Figure 12-4. Initially, all the bitumen would be on the water surface. Evaporation would begin immediately and account for approximately 17% of the loss of volume after 15 days. The diluted bitumen would disperse in all directions, but the vast majority is predicted to move southwest through Squally Channel and beyond. The first shoreline contact would occur at Fin Island eight hours after the spill. With no mitigation, the diluted bitumen is predicted to move in multiple directions: north along the western shores of Hawkesbury Island, west through Otter Channel and up Principe Channel and down Estevan Sound, northwest up Grenville Channel, and south through Squally Channel to Campania and Caamaño Sounds reaching Rennison Island.

April 2010 FINAL Page 12-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

100

75

% 50 Water Surface Water Column Ashore Evaporated 25

0 0 2 4 6 8 10 12 14 Days After Release

Figure 12-4 Mass Balance Estimates for a Hypothetical 36,000 m3 Diluted Bitumen Spill at Wright Sound during Summer

Table 12-5 Mass Balance Estimates for a Hypothetical 36,000 m3 Diluted Bitumen Spill at Wright Sound during Summer

Environmental Compartment Days After Spill Water Surface Evaporated Ashore Water Column (%) (%) (%) (%) 1 74 8 17 1 2 53 10 34 3 3 45 11 40 3 4 33 12 51 4 5 26 13 56 5 6 21 14 60 5 7 16 14 64 6 8 9 15 71 6 9 6 16 72 6 10 4 16 74 6 11 3 16 75 6

Page 12-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-5 Mass Balance Estimates for a Hypothetical 36,000 m3 Diluted Bitumen Spill at Wright Sound during Summer (cont’d)

Environmental Compartment Days After Spill Water Surface Evaporated Ashore Water Column (%) (%) (%) (%) 12 1 17 76 6 13 1 17 76 6 14 1 17 76 6 15 1 17 76 6

12.7.2 Mitigation Key mitigations, assumptions, safety, initial response objectives and activities, protection of environmentally sensitive areas and species, containment and control strategies and cleanup responses are described in Table 12-6.

Table 12-6 Response for Wright Sound Example

Activity Description Assumptions Safe to commence initial response operations The cause of the spill has no effect on the size or duration of the spill event No injuries are associated with the incident Safety All personnel accounted for and no injuries. Emergency responders assure safety, PPE and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and response actions. Secure spill source. Contain and recover oil on water. Protect priority sensitive areas. Coordinate response actions with Unified Command. Procedures to stop a The vessel Captain decides whether the a spill (or risk of fire and discharge explosion) may be contained by keeping the ships interlocked; if the ships are separated, attempts to contain the spill and stop sea water entering the cargo tanks and if possible begins tank-to-tank transfers to remove oil from damaged tanks. Emergency lightering pumps on escort tugs may be used to assist tank to tank transfers on the ship, or to begin offload of cargo from damaged tanks to internal tanks on tugs or to oil barges. Notifications The Captain notifies the RO Dispatcher, CCG and Northern Gateway, as dictated in the Marine OSRP. The RO Dispatch notifies the RO Duty Officer, Spill Response Team and Terminal Supervisor. The OSC requests notification be made (or confirmed) to CCG, Environment Canada and British Columbia Ministries. Given the size of the spill, Northern Gateway mobilizes Tier 3 international resources from Alaska, eastern Canada, OSR Southampton (UK), and US West Coast.

April 2010 FINAL Page 12-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-6 Response for Wright Sound Example (cont’d)

Activity Description Fire prevention and No ignition source or open lights are present. In the event of fire, the control Captain and escort tugs activate the Marine Fire Response Teams and notify the Kitimat Fire Department. Discharge tracking Visual assessment is maintained by crew on vessel and responders. The OSC dispatches a helicopter or aircraft for overflight documentation, tracking and initial guidance to on-water containment and recovery operations. In addition, pathways begin using real-time weather and oceanographic conditions, and oil weathering properties. Oil pathway examples are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground- truth for forecasting. Protection of Responders refer to the Marine OSRP and to pertinent GRPs to identify environmentally priority sensitive areas near in the pathway of the diluted bitumen. Targeted sensitive areas priority protection sites, subject to Unified Command approval, are: Fin Island: Curlew, Hawk, and Brant bays Gil Island: Fisherman Cove and Turtle Point Conservancy Gill Island: Crane Bay Ashdown Island: Steller sea lion haulout (south shore) Hinton Island (inlet) and Tuwartz Inlet-Narrows Major salmon rivers at risk of tidal flooding Boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command. Containment and Initial containment is provided around the vessel through deployment of control strategies boom aboard the escort tugs. Response teams from RO bases are deployed to add additional secondary containment boom Floating storage barges and/or other tankers are dispatched to site to provide capacity for emergency lightering operations. Oil recovery strategies Skimmers and pumps on the escort tugs and from RO bases are used to recover oil from inside containment boom. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is limited from reaching sensitive shorelines by directing sweep and recovery operations from overflights. Re-direction boom may be used to intercept and divert oil to the shoreline for recovery along relatively non-permeable shorelines. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline. Dispersant stockpile sites and application aircraft are mobilized from the locations in the US.

Page 12-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-6 Response for Wright Sound Example (cont’d)

Activity Description Shoreline cleanup SCAT teams are deployed for assessment and cleanup recommendations. strategies Cleanup endpoints are defined and submitted to CCG for oiled shore types. Primary oiled shoreline types, based on trajectories, consist of rock cliff and rock with gravel beach. Oiled segment cleanup undertaken as defined in the IAP based on SCAT team recommendations. Recommended cleanup techniques for the three primary shoreline types are: Rock cliff (spot washing, natural cleaning) Rock with gravel beach (spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas) Rock, sand, and gravel beach (spot washing and vacuum, flushing and recovery, tilling, berm relocation, manual cleaning for small areas) Segments inspected for meeting endpoints and signed-off, if so. Recovered oil transfer All wastes are segregated into oiled versus non-oiled and liquid versus solid. and storage Recovered oily water mixture is pumped to available tanks on vessels, barges, bladders and to shore. Oily liquids are transferred from floating storage to the 10,000 m3 oil tank at the Kitimat Terminal. Oily solids are handled in specified drums. Wildlife protection Wildlife at risk in the immediate spill area are assessed. Wildlife hazing kit is readied in case of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate capture and hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area. Meteorological conditions will strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures would be based on hydrocarbon behaviour and distribution, degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation for a spill is immediate deployment of containment booms around the tanker, using equipment on the escort tugs. In this summer mass balance example for Principe Channel, it is assumed that wind during the 15-day period blows mostly from the south-southwest at speeds of 0.5 to 10.5 m/s, with two brief periods on days 6 to 7 and 11 to 12 where it blows from the north. Wind blowing over the water surface would generate waves that would mix some bitumen into the upper water column and generate surface currents that transport floating bitumen. The currents are driven by a combination of tides, freshwater input and winds. This example is assumed to occur at the end of a flood tide, with modeled current speeds of 2 to 26 cm/s. The example indicates that, under these conditions, responders would have eight hours to place initial protection boom before oil begins to strand on the shoreline. Overflights and bitumen tracking would confirm movement and be used to prioritize protection sites. Sensitive areas along the unmitigated pathway include: shorebird habitat at Fin Island

April 2010 FINAL Page 12-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Fisherman Cove, where shorebirds congregate, and Turtle Point Blackfly Point (Gil Island), a salmon-bearing river Aboriginal fisheries at Sainty and Stephens Point on Promise Island a salmon-bearing river at Tuwartz Inlet salmon-bearing rivers at Crane Bay MacDonald Bay, the opening to a salmon-bearing river prime shorebird habitat at Skinner Islands a salmon-bearing river northwest of Fawcett Point Ashdown Island, a sea lion haulout and habitat for shorebirds a salmon-bearing river at Maple Point the eastern shore of Casanave Passage, which includes Kayel Aboriginal community the eastern shore of Campania Island where Squally Channel meets Campania Sound the western shore of Campania Island, which includes access to a salmon-bearing river and habitat for marine birds an Aboriginal fishery and shorebird habitat at Otter Channel the mouth of a salmon river midway between Wilman Point and Dillon Bay southeastern shores of Dewdney and Glide Islands Ecological Reserve, with a focus on the Porter Islands Glide Islands, east of Estevan Reef rockfish conservation areas between Banks and Trutch islands an Aboriginal fishery at Deer Point the mouth of a salmon river at Ulric Point

12.7.3 Potential Effects on Key Resources at Risk Unmitigated effects described below illustrate conservative estimates of potential situations for the biophysical and human environment. Booming around tankers would contain some of the oil, skimmers and booms would be used to remove additional oil, and exclusion or re-direction booming at sensitive areas would protect many areas described below.

12.7.3.1 Unmitigated Effects on the Biophysical Environment The aquatic organisms most vulnerable to a spill in Wright Sound during the summer are those that use shoreline habitat. In addition, attention would be focused on species of conservation concern (e.g., Marbled Murrelet, Surf Scoter, killer whale and humpback whale) that are in the area at the time of a spill. The affected shore would be predominately rock cliff, rock and coarse-grained (gravel) beaches, and

Page 12-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning rock, sand, and gravel mixed beaches. The latter two have the potential for penetration and remobilization of oil where oil might persist. An unmitigated spill might reach many sensitive and commercially IAs distributed over the 204 km of shoreline where diluted bitumen might strand. Diluted bitumen on the water surface, dispersed in water and coating the shoreline would result in short- term effects on water quality and potentially longer-term effects on sediment quality. Organisms in contact with shoreline oil would include rockweed, kelp and other algae and intertidal marine invertebrates. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Marine mammals and birds tend to follow prey. During this time of year salmon would be migrating, which could increase the potential for presence of predators. Animals could inhale or ingest oil from cleaning their fur or feathers. Oiled fur or feathers pose the risk of hypothermia. Marine mammals would be most vulnerable when oil is on the surface, mostly in the first eight days if unmitigated, and in narrow channels where the confined nature of the area might limit evasion. Terrestrial mammals such as river otters, black-tailed deer, bears and wolves, or birds such as ravens or crows, which feed and scavenge along the shoreline could come into contact with stranded oil.

12.7.3.2 Unmitigated Effects on the Human Environment Bitumen reaching intertidal and shoreline regions l could affect heritage resources and traditional marine uses. Aboriginal groups would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. A major (36,000 m3) spill at Wright Sound during summer would be likely to affect current and future well-being, and the ability of Aboriginal groups to exercise their rights, mainly through effects on the marine food resources. In addition to mortality of fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting (for example salmon moving into the area to spawn). Salmon typically return to spawn in July, and this would also be the prime Aboriginal commercial and recreational fishing season for many species. Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook). There are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. As information regarding traditional uses in these areas has not yet been provided, it is difficult to generalize the effects, which, in addition to harvesting of food, could include areas of cultural and sacred importance or periodic habitation areas. Diluted bitumenl might affect heritage resource sites through contamination or sites could be damaged by cleanup activities, although there is little documentation of heritage resources in the area. A spill at Wright Sound would have effects on non-traditional marine uses at the marinas at Stephens Point since summer is typically a peak recreational season. The likely effects would be aesthetic disturbances and restricted access to shorelines and marinas during the cleanup. The channels are wide enough to allow vessel traffic to avoid the spill; however, vessels and marine infrastructure (e.g., mooring buoys at Stewart Narrows) in contact with oil would be fouled.

April 2010 FINAL Page 12-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Given that diluted bitumen could reach 240 km of shoreline, many communities and Aboriginal reserves would likely be affected. Communities in the area could experience temporary disruption of vessel traffic, loss of local fish and shellfish resources over at least one season (e.g., from salmon fry mortality or contamination of shellfish). As noted earlier, Northern Gateway has offered coastal Aboriginal organizations the option of owning and directly participating in the RO for emergency events. Assuming that this is the case, trained responders, vessels and equipment from the coastal communities would be engaged immediately in the response.

12.8 Example 4: Wintertime Spill at Ness Rock in Caamaño Sound This hypothetical spill example simulates a spill of 10,000 m3 of diluted bitumen in Caamaño Sound in February as a result of tanker grounding on Ness Rock. It is assumed the bitumen is spilled over a period of 13 hours, with approximately 20 % being released within the first hour and the remainder being slowly released over the next 12 hours. Strong southeast winds predominate during the winter months in southern Hecate Strait. Modelled current speeds are 3 to 120 cm/s for the assumed conditions, and are dominated by wind forces. Caamaño Sound is located in southern Hecate Strait, an area exposed to strong winter storms. During successive winter storms, high seas and swells, combined with strong crosswinds and currents, create difficult navigational conditions in Caamaño Sound. Caamaño Sound provides direct access to Queen Charlotte Sound and the open sea for vessels coming from Douglas Channel via Wright Sound, Lewis Passage, Squally Channel and Campania Sound. The indicated navigable channel has a minimum width of 2.5 nm.

12.8.1 Spill Characteristics Characteristics for the hypothetical spill example are based on unmitigated results (no allowance for a containment boom or oil removal). As a result, the predicted quantities and distribution of diluted bitumen are conservative. Winds and currents would transport bitumen north and east towards Dewdney and Trutch Islands for the first 22 hours. Bitumen would first reach Dewdney Island approximately 22 hours after the spill. By the end of Day 3, diluted bitumen would have stranded along the western coastline of Dewdney and Trutch Islands and the southern coast of Banks Island. By this time, the bitumen would have split into two patches, one moving north through Principe Channel and the majority travelling up the west coast of Banks Island. At the end of Day 5, bitumen would be stranded on the western coast of Pitt Island and much of the southwest coast of Banks Island. After 15 days, bitumen would have stranded on portions of the entire west coast of Banks Island. The mass balance analysis shows the predicted distribution of diluted bitumen over a 15-day period (see Table 12-7 and Figure 12-5). During the first day, almost all the bitumen would remain on the water surface except for lighter fractions that evaporate and a small amount that would disperse in the water column. Evaporation would begin immediately resulting in approximately 13% volume loss by Day 15. The first shoreline contact would occur at 22 hours. By the end of Day 15, less than 0.5% of the product

Page 12-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning would remain on the surface and approximately 84% would be ashore. Approximately 3% of the oil would be naturally dispersed into the water column. With no mitigation, the diluted bitumen would move as far north as the top of Banks Island.

Table 12-7 Mass Balance for a Hypothetical 10,000 m3 Diluted Bitumen Spill at Caamaño Sound during Winter

Amount in Each Environmental Compartment 3 Days After (m ) Spill Water Surface Ashore Evaporated Water Column (%) (%) (%) (%) 0.25 99.5 0 0.2 0.3 0.5 99.2 0 0.4 0.4 1 98 0.1 0.9 1 3 8 86 4 2 5 3 88 6 3 15 0.4 83.7 13.1 2.8

100

75

% 50 Water Surface Water Column Ashore Evaporated 25

0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Days After Release

Figure 12-5 Mass Balance for a Hypothetical 10,000 m3 Diluted Bitumen Spill at Caamaño Sound during Winter

April 2010 FINAL Page 12-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.8.2 Mitigation Key mitigations, assumptions, safety, initial response objectives and activities, protection of environmentally sensitive areas and species, containment and control strategies and cleanup responses are described in Table 12-8.

Table 12-8 Response for Caamaño Sound Example

Activity Description Assumptions Safe to commence initial response operations. The cause of the spill has no effect on the size or duration of the spill event. No injuries are associated with the incident. Safety All personnel accounted for and no injuries. Emergency responders assure safety, PPE and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and response actions. Secure spill source. Contain and recover bitumen on water. Protect priority sensitive areas. Coordinate response actions with Unified Command. Procedures to stop The vessel Captain identifies the compromised tank(s) and begins tank-to-tank a discharge transfers to remove oil from damaged tank(s). When tugs and other response vessels arrive, emergency lightering pumps on response vessels are used to begin offload of cargo from damaged tank(s) to internal tanks on response vessels, barges or other vessels. Notifications The Captain notifies the RO Dispatcher, CCG and ENGP (Terminal Control Room), as dictated in the Marine Area OSR Plan. The RO Dispatch notifies the RO Duty Officer, ENGP Spill Response Team and Terminal Supervisor. The ENGP, OSC requests notification be made (or confirmed) to Coast Guard, Environment Canada, British Columbia Ministries, etc. Fire prevention No ignition source or open lights are present. In the event of fire, the Captain and control activates the shipboard Marine Fire Response Team and notifies the CCG. Discharge tracking Visual assessment is maintained by crew on vessel, responders and shore personnel. The OSC dispatches a helicopter or aircraft for overflight documentation, tracking and initial guidance to on-water containment and recovery operations. In addition, oil trajectory modelling begins using real-time weather and oceanographic conditions, and oil weathering properties. Trajectory models are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground-truth for forecasting.

Page 12-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-8 Response for Caamaño Sound Example (cont’d)

Activity Description Protection of Responders refer to the Marine Area OSR Plans and to pertinent GRPs to environmentally identify priority sensitive areas near or in the l pathway of the moving diluted sensitive areas bitumen. Targeted priority protection sites, subject to Unified Command approval, are (from OWA Atlas Maps 8-3 and 8-6; CCAA Maps C-10 to -13, P-1 through -3): Inlets along Dewdney and Glide islands (ecological reserve): Murray Anchorage, Langley Passage Banks Island: Deer Point, Keecha Point, Kooryet Island (First Nations fisheries) Pitt Island: Mink Trap Bay, Patterson Inpet (First Nations fisheries) Boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command. Dispersant use is identified as a viable protection strategy. Containment and Initial containment is provided around the tanker through deployment of boom control strategies from first response vessels and/or aboard the escort tug(s). Response teams from RO bases are deployed to add additional secondary containment boom. Floating storage (barges and/or other tankers) is dispatched to site to provide capacity for emergency lightering operations. Oil recovery Skimmers and pumps on the escort tug(s) and from RO response vessels are strategies used to recover oil from inside containment boom. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is minimized from reaching shorelines by directing sweep and recovery operations from overflights. Re-direction boom may be used to intercept and divert oil to the shoreline for recovery along relatively non-permeable shorelines. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline. Shoreline cleanup SCAT team(s) is deployed for assessment and cleanup recommendations. strategies Cleanup endpoints are defined and submitted to CCG for oiled shore types. Primary oiled shoreline types, based on trajectories, consist of rock cliff and rock with gravel beach. Oiled segment cleanup is undertaken as defined in the IAP based on SCAT team recommendations. Recommended cleanup techniques for the three primary shoreline types are: Rock cliff: spot washing, natural cleaning Rock with gravel beach: spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas Rock, sand, and gravel beach: spot washing and vacuum, flushing and recovery, tilling, berm relocation, manual cleaning for small areas Segments inspected for meeting endpoints and signed-off, if so.

April 2010 FINAL Page 12-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-8 Response for Caamaño Sound Example (cont’d)

Activity Description Recovered oil All wastes are segregated into oiled vs. non-oiled and liquid vs. solid. transfer and Recovered oily water mixture is pumped to available tanks on vessels, barges, storage bladders and to shore. Oily liquids are transferred from floating storage to onshore tanks for additional oil-water separation and recycling. Oily solids are handled in specified drums. Wildlife protection Wildlife at risk in the immediate spill area are assessed. Wildlife hazing kits are readied in case of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate capture / hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area. Meteorological conditions will strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures would be based on hydrocarbon behaviour and distribution, degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation for any spill is immediate deployment of containment booms around the tanker, using equipment on the first response vessels. In this winter mass balance example for Ness Rocks, winds are variable from the south to the southeast at speeds varying from 4.5 to 7.5 m/s. In this region, the currents are driven by the combination of tides, fresh water input, and winds. Wind blowing over the water surface would generate waves that mix some diluted bitumen into the upper water column and surface currents that would transport floating bitumen. Given these conditions, responders would have 22 hours to effect dispersant application operations or to place exclusion or re-direction booms at sensitive areas before first oil contact with the shoreline.

12.8.3 Potential Effects on Key Resources at Risk from a Spill in Caamaño Sound Unmitigated effects described below illustrate predicted serious situations for the biophysical and human environment. Booming around tankers would contain some of the oil, skimmers and booms would be used to remove additional oil, and exclusion or re-direction booming at sensitive areas would protect many areas described below.

12.8.3.1 Unmitigated Effects on the Biophysical Environment The Ness Rock hypothetical spill pathway example is set in an open water area approximately 12 km from any major landmass. For the first 24 hours after the spill almost all the oil would remain on the water surface. The area of exposure risk during the first 24 hours would be the oil water interface and oil air interface in the areas covered by the oil slick. Aquatic organisms with physiological and biological characteristics that require them to use these habitats are at risk of exposure. Several species of marine birds, marine mammal and plankton fall within this category. Although an aquatic organism may be exposed to oil, the consequence of the exposure is related to the organism‘s vulnerability to hydrocarbon exposure. By the third day approximately 86% of the oil would be stranded on shorelines to the northeast

Page 12-30 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning of the hypothetical spill site. From this point on the organisms most vulnerable to exposure would be those that use shoreline habitat. Marine mammals and birds tend to follow prey locally but also undergo seasonal migrations. Caamaño Sound is known to be feeding habitat for several species of marine birds and marine mammals. In general winter distributions of marine mammals in the OWA are poorly understood. The acute exposure potential for cetaceans is limited to the period when oil is on the water surface. For this hypothetical example the exposure window is mostly limited to the first three days. It is commonly thought that most baleen whales undergo migrations between northern summer feeding habitat and southern wintering areas. For this reason the potential acute effects of a winter oil spill in the OWA on baleen whales would be limited. Toothed whales and dolphins are known to be widely distributed throughout the OWA and Caamaño Sound is thought to be an in important area for killer whales. Diving birds and marine mammals could come in contact, inhale or ingest oil, from cleaning fur or feathers, and may be at risk of hypothermia. After the third day, the exposure risk would be greater for species that use shoreline habitats. This would include shore birds, seals, sea lions, terrestrial wildlife, invertebrates and marine vegetation. The first place where oil contacts shoreline in this pathway example is the Dewdney and Glide Islands Ecological Reserve. The ecological reserve contains nesting habitat for several species of birds, including the Cassin‘s auklet, which nest on Glide Islands. In this hypothetical spill pathway example North Danger Rocks, one of three rookeries sites used by Stellar Sea lions, would be oiled by Day 15. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Unmitigated the oil would continue to migrate north and oil shorelines area along the west coast of Banks and the west coast of Pitt Island. Diluted bitumen on the water surface, dispersed in water and coating the shoreline would result in short term effects on water quality and potentially longer term effects on sediment quality. There are several Rock Fish Conservation Areas along the hypothetical spill pathway. Short term effects on water quality could have greater effects on rock fish given the expected concentration in these areas.

12.8.3.2 Unmitigated Effects on the Human Environment The area of this hypothetical mass balance example is considered remote and there is relatively little anthropogenic activity in the area. This is especially true during the winter months as the area is exposed to strong winter storms. There are First Nations reserve lands located along both sides of Principe Channel, on Banks Island and Pitt Island. Several of these reserve lands are associated with salmon rivers. Aboriginal people would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. Bitumen in Caamaño Sound during winter would be likely to affect current and future well-being, and the ability of First Nations to exercise their Aboriginal rights, mainly through effects on the marine food resources. Since no consultations have occurred regarding traditional uses in these areas, it is difficult to generalize the effects, which, in addition to harvesting of food, could include areas of cultural and sacred importance or periodic habitation areas. A spill might affect heritage resource sites through contamination with oil or

April 2010 FINAL Page 12-31

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning through physical damage associated with cleanup, although there is little documentation of heritage resources in the area. In addition to mortality of fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting. Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook). There are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. This area is especially productive for the red sea urchin dive fishery. A spill at Caamaño Sound would have effects on non-traditional marine uses. There are aquacultures sites off the north side of Anger Island. The likely effects could include oiled infrastructure and tainted product. This would have economic considerations. As noted earlier, Northern Gateway has offered coastal Aboriginal organizations the option of owning and directly participating in the RO for emergency events. Assuming that this is the case, trained responders, vessels and equipment from the coastal communities would be engaged immediately in the response.

12.9 Example 5: Summer Spill at Butterworth Rocks in North Hecate Strait This hypothetical example simulates 10,000 m3 of synthetic oil in North Hecate Strait in July, as a result of a tanker grounding at Butterworth Rocks. It is assumed the oil is spilled over a period of 13 hours, with approximately 20 % being released within the first hour and the remainder being slowly released over the next 12 hours. The mass balance example is based on typical summer environmental conditions for North Hecate Strait. Butterworth Rocks was chosen due to its proximity to the Northern Approach. The water west of Butterworth Rocks is 5.5 km wide, with charted depths of 36 to 100 m. Butterworth Rocks is 9.3 km south of the pilot boarding area.

12.9.1 Spill Characteristics Characteristics for the hypothetical example are based on unmitigated results (i.e., no allowance for a containment boom or oil removal). As a result, the predicted quantities and distribution of hydrocarbons are conservative. For this mass balance example winds during the first two days are generally weak and blow mostly from the west at speeds of 2.0 to 5.4 m/s. During Day 2, winds shift to the southeast and increase to 8.4 m/s during Day 3. Calmer winds from the north then prevail through Day 5. Wind blowing over the water surface generates waves that mix part of the oil on the water surface into the upper water column. Wind blowing over the water also generates surface currents that transport floating oil. In this region, currents are driven by a combination of tides, fresh water input, and winds. Modelled current speeds are 2 to 92 cm/s for the assumed conditions, and are dominated by tidal forcing in the first two days, and then by winds on the third and fourth day.

Page 12-32 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Over the first 26 hours, winds and currents would transport oil east towards the Tree Nob Group just north of Stephens Island. Synthetic oil would first reach Stephens Island approximately 30 hours after the spill. By the end of Day 2, synthetic oil would strand along the northern tip of Stephens Island and most of the Tree Nob Group. Higher winds during Day 3 would transport some remaining oil to the northwest, narrowly missing the Dundas Islands. At the end of Day 5, synthetic oil would strand on the entire Tree Nob Group, the northern coast of Stephens Island and some islets off the south coast of Melville Island. The mass balance analysis shows the predicted distribution of synthetic oil over a five-day period (see Table 12-9 and Figure 12-6). Initially all the oil would be on the water surface. Evaporation would begin immediately, resulting in approximately 25% volume loss by Day 5. The first shoreline contact would occur at 26 hours. By the end of Day 5, none of the product would remain on the surface and approximately 33% of the volume would be ashore. Approximately 41% of the oil would be naturally dispersed into the water column by wave action.

100

75 Water Surface Water Column Ashore Evaporated % 50

25

0 0 1 2 3 4 5 Days After Release

Figure 12-6 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill at Butterworth Rocks during Summer

April 2010 FINAL Page 12-33

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-9 Mass Balance Estimates for a Hypothetical 10,000 m3 Synthetic Oil Spill at Butterworth Rocks during Summer

Amount in Each Environmental Compartment 3 Days After (m ) Spill Water Surface Ashore Evaporated Water Column (%) (%) (%) (%) 0.25 86 0 11 3 0.5 81 0 14 5 1 67 0 20 13 3 2 33 25 40 5 0 33 26 41

12.9.2 Mitigation Key mitigations, assumptions, safety, initial response objectives and activities, protection of environmentally sensitive areas and species, containment and control strategies and cleanup responses are described in Table 12-10.

Table 12-10 Response for Butterworth Rocks Example

Activity Description Assumptions Safe to commence initial response operations. The cause of the spill has no impact on the size or duration of the spill event. No injuries are associated with the incident. Safety All personnel accounted for and no injuries. Emergency responders assure safety, proper personal protective equipment (PPE) and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and spill response actions. Secure spill source. Contain and recover oil on water. Protect priority sensitive areas. Coordinate response actions with Unified Command. Procedures to The vessel Captain identifies the compromised tank(s) and begins tank-to-tank stop a transfers to remove oil from damaged tank(s). It is assumed that tethered escort discharge tugs are not used in open waters. When tugs and other response vessels arrive, emergency lightering pumps on response vessels are used to begin offload of cargo from damaged tank(s) to internal tanks on response vessels, barges, or other vessels. Notifications The Captain notifies the RO Dispatcher, CCG and Northern Gateway (Terminal Control Room), as dictated in the Marine Area OSR Plan. The RO Dispatch notifies the RO Duty Officer, Northern Gateway Spill Response Team and Terminal Supervisor. The Northern Gateway, OSC requests notification be made (or confirmed) to Coast Guard, Environment Canada, BC Ministries, etc. Fire prevention No ignition source or open lights are present. In the event of fire, the Captain and and control activates the shipboard Marine Fire Response Team and notifies the CCG.

Page 12-34 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-10 Response for Butterworth Rocks Example (cont’d)

Activity Description Discharge Visual assessment is maintained by crew on vessel, responders and shore tracking personnel. The OSC dispatches a helicopter or aircraft for overflight documentation, tracking and initial guidance to on-water oil spill containment and recovery operations. In addition, oil trajectory modelling begins using real-time weather and oceanographic conditions, and oil weathering properties. Trajectory models are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground-truth for forecasting. Protection of Responders refer to the Marine Area OSR Plans and to pertinent GRPs to identify environmentally priority sensitive areas near in the spill pathway. Targeted priority protection sites, sensitive areas subject to Unified Command approval, are (from OWA Atlas Map 9-2): First Nations sites in the Tree Nob Group and on Melville and Stephens islands Sea lion haulouts Boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command. Dispersant use is identified as a viable protection strategy. Containment Initial containment is provided around the vessel through deployment of boom from and control first response vessels and/or aboard the escort tug(s). strategies Response teams from RO bases are deployed to add additional secondary containment boom Floating storage (barges and/or other tankers) are dispatched to site to provide capacity for emergency lightering operations. Oil recovery Skimmers and pumps on the escort tug(s) and from RO response vessels are used strategies to recover oil from inside containment boom. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is minimzed from reaching shorelines by directing sweep and recovery operations from overflights. Re-direction boom may be used to intercept and divert oil to the shoreline for recovery along relatively non-permeable shorelines. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline. Shoreline SCAT team(s) deployed for assessment and cleanup recommendations. cleanup Cleanup endpoints defined and submitted to CCG for oiled shore types. Primary oiled strategies shoreline types, based on trajectories, consist of rock cliff and rock with gravel beach. Oiled segment cleanup undertaken as defined in the IAP based on SCAT team recommendations. Recommended cleanup techniques for the three primary shoreline types are: Rock cliff: spot washing, natural cleaning Rock with gravel beach: spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas Rock, sand, and gravel beach: spot washing and vacuum, flushing and recovery, tilling, berm relocation, manual cleaning for small areas Segments inspected for meeting endpoints and signed-off, if so.

April 2010 FINAL Page 12-35

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-10 Response for Butterworth Rocks Example (cont’d)

Activity Description Recovered oil All spill response wastes are segregated into oiled vs. non-oiled and liquid vs. transfer and solid. storage Recovered oily water mixture is pumped to available tanks on vessels, barges, bladders and to shore. Oily liquids are transferred from floating storage to onshore tanks for additional oi-water separation and recycling. Oily solids are handled in specified drums. Wildlife Wildlife at risk in the immediate spill area are assessed. Wildlife hazing kits are protection readied in case of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate capture / hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area. Meteorological conditions will strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures would be based on hydrocarbon behaviour and distribution, degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation is immediate deployment of containment booms around the tanker, using equipment on the escort tugs. In this summer mass balance example for Butterworth Rocks, wind during the five day period would blow from the west and then southeast at speeds up to 8 m/s, which would generate waves that mix some oil into the upper water column and generate surface currents that would move floating oil. In this region, currents are driven by a combination of tides, fresh water input and wind. Modeled current speeds are in the range of 2 to 92 cm/s for the assumed conditions, dominated by tidal forcing in the first two days, then driven mostly by winds in the next two days. Given these conditions, responders would have 26 hours to effect dispersant application operations or to place exclusion or re-direction booms at sensitive areas before first oil contact with the shoreline.

12.9.3 Potential Effects from a Spill at Butterworth Rocks Unmitigated effects described below illustrate predicted serious situations for the biophysical and human environment. Booming around tankers would contain some of the oil, skimmers and booms would be used to remove additional oil, and booming at sensitive areas would protect many areas described below.

12.9.3.1 Unmitigated Effects on the Biophysical Environment The hypothetical Butterworth Rocks mass balance example is set in an open water area approximately 35 km west of Prince Rupert. After 24 hours, approximately 67% of the synthetic oil would still be on the surface, with 20% evaporated and 13% dispersed into the water column. The area of exposure risk during the first 24 hours would be the oil water interface and oil air interface in the areas covered by the surface oil. Aquatic organisms with physiological and biological characteristics that require them to use these habitats are at risk of exposure. Several species of marine birds, marine mammal and plankton fall within this category. Although an aquatic organism might be exposed to oil, the consequence of the exposure is related to the organism‘s vulnerability to hydrocarbon exposure. By the third day less than 2% of the oil

Page 12-36 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning would remain on the surface, 40% would be dispersed in the water column, 25% would have evaporated and 33% would be stranded on shore, almost entirely within the Tree Nob Group. This area has strong tidal mixing which results in high aquatic productivity. Given the large percentage of synthetic oil that would disperse in the water column, short-term effects on water quality and potentially longer-term effects on sediment quality are possible. Changes in water column quality could occur and affect subtidal species. This area has extensive kelp beds that could be affected by temporary exposure to synthetic oil or changes in water quality. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. This area is also known to have abundant shrimp populations. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Marine mammals and birds tend to follow prey locally, but also undergo seasonal migrations. Food abundance for marine mammals and birds is high during the summer due to increased ocean productivity. Several species of marine mammals and birds are known to inhabit areas that could be affected. This includes Scoters, Northern Resident killer whales, and humpback whales. The acute exposure potential for cetaceans is limited to the period when oil is on the water surface. For this hypothetical example, the exposure window is limited to the first three days. Toothed whales and dolphins are known to be widely distributed throughout the OWA. Diving birds and marine mammals could come in contact, inhale or ingest oil, from cleaning their fur or feathers, and might be at risk of hypothermia. After the third day, the exposure risk would be greater for species that use shoreline habitats. This would include shore birds, seals, sea lions, terrestrial wildlife, invertebrates and marine vegetation. The first place where oil would contact the shore in this mass balance example is the Tree Nob Group of the proposed Stephens Islands British Columbia Parks Conservancy. Unmitigated, it is predicted that most of the islands making up the Tree Nob Group and parts of Stephens Island would be oiled.

12.9.3.2 Unmitigated Effects on the Human Environment A hydrocarbon spill could affect heritage resources and traditional marine uses in the intertidal and shoreline regions. The area of this hypothetical mass balance example is within close proximity to Prince Rupert and several First Nations communities. During summer there would be an expected peak in human use within this area. There are First Nations reserve lands located near the hypothetical spill site. Aboriginal people would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. Effects of an example spill on human health are discussed in the section on Human Health Risk Assessment (Section 11.3). Since no consultations have occurred regarding traditional uses in these areas, it is difficult to generalize about effects, which, in addition to harvesting of food, could include areas of cultural and sacred importance or periodic habitation areas. A spill might affect heritage resource sites through contamination with oil or through physical damage associated with cleanup, although there is little documentation of heritage resources in the area.

April 2010 FINAL Page 12-37

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

In addition to mortality of fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting. Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook). This area is also productive for the shrimp, red urchin and crab fishery. There are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. A summer spill of synthetic oil at Ness Rock would likely have effects on non-traditional marine uses. This would include potential short-term access limitation to pleasure boaters, kayakers and recreational fishers. Ecotourism and sports fishing outfitters from Prince Rupert could also be affected. As noted earlier, Northern Gateway has offered coastal Aboriginal organizations the option of owning and directly participating in the RO for emergency events. Assuming that this is the case, trained responders, vessels and equipment from the coastal communities would be engaged immediately in the response.

12.10 Example 6: Medium Size Diluted Bitumen Spill at Kitimat Terminal This hypothetical example simulates the maximum discharge of diluted bitumen because of a loading arm and containment boom failure during transfer operations, while loading a tanker at the marine terminal in June. The tanker would be pre-boomed and most of the diluted bitumen would remain within the primary boom. However, should the boom become ineffective (i.e., mechanical failure or wave height exceedance) this example reflects the unmitigated fate of hydrocarbons over 10 hours (i.e., diluted bitumen without a boom in place). The example takes place in June under inflow wind and ocean conditions, the time of year when the greatest range of marine organisms are likely to come in contact with a spill (see Section 10). Typical winds at that time of year are from the south-southwest at speeds varying from 4 to 12 m/s, generating waves that could mix some of the oil into the upper water column. The model assumes current speeds in the range of 5 to 40 cm/s. Kitimat Arm is approximately 3 km wide, with water depths greater than 180 m. Five parks (Nalbeelah Creek Wetlands Park, Kitimat River Park, Dala-Kildala Rivers Estuaries Park, Eagle Bay Park and Coste Rocks Park) and Jesse Falls Protected Area are adjacent to Kitimat Arm. Most of these include a small portion of associated protected marine habitat.

12.10.1 Spill Characteristics In this hypothetical example, winds and currents would transport oil north and east up Kitimat Arm over the first 10 hours. Oil would reach the western shores of Kitimat Arm almost immediately along a 500 m section of shoreline adjacent to the north side of the terminal. After three hours, is predicted to reach the eastern shores of the Arm at Kitamaat Village and after four hours, oil is predicted to be stranded from the marina north of Kitamaat Village to the southern marina at the end of Haisla Avenue, with a total affected shoreline of 2.5 km. No oil is predicted to remain on the water surface after 9 hours. The mass balance analysis shows the predicted distribution of diluted bitumen and its fate over the next 10 hours (see Table 12-11 and Figure 12-7). Initially, the oil is predicted to be on the water surface and then move to the shore adjacent to the terminal. Evaporation into the air would begin immediately. With

Page 12-38 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning no mitigation, the oil would move northeastward and make contact with the estuary and marinas at Kitamaat Village after three hours. After 10 hours, no oil is predicted to be on the surface; approximately 232 m3 (93%) would be on the shoreline, 6 m3 (2%) would be dispersed into the water column and 12 m3 (5%) would be evaporated.

Figure 12-7 Mass Balance for a Hypothetical 250 m3 Diluted Bitumen Spill at the Terminal During Summer

Table 12-11 Mass Balance for a Hypothetical 250 m3 Diluted Bitumen Spill at the Terminal during Summer

Hours Amount in Each Environmental Compartment After Spill Water Surface Ashore Evaporated Water Column (%) (%) (%) (%) 1 69.2 27.2 2.4 1.2 2 67.2 27.2 4 1.6 3 19.6 73.6 4.4 2.4 4 3.2 90 4.4 2.4 5 2.4 90.4 4.8 2.4 6 1.6 91.2 4.8 2.4

April 2010 FINAL Page 12-39

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-11 Mass Balance for a Hypothetical 250 m3 Diluted Bitumen Spill at the Terminal during Summer (cont’d)

Hours Amount in Each Environmental Compartment After Spill Water Surface Ashore Evaporated Water Column (%) (%) (%) (%) 7 1.2 91.6 4.8 2.4 8 0.4 92.4 4.8 2.4 9 0 92.8 4.8 2.4 10 0 92.8 4.8 2.4

12.10.2 Mitigation Key actions for emergency response are described in Table 12-12, including assumptions, safety, initial objectives and activities, protection of environmentally sensitive areas and species, containment and control strategies, and cleanup responses.

Table 12-12 Emergency Response for Diluted Bitumen

Activity Description Assumptions Safe to commence initial response operations. The cause of the spill has no affect on the size or duration of the spill event. No injuries are associated with the incident. Safety All personnel accounted for and no injuries. Emergency responders assure safety, proper personal protective equipment (PPE) and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and spill response actions. Secure spill source. Contain and recover oil on water. Protect priority sensitive areas. Coordinate response actions with Unified Command. Procedures to The Person In Charge (PIC) of transfers actuates the emergency shutdown to stop the stop a transfer pump and shuts all valves. Shore personnel confirm manifold is shut down. discharge Notifications The PIC notifies the Control Room. Control Room notifies Spill Response Team and Terminal Supervisor. Supervisor requests notification be made to Coast Guard, Environment Canada, BC Ministries, etc. Fire prevention No ignition source or open lights are present. In the event of fire, the Terminal and control Supervisor would activate the Fire Response Team and notify the Kitimat Fire Department. The vessel and its crew would also assist with any firefighting efforts on the water.

Page 12-40 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-12 Emergency Response for Diluted Bitumen (cont’d)

Activity Description Discharge Visual assessment is maintained by terminal staff and crew on vessel, responders, and tracking shore personnel. As deemed necessary and appropriate, Terminal Supervisor would dispatch a helicopter or aircraft for overflight documentation, tracking, and initial guidance to on-water oil spill containment and recovery operations. In addition, oil trajectory modelling begins using real-time weather and oceanographic conditions, and oil weathering properties. Trajectory models are used to forecast movement direction and timing and are updated and re-run as overflight information becomes available as ground-truth for forecasting. Protection of Responders refer to the Terminal and Marine Area OSR Plans and to pertinent GRPs environmentally to identify priority sensitive areas near the terminal or in the spill pathway. Targeted sensitive areas priority protection sites, subject to Unified Command approval, are: Minette Bay and Kitimat River estuary, Kitamaat Village, Bish Cove Kitimat Docks and Wharves Boom and boom boats are dispatched to implement protection strategies at top priority protection sites, as directed by Unified Command. Containment The vessel is pre-boomed and contains most, if not all, spilled oil. and control Response teams are deployed to add additional secondary containment boom strategies Tanker and shore personnel drain remainder of damaged hose, secure ends, and disconnect. Oil recovery Skimmers and pumps at terminal are used to transfer recovered product to available strategies tanks at terminal. RO initiates sweep and recovery operations for oil observed outside of containment. Oil is minimized from reaching shorelines by directing sweep and recovery operations from overflights. Oil skimmers recover oil from beach cleaning operations. Dispersants Application for use of chemical dispersant operations is initiated. Potential target areas are those in which oil on the water surface is not contained, with water depths greater than 10 m, and located more than 500 m from the shoreline. Shoreline SCAT team(s) deployed for assessment and cleanup recommendations. cleanup Cleanup endpoints defined and approved by Unified Command for oiled shore types. strategies Primary oiled shoreline types, based on trajectories, will consist of rock cliff and rock with gravel beach. Oiled segment cleanup undertaken as defined in the IAP based on SCAT team recommendations (manual, mechanical, etc.).. Recommended cleanup techniques for the two primary shoreline types are: Rock cliff: spot washing, natural cleaning Rock with gravel beach: spot washing and vacuum, flushing and recovery, tilling, manual cleaning for small areas Segments inspected for meeting endpoints and signed-off, if so.

April 2010 FINAL Page 12-41

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-12 Emergency Response for Diluted Bitumen (cont’d)

Activity Description Recovered oil All spill response wastes are segregated into oiled vs. non-oiled and liquid vs. solid. transfer and Recovered oily water mixture is pumped to available tanks on vessel, barges, bladders, storage and to shore. Oily liquids are transferred from floating storage to available tanks at the Kitimat Terminal. Oily solids are handled in specified drums. Wildlife Wildlife at risk in the immediate spill area is assessed. Wildlife hazing kit is readied in protection case of need. Bird capture/rehabilitation unit at the terminal is prepared pending field assessment. Appropriate Capture / Hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area.

Meteorological conditions would strongly influence how hydrocarbons disperse and weather on the water. Mitigation measures would be based on hydrocarbon chemistry, behaviour and distribution, and degree of physical disturbance that could result from response methods and established remediation criteria. The key mitigation is immediate deployment of containment booms around a tanker being loaded for oil, using equipment on the harbour tugs and the tanker itself. For the diluted bitumen example, wind during the 10-hour period is predicted to blow from the south- southwest at speeds up to 12 m/s, which would generate waves that mix some oil into the upper water column and generate surface currents that would move floating oil. In this region, currents are driven by a combination of tides, fresh water input and winds. For these example the current are modelled as 5 to 40 cm/s. Given these conditions, responders would have three hours to place exclusion booms around sensitive habitats and infrastructure such as: the Kitimat River estuary (discharge of fresh water into the salt water would likely prevent oil migration into high flow area of the estuary) sand flats on the northern shores of Kitamaat Village spawning river running from sand flats through Kitamaat Village marinas and boat launches north and south of Kitamaat Village

12.10.3 Potential Effects on Key Resources at Risk from a Diluted Bitumen Spill at the Terminal Unmitigated effects described below illustrate predicted worst-case situations for the biophysical and human environment. Booming around vessels would contain some of the oil, skimmers and booms would be used to remove additional oil and exclusion booming around sensitive areas and infrastructure would protect many areas described below.

Page 12-42 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.10.3.1 Unmitigated Effects on the Biophysical Environment The aquatic organisms most vulnerable during summer would be those that use shoreline habitat, as discussed generally in Section 10. In addition, attention would focus on species of conservation concern (e.g., Marbled Murrelet, Surf Scoters, killer whales and humpback whales) that are in the area. The affected shore would be predominately coarse-grained beaches with potential for penetration and remobilization of oil and sand flats where oil would persist. The oil would immediately reach the western shores of Kitimat Arm, and after flowing northeast, it would reach the eastern shores of Kitamaat Village. The estuary at the northern border of the community is prime fish habitat and leads to a salmon spawning river. South of the marina and boat launch is shorebird habitat (e.g., Marbled Murrelet). Diluted bitumen on the water surface, dispersed in water and coating the shoreline would result in short- term effects on water quality and potentially longer-term effects on sediment quality. Organisms in contact with shoreline oil would include rockweed, kelp and other seaweeds, intertidal marine invertebrates and waterbirds. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. Marine vegetation typically recovers relatively quickly once the oil is removed. Marine mammals and birds tend to follow prey. During this time of year, salmon would be migrating into Douglas Channel and other areas. This could increase the potential for presence of predators. Animals could inhale or ingest oil from cleaning their fur or feathers. Oiled fur or feathers pose the risk of hypothermia. A spill could have large effects on the Marbled Murrelet population because the birds moult during summer and cannot fly to escape the oil. Marine mammals would be most vulnerable while oil is on the water surface (first 48 hours), because the channel is relatively confined in the potentially affected area. Wildlife such as river otters, black-tailed deer, bears, wolves, ravens and crows that feed and scavenge along the shoreline could be exposed to stranded oil.

12.10.3.2 Unmitigated Effects on the Human Environment Heritage resources and traditional marine uses in the intertidal and shoreline regions could be affected during summer. Marine food resources could be affected, presenting a temporary limitation to the ability of Aboriginal groups to exercise their fishing and harvesting rights. In addition to mortality of some fish and invertebrates, there could be fisheries closures due to contaminant levels, conservation concerns or tainting (for example salmon moving into the area to spawn). Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook), there are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, sea cucumber, octopus, geoduck, horse clam, red sea urchin, herring and octopus. Aboriginal groups would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. In this example, Aboriginal groups at Kitamaat Village, which is home to the Haisla people, would be particularly vulnerable. Because detailed information regarding traditional use in these areas has not yet been provided, conclusions regarding effects on harvesting and cultural resources have not been reached. A spill might affect heritage resource sites through contamination with oil or through physical damage associated with cleanup, although there is little documentation of heritage resources in the area.

April 2010 FINAL Page 12-43

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

There would be effects on non-traditional marine uses at the marinas because summer is typically a peak recreational season. The possible effects would be aesthetic disturbances and restricted access to shorelines and marinas during the cleanup. The channel is wide enough to allow vessel traffic to avoid a spill; however, vessels and marine infrastructure in contact with oil would be fouled and require cleanup and decontamination. Given that, in this example, diluted bitumen would move directly toward the shores of the community, the area could experience temporary disruption of vessel traffic, as well as loss of local fish and shellfish resources over one season or more (e.g., from salmon mortality or contamination of shellfish). If, as proposed by Northern Gateway, participating coastal Aboriginal groups have direct ownership or involvement in a locally-based RO for emergency events, trained responders, vessels and equipment from the coastal communities would be engaged.

12.11 Example 7: Medium Size Condensate Spill at the Terminal The second hypothetical example simulates the maximum discharge of condensate during transfer operations while unloading a tanker at the marine terminal in June. The tanker would not be pre-boomed and most of the condensate would evaporate within minutes. This example shows the fate after a 220- second spill period at thirty minutes after the event. June is when a release would interact with the greatest range of aquatic organisms (see Section 10) and represents wind and currents characteristic of summer inflow conditions. During the modelled 30-minute period, winds blow from the south-southwest at a speed of 9 m/s, and would generate waves that could mix some of the condensate into the upper water column. Modelled current speeds are in the range of 5 to 40 cm/s for the assumed conditions. Kitimat Arm is approximately 3 km wide, with water depths greater than 180 m. Five parks (Nalbeelah Creek Wetlands Park, Kitimat River Park, Dala-Kildala Rivers Estuaries Park, Eagle Bay Park, Coste Rocks Park,) and Jesse Falls Protected Area are adjacent to Kitimat Arm. Most of these include a small portion of associated protected marine habitat.

12.11.1 Spill Characteristics In this example, condensate would almost immediately reach the western shores of Kitimat Arm and cover a 500 m section of shoreline on the north side of the terminal. From there, the condensate is predicted to move north and east up Kitimat Arm toward Kitamaat Village. The maximum area affected is predicted to be 0.31 km2 within 17 minutes after the spill. Winds and currents transport condensate a maximum of 700 m north and east of the terminal. After 30 minutes, all condensate on the water of on the surface of the shorelines would be completely evaporated and dispersed. Condensate that penetrates into the beach sediment is predicted to evaporate or disperse within two tidal cycles (25 hours). The mass balance developed for a condensate spill and its fate over 30 minutes is shown in Table 12-13 and Figure 12-8. Initially, the condensate would spread over the water surface and then move to the shore adjacent to the terminal. Evaporation would begin immediately. With no mitigation, the condensate is predicted to move northwestward along the shore, where it would continue to evaporate and disperse. After 30 minutes, none of the condensate is predicted to be on the surface; approximately 13% would be on the shoreline, 36% would be dispersed into the water column and 51% would be evaporated.

Page 12-44 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-13 Mass Balance Estimates for a Hypothetical 250 m3 Condensate Spill at the Terminal during Summer

Amount in Each Environmental Compartment Minutes Evaporated into After Spill Water Surface Ashore Air Water Column (%) (%) (%) (%) 5 64.4 24.8 7.2 3.6 10 38.4 19.2 28.4 14 15 18.4 14.8 44.4 22.4 20 5.6 13.6 49.6 31.2 25 0 13.2 50.8 36 30 0 12.8 51.2 36

Figure 12-8 Mass Balance Estimates for a Hypothetical 250 m3 Condensate Spill at the Terminal during Summer

April 2010 FINAL Page 12-45

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.11.2 Mitigation Key response measures are listed in Table 12-14.

Table 12-14 Emergency Response for a Condensate

Activity Description Assumptions Safe to commence initial response operations. The cause of the spill has no affect on the size or duration of the spill event. No injuries are associated with the incident. Safety All personnel accounted for and no injuries. Emergency responders assure safety, proper personal protective equipment (PPE) and initial deployment of response equipment. Vapour concentrations are not an issue with prevailing wind. Initial objectives Establish safety requirements for operational and spill response actions. Secure spill source. Contain and recover condensate on water. Protect priority sensitive areas. Coordinate response actions with Unified Command. Procedures to The Person In Charge (PIC) of transfers actuates the emergency shutdown to stop the stop a transfer pump and shuts all valves. Shore personnel confirm manifold is shut down. discharge Notifications The PIC notifies the Control Room. Control Room notifies Spill Response Team and Terminal Supervisor. Supervisor requests notification be made to Coast Guard, Environment Canada, BC Ministries, etc. Fire prevention No ignition source or open lights are present. In the event of fire, the Terminal and control Supervisor would activate the Fire Response Team and notify the Kitimat Fire Department. The vessel and its crew would also assist with any firefighting efforts on the water. Discharge Visual assessment is maintained by crew on vessel, responders, and shore personnel. tracking As deemed necessary and appropriate, Terminal Supervisor would dispatch a helicopter or aircraft for overflight documentation, tracking, and initial guidance to on- water oil spill containment and recovery operations. Protection of Responders refer to the Terminal and Marine Area OSR Plans and identify priority environmentally sensitive areas near the terminal. Targeted priority protection sites, subject to Unified sensitive areas Command approval, are: Minette Bay and Kitimat River estuary Kitamaat Village Bish Cove Kitimat Docks and Wharves seasonally important habitat that might lie within the spill pathway. Containment Response teams are deployed to add additional secondary containment boom and control Tanker and shore personnel drain remainder of damaged hose, secure ends and strategies disconnect.

Page 12-46 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

Table 12-14 Emergency Response for a Condensate (cont’d)

Activity Description Condensate Skimmers and pumps at terminal are used to transfer recovered product to available recovery tanks at terminal. strategies RO initiates sweep and recovery operations for condensate observed outside of containment. Minimize condensate from reaching shorelines by directing sweep and recovery operations from overflights. Shoreline SCAT team(s) deployed for assessment and cleanup recommendations. cleanup Cleanup endpoints defined and approved by Unified Command for oiled shore types. strategies Oiled segment cleanup undertaken per SCAT team recommendations (manual, mechanical, etc.). Segments inspected for meeting endpoints and signed-off, if so. Recovered All spill response wastes are segregated into oiled vs. non-oiled and liquid vs. solid. condensate Recovered oily water mixture is pumped to available tanks on vessel, bladders, and transfer and shore. storage Oily solids are handled in specified drums. Wildlife Wildlife at risk in immediate spill area is assessed. Wildlife hazing kit is readied in case protection of need. Bird capture/rehabilitation unit is prepared pending field assessment. Appropriate Capture / Hazing permits are submitted through Unified Command. Response activities in the area should reduce the likelihood of wildlife congregation in area.

Meteorological conditions would strongly influence how condensate disperses and degrades on the water. Condensate is a non-persistant oil and may naturally degrade within 24-hours except from large volumes under calm marine conditions. Response strategies and tactics must factor in safety and the short environmental exposure time. In this summer example, wind during the 30-min period is predicted to blow from the south-southwest at speeds of 9 m/s, which would generate waves that would mix some oil into the upper water column and generate surface currents that would move condensate on the water surface. In this region, currents are driven by a combination of tides, fresh water input and winds, and for this example are modelled at 5 to 40 cm/s. Given these conditions, responders would likely prioritize placement of exclusion booms around sensitive habitats and infrastructure. However, in this example the effects from a condensate spill are predicted to be limited to the shores approximately 700 m from the Kitimat Terminal.

12.11.3 Potential Effects on Key Resources at Risk Unmitigated effects described below illustrate predicted serious situations for the biophysical and human environment. Exclusion booming around sensitive areas and infrastructure would protect many areas described below.

April 2010 FINAL Page 12-47

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning

12.11.3.1 Unmitigated Effects on the Biophysical Environment The aquatic organisms most vulnerable during summer would be those that use shoreline habitat, as discussed generally in Section 10. In addition, attention would be focused on any species of conservation concern (e.g., Marbled Murrelet, Surf Scoter, killer whale and humpback whale) that are in the area. The affected shore would be predominately coarse-grained beaches with potential for penetration and remobilization of oil and sand flats where oil would persist. Hydrocarbons would immediately reach the western shores of Kitimat Arm, then spread northwest along the same shore. Condensate on the water surface, dispersing in water and coating the shoreline, would result in short-term effects on water quality and potentially longer-term effects on sediment quality. Organisms in contact with shoreline oil would include rockweed, kelp and other seaweeds, intertidal marine invertebrates and water birds. Intertidal vegetation is particularly important habitat for juvenile salmon migrating near shores and for invertebrate and fish larvae that live among and feed on aquatic plants. Marine vegetation and invertebrates typically recover relatively quickly once the oil is removed. Marine mammals and birds tend to follow prey. During summer, salmon would be migrating into Douglas Channel and other areas. This could increase the potential for the presence of predators. Animals could inhale or ingest condensate from cleaning their fur or feathers. Oiled fur or feathers pose the risk of hypothermia. A spill could have large effects on the Marbled Murrelet population because the birds moult during summer and cannot fly to escape the oil. Marine mammals would be most vulnerable while oil is on the surface in the first 30 minutes because the confined channel limits oil free areas as the whales breach for air. For the two days following the spill (until the tidal cycle acts to disperse the remaining condensate), wildlife such as river otters, black-tailed deer, bears, wolves ravens or crows that feed and scavenge along the shoreline could come into contact with stranded condensate.

12.11.3.2 Unmitigated Effects on the Human Environment Heritage resources and traditional marine uses in the intertidal and shoreline regions could be affected during summer. Potential effects on marine food resources could present a temporary limitation to the ability of Aboriginal groups to exercise their fishing and harvesting rights. In addition to mortality of some fish and invertebrates, fisheries could close due to contaminant levels, conservation concerns or tainting (for example salmon moving into the area to spawn). Economic returns are greatest for salmon (pink, chum, coho, sockeye and chinook), but there are also fisheries for Pacific halibut, several species of groundfish, shrimp, prawn, octopus, red sea urchin and octopus. Aboriginal groups would be particularly sensitive because of their long association with and dependence on the sea for food, transportation, social and ceremonial purposes. In this example, the shores across from Kitamaat Village would be particularly vulnerable. Because detailed information regarding traditional use in these areas has not yet been provided, conclusions regarding effects on harvesting and cultural resources have not been reached. Heritage resource sites might be affected through contamination with oil or through physical damage associated with cleanup activities, although there is little documentation of heritage resources in the area. A condensate spill would have effects on non-traditional marine uses at the marinas since summer is typically a peak recreational season. The possible effects would be aesthetic disturbances and restricted access to shorelines, the Kitamaat Village marina, and the

Page 12-48 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 12: Mass Balance Examples for Response Planning terminal during the cleanup. The channel is wide enough to allow vessel traffic to avoid the spill; however, log snags, vessels and marine infrastructure in contact with oil would be fouled. Given that condensate is predicted to stay near the shores of the terminal for at least two days, the area could experience temporary disruption of vessel traffic, loss of local fish and shellfish resources over one recruitment season (e.g., from salmon fry mortality or contamination of shellfish). As noted earlier, Northern Gateway has offered coastal Aboriginal groups the option of directly participating in the RO for emergency events. If this is the case, trained responders, vessels and equipment from the coastal communities would be engaged.

12.12 Summary Table 12-15 summarizes the likely fates of hydrocarbons from the mass balance examples. Examples for the summer typically result in more hydrocarbons retained at the shore after the first 8 to15 days than the example for the winter, where more hydrocarbons are retained in the water column after the first three days, primarily due to wind and current conditions at that time of year.

Table 12-15 Summary of Mass Balances for Hypothetical Spill Examples

Example Water Evaporated Retained at Shore (%) (%) (%) Grounding incident at Emilia Island in the Surface: none 3 winter with 10,000 m synthetic oil spill: Water Column: 41 22 37 fate by day 3 Grounding incident at Principe Channel in Surface: none 3 the summer with 10,000 m diluted Water Column: 2 15 83 bitumen spill: fate by day 8 Collision incident at Wright Sound in the Surface: <1 3 summer with 36,000 m diluted bitumen Water Column: 6 17 76 spill: fate by day 15 Grounding incident at Ness Rock in the Surface: <1 3 winter with 10,000 m diluted bitumen spill: Water Column: 3 13 84 fate by day 15 Grounding incident at Butterworth Rocks Surface: none 3 in the winter with 10,000 m synthetic oil Water Column: 41 25 33 spill: fate by day 5 Loading arm accidental release at the Surface: none Kitimat Terminal in the summer with 250 Water Column:2 5 93 m3 diluted bitumen spill: fate by hour 10 Loading arm accidental release at the Surface: none Kitimat Terminal in the summer with 250 Water Column:36 51 13 m3 condensate spill: fate after 30 minutes.

NOTE: Results from the mass balance examples consider natural attenuation processes and do not take into account emergency response actions.

April 2010 FINAL Page 12-49

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

13 Risk Assessment Related to Hydrocarbons in the Marine Environment

13.1 Introduction This section summarizes the results of the risk assessment for a hypothetical spill of diluted bitumen in the marine environment during shipping in Wright Sound. Detailed results are provided in the Risk Assessment for Accidental Hydrocarbon Spills at the Kitimat Terminal and in the CCAA TDR (Stephenson et al. 2010a). The risk assessment was undertaken to increase understanding of the long-term or chronic risk that hydrocarbon spills might pose to the ecological health of the marine environment in Wright Sound and Douglas Channel, or to human consumers of marine resources. Risk assessment is a process that characterizes the nature and magnitude of health risks to ecological receptors (e.g., birds, fish, wildlife) and to humans (e.g., residents, workers, recreational visitors) from exposure to chemical contaminants and other stressors that might be present in the environment (US EPA 2009). In general terms, risk depends on: how much of each chemical is present in environmental media (e.g., water, sediment) how much contact (exposure) ecological receptors or humans have with the contaminated environmental media how much bioaccumulation of each chemical the ecological receptors experience how much exposure ecological receptors or humans have to chemicals that were bioaccumulated by other living organisms that they consume the inherent toxicity of the chemical

13.1.1 Hypothetical Example For this risk assessment, only the Wright Sound example was considered (a 36,000 m3 diluted bitumen spill during the summer). This example was investigated, as the large volume and relative persistence of diluted bitumen allows for a conservative assessment. It should be noted that results could vary according to several factors, including but not limited to, environmental and meteorological conditions, seasonality, and location. The risk assessment does not attempt to provide a quantitative assessment of cumulative environmental effects (i.e., risk) arising from the construction, operation, or decommissioning of other projects that might presently be in the planning stages, since data required to evaluate such projects quantitatively are not available. For the example considered, it was assumed that 36,000 m3 is released over a period of 13 hours following a VLCC collision in Wright Sound during the summer. Modelling predicts that diluted bitumen

April 2010 FINAL Page 13-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment first strands on Fin Island, with further stranding on Farrant Island, Gil Island, and in the region of Hartley Bay eventually reaching Cridge Passage, Lewis Passage, Squally Channel, Grenville Channel, Whale Channel, Campania Island, and Caamaño Sound. According to the modelling summary, by the end of day 15, diluted bitumen is predicted to be present within the environment as follows (see Section 12.7 for details): approximately 1% (196 m3) is left on the water surface approximately 6% (2,199 m3) has entered the water column approximately 17% (6,233 m3) has evaporated approximately 76% (27,372 m3) has stranded along shorelines Information from the mass balance modelling, in combination with a separate toxicity model (Stephenson et al. 2010a) were used in this risk assessment to predict the fate of diluted bitumen within the marine water, intertidal, and subtidal environments.

13.1.2 Assessment Area The highest level of chemical emissions to the marine environment in Wright Sound would be deposited in the Wright Sound and Douglas Channel areas, extending to the Grenville Channel, Verney Passage, Whale Channel, Otter Channel, Nepean Sound, Squally Channel, Campania Sound, Caamaño Sound, and Estevan Sound (see Example 3, Section 12.7). The location of Example 3 in Wright Sound is identified on Figure 12-1. The concentrations of chemicals of potential concern (COPC) in water and subtidal sediment were determined using a marine water quality model (MWQM) and a marine sediment quality model (MSQM), respectively. The water column and subtidal sediment compartments were considered as a potential marine ecological resource location, where marine biota identified as key indicators (KIs) might be exposed to COPC in the environment. The receptors include marine algae, benthic invertebrates, fish, marine birds, shore birds, marine mammals, and semi-aquatic mammals inhabiting the shoreline. Risks to receptors from exposure to COPC were considered collectively to determine whether there might be adverse environmental effects from the diluted bitumen example in Wright Sound on the marine water and sediment environment.

13.2 Ecological Risk Assessment Ecological risk assessment (ERA) has been defined as a process that evaluates the likelihood that adverse ecological effects might occur or are occurring as a result of exposure to one or more ecological stressors (US EPA 1998). Typically, ERAs are developed to evaluate chemical and non-chemical stressors and to support appropriate environmental decision-making. Ecological stressors might be chemical, physical or biological. This ERA was based on one hypothetical example in Wright Sound (Example 3, see Section 12.7). The potential stressors are hydrocarbons that might be released into the environment under such circumstances.

Page 13-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

The following three key factors must be present for an ecological risk to be realized: presence of a chemical or chemicals that might be an environmental hazard environmental pathways that might cause receptor organisms to be exposed to the chemicals a receptor organism present at the right time and place to be exposed to the chemicals However, even when all three factors are simultaneously present, considerations such as chemical concentrations, duration of exposure, and the specific exposure pathways determine whether there might be an adverse environmental effect.

13.2.1 Methodology for the ERA To quantify the potential for ecological risk, a standard ERA methodology was followed (US EPA 1998; CCME 1996, 1997). This methodology includes: problem formulation (including resource identification) exposure assessment hazard assessment risk characterization discussion of certainty and confidence in the predictions statement of conclusions The approach to completing an ERA for the Wright Sound example consisted of the following steps: identification of an appropriate assessment area development of example identification of chemicals of potential concern (COPC) modelling the likely fate of COPC in the marine environment completing the exposure and hazard assessments estimating the level of risk to ecological receptors, and the degree of confidence in these estimates Risk assessments are conducted using conservative assumptions, which tend to overestimate exposure and risk, and are intended to be iterative or tiered in application. Therefore, if a preliminary ERA finds that wildlife exposures to COPC are below levels considered likely to cause unacceptable risk, it is unlikely that adverse effects would occur. However, finding in a preliminary ERA that wildlife exposures might exceed levels considered safe does not necessarily mean that adverse effects would occur. Rather, a more detailed analysis would be advisable to determine more precisely whether the preliminary conclusion withstands more rigorous and less conservative analysis.

13.2.1.1 Fate of Stranded Diluted Bitumen in the Marine Intertidal Environment Modelling predicts that 36,000 m3 of diluted bitumen in Wright Sound first reaches Fin Island within approximately 8 hours, and then strands on southern Farrant Island, northwestern Gil Island, and in the region of Hartley Bay 13 to 24 hours post-spill (Section 12.7). In a few days, diluted bitumen reaches Cridge Passage, Lewis Passage, Squally Channel, Grenville Channel, Whale Channel, Campania Island, and Caamaño Sound. By the end of 15 days, approximately 76% has stranded along 240 km of shorelines (Section 12.7).

April 2010 FINAL Page 13-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

The fate and potential effects of stranded diluted bitumen in the marine intertidal environment were evaluated qualitatively with reference to the experience from EVOS.

13.2.1.2 Fate of Diluted Bitumen Constituents in the Marine Water and Subtidal Sediment Environments A variety of liquid hydrocarbons will be transported by project-related tankers, including diluted bitumen, synthetic oil and condensate. Representative samples of each hydrocarbon were obtained from Enbridge and were analyzed by the Research and Productivity Council laboratory in Fredericton, NB. Analytical results for these representative samples are presented in Stephenson et al. 2010b. Chemical screening was based upon chemical analyses of representative hydrocarbon samples, followed by an analysis that takes into consideration the relative solubility and volatility of various hydrocarbon and inorganic element fractions. Substances not detected in the representative hydrocarbon samples (i.e., not detected in diluted bitumen) were not carried forward. For example, since the benzo(a)pyrene was not detected in the analysis of the representative hydrocarbon sample, it was not considered further in the risk assessment. Trace elements were generally not detected, or were present at low concentration (less than 1 mg/kg) in the liquid hydrocarbon samples. The diluted bitumen sample generally contained the highest concentrations of trace elements including vanadium, nickel, calcium and sodium (in order of decreasing concentration). In water, trace elements are typically encountered as dissolved ionic species, which might associate with other organic or inorganic species, or bind to solids. Most trace elements exhibit low vapour pressure (an indication of their low potential for volatilization). As such, they were assumed to dissolve readily within the water column rather than to evaporate. A screening calculation was carried out to estimate the possible trace element concentrations dissolved in seawater beneath the surface oil, assuming that the trace elements dissolved rapidly and dispersed following an assumed volume of 36,000 m3 of diluted bitumen. Based on the very low concentrations resulting from this calculation, trace elements were not carried forward in the risk assessment (Stephenson et al. 2010b). Organic compounds detected in the liquid hydrocarbon samples include petroleum hydrocarbons (e.g., benzene, toluene, ethylbenzene and xylenes (BTEX), total petroleum hydrocarbon (TPH) fractions, fractionated according to the number of carbon atoms present, and asphaltenes), selected polycyclic aromatic hydrocarbons (PAH), phenolic compounds, and two volatile organic compounds (VOC; 1,2,4- trichlorobenzene and 1,3,5-trimethylbenzene). Analytical results revealed important differences in the respective chemical composition of condensate, synthetic oil, and diluted bitumen samples obtained from Enbridge. Petroleum hydrocarbons, as a chemical class, were the largest contributors, with PAH, phenolic compounds and VOC contributing little to the total mass (less than 1%). The diluted bitumen sample was made up primarily of longer aliphatic petroleum hydrocarbons (i.e., aliphatic fractions greater than C12-C16, greater than C16-C21, and greater than C21-C32) as well as asphaltenes. These fractions exhibit low or negligible vapour pressures. Approximately 5 to 10% of the diluted bitumen sample would be considered highly volatile (in relation to naphthalene). The volatile components would be expected to evaporate, leaving less volatile constituents within the surface oil. Modelling indicates that by the end of 15 days following the hypothetical release approximately 17% of the diluted bitumen has evaporated (Section 12.7).

Page 13-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Organic compounds carried forward as COPC include BTEX and TPH fractions. In addition, eight PAH compounds, three phenolic compounds and two VOC were carried forward. The complete list of COPC carried forward is presented in Table 13-1.

Table 13–1 COPCs from a Hypothetical Diluted Bitumen Example to the Marine Environment in Wright Sound

BTEX Compounds PAH Compounds Benzene 1-Methylnaphthalene Toluene 2-Methylnaphthalene Ethylbenzene Anthracene Xylenes (total m,o,p) Benzo(a)anthracene TPH Fractions Fluorene Aromatics Naphthalene >C8 – C10 Phenanthrene >C10 – C12 Pyrene >C12 – C16 Phenolic Compounds >C16 – C21 2.4-dimethylphenol >C21 – C32 2,4-dinitrophenol Aliphatics Phenol >C6 – C8 VOC >C8 – C10 1,2,4-trichlorobenzene >C10 – C12 1,3,5-trimethylbenzene >C12 – C16 >C16 – C21 >C21 – C32

Loadings of COPC to the marine water and subtidal sediment environments in Wright Sound were estimated using two independent mass balance (compartment) models. The MWQM estimates the fate of COPC in the marine environment. The MSQM estimates the deposition of COPC from the water column into marine subtidal sediment. The results of the two models were used to calculate exposure point concentrations (EPC) for COPC in water and subtidal sediment, to which community-level indicators are likely to be exposed. These exposures were used as the basis for ERA calculations. The overall objective of this process was to evaluate the likely fate and chronic environmental effects of COPC in the marine environment. The potential chronic accumulation of individual COPC from subtidal sediment to biota, such as marine invertebrates, was also considered throughout the modelling period, in order that risk to human consumers of seafood can be considered. Both models were created using Stella® v8.1.1 modelling software. The MWQM simulations extend over a 360-hour period, and consider the potential acute toxicity of dissolved COPC in the water column to community-level marine receptors (e.g., fish and zooplankton). Mass balance compartment models can provide explicit estimates of the fluxes of contaminants across

April 2010 FINAL Page 13-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment boundaries or the masses within compartments as a function of time. In the MWQM, the compartments represent specific states of the hydrocarbon (i.e., surface oil, emulsions, globules and droplets) or its flux to environmental media (i.e., volatilization to the atmosphere, dissolution in water, impingement on intertidal sediment, and deposition to subtidal sediment). Additionally, the mass of each COPC within each compartment changes over time in response to processes representing fluxes between compartments. The model was implemented to individually model the fate of up to 32 COPC in seawater. The following processes are responsible for driving the flow of mass (the distribution and fate of COPC) in the MWQM: Spreading of the surface oil and plume. Spreading uses information and equations provided by Lehr et al. (1984). The area of affected water (i.e., the plume) is tracked by estimating the area of the surface oil at a series of time steps and using this information to calculate the growth or shrinkage of the surface oil during each time step. Evaporation of the surface oil. The evaporation is considered through the evaporation of its individual components according to their respective volatilities (i.e., vapour pressures). Relatively volatile components evaporate quickly, whereas components with low vapour pressures are not expected to evaporate in appreciable amounts. The evaporation rate of individual COPC was determined using Raoult‘s law, air-water exchange velocity and the ideal gas law as described by Schwarzenbach et al. (2002). Emulsification of the surface oil. The emulsification is based on the properties of the hydrocarbon and its propensity to form emulsions. First, the stability of a water-in-hydrocarbon mixture was described according to four classes (entrained, unstable, mesostable and stable) using a numerical index presented by Fingas and Fieldhouse (2004). Only mesostable and stable mixtures are carried forward as emulsions. The emulsification rate is estimated using equations by Fingas and Fieldhouse (2004). Liquefaction of a mesostable emulsion (if applicable). As mesostable emulsions have shorter lifespan than stable emulsions (Fingas and Fieldhouse 2004) the hydrocarbon present within a mesostable emulsion was allowed to liquefy back into the surface oil. The liquefaction rate was derived using first-order kinetics and information provided by Fingas and Fieldhouse (2004). Impingement of the surface oil or emulsions to the intertidal sediment shoreline. The impingement of hydrocarbon onto intertidal shoreline was considered through hydrocarbon-holding capacity calculations as reported by Gundlach (1987), Reed et al. (1989) and Cheng et al. (2000) as described in Etkin et al. (2008). This approach defines the maximum hydrocarbon-holding capacity as consisting of the maximum surface loading and the maximum sub-surface loading. Additionally, it considers the effects of shoreline morphology, as well as the characteristics of the hydrocarbon (through viscosity) on the impingement process (Etkin et al. 2008). Dispersion of hydrocarbon droplets in the water column. Following entrainment within the water column, small hydrocarbon droplets might remain submerged for a certain period of time, whereas larger hydrocarbon droplets would quickly resurface. The maximum diameter of hydrocarbon droplets that might remain submerged within the water column was determined using Stoke‘s law (Schwarzenbach et al. 2002). The dispersion rate of these small hydrocarbon droplets within the water column was then investigated through their entrainment rate using equations and information presented in Delvigne and Sweeney (1988), Korotenko et al. (2000), French-McCay (2004) and Reed

Page 13-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

et al. (2004) which take into account the dissipated wave energy, the fraction of the sea water surface covered by oil, and the fraction of sea surface covered by breaking waves. Settling of hydrocarbon droplets to subtidal sediment. Two processes are responsible for driving the flow of mass in the settling of hydrocarbon droplets to subtidal sediment: sorption of the hydrocarbon droplets to suspended sediment and deposition of suspended particles with sorbed hydrocarbon into the sediment layer. Dissolution of hydrocarbon droplets dispersed within the water column. The dissolution of hydrocarbon droplets is considered through the dissolution of its individual components according to their respective solubilities in water. Relatively soluble components might be expected to dissolve readily, whereas components with low solubilities are not expected to dissolve in appreciable amounts but might be present as undissolved forms within the medium. The effective solubility of individual COPC was determined using Raoult‘s law and equations provided by Schwarzenbach et al. (2002). Sedimentation of dissolved COPC to subtidal sediments. Two processes were considered in driving the flow of mass in the sedimentation of dissolved COPC to subtidal sediment: partitioning of COPC between water and suspended sediment (through the water-sediment partition coefficient); and deposition of suspended particles and sorbed COPC into the sediment layer. Sinking of the surface oil. As the hydrocarbon present in the surface oil weathers, its density increases. If the density of the hydrocarbon in the surface oil exceeds that of seawater, the surface oil is assumed to form globules that would sink to the subtidal sediment. Degradation of COPC in the water column. Some COPC might persist for long periods of time (e.g., high molecular weight PAHs), whereas others degrade more rapidly (e.g., low molecular weight hydrocarbons such as benzene). The degradation of COPC in the environment was therefore considered on an individual COPC basis. The half-lives of individual COPC in water as reported by Mackay et al. (2000) were converted to first-order rate constants and degradation rates was simulated assuming first order kinetics. Other processes, such as evaporation of COPC dissolved in water, present in emulsified hydrocarbon, or impinged at the intertidal shoreline, and decomposition of COPC in hydrocarbon and hydrocarbon emulsion as a result of microbial action or photolysis, would also act in a minor way in the environment to reduce the concentrations of COPC in water and sediment. These processes were not included in the MWQM. This simplifying step in model development is justified by the probability that clean-up activities would be initiated rapidly, resulting in the recovery of hydrocarbons that impinge on shorelines, or are floating on the water as emulsified or recoverable material. In addition, flushing and dispersion were not considered in this model, although they would result in the dilution of affected water and reduce the estimated hydrocarbon concentrations dissolved in water. Omitting these processes is conservative and results in a model that tends to overestimate dissolved COPC concentrations in water, and subsequently in sediment. When hydrocarbon droplets are no longer predicted to be present in the water column, the modelling run for the MSQM is initiated to evaluate the potential behaviour and fate of COPCs in sediment over a 20- year period following deposition.

April 2010 FINAL Page 13-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Marine Sediment Quality Model The total loadings of COPCs to subtidal sediment are assumed to distribute evenly over the area of affected sediment, which is based on the area of water affected by the surface oil and horizontal spreading within the water column along the axis of tidal movements. Additional spreading lateral to the axis of tidal movements is not considered because this would increase the area of sediment over which deposition would occur, thereby decreasing the COPC concentrations in subtidal sediment. Omitting this process in the estimation of the affected sediment area is conservative and results in an overestimate of the average COPC concentration in subtidal sediment. The MSQM was adapted from the model of Kansanen and Seppälä (1992). This model simulates contaminant profiles in subtidal sediments and computes the concentrations by solving a one-dimensional mass balance equation. Downward advection of contaminants as a result of fresh sediment deposition at the top of the profile was simulated, as were sediment mixing in the upper layers (representing bioturbation or other mixing processes) and radioactive decay or chemical decomposition. The model was implemented to individually model the fate of up to 32 COPC in a 20 cm sediment profile, sectioned into 100 slices each 2 mm thick, over a period of up to 20 years. As implemented here, processes of bioturbation, availability of oxygen (or oxygen donors such as sulphate) for chemical decomposition, and water content of subtidal sediment use a negative exponential function related to depth in the sediment profile. Data describing the sedimentation rate and characteristics of subtidal sediment in Kitimat Arm and Douglas Channel were obtained from papers compiled in Macdonald (1983). The model was calibrated against data for PAHs in a sediment core from Kitimat Arm reported by Cretney et al. (1983). Benthic invertebrates were assumed to inhabit the surface 5 cm of sediments. The MWQM simulations extended over a period of up to 360 hours, taking into consideration the potential acute effects of dissolved COPC in the water column. The COPC loadings to subtidal sediments were used as input to the MSQM to evaluate the potential for the pulse of COPC to have chronic adverse effects on subtidal sediment infauna and epifauna over a period of up to 20 years. The potential accumulation of COPC in subtidal benthic invertebrates such as crab or bivalves was considered throughout the modelling period, and was used as the basis for evaluating risks to human health through consumption of seafood (i.e., molluscs and crustaceans) that might be contaminated due to exposure to subtidal sediment.

13.2.2 Exposure Assessment The exposure assessment evaluates the extent to which a selected receptor will be exposed to COPCs. To determine the total exposure of a selected receptor to COPC, potential exposure routes are simultaneously considered for COPC modelled in the marine environment (Figure 13-1). The exposure routes include direct exposure to COPC in water and sediment for community-level receptors (marine algae, fish and invertebrates) In all model compartments, the potential risk of environmental effects was evaluated for the following aquatic and sediment community-level receptors: acute effects to marine algae, invertebrates, and fish exposed to COPC in water chronic effects to invertebrates exposed to COPC in subtidal sediment

Page 13-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Figure 13-1 Conceptual Exposure Model for Marine Key Indicators (11x8.5 figure “Annick’s figure”)

April 2010 FINAL Page 13-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

In addition, the potential acute and chronic effects of, and time to recover from, hydrocarbon compounds stranded on shorelines are evaluated in a qualitative manner for a variety of receptors, including algae, mussels, fish, birds and mammals, with reference to the experience from EVOS.

13.2.3 Hazard Assessment The hazard assessment identifies the level (exposure concentration or receptor dose) at which COPC have the potential to cause effects to the receptor‘s health. For a receptor, the threshold concentrations or doses above which an adverse effect might be expected have been defined based on reports in the primary literature, under regulatory sponsorship in various jurisdictions, and by regulatory agencies directly. These threshold concentrations are referred to as community benchmarks (for community-level receptor). In selecting appropriate effect magnitude benchmarks, preference was given to those derived through comprehensive review of effects-based studies. Guidelines such as the CCME water and sediment quality guidelines for the protection of aquatic life (CCME 2002, 2007) are relevant in a generic context, and are used to screen COPC concentrations in water and sediment to determine whether an ERA is required. However, the CCME guidelines are highly conservative in nature and are not suitable as effect magnitude benchmarks. The potential for adverse effects was quantified by comparing the estimated environmental media concentration of each COPC to the benchmark for that COPC. The quotient of the two ([COPC concentration]/[benchmark]) is referred to as a hazard quotient (HQ). Acute and/or chronic hazard indices (HI) were derived for chemicals with similar modes of action and target organs by summing the HQ of individual COPC. This methodology follows that of Di Toro and co-workers (Di Toro et al. 2000, Di Toro and McGrath 2000). Hazard indices were calculated separately for PAH compounds and petroleum hydrocarbons (TPH fractions and BTEX compounds). These groups were treated separately to avoid ―double counting‖ PAHs, which are also included in the analysis of bulk petroleum hydrocarbon fractions. An acute or chronic HI value less than 1.0 indicates that the exposure concentration is less than the threshold of toxicity for the chemical class evaluated. Given the conservative approach to the estimation of exposure and selection of benchmarks, a chronic HI less than 1 is not expected to result in adverse effects and no further assessment is required. Since the acute to chronic ratio for toxicity of non- polar narcotic substances has a value of approximately 5 (Di Toro et al. 2000), a chronic HI value greater than 5 may also be used as an indicator of potential acute effects.

13.2.4 Results of the ERA

13.2.4.1 Acute and Chronic Effects of Stranded Diluted Bitumen in the Marine Intertidal Environment The potential acute and chronic effects of a large hydrocarbon spill are best evaluated with reference to EVOS. Similarities between EVOS and this example include the volume of hydrocarbon, aspects of the receiving environment, and human use of the receiving environment. EVOS became the most studied large oil spill in history. Harwell and Gentile (2006) provide an overview of the ecological significance of residual exposures and effects of EVOS on wildlife and marine ecological receptors during the 17 years following the incident. Extensive and detailed information on the environmental effects of EVOS is also

Page 13-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment available from the Exxon Valdes Oil Spill Trustees (the Trustees 2010, Internet site). Although no two events of this nature are the same (i.e. this is due to differences in hydrocarbon types and quantities released in the receiving environments, the variables of season and weather that influence the fate of oil, and the capacity and training of response teams), information from EVOS and associated environmental monitoring was used to illustrate the types of environmental effects that have been observed and to corroborate the results of the ERA for the example in the CCAA. Following EVOS, it was reported that approximately 780 km of shoreline among the islands of Prince William Sound, and 1,315 km of shoreline in the Gulf of Alaska were oiled to some degree, and this oil represented about 50% of the total volume. Most of the stranded oil subsequently dispersed back into the ocean during the three years following the spill. Overall it is estimated that approximately 20 to 30% of the oil quickly evaporated or was otherwise diluted and dispersed; 14% was recovered and disposed; 13% was deposited in subtidal sediments (mostly as highly weathered residuals of oil previously stranded on shore); and 2% of the initial volume remained on beaches in Prince William Sound (Harwell and Gentile 2006). Estimates of the linear distance of oiled shoreline in Prince William Sound were approximately 780 km in 1989, 420 km in 1990, 96 km in 1991, and 10 km in 1992 (Neff et al. 1995). On an areal basis, it was estimated that initially between 2 and 10 km2 of beach was oiled. By 2001, it was estimated (Short et al. 2004) that only 11.3 ha (0.1 km2) of subsurface oil contamination remained. Modelling for the Wright Sound example indicates that approximately 27,372 m3 of diluted bitumen would strand along shorelines. Diluted bitumen first reaches Fin Island in approximately 8 hours and subsequently impinges on Farrant Island, Gil Island, the region of Hartley Bay, Cridge Passage, Lewis Passage, Squally Channel, Grenville Channel, Whale Channel, Campania Island, and Caamaño Sound. Altogether approximately 240 km of shoreline is affected over a period of 15 days. Since the diluted bitumen does not readily evaporate, and might be moved around by the tide, it should be assumed that the entire intertidal zone would be oiled along the shoreline areas exposed to oil. The stranded diluted bitumen can be expected to coat rocks, rockweed, and sessile invertebrates in the affected area, and some of the diluted bitumen might find its way deeper into gravel or rocky intertidal substrates. However, the stranded bitumen would not be uniformly distributed on shorelines and a small portion of shorelines might be heavily oiled, whereas less affected shorelines would be more prevalent. Comparison of the hypothetical example in Wright Sound to EVOS would suggest that more oil would be distributed along a shorter length of shoreline, on average, than was the case for EVOS. Sensitivity mapping indicates that the shorelines of the islands in Wright Sound offer a considerable area of coarse- grained beaches, where oil could both penetrate and be remobilized, but relatively few areas with potential for long oil residency. On the other hand, Wright Sound is more protected than Prince William Sound (i.e., typical fetch distances are shorter) suggesting that the effects of winter storms acting to remobilize stranded oil would be lower. Taking these factors into consideration, it is reasonable to assume that in the absence of human intervention, most of the stranded oil would be weathered or dispersed to the marine environment within 3 to 5 years. This process might represent an important secondary source of hydrocarbon contamination for offshore or subtidal sediments, but the weathered hydrocarbons themselves would have lower toxicity than fresh product. The following receptors are considered, and are intended to be broadly representative of all potentially affected wildlife and marine resources (Table 13-2). The potential environmental effects on marine water

April 2010 FINAL Page 13-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment quality and subtidal sediment quality, including its potential effects on phytoplankton, zooplankton, marine plants, benthic invertebrates and fish, will be addressed in Section 13.2.4.2.

Table 13–2 Receptors for the Marine Environment in Wright Sound

Receptor Other Wildlife and Resources Marine Water Quality Phytoplankton, zooplankton, marine plants, benthic invertebrates and fish Subtidal Sediment Quality Marine plants, benthic invertebrates and fish Intertidal Sediment Quality Marine plants, benthic invertebrates and fish Spotted Sandpiper Other shorebirds (including the Black Oystercatcher) Surf Scoter Other sea ducks Marbled Murrelet Other fish-eating seabirds Bald Eagle Other raptors Mink Other shoreline-dwelling mammals Sea Otter River Otter Steller Sea Lion Other pinnipeds Harbour porpoise Other cetaceans

Intertidal Sediment Quality The Trustees report that sediments and intertidal communities are still recovering. Approximately 350 km of shoreline was considered to be heavily oiled, and both the spill and the subsequent clean-up activities had significant effects on the flora and fauna of the intertidal zone. The intertidal sediments captured approximately 40 to 45% of the oil, and most of these beaches were cleaned using a variety of methods. By 1992, approximately 10 km of beaches remained uncleaned. Today, remaining oil is limited to a few small areas, in protected bays and beaches (the Trustees 2010, Internet site). Initial effects on the flora and fauna of the intertidal zone occurred at all tidal levels and in all types of habitats. Dominant species of algae and invertebrates including common rockweed, limpets, barnacles, mussels, periwinkles and oligochaete works were directly affected, and intertidal fish also showed lower density and biomass at oiled sites relative to reference sites in 1990. By 1991, in the lower and middle intertidal zones, algal coverage and invertebrate abundance on oiled rocky shores had returned to conditions similar to those observed in unoiled areas. In the upper intertidal zone, recovery efforts initially eliminated the rockweed (Fucus) canopy, leading to desiccation and cascading effects on other members of the community. As of 1997, rockweed had not yet fully recovered on shores oriented towards direct sunlight, but in many locations, recovery was substantial. It is undetermined how much recovery has occurred in these locations since 1997, as further work has not been conducted. Harwell and Gentile (2006) concluded that the adverse effects on intertidal habitats from EVOS were spatially extensive and ecologically significant. However, whether cleaned or not, the intertidal communities had substantially recovered within 5 years, and that after 10 years the effects were no longer ecologically significant.

Page 13-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Effects of EVOS on Fish and Shellfish Based on information reported by the Trustees (2010, Internet site), as well as Harwell and Gentile (2006), Pacific herring spawned in intertidal and subtidal habitats in Prince William Sound shortly after EVOS, and both lesions and elevated hydrocarbon levels were documented in some Pacific herring from the oiled areas at that time. Subsequently, a higher level of egg mortality and embryonic deformity was documented in the 1989 year class. These effects would be consistent with our understanding of hydrocarbon toxicity to fish; however, links to population level effects were never clearly established. Subsequently (four years after EVOS), Pacific herring populations in Prince William Sound collapsed. No clear link between EVOS and this collapse has been established, but until the population returns to pre- spill levels, the Trustees will continue to list Pacific herring as ―recovering‖ (the Trustees 2010, Internet site). Harwell and Gentile (2006) argue that the collapse of the population was likely caused by factors other than EVOS, and that the effects of EVOS on Pacific herring likely lasted no longer than 1 or 2 years. For sockeye and pink salmon, the direct effects of EVOS appear to have been negligible, although effects on these populations were caused by management responses. For sockeye salmon, a closure of the fishery due to concern about contaminated fish reaching markets appears to have led to over-escapement, and too strong a year-class of juvenile salmon in the freshwater habitat. This in turn resulted in juvenile fish returning to the sea in poor condition, with a subsequent decline in adult returns. However, the growth of juvenile sockeye salmon in fresh water habitat returned to normal with 3 years, and the Trustees have concluded that the population has recovered (the Trustees 2010, Internet site). For pink salmon, there is no convincing evidence of effects of EVOS (Harwell and Gentile, 2006). Mussels are a keystone species in the nearshore environment, and due to their filter-feeding habit and poor biochemical ability to metabolize or excrete hydrocarbons, tend to accumulate PAHs and other petroleum-derived substances. After the spill, hydrocarbon concentrations in mussels increased rapidly to high levels in oiled areas, and tissue concentrations remained high in areas where substantial amounts of oil were trapped in the sediments. However, there do not appear to have been ecologically significant effects of hydrocarbon exposure on the mussels themselves (Harwell and Gentile 2006). Mussels have been cited by the Trustees as being a pathway of exposure for several other animals, including birds and mammals, and for this reason, mussels continue to be listed as ―recovering‖, even though the areas affected are extremely limited (the Trustees 2010, Internet site). Harwell and Gentile (2006) argue that mean PAH levels in oiled and un-oiled sites were not significantly different after 1992, and that there was no EVOS-derived risk to wildlife from consumption of mussels as early as 1993. Therefore, environmental effects on mussels, as well as food-chain effects of hydrocarbons concentrated by mussels, are unlikely to be significant for more than 5 years.

Shorebirds The Trustees evaluated the effects of EVOS on black oystercatchers, a shorebird that favours rocky shorelines (the Trustees 2010, Internet site). Only nine carcasses of adult oystercatchers were recovered following EVOS, although the actual number of mortalities might have been several times higher. In addition to direct mortalities, breeding activities were disrupted by the spill and cleanup activities. The Trustees consider the black oystercatcher to be recovering. Andres (1994) reported that nesting sites were

April 2010 FINAL Page 13-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment recolonized rapidly in the early years after after cleanup activities ceased. Harwell and Gentile (2006) consider that no ecologically significant effects remained in 2006, and that the population had substantially recovered within 5 years. The spotted sandpiper is considered a good representative of the shorebirds because it exhibits characteristics of a variety of shorebirds, in addition to those of the black oystercatcher. It forages in the intertidal zone, feeding on marine invertebrates, particularly molluscs and crustaceans. It hunts through the intertidal area, searching for food visually, often so close to the water's edge it has to fly up to avoid crashing surf. Environmental effects on spotted sandpiper and other shorebirds are likely to be of similar magnitude and duration to those for the black oystercatcher.

Surf Scoter The surf scoter was selected as a representative sea duck. The Trustees evaluate the effects of EVOS on Barrow‘s goldeneye and harlequin duck (the Trustees 2010, Internet site). Sea ducks generally were vulnerable to acute mortality following EVOS, and ducks comprised approximately 25% of the carcasses recovered in Prince William Sound. Acute mortality of Barrow‘s goldeneye was likely in the thousands, and mortality of harlequin duck was likely around 1,000. Both species are still considered by the Trustees to be recovering (the Trustees 2010, Internet site). Harwell and Gentile (2006) review the current status of harlequin duck. While the data are equivocal, they conclude that recent data indicate that the population returned to pre-spill levels or higher by the mid- to late-1990s, and that the exposure of these ducks to hydrocarbons through consumption of mussels is not ecologically significant. Environmental effects on sea ducks are likely to persist for up to 10 years.

Marbled Murrelet The marbled murrelet was selected as a representative seabird. The Trustees provide information on a number of fish-eating seabirds including marbled murrelet, Kittlitz‘s murrelet, pigeon guillemot, cormorants, common murre and common loon. Of these birds, the recovery status of the marbled murrelet and Kittlitz‘s murrelet are listed as ―recovery unknown‖, while the status of the pigeon guillemot is listed as ―not recovering‖. The other birds are considered to have recovered. Thousands of seabirds died following EVOS, the majority being killed by the initial oiling. In addition to direct mortality, breeding and foraging activities were likely disrupted to varying extents by cleanup activities. In contrast to the conclusions of the Trustees, Harwell and Gentile (2006) concluded that although marbled murrelet did experience ecologically significant effects following EVOS, factors other than EVOS (including the abundance of Pacific herring) contributed to the general decline of this species throughout the region. Harwell and Gentile (2006) conclude that it is more likely that the marbled murrelet and pigeon guillemot populations at present have no detectable ecologically significant effects from EVOS. It is not likely that environmental effects would persist for more than 10 years.

Bald Eagle Bald eagle was selected as a representative raptor. The Trustees considered the environmental effects of EVOS on bald eagle, reporting that a total of 151 eagle carcasses were recovered from the area of the oil spill, and that about 250 bald eagles are thought to have died as a result of the spill (the Trustees 2010,

Page 13-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Internet site). However, reproductive success was back to normal in 1990 and 1991, and the population had returned to or exceeded its pre-spill level by 1995, and is considered to have recovered. Harwell and Gentile (2006) reach similar conclusions. Environmental effects on bald eagle are unlikely to persist for more than 5 years.

Mink The mink was selected as a representative mammal likely to frequent the shoreline, with a high probability of encountering and/or ingesting oil. The mink is not reported on by the Trustees, although they do report on the river otter (the Trustees 2010, Internet site). Twelve river otter carcasses were found following EVOS, although the actual total mortality is not known. Studies between 1989 and 1991 identified biochemical effects, reduced body size, and increased home range size in otters in oiled areas of the sound, and some of these effects persisted until 1996. The river otter is now considered by the Trustees to be recovered (the Trustees 2010, Internet site). Harwell and Gentile (2006) concur with this conclusion. Environmental effects on mink might persist from 5 to 10 years before recovery is complete.

Sea Otter Sea otters spend their life at sea, on the water surface, and are listed as ―special concern‖ under COSEWIC, and on Schedule 1 of SARA. Since the species was extirpated and subsequently re-introduced to British Columbia, it has repopulated 25 to 33% of its historical range. Sea otters are not known to be present in the CCAA, as reported in field studies (Marine Mammals TDR) but could re-colonize the area at some time in the future. Therefore they are considered here as a special case. The Trustees report that hundreds of sea otters became coated with oil in the days following EVOS, and 871 carcasses were collected. As many as 2,650 sea otters might have died. Relatively poor survival of sea otters appears to have persisted for over a decade, and the Trustees list the sea otter as ―recovering‖ (the Trustees 2010, Internet site). The lack of recovery might reflect the extended time required for population growth in a long-lived mammal with a low reproductive rate, but chronic exposure to hydrocarbons is also cited as a possible factor. Food limitation does not appear to be a factor. Harwell and Gentile (2006) discuss more complex and uncertain examples. They conclude that it is difficult to support the conclusion that sea otters were being affected by residual exposure to hydrocarbons from EVOS, even during the time period only 2 or 3 years following the spill. Declines in sea otter populations in large parts of western Alaska might be due to increased predation by killer whales, which are in turn linked to declines in Steller sea lion and harbour seal populations and ultimately to declines in fish populations; these are considered to be phenomena taking place over longer time scales and over a wider area than the environmental effects of EVOS. Harwell and Gentile (2006) conclude that there are no continuing ecologically significant effects of EVOS on sea otters, except possibly in one heavily oiled area on northern Knight Island. Thus, environmental effects on sea otters are not likely to persist for more than 5 to 10 years.

Steller Sea Lion The Steller sea lion was selected as representative of pinnipeds generally. The Trustees do not report specifically on Steller sea lion, but do report on harbour seals (the Trustees 2010, Internet site). Estimated

April 2010 FINAL Page 13-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment mortality of harbour seals was approximately 300 individuals following EVOS. Populations in Prince William Sound were declining before EVOS, likely in relation to declines in the availability of high quality forage fish such as Pacific herring and capelin, but also potentially in relation to predation, subsistence hunting and mortality due to interactions with commercial fisheries. Seal counts stabilized between 1996 and 2005, and increased between 2001 and 2005, however, from 1990 to 2005, seal numbers decreased more rapidly at reference sites than at oiled sites. The Trustees consider the harbour seal population to have recovered (the Trustees 2010, Internet site). Harwell and Gentile (2006) reach similar conclusions to those reached by the Trustees, and note similarities between population trends for harbour seals and Steller sea lion, which have also declined generally in the Gulf of Alaska over the past 25 years. Overall, it would appear that for both harbour seal and Steller sea lion, the effects might persist for 5 to 10 years.

Harbour Porpoise The harbour porpoise was selected as representative of cetaceans generally, and toothed whales in particular, including killer whale (orca), Dall‘s porpoise and Pacific white-sided porpoise, all of which might frequent the CCAA. The Trustees focus on the effects of EVOS on killer whale (the Trustees 2010, Internet site). No carcasses of any resident whales were discovered following EVOS, however, the mortality rates for pods resident in the area were higher (approximately 20% in 1989 and 1990 compared with an expected natural mortality rate of about 2%). Transient killer whales also appear to have had higher mortality rates in the area. The Trustees consider that killer whales might have inhaled petroleum or petroleum vapours, and might have eaten contaminated fish or harbour seals. Killer whales are still listed as ―recovering‖ (the Trustees 2010, Internet site). Harwell and Gentile (2006) concur with the conclusion that killer whales were acutely affected by EVOS. However, they question whether factors other than residual hydrocarbon contamination might account for the slow recovery of this species. The intensively studied AB pod of whales was subject to mortality arising from conflicts with fisheries prior to EVOS, losing key members. The failure of this pod to recover following EVOS appears to have been compensated for by other pods as the total population of resident killer whales in Prince William Sound increased from 117 in 1988 to 155 in 2003. Harwell and Gentile (2006) conclude that there might be local-scale ecologically significant effects on the AB pod, but that the larger Prince William Sound population of killer whale pods, both resident and transient, shows no signs of short- or long-term effects from EVOS. As for other marine mammals, environmental effects might persist for 5 to 10 years.

13.2.4.2 Acute and Chronic Effects in the Marine Water and Subtidal Sediment Environments For a hypothetical example resulting in the release of 36,000 m3 of diluted bitumen to the marine environment in Wright Sound, modelling indicates that approximately 6% of the diluted bitumen would naturally disperse into the water column. The MWQM contains various compartments used to describe the fate of the liquid hydrocarbon once it enters the water column. These compartments represent environmental media as well as states of liquid hydrocarbon and are used to characterize the risk to water and subtidal sediment community resources. Concentrations of COPC would not be uniformly distributed within the CCAA. Areas of high effect would constitute only a small portion of the total affected area, whereas areas of low effect would be more prevalent. As such, for the purpose of this risk assessment, an

Page 13-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment intermediate approach was considered based on instantaneous and uniform distribution of COPC over an affected area. To evaluate potential risk of acute adverse effects on aquatic biota such as fish or plankton, HI were quantified using the estimated concentration of COPC dissolved in water uniformly dispersed over the affected water area, with highest concentrations expected to occur while hydrocarbon droplets are still present within the water column and thus available for dissolution. Potential risk of chronic adverse effects on subtidal sediment biota was quantified using the estimated concentration of COPC in subtidal sediment uniformly dispersed over the affected sediment area. For this example, the HI for PAH at the water community level is well below a threshold that would indicate a potential risk of acute adverse effects on aquatic biota (Figure 13-2). However, the HI for hydrocarbon compounds (BTEX and TPH) briefly exceed the threshold that indicates a potential risk of acute adverse effects immediately following the hypothetical spill (Figure 13-2). Exceedance of the hydrocarbon compounds acute threshold (effectively a 96-hour LC50 value) was only observed for a period of 9 hours whereas acute lethality might be expected if the acute threshold is exceeded for a period of 96 hours. Values of HI for hydrocarbon compounds slightly exceeded the acute threshold between 3 and 12 hours following the hypothetical spill, with the period of exceedances corresponding to the ongoing release of diluted bitumen over a period of 13 hours. The main contributors to this HI were the BTEX and other volatile hydrocarbons, which rapidly evaporate from the water surface. Once these substances have been lost from the hydrocarbon mixture, the potential for acute toxicity due to dissolved hydrocarbons disappears due to the low solubility of the residual hydrocarbons. Values of HI for hydrocarbon compounds consequently decrease after the first day, reaching HI values less than 10% that of the acute threshold approximately 70 hours after the start of the release. As such, although acute lethality due to dissolved hydrocarbons could potentially be observed in some relatively small and confined areas, it is not expected to be widespread. This calculation assumes that aquatic biota would be continuously exposed to COPC even though exposure at any single location might not be continuous due to the movement of the affected water through tidal and wind forcing. Additionally, dispersion due to tidal cycling and flushing was not considered within the MWQM even though it is expected to occur over time, thereby further reducing COPC concentration within the water column. For the subtidal sediment community, the HI for PAH and hydrocarbon compounds were well below thresholds that indicate a potential risk of chronic adverse effects on subtidal benthic invertebrates (Figure 13-3). Hydrocarbons present within subtidal sediment are assumed to be distributed evenly over the area of affected sediments. The area of affected sediment isestimated from the area of affected water and assumed horizontal spreading caused by tidal motionl; however, lateral spreading is not considered. Hydrocarbons dissolved in the water column or present within hydrocarbon droplets are assumed to disperse beyond the area of affected sediment. Given the conservative approach to the estimation of HI at the subtidal sediment community level, and that HI for both PAH and hydrocarbon compounds were well below thresholds that indicate a potential risk of chronic adverse effects, it is unlikely that this hypothetical diluted bitumen example would result in chronic effects on benthic invertebrates in the subtidal environment.

April 2010 FINAL Page 13-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Figure 13-2 Acute Hazard Index at the Water Community-Level for a 36,000 m3 Hypothetical Diluted Bitumen Example in Wright Sound during Summer

Figure 13-3 Chronic Hazard Index at the Subtidal Sediment Community-Level for a 36,000 m3 Hypothetical Diluted Bitumen Example in Wright Sound during Summer

Page 13-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

13.2.5 Prediction Confidence Prediction confidence for the risk assessment was based on the following seven major aspects: selection of COPC selection of species environmental fate modelling resource exposure to COPC and HQ calculations chemical speciation and bioavailability reference to environmental effects and recovery observed following EVOS The selection of COPC was based on chemical analyses of representative hydrocarbon samples. Substances that were not detected in the representative hydrocarbon samples were not carried forward. Organic compounds carried forward as COPC include: BTEX and TPH compounds, detected PAH compounds, detected phenolic compounds, and two VOCs. Trace elements were determined to be of negligible concern based on their low concentration and their potential for dilution within the water column. The species selected for the risk assessment include generic water and sediment community receptors which are considered to be representative of sensitive species within the community, as well as major commercial species (such as Dungeness crab). Environmental fate models use average environmental data and simplified representations of the environment and processes involved to model the fate of COPC. Although the overall model structures are reliable, the quality of many of the parameter values describing the environmental fate and partitioning (i.e., distribution) of COPC varies. In such situations, conservative assumptions were invoked, which tends to overestimate environmental COPC concentrations, exposure of species, and likely environmental effects. The water and sediment quality modelling carried out focuses on the fate of hydrocarbons that are dispersed into the water column during and immediately following the hypothetical spill. However, experience from EVOS would suggest that a large part of the oil that is stranded on shore would subsequently be remobilized and redistributed to the water column and subtidal sediments. This oil would be considerably weathered in comparison with the freshly released oil, and the weathering process would have resulted in the loss of the more volatile, water soluble and acutely toxic components of the mixture. In addition, any recovery of stranded hydrocarbon would also reduce the potential for remobilization of hydrocarbons in the months and years following the spill. As a result, although the total hydrocarbon concentrations present in water and sediment might be underestimated in the long-term (i.e., between 6 months and 3 years following the hypothetical spill)6, it is highly unlikely that conditions of acute or chronic toxicity would occur in either the water or the sediment compartment. For assessing the maximum potential exposure to COPC, HQ and HI were calculated for the water community and sediment community species. This assumption is generally conservative because it might overestimate the exposure of a species to COPC. Hydrocarbons and many marine fauna (e.g., fish, crab) are both independently mobile, and continuous exposure of a species to COPC is unlikely during the

6 If a spill occurred in summer, there could be up to six months before the more severe winter storms remobilized oil from beaches. It is assumed that the majority of residual oil would be remobilized within 3 years of the spill (Neff et al. 1995, Page et al. 1999).

April 2010 FINAL Page 13-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment acute exposure phase (involving COPC in water). The flora and fauna of the intertidal zone, as well as birds and mammals contacting oil at the water surface would be most affected by the acute phase of the spill example. In the chronic exposure phase (involving COPC in sediment) for the examples investigated, shoreline and subtidal sediments might be affected, but the affected area is small in comparison with the areas of shoreline and subtidal sediments in the overall CCAA. While acute and chronic adverse effects are likely to occur within these relatively small areas of shoreline, these effects are not likely to be significant at the scale of the CCAA. The Marine ERA is built on measured data to the extent practical, with conservative assumptions used in the exposure and hazard assessments. Therefore, it is unlikely that the risk of adverse effects on the marine environment is underestimated.

13.3 Human Health Risk Assessment A human health risk assessment (HHRA) is a process for estimating the nature and likelihood of adverse health effects in humans who might be exposed to chemicals in contaminated environmental media. The HHRA follows processes similar to those in the ERA to address questions such as: What types of health problems might be caused by environmental stressors such as hydrocarbons? Whether some people are more likely to be exposed to environmental stressors because of factors such as where they work, where they live, or what they like to eat? What is the risk that people would experience health problems when exposed to different levels of environmental stressors? Based upon the short-term and long-term fate of hydrocarbons associated with the hypothetical example, potential risks to human receptors as a result of consuming country foods from Wright Sound and Douglas Channel (i.e., crab or shellfish from the subtidal environment) were evaluated. Human health risk assessment (HHRA) involves consideration of both the toxic properties of chemicals and the levels of exposures among people. This assessment is based on the prescriptive protocols outlined primarily by Federal Contaminated Site Risk Assessment in Canada: Part V (Health Canada 2007a) with additional guidance from Health Canada (2007b, c and d), Health Canada (1994), and US EPA Human Health Risk Assessment Protocol (US EPA 2005). The assessment focuses on physiological based health effects as a cause of the toxico-chemical interactions with the human body and its processes.

13.3.1 Methods for the HHRA The principal elements of this assessment include: problem formulation toxicity assessment exposure assessment risk characterization discussion of certainty and confidence in the prediction statement of conclusions

Page 13-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

HHRA for 36,000 m3 of diluted bitumen (Example 3 as described in Section 12.7) is based on the results of the fate modelling of the ERA (i.e., predicted concentration of contaminants in the tissue of biota from the subtidal environment) and the application of those results to the human health exposure risk model. Example 3 describes a situation for the largest volume of hydrocarbons that is considered in the ESA. The example and the COPCs are the same as those identified in the ERA. The focus of the assessment was on First Nation populations, particularly because of the proximity of many First Nation communities to Wright Sound, and because they would be expected to have higher consumption rates of locally harvested seafoods than the general Canadian population. There was no guidance available regarding the consumption rates for crustacean and mollusc seafood for First Nation populations. It was assumed that the consumption rates of crustacean and mollusc for a toddler and adult are 23.75 g/d and 55 g/day, respectively. This was based on the assumption that the rate of consumption of shellfish was half the total fish consumption rate for First Nation populations as presented in Health Canada (2007a) (i.e., 95 g/d for toddlers and 220 g/d for adults). Shellfish consumption was then equally apportioned between crustacean and mollusc consumption (i.e., ¼ of the total for each).

13.3.2 Results of the HHRA A significant adverse effect to health would be assumed to occur if the estimated exposure to a COPC exceeds the human health risk criteria established by regulatory bodies, including Health Canada (2007c), the US EPA (2010) and CCME (2007). For non-carcinogenic exposures, this is evaluated using a risk estimate referred to as a hazard quotient (HQ) while an incremental lifetime cancer risk (ILCR) is estimated for carcinogenic exposures. Based on Health Canada (2007b), an ILCR greater than 10-5 (i.e., 1 in 100,000) is considered to represent an unacceptable level of risk, while an HQ of greater than 0.2 (i.e., a total exposure that is greater than 20% of the reference exposure limit) for non-carcinogenic compounds is considered to represent an unacceptable risk. A toddler was used as the target receptor for non- carcinogenic assessments due to their high consumption rates relative to body weight. Carcinogenic risks were evaluated for the adult receptor due to the longer exposure durations and the latency periods for the development of tumours. In Section 13.2.4.2, the chronic effects of a hypothetical large-scale spill example in Wright Sound (Example 3) were assessed for the subtidal sediment community (benthic invertebrates). Subsequently, tissue COPC concentrations of benthic invertebrates (molluscs and crustaceans) that inhabit the affected subtidal sediment were predicted as a function of time for a period of up to 20 years. These tissue concentrations are the inputs into exposure modelling to estimate daily intakes for people who might consume crustacean or mollusc seafoods from the subtidal environment. It was assumed for the purposes of this assessment that the marine fish species would be less likely to be affected by hydrocarbons because they would have lower tissue hydrocarbon concentrations due to their greater capacity to metabolize hydrocarbons (CCME 1999a, b). Marine fish species are also highly mobile and have large home ranges compared to the duration and localized nature of the event. Benthic invertebrates are more likely to accumulate hydrocarbons due to their smaller areas of habitation, and longer contact time with the hydrocarbons in the subtidal sediments.

April 2010 FINAL Page 13-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

13.3.2.1 Protection of Human Health Following major oil spills (e.g., the Exxon Valdez in Alaska, the Sea Empress in Wales, and the Braer in the Shetland Islands), public advisories are commonly issued regarding the consumption of fish, crustaceans and molluscs from the marine environment. A review of recent oil spills (Yender et al. 2002) shows that advisories have ranged in duration from 15 days to more than 6 years (Table 13-3). Additionally, the aquatic community and associated fish as well as the intertidal community and associated invertebrates are typically monitored following a spill, until baseline conditions are restored.

Table 13-3 Fisheries Closures Following Recent Oil Spills

Spill Name/Location Oil Type/Volume Spill Conditions Closures M/V New Carissa / Near Two bunker oils and two Oil released in the surf Bivalves: 21 days, Coos Bay, OR marine diesels / 70,000 zone (> 5 m waves) longer adjacent to the 4 Feb 1999 gallons over several weeks vessel M/V Kure / Humboldt Intermediate fuel oil 4 days of sheens in bay; Mariculture oyster, Bay, CA (IFO 180) / 4537 gallons light shoreline oiling crabs: 49 days 5 Nov 1997 T/V Julie N / Portland, IFO 380 and No. 2 fuel Heavy shoreline oiling in Shellfish: 15 days ME oil / 180,000 gallons Fore River and Casco 27 Sept 1996 total Bay M/T Provence / Heavy fuel oil No. 6 Released in Piscataqua None Piscataqua River, NH (API 6.2) / ~880 gallons River, most of the oil 2 July 1996 sank T/V Sea Empress / Forties light crude/ Severe weather; Marine finfish: 82 days Milford Haven, Wales Heavy fuel oil #6 / extensive use of Whelk & crustaceans: 15 Feb 1996 21,274,000 gallons total dispersants 183 days Cockles: 125 days Mussel: 8-19 months T/B North Cape / Block Home heating oil No. 2 / Gale-force winds, Finfish and bivalves: 73 Island Sound, RI 828,000 gallons release in surf zone, 6 days 19 Jan 1996 to 7 m waves, naturally Lobsters: 75-155 days dispersed T/V Braer / Shetland Gullfaks light crude / Hurricane-force winds; Wild finfish: 2 months Islands 25,000,000 gallons release in surf zone, Farmed salmon: 12 5 Jan 1993 naturally dispersed months Burrowing lobster: > 6 years T/V Exxon Valdez / Prudhoe Bay crude / Over 700 km of Herring and salmon: Prince William Sound, 11,000,000 gallons shoreline oiled entire season AK Advisories on bivalves 24 Mar 1989 in 4 subsistence harvest areas

SOURCE: Yender et al. 2002

Page 13-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

The Exxon Valdez Oil Spill Trustee Council (the Trustees) provide information on the effects of the hydrocarbons on commercial fisheries and fishery closures (the Trustees 2010, Internet site). Commercial fishing was injured as a result of direct effects to commercial fish species and through subsequent emergency fishery closures. Fisheries for salmon, herring, crab, shrimp, rockfish and sablefish were closed in 1989 throughout Prince William Sound, Cook Inlet, the outer Kenai coast, Kodiak and the Alaska Peninsula. Shrimp and salmon commercial fisheries remained closed in parts of Prince William Sound through 1990. However, no spill-related district-wide fishery closures related to oil contamination have been in effect since 1989. This fact illustrates that hydrocarbons generally did not accumulate and persist in fish or shellfish tissues following EVOS, and that risk to human consumers of seafoods, including fish, crustaceans and molluscs, was generally manageable and of short duration. Mussels and other shellfish from the intertidal zone might represent an exception to this general observation (Harwell and Gentile 2006). However, following EVOS, hydrocarbon residue levels in mussels declined steadily so that by 2002, even at sites that were initially heavily oiled, the concentrations of PAH in mussel tissues were generally not significantly different from background levels measured at unaffected sites. Hydrocarbon residues are not likely to persist in mussel tissues for more than 3 to 5 years at most locations, although they might persist for up to 10 to 12 years at some heavily oiled or sheltered locations. These cases are likely to be isolated exceptions to a general pattern of rapid recovery. Protection of preferred harvesting locations through use of booming would help to reduce the risk of oiling and long-term effects7. After the Braer oil spill, which released 84,700 tonnes of light Norwegian crude oil off the Shetland Islands, UK, in January 1993, a fisheries exclusion zone was quickly established. Contamination effects were observed at salmon farms located nearby. Although no mortality of caged fish was observed, these fish became heavily contaminated and tainted with hydrocarbons including PAH. Some fish close to being marketed were destroyed, but younger fish were grown to market size and were found to have cleared themselves from taint, and to have typical background PAH concentrations by the second half of 1993. The exclusion zone for farmed salmon was lifted in December 1993. Wild fish were also contaminated, but at lower levels than seen in caged fish, presumably because they did not remain confined to areas of high contamination. The exclusion zone for wild fish species was lifted in April 1993. Shellfish, including crustaceans and molluscs, were initially more contaminated than wild fish, due to their more intimate contact with sediment and to their slower elimination rate for hydrocarbons. Exclusion orders were lifted for most shellfish between September 1994 and February 1995. An exclusion order remained in place for some Norway lobster and for some mussels until March 2000. The Sea Empress released 72,000 tonnes of light crude oil at the entrance to Milford Haven, Wales, UK in February 1996. Soon after the incident, concentrations of hydrocarbons were elevated in molluscs, and less so in crustaceans and fish, over a substantial area. Fisheries were closed on a precautionary basis over an area of 2,100 km2, and were lifted progressively based upon the results of taste testing (for taint) and chemical analysis (for hydrocarbons). The restrictions on fin fish were lifted within 3 months, followed

7 As part of the spill response planning, coastal Aboriginal communities and directly affected public stakeholders will be engaged in developing site specific spill response plans that would prioritize protection of environmentally- sensitive areas, including harvesting areas.

April 2010 FINAL Page 13-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment by crustaceans and some molluscs within 8 months. Mussels remained under a consumption advisory until June 1997 (Edwards and White 1999). Advisories regarding consumption of pelagic, bottom-dwelling and anadromous fish, and invertebrates from open water areas and subtidal sediments would probably be of relatively short duration (not more than one year). It is expected that the intertidal community and associated invertebrates (i.e., mussels) could be affected such that advisories against consumption of some seafoods could persist for 3 to 5 years or longer in some sheltered areas. As such, human health is assumed to be protected through the publication of advisories regarding the consumption of fish, crustaceans and molluscs, followed by biological, taint and chemical monitoring to confirm that baseline conditions are restored at the aquatic and intertidal community levels.

13.3.2.2 Diluted Bitumen in the Marine Environment For the hypothetical accidental release of 36,000 m3 of diluted bitumen in Wright Sound, predicted HQ values for a 20-year period for PAH, BTEX and TPH exposures through the ingestion of mussels, crabs and other type of seafood from the subtidal environment were orders of magnitude below thresholds that indicate a potential risk of chronic adverse effects. For illustration, Figures 11-4 and 11-5 show the HQ and ILCR values for benzene as a function of time. Figure 11-6 shows the HQ values for the suite of C>16 to C21 aromatic hydrocarbons (which would include the 4- and 5-ring high molecular weight PAH compounds). Figure 11-7 shows the ILCR values for benzo(a)anthracene, a representative carcinogenic PAH compound. The ILCR values for benzo(a)anthracene, benzene, and other carcinogenic hydrocarbons are many orders of magnitude lower than the threshold value of 1.0 x 10-5. Likewise, the HQ values for all of the hydrocarbon fractions are much lower than the threshold value of 0.2. The highest HQ is estimated for C>21 to C32 aromatic hydrocarbon compounds, and is less than 0.001 throughout the first year. For most other compounds, HQ values are less than 0.00001 at all times. These calculations assume that the a person‘s seafood intake is sourced exclusively from within the affected subtidal sediment area as defined in Section 13.2.1.2, and that his/her daily consumption of seafood would be continuous for 20 years after the event. Hydrocarbon concentrations in mussels that could be harvested along the shoreline are less amenable to forecasting, since effects would be highly site-specific. As noted above, mussel tissues would generally return to levels similar to background levels within 3 to 5 years, although at heavily oiled and sheltered sites, effects might persist for 10 to 12 years. It is assumed that monitoring would identify areas where the harvesting of shellfish from the intertidal zone would not be recommended, based upon consideration of potential human health effects.

13.3.3 Prediction Confidence The HHRA, and thus this assessment, is highly dependent on the results of predictive modelling. The modelling involves predicting the concentration of hydrocarbon chemicals in the surface water following a hypothetical release and their transfer to subtidal sediment. Subsequently concentrations within the various crustaceans and molluscs that inhabit the affected subtidal sediment are predicted as a function of time. Predictions are based on computerized mathematical hydrocarbon fate and bioaccumulation models.

Page 13-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

These approaches are well developed and are generally accepted by the regulatory agencies that have jurisdiction over this assessment. Predictions on effects to human health have a high level of confidence. Table 13-4 summarizes the assumptions used in the assessment. It shows that the assumptions followed are likely to be neutral or to overestimate exposures, and therefore provide a medium level of confidence to the conclusion of no adverse effects on human health.

Table 13-4 Analysis of Assumptions Used in the HHRA

Over- or Under- Assumption Justification Estimated Opinion Surface water See Section 12 Neutral Reasonable average exposure modelling was concentrations at receptor locations conservative in were used in the comparison to developing COPC guidelines for the protection of human concentrations at the health individual receptor locations Bioaccumulation Basic bioaccumulation Over- estimated Assumed that all bioaccumulation in the modelling was model tissues of the shellfish would conservative in accumulate considering conservative developing COPC biological mechanisms factors concentrations in shellfish Guidelines that Conservative Neutral Guidelines were obtained from exposures were assumption appropriate regulatory agencies and compared to were accorded relevance in, the following protective of human order of priority: Health Canada, then health British Columbia, CCME, Alberta, Ontario and Texas. All potential receptors Information used in Neutral Age groups to be addressed are those were considered for the HHRA approach specified by Health Canada (2007b): the risk assessment for receptor selection toddlers (7 months to 4 years) was chosen in adults (20+ years) accordance with First Nation populations were also protocols provided by considered. Health Canada (2007b) All potential exposure Information used in Neutral Operable and inoperable exposure pathways were the HHRA approach pathways should be identified and a considered for human for exposure pathway rationale provided for pathways receptors at the site selection was deemed inoperable at the subject site selected in accordance with protocols provided by Health Canada (2007b)

April 2010 FINAL Page 13-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Benzene

1.8E-08

1.6E-08

1.4E-08

1.2E-08

1.0E-08

Mollusc ingestion 8.0E-09

Crustacean ingestion HazardQuotient

6.0E-09

4.0E-09

2.0E-09

0.0E+00 0.08 0.25 0.42 0.58 0.75 0.92 2 4 6 8 10 12 14 16 18 20 Years after spill event

Figure 13-4 HQ Estimates for Benzene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer

Page 13-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Benzene

3.50E-13

3.00E-13

2.50E-13

2.00E-13

Mollusc ingestion 1.50E-13 Crustacean ingestion

Incremental LifetimeCancer Risk 1.00E-13

5.00E-14

0.00E+00 0.08 0.25 0.42 0.58 0.75 0.92 2 4 6 8 10 12 14 16 18 20 Years after spill event

Figure 13-5 ILCR Estimates for Benzene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer

April 2010 FINAL Page 13-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

C>16 - C21 Aromatics

4.5E-04

4.0E-04

3.5E-04

3.0E-04

2.5E-04

Mollusc ingestion 2.0E-04

Crustacean ingestion HazardQuotient

1.5E-04

1.0E-04

5.0E-05

0.0E+00 0.08 0.25 0.42 0.58 0.75 0.92 2 4 6 8 10 12 14 16 18 20 Years after spill event

Figure 13-6 HQ Estimates for C>16-C21 Aromatics – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer

Page 13-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Benzo(a)anthracene

1.20E-09

1.00E-09

8.00E-10

6.00E-10 Mollusc ingestion Crustacean ingestion

4.00E-10 Incremental LifetimeCancer Risk

2.00E-10

0.00E+00 0.08 0.25 0.42 0.58 0.75 0.92 2 4 6 8 10 12 14 16 18 20 Years after spill event

Figure 13-7 ILCR Estimates for Benzo(a)anthracene – Hypothetical 36,000 m3 Diluted Bitumen Spill in Wright Sound During Summer

13.4 Follow-up and Monitoring Northern Gateway has invited coastal Aboriginal groups to participate in the establishment and monitoring of permanent environmental monitoring transects and sites. Information from these monitoring programs would be used to obtain information on baseline environmental conditions, including the quantity and quality of important harvested species and existing environmental quality. Offers to undertake a monitoring program have been extended to several coastal Aboriginal communities (Gitga‘at, Kitkatla and Lax Kw'alaams) as part of ongoing engagement activities, and similar offers will be extended to other coastal communities as engagement progresses. The following types of monitoring programs might be implemented, as appropriate, to evaluate potential chronic environmental effects: Water quality monitoring: daily to weekly measurements of BTEX, TPH and PAH compounds in the marine water receiving environment, representing near-field, far-field and reference areas, until it is confirmed that baseline conditions are restored.

April 2010 FINAL Page 13-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 13: Risk Assessment Related to Hydrocarbons in the Marine Environment

Sediment quality monitoring: annual measurements of BTEX, TPH and PAH compounds in the marine sediment receiving environment, representing near-field, far-field and reference areas, until it is confirmed that baseline conditions are restored. Biological monitoring: shoreline flora and fauna including the rockweed community and associated invertebrates should be monitored initially on a monthly basis, switching to an annual basis after the first year, until it is confirmed that baseline conditions are restored. Chemical monitoring: hydrocarbon residues in edible fish and shellfish (crustacean and mollusc) tissues from affected and non-affected areas should be monitored initially on a monthly basis, switching to a semi-annual basis after the first year, until it is confirmed that baseline conditions are restored. Taint monitoring: edible fish and shellfish from affected and unaffected areas should be monitored for taint by a trained taste panel, initially on a monthly basis, switching to a semi-annual basis after the first year, until it is confirmed that taint is no longer detectable. It would be feasible for the water, sediment, biological, chemical and taint monitoring to be undertaken by coastal Aborginal communities or a community-based organization, particularly if the community was involved in the establishment and monitoring of baseline environmental monitoring transects and sites.

Page 13-30 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment

14 Vapour Cloud Simulations and Risk Assessment The potential for a vapour cloud to form in the event of an accidental release and the associated risk to human populations was examined. Of the three hydrocarbon products that will be transported as part of the Project, only condensate (CRW) has the properties necessary to form a vapour cloud. Spill probabilities for condensate were determined as part of the QRA. The following section is a summary of the vapour cloud work completed for TERMPOL (Bercha 2010). The analytical work was based on accepted and state-of-art methods of consequence evolution, consequence analysis, and risk analysis in the following principal areas: Hazard scenario definition Dispersion and fire and explosion analysis Ignition frequency evaluation Risk analysis The risk analysis section on the vapour cloud analysis is considered preliminary as detailed engineering and staffing requirements at the terminal have not been completed.

14.1 Hazard scenario definition The initial hazard and its general location, a marine tanker condensate spill, was defined based on the QRA and TERMPOL requirements. Based on the QRA, three scenarios were identified: A grounding incident which results in a condensate spill A two ship collision incident which results in a condensate spill A spill at the terminal during discharge operations. Following the initial definition, it was necessary to estimate the credible worst case location of spill occurrences for each of the three specified scenarios. This was carried out by an inspection of satellite, bathymetric, and route data. Essentially, the hazard scenario definition consists of the identification of the initiating accident and its spatial and temporal characteristics. This includes the spill location and release rate of the spill. For marine liquid hydrocarbon spills initiated by accidents such as groundings or collisions, there is generally an initial, nearly instantaneous release of volume, followed by a continuous decaying discharge for a period of time following the initial release. The behaviour of the spill is dependent on the material properties of the spill substance, including both hydrodynamic and vaporisation characteristics, as well as the likely ambient conditions – in this case, associated with the credible worst case for spill and evaporate transport. Release volumes were taken from the QRA (DNV 2010). A 10,000 m3 spill resulting from a grounding incident is considered a credible worst case (DNV 2010), and the probability of such a release exceeding 10,000 m3 is very low: less than 3% for the condensate tankers (Suezmax or smaller). A 28,980 m3 spill resulting from a two ship collision incident involving a Suezmax condensate tanker is considered a

April 2010 FINAL Page 14-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment credible worst case, with the probability of a release exceeding that volume being 10% (DNV 2010). Spill scenarios used in the analysis and their general locations are outlined below and in Table 14-1:

Table 14-1 Hazard scenario summary

Release Scenario Description CRW Density Initial Continuous Duration (kg/m3) (m3) (m3/hr) (hr) Scenario 1 Wright Sound collision (WSC) 800 7,245 1,811 12 Scenario 2 Douglas Channel grounding 800 1,896 632 12 on Halsey Point (WSG) Scenario 3 Kitimat Terminal (KTW) 800 250 n/a n/a unloading spill

In a sensitivity analysis for Scenario 1 carried out to identify worst case conditions, it was found that the maximum extent of the lower flammability limit (LFL) isopleths is associated with the low wind speeds and high ambient and water temperatures, with a maximum extent of ½LFL occurring for the stable atmosphere, low wind speed, high temperature out to a distance of 1.53 km. Scenario 1 represents the largest release, resulting from a mid-channel collision of the CRW tanker with a sufficiently large vessel to cause the specified spill. Dispersion analysis carried out for worst case conditions indicates the dispersion time plans and profiles. For the given conditions, the maximum extent of the flammable zone out to 1.7 km occurs at approximately 8 minutes, and thereafter, due to dispersion, the flammable zone decreases, to one of approximately 1.0 km after 60 minutes. The vapour cloud is not predicted to reach the community of Hartley Bay, based on a credible worst case scenario of a mid channel collision which results in an unlikely large condensate spill. Scenario 2 involves the grounding on Halsey Point and resulting spill of condensate in the vicinity of Hartley Bay. Based on the sequential dispersion isopleths, the maximum extent of the flammable (LFL) zone occurs at 5 minutes out to 1.1 km. Scenario 3 is characterized by a 250 m3 instantaneous spill of condensate occurring at an unloading arm at one of the terminal berths. Based on the time dispersion snapshots for the critical concentration levels associated with this scenario, the maximum flammable zone occurs after 1 minute out to a distance of approximately 200 m.

14.2 Dispersion, Fire and Explosion Analysis The fire and explosion modeling was carried out for the worst case vapour cloud flammability conditions for each of the scenarios, together with a subsequent pool fire. Table 14-2 summarizes the results of the fire and explosion analysis for each of the scenarios, giving the maximum distances to critical isopleths associated with significant damage criteria.

Page 14-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment

Table 14-2 Fire and explosion analysis results summary

Max Isopleth Distance (m) Explosion Pool Fire Thermal Flash Fire (20 min) Explosion Radiation LFL Thermal Radiation Overpressure Overpressure (kW/m2) (ppm) (kW/m2) (kPa) (kPa) Scenario 20 10 5 12000 170 70 20 70 35 25 70 35 25 Wright 417 433 578 1,653 746 754 758 182 309 388 1,735 1,862 1,941 Sound collision Douglas 261 272 375 1,073 504 510 512 134 231 289 1,107 1,204 1,262 Channel grounding Kitimat 90 98 146 248 179 183 185 73 124 155 221 272 303 Terminal unloading

14.3 Ignition Frequency Evaluation In the ignition analysis, the evolution of consequences following the occurrence of the spill, and its probable exposure to ignition sources is evaluated. The results of the evaluation by the relative likelihood of different damaging consequences, including pool fires, flash fires, and explosions. Because of the nature of the vapour cloud components, dispersion without ignition is not considered a damaging consequence. The probability of different outcomes from the spill is evaluated using an event tree, shown in Figure 14-1. As can be seen, starting at the left side, for each of the scenarios, the likelihood of ignition, given the occurrence of the spill, is estimated. The highest likelihood of ignition, of 0.6 or 60%, has been assigned to the collision scenario because of the likely high impact energies that could generate auto- ignition in the collision event, followed by escalating failures, including electrical equipment that may likely produce sparks. Next, for scenario 1, it is estimated that given ignition, it can be immediate or delayed. As can be expected, immediate ignition is more likely than delayed ignition. The collision itself and its immediate aftermath is likely to cause sufficient sparking or fires to ignite the evaporating vapour cloud. Next, in the instance of immediate ignition, a fire would occur in the early stages of the collision aftermath. In the case of delayed ignition, with a 20% likelihood given that ignition does occur, the most likely outcome is a flash fire, which occurs when an ignition source is introduced into the vapour cloud, resulting in a wave front radiating from the ignition source in all directions throughout the flammable zone. Explosion, in the instance of sufficiently high concentration could occur, but has been deemed to occur at the very low probability of 1%.

April 2010 FINAL Page 14-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment

Scenario Ignition Timing Consequence Ratio of Occurrence (ROO)

Immediate WSC 0.80 Early PoolFire 0.480 Pool Fire 0.600 WSC Ignition WSG 0.20 0.060 0.300 WSG 1 - WSC 0.60 KTW 0.50 0.100 0.200 KTW 2 - WSG 0.30 Flash Fire 3 - KTW 0.20 inert expl proof WSC 0.99 Flash Fire 0.119 WSG 0.95 0.228 1 - CRW-WSC KTW 0.90 0.090 2 - CRW-WSG Delayed 3 - CRW-KTW WSC 0.20 Explosion WSG 0.80 WSC 0.01 Explosion 0.001 KTW 0.50 WSG 0.05 0.012 KTW 0.10 0.010 Non Ignition 1 - WSC 0.40 Spill / Dispersion 0.400 WSC 2 - WSG 0.70 0.700 WSG 3 - KTW 0.80 0.800 KTW

WSC – Wright Sound Collision , WSG – Douglas Channel Grounding, TTW- Kitimat Terminal Workers

Figure 14-1 Event tree For the case of the second scenario, the grounding, ignition is much less likely as there is unlikely to be significant above-water impact to cause auto-ignition. Accordingly, a 30% probability has been assigned to scenario 2, the grounding. Auto-ignition, therefore, is also unlikely. In the event of an ignition, it is likely to be a delayed one, such as the vapour cloud reaches an ignition source such as a fishing vessel. Scenario 3, the terminal spill, has the lowest probability of ignition because, the proposed Kitimat Terminal will have explosion proof electrical fittings, forbid open fires (such as welding) during loading and unloading, and exercise other precautions that would significantly reduce the likelihood of ignition of any vapours of flammable concentrations at the terminal. Given that ignition does occur, at the relatively low likelihood of 20%, it is estimated that it is equally likely that it is immediate or delayed. In the case of explosion, however, it is significantly more probable due to the potential confining spaces that could result in vapour pockets of high concentration. Accordingly, an explosion, given delayed ignition, has been assigned a probability of 0.1 or 10%. The resultant probabilities are relative probabilities of occurrence or ratios of occurrence (ROO) are those given in the final column of each of the consequence streams.

14.4 Risk Analysis Individual specific risks (ISR), the risks to a specific individual considering the time spent at the designated location indoors and outdoors, have been quantified in the form of risk transects for each of the three scenarios. Table 14-3 summarizes the results from these individual risk transects. The maximum ISR for Scenarios 1 and 2 is less than 1 in 1 million, while the maximum ISR for the receptors nearest to the source effectively ranges between zero for Scenario 1 and 1 in 10 million per year for Scenario 2.

Page 14-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment

Given a threshold of significance of 1 in 1 million per year, risk for both scenarios 1 and 2 is insignificant. In both Scenarios 1 and 2 the vapour cloud did not reach Hartley Bay.

Table 14-3 Summary of individual risk assessment results

Acceptability ISR for Northern Scenario Maximum ISR Nearest Receptor Public Gateway Workers Wright Sound 4 x 10-7 0 Insignificant Insignificant collision Douglas Channel 3 x 10-7 1 x 10-7 Insignificant Insignificant grounding Kitimat Terminal 8 x 10-5 8 x 10-5 Grey Grey unloading

Scenario 3 ISR are higher, showing a level of 8 in 100 thousand per year as a maximum and the same for the nearest receptors, since receptors could be located anywhere up to the location of the spill. In terms of public ISR acceptability, levels below 1 in 10 thousand are considered tolerable but require mitigation to ensure risk is As Low as Reasonably Practicable (ALARP). The acceptability of these risk levels for Northern Gateway personnel depends on the interpretation of worker individual risk thresholds. In general, risk thresholds for operator or project owner employees are relaxed to be somewhat higher than those for third parties or members of the public. Regardless of where these thresholds are set, the risks to personnel at the terminal require mitigation to ensure the risk is ALARP. Collective risks were evaluated for the terminal only. Collective risks are best depicted on a risk spectrum or frequency–number (F-N) curve such as that shown in Figure 14-2. The risk spectrum solid line is above the lower and below the upper risk threshold shown on this spectrum, indicating that risks are in the acceptable region, but requiring risk mitigation. It was shown that risks from the collision, Scenario 1, and the grounding, Scenario 2, are in the insignificant region for any receptors, while those associated with the terminal spill fall into the grey region requiring mitigation. Thus, terminal spill risks require mitigation to reach ALARP, while the other two scenarios do not require additional mitigation. Mitigation at the terminal will include ignition source reduction, closed loading, and limiting personnel access to the berth area during cargo transfer.

April 2010 FINAL Page 14-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 14: Vapour Cloud Simulations and Risk Assessment

Terminal Collective Risk Spectrum

1.00E-03

1.00E-04

1.00E-05

Annual Chance of N or More Casualties More or N of Chance Annual 1.00E-06

1.00E-07 1 10 100 1000 Number of People (N)

Figure 14-2 Terminal collective risk spectrum

Page 14-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

15 Mitigation and Amelioration Northern Gateway is committed to reducing project risk as described below.

15.1 Shipping Risk Mitigation Northern Gateway is committed to confirming that tankers transporting condensate, diluted bitumen and synthetic oil to and from the Kitimat Terminal adhere to high safety standards and operate in an environmentally responsible manner. The safe passage of marine traffic will be achieved through a comprehensive strategy that brings together highly trained professionals, technology and planning. The strategy includes the following: All vessels navigating within Canadian waters and calling at the Kitimat Terminal must be in full compliance with all relevant shipping regulations and safety standards required under the Canada Shipping Act. Tankers calling on the Kitimat Terminal will have double hulls, double bottoms and separate tanks for ballast to prevent ballast seawater from coming in contact with hydrocarbon products. Ships officers and crew will be certified in accordance with the IMO International Convention on Standards of Training, Certification and Watchkeeping for Seafarers, 1978 and the 1995 amendments to this convention. Full bridge simulations have be done, and will be continued with pilot and tug crew training to identify navigational risks and assess potential emergency situations (such as engine or steering failures), as well as recovery from these situations (e.g., with assistance from escort tugs). As ships approach or leave the CCAA, experienced coastal pilots with local knowledge will board the tankers at designated locations. Pilots will be equipped with independent navigational systems and will provide guidance to the ship‘s master during the voyage into and out of the Kitimat Terminal. All laden tankers in the CCAA will be accompanied by two escort tugs with one tug attached (tethered) to the tanker, and the second tug in close escort. Ballasted tankers within the CCAA will be accompanied by a close escort tug. In the OWA, all tankers (laden and ballasted) will be accompanied by one close escort tug between the pilot boarding station (at Triple Island, or at potential stations at Caamano Sound and Browning Entrance) and the CCAA. Escort tugs will also be available for emergency (ocean rescue) response outside the CCAA. All tugs will be equipped with emergency response equipment and would act as the first response to a hydrocarbon spill. Safe transit speeds for tankers and tugs underway will be identified in Northern Gateway‘s Port Information Book to limit shoreline disturbances and reduce the likelihood of navigational incidents. Speed restrictions would also allow a tethered escort tug to arrest a ship in the event of a complete engine failure.

April 2010 FINAL Page 15-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

Tanker speeds should be adjusted in Wright Sound when traffic is present to avoid collision. As briefly discussed and discussed in more detail in other TERMPOL documents, tankers will modify their speed in certain areas that are known to be more challenging to navigation or for environmental reasons such as the presence of marine mammals. Radar will be installed along important sections of the ship routes to monitor all marine traffic and provide additional information to the MCTS centre and to pilots. Operational safety limits will be established to cover visibility, wind and sea conditions. Northern Gateway is also committed to environmental protection measures to safeguard coastal resources from potential maritime accidents, as demonstrated through the following examples: Tanker vetting criteria for all tankers calling on the Kitimat Terminal will conform to the highest standards of tanker construction, maintenance, onboard navigation and communication equipment and response capability. Northern Gateway will decline the nomination of tankers with cargo tank arrangement extending the width of the tanker (minus the ballast tanks). Only tankers with centreline partitions will be used for transporting hydrocarbons to and from the Kitimat Terminal. Increasing the number of tanks and reducing the volume per tank will limit the cargo spilled if a tank is penetrated. Operational environmental limits will be identified in Northern Gateway‘s Terminal/Transhipment Site Operations Manual (TOM) for tanker and cargo handling at berth. Terminal personnel will be highly skilled and trained to deal effectively with operational incidents including the use of an Emergency Shutdown System. Crews of the escort tugs and other support vessels (e.g., harbour tugs) will have extensive training in emergency response. The escort tugs and harbour tugs will carry response equipment that meets the requirements for a Tier 1 response (see Section 5.5). This equipment will include containment boom, anchors, skimmer system and temporary storage tanks for recovered oil. All tankers calling on the Kitimat Terminal will be required to have a Shipboard Oil Pollution Emergency Plan (SOPEP) under Regulation 26 of Annex I to the International Convention for Prevention of Pollution from Ships (MARPOL 73/78). All tanker operators are required to have a contractual agreement with a certified Oil Spill Response Organization, and to carry suitable insurance coverage under the Civil Liability Convention. Response equipment will be stored at several locations within the CCAA and OWA. Locations will be determined as part of the response planning process and through discussions with participating groups and the federal and provincial government. Locally based staff will be trained in the effective deployment of response equipment.

Page 15-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

Northern Gateway will, in consultation with participating groups, define areas of particular ecological sensitivity and assess whether installation of permanently located quick deployment first response systems is viable for the most critical of these locations. Each tanker berth will be equipped with a containment boom, which will be deployed during all loading operations for oil. The containment boom will extend from shore, out around the tanker and back to shore. The containment boom will not be used during condensate off-loading, both because condensate dissipates quickly, and to reduce the risk of ignition. Specific mitigation relevant to TERMPOL 3.15 is provided in greater detail below.

15.1.1 Escort Tugs The tug plan for Northern Gateway will be as follows: All laden tankers will have a close escort tug between the pilot boarding stations at Triple Islands, Browning Entrance and Caamaño Sound and the Kitimat Terminal (Segments 1, 2, 3, 4a, half of 4b, 6 and 7). In addition all laden tankers will have a tethered escort tug between Browning Entrance and Caamaño Sound and the Kitimat Terminal (i.e., throughout the CCAA, Segments 1, 2, 3, 6 and 7). All tankers in ballast will have a close escort tug between the pilot boarding stations at Triple Islands, Browning Entrance and Caamaño Sound and the Kitimat Terminal (Segments 1, 2, 3, 4a, half of 4b, 6 and 7). The action taken by an escort tug boat will depend on instructions from the captain and pilot onboard the tanker and will vary with the position of the tanker and the nature of the unfolding incident. The four basic operations are listed below: Brake – Arrest - slow the tanker as fast as possible Steer-Brake - steer the vessel on a safe course, and at the same time apply braking forces Steer - tug acts like the rudder of the ship and steers the tanker on a safe course Assisted (―U‖)-turn – Brake - tug will turn the tanker 180º or more to reduce speed and or avoid grounding

15.1.2 Improvements to Navigational Aids In response to the recent interest in the development of marine projects in the Kitimat and Prince Rupert areas, the CCG conducted Level of Service (LOS) reviews of Aids to Navigation in the region. The objectives of these reviews are to analyze the existing aids to navigation systems and recommend improvements that will enhance safety and reliability of the navigation systems. In addition, the BC Coast Pilots have developed a list of navigation aid additions and improvements for the principal shipping routes to Kitimat Terminal.

April 2010 FINAL Page 15-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

In addition to making recommendations on any shortfalls in the current systems, the reviews also identify any redundancies or unnecessary aids to navigation. The review has recommended several additional navigational aids in areas as follows (Source: TERMPOL 3.5): Shallows in Caamaño Sound Lewis Passage The additional navigational aids are assessed to have a low risk reducing effect in the Confined Channel Assessment Area and are only relevant for powered grounding. Navigational aids, however, are considered to have a medium risk reducing effect in the area of Lewis Passage and Caamaño Sound. It should be noted that increased aids to navigation was cited as an important potential risk reducing measure in interviews with local stakeholders. The aids should have a positive risk on reducing incidents not only involving tankers travelling to and from the Kitimat marine terminal, but also other marine traffic in the area of the three routes. The final definition of navigation aid improvements will be made in consultation with The CCG and industry stakeholders during the project design phase.

15.1.3 Electronic Chart Display and Information System (ECDIS) IMO is working on guidelines making the installation of ECDIS (Electronic Chart Display and Information System), with back-up mandatory on all tankers. One study indicates that the installation of ECDIS, with approved charts, may reduce the risk of powered grounding by up to 36% (source: MSC 81), while another indicates a reduction of the frequency of powered groundings by 11 % to 38 % (source: NAV 52). Both these studies compare situations with and without ECDIS. Many tankers already have ECDIS. It is likely that the accident statistics already partly includes the effect of ECDIS, and therefore the net additional effect of a general requirement for ECDIS for the tankers calling at the Kitimat Terminal is difficult to estimate. However, for an individual tanker otherwise without ECDIS installed, the installation of ECDIS is expected to reduce the probability of powered grounding by some 30 %, and the total probability of an oil spill by some 15 to 20 %. Although it is not required in SOLAS to have ECDIS installed on existing tankers until 1 July 2015, many ships have already installed the system. SOLAS requires ECDIS to be fitted on new tankers constructed on or after 1 July 2012. DNV recommended that ECDIS is installed on the tankers calling at the Kitimat Terminal, and this recommendation will be incorporated into the ship vetting process.

15.1.4 Improvements to Vessel Traffic Service (VTS) The existing Vessel Traffic Service (VTS) system on Canada‘s west coast involves reporting requirements for vessels over a certain size at designated call-in points. The existing VTS system on the north coast focuses heavily on marine traffic within the Inner Passage route, due to its historic and continued use as a marine transportation corridor.

Page 15-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

Today, however, larger vessels travelling the coast and to and from Kitimat are more often using Hecate Strait and the wider Outside Passage. Based on discussions in TERMPOL 3.5 & 3.12, additional calling- in points provided within the Outside Passage would enhance the effectiveness of the current VTS system and increase navigational safety in the area. Similarly marine radar coverage relayed to Prince Rupert VTS at certain points such as Wright Sound could greatly enhance both the overall VTS capability and the navigational safety of all vessels transiting the Confined Channel Assessment Area. The risk reducing effect of the enhanced VTS system is assessed to be low for the proposed vessels with the current level of traffic. The main risk reducing effect of the VTS will be on collisions which already have a low risk of occurrence given the relatively low traffic density. Improvements such as the installation of radar coverage proposed to Northern Gateway to augment VTS systems has a risk reduction effect on all shipping in the area and therefore the overall risk reducing effect would be greater than the effect for this project alone.

15.1.5 Traffic Separation In many coastal areas traffic separation schemes have been implemented in order to reduce the risk of collisions. The traffic separation schemes are similar to roads on land where inbound traffic would sail on one side and outbound traffic on the other. In this way one would reduce the risk for head on collisions. The collision risk for the proposed tankers is assessed to be low. Therefore, the effect of implementing the traffic scheme would also be low, and the potential effect on oil spill risk very limited. However a traffic separation scheme would make it easier for small recreational crafts in the area to keep out of the way of passing larger vessels as they would know which side the tankers would transit. Given the low cost of implementation and that traffic separation would have a positive effect for all traffic in the area, it is recommended that traffic separation be assessed for the area.

15.2 Terminal Risk Mitigation Measures taken to prevent spills will include: designing the marine terminal to comply with applicable codes and standards and adequate monitoring systems developing materials specifications for components to be used in the marine terminal developing a plan that describes the design process, specifies the need for constructability and operability assessments, and outlines the methodology for design review and verification, including the process for checking produced documents (i.e., the engineering quality control process) allowing only tankers that are designed for closed loading systems with backup audible and visual alarm to call at the terminal to enable operations to be shut down before a potential tank overfill (The International Safety Guide for Oil Tankers and Terminals, and International Chamber of Shipping 2006)

April 2010 FINAL Page 15-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

Particular attention will be paid to implementing industry-best practices and mechanical measures to avoid spills. During operations, the following measures will be taken at the marine terminal: operational procedures will be in place for unloading and loading liquid hydrocarbons, including using trained personnel to monitor tanker loading and unloading operations fail-safe valves for connecting loading arms to the oil tanker‘s manifold automatic shut-off valves and emergency release coupling will be installed on loading and unloading arms secondary containment for loading and unloading arms and associated fittings a containment boom will be placed at each tanker berth and deployed during all oil loading operations. The containment boom will extend from shore, out around the tanker and back to shore. Because condensate dissipates quickly, the containment boom will not be used during condensate off- loading. electronic sensors on arms to monitor position and engage the automatic shut-off valve, if the oil tanker moves beyond the safety limit of the loading arm envelope during transfer of liquid hydrocarbons a mooring load monitoring system to mitigate mooring lines overloading and minimise ship movements at berth a dock monitoring system which will provide data on the ships speed of approach during berthing. an alarm system will provide a warning and automatic shutdown time of approximately 47 s for oil and 60 s for condensate if a release is detected use of adequate number and capacity of tugs during berthing and unberthing of oil tankers an exclusion zone or navigational restrictions will be in place while tankers are at berth wind speed monitoring during loading and unloading operations, with cargo transfer stopped when weather limitation criteria are reached The following measures will also be implemented for detection, containment and cleanup: leak detection system continuous system monitoring through the Supervisory Control and Data Acquisition (SCADA) system routine visual inspection routine checks of valves conformance to design codes emergency response plans capability to immediately deploy containment and response equipment at the marine terminal emergency response equipment

Page 15-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 15: Mitigation and Amelioration

emergency response training establishment of trained response crews Additional details on these procedures and equipment are provided in TERMPOL Study 3.10, 3.11 and 3.13. Prevention and emergency response from oil handling facilities is strictly regulated in Canada. An oil pollution emergency plan, pursuant to the Canada Shipping Act, will be in place for the marine terminal.

15.2.1 Closed Loading (with Vapour Return System) Closed loading is a design feature that is in addition to the many cargo monitoring systems on board modern tankers, and the many spill prevention measures now built into tankers including a deck containment system in place while in port. Any oil spilled by overloading the cargo tanks, and for some reason not collected by the vapour return system would be captured by the deck containment system and stored in the vessels slop tanks. The Kitimat Terminal is to be designed with a vapour emissions control system in order to collect the vapours that are displaced from the cargo tanks during loading . If the cargo tanks were to be accidentally overfilled the excess oil can be redirected into alternate (empty) ships tanks by the vapour control system thereby eliminating the risk for oil spill.

15.2.2 Ignition Source Prevention and Suppression The Kitimat terminal will employ state-of-art ignition source prevention and suppression. This will include explosion proof electrical facilities, and full range of work place ignition protection practices. These measures will reduce ignition probability in the instance of a spill and ensuing vapour cloud.

April 2010 FINAL Page 15-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 16: Conclusions and Recommendations

16 Conclusions and Recommendations Historical spills demonstrate that oil and condensate can adversely affect the biophysical and human environments, and these potential effects are outlined in the assessments above. As a result, prevention is a key priority for Northern Gateway and has been the focus of planning for the marine transportation component of the Project. The Project will incorporate the most modern technology and most stringent requirements for tankers. All tankers will be required to have double hulls, double bottoms and separate ballast tanks and must observe speed restrictions and operational safety limits. Tanker operators will have the highest level of mariner training, and local pilots will board tankers in the CCAA for additional navigational aid. In addition, radar will be installed along the Northern and Southern Approaches, and laden tankers will have tethered and close escort tugs in Principe Channel, Caamano sound, Douglas Channel and Kitimat Arm to reduce the risk of an accident. As outlined in Sections 4 and 5, the likelihood of a spill occurring in the CCAA or OWA is low (Figure 16-1). Northern Gateway will have response capabilities in place that exceed regulatory requirements. Equipment and logistics support and a detailed marine OSRP with identified priority response areas will be in place prior to any handling of bulk oil. First response stations, local personnel, and equipment will be located at the proposed Kitimat Terminal and throughout the CCAA to enable rapid response. First response stations will include boom, response vessels, skimmers, pumps, temporary storage, shoreline cleaning equipment, and equipment to respond to affected wildlife. The marine OSRP for the Project and vessel-specific SOPEPs will be integrated with provincial and CCG response and contingency plans. The potential effects of a hydrocarbon spill depend on numerous factors, including type of oil, size, time of year and type of environmental receptor. Effects are largely reversible with appropriate cleanup strategies and natural recovery. Spills to the marine environment would result in hydrocarbons being released into the water, spreading onto the shore and evaporating into the air. Condensates evaporate quickly, typically producing short- lived toxic effects. Persistent oils (e.g., diluted bitumen, synthetic oil) can also result in immediate adverse effects, mainly through direct contact (e.g., coating, smothering), but also tend to cause more chronic (long-term) toxic effects. Organisms likely to come in direct contact with spilled oil include marine birds, fish, marine mammals, and intertidal invertebrates and vegetation. Terrestrial biota along the shoreline might also come in contact with spilled oil. Historical marine spills have shown that a wide range of organisms might be adversely affected. Adverse effects have been documented for shoreline vegetation, intertidal invertebrates, fish, waterbirds and furred aquatic and semi-aquatic mammals. Depending on spill size, human activities (e.g., traditional harvesting, subsistence, commercial and recreational activities) are also likely to be adversely affected.

April 2010 FINAL Page 16-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 16: Conclusions and Recommendations

Accumulated frequency of spills exceeding a certain size

0,014

0,012

0,01

0,008 Unmitigated Mitigated

0,006

0,004 Annual frequency [per year] frequency Annual

0,002

0 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Spill volume [m3]

Figure 16-1 Accumulated frequency of spills exceeding a certain size; Unmitigated / Mitigated Northern Gateway is committed to prevention, detection and mitigation measures to prevent spills from occurring and would take action so that any potential effects are reduced or remediated, as well as appropriately remunerating affected individuals. Considering the stringent tanker requirements, use of tethered and close escort tugs for laden tankers in the CCAA, and other mitigation and response measures proposed for the Project, the probability of a spill in the CCAA or OWA is considered low.

Page 16-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

17 References

17.1 Literature Cited Ackerman, R.G. and H. Heras. 1992. Tainting by short-term exposure of Atlantic salmon to water soluble petroleum hydrocarbons. Proceedings of the 15th Arctic and Marine Oil Spill Program Technical Seminar, Ottawa. Environment Canada. 757-762. Agler, B. A., P. E. Seiser, S. J. Kendall, and D. B. Irons. 1994. Marine bird and sea otter populations of Prince William Sound, Alaska: population trends following the Exxon Valdezoil spill. EVOS Restoration Final Report, Proj. 93045. U.S. Fish Wildl. Serv., Anchorage, Alas. 51pp. Ainley, D.G., C.R. Grau, T.E. Roudybusk, S.H. Morrell and J.M. Utts. 1981. Petroleum ingestion reduces reproduction in Cassin‘s auklets. Marine Pollution Bulletin 12: 314-31. Alaska Department of Fish and Game. 1985. Dungeness Crab, Cancer magister. In Alaska Habitat Management Guide, Southcentral Region, Volume 1: Life Histories and Habitat Requirements of Fish and Wildlife. Juneau, Alaska: Alaska Department of Fish and Game, Division of Habitat. 379-385. Albers, P.H. 1990. Oil spills and the environment: a review of chemical fate and biological effects of petroleum. In: The Effects of Oil on Wildlife: Research, Rehabilitation, and General Concerns. Proceedings from The Oil symposium, 16 October 1990, Herndon, Virginia. Hanover, PA. The Sheridan Press: 1-12 Albers, P.H. 1983. Effects of oil on avian reproduction: a review and discussion. In: The Effects of Oil on Birds: Physiological Research, Clinical Applications & Rehabilitation. Wilmington, DE: TriState Bird Rescue and Research. 78-97. Albers, P.H. 1980. Transfer of crude oil from contaminated water to bird eggs. Environmental Research 22:307–314. Albers, P. H., 1978. Effects of petroleum on different stages of incubation in bird eggs. Bulletin of Environmental Contamination and Toxicology 19:624-630. Albers, P.H. and R.C. Szaro. 1978. Effects of No. 2 Fuel Oil on Common Eider Eggs. Marine Pollution Bulletin 9: 138-139. Cited in Bernatowicz, J.A., P.F. Schempf, and T.D. Bowman. 1996. Bald eagle productivity in south-central Alaska in 1989 and 1990 after the Exxon Valdez oil spill. In S.D. Rice, R.B. Spies, D.A. Wolfe, and B.A. Wright. (ed.). Proceedings of the Exxon Valdez Oil Spill Symposium. American Fisheries Society Symposium 18. Bethesda, MD. American Fisheries Society. 785-797 Albers, P. H., and T. R. Loughlin. 2003. Effects of PAHs on marine birds, mammals, and reptiles. In P.E.T. Douben (ed.). PAHs: An ecotoxicological perspective. John Wiley and Sons, London. 243- 261.

April 2010 FINAL Page 17-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Andres, B. A. 1994. The effects of the Exxon Valdez oil spill on Black Oystercatchers breeding in Prince William Sound, Alaska. Final Report, Exxon Valdez Oil Spill Public Information Office, Anchorage, AK. Andres, B.A. 1998. Black oystercatcher (Haematopus bachmani). Exxon Valdez Oil Spill Trustee Council Restoration Notebook. 8 pp. Andres, B. A. 1999. Effects of persistent shoreline oil on breeding success and chick growth in Black Oystercatchers. Auk. 116(3):640–650. Ang Jr. P.O. 1991. Natural dynamics of a Fucus distichus (Phaeophyceae, Fucales) population: reproduction and recruitment. Marine Ecology Progress Series 78:71-85. Armstrong, D.A., P.A. Dinnel, J.M. Orensanz, J.L. Armstrong, T.L. McDonald, R.F. Cusiamo, et al. 1995. Status of selected bottomfish and crustacean species in Prince William Sound following the Exxon Valdez oil spill. In P. G. Wells, J. N. Butler and J. S. Hughes (ed.). Exxon Valdez oil spill: fate and effects in Alaskan waters. Philadelphia: ASTM. ASL Environmental Sciences Inc.. (ASL) 2010a. Marine Physical Environment Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. ASL Environmental Sciences Inc. (ASL). 2010b. Weather and Oceanographic Conditions at Sites in the CCAA and in Queen Charlotte Sound, Hecate Strait and Dixon Entrance Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. ASL Environmental Sciences. 2009. Marine Physical Environment Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Calgary, AB. Baca, B.J. and C.D. Getter. 1984. Toxicity of Oil and Chemically Dispersed Oil to the Seagrass Thalassia testudinum. In Oil Spill Chemical Dispersants: Research, Experience and Recommendations. A Symposium Sponsored by ASTM Committee F-20 on Hazardous Substances and Oil Spill Response. Philadelphia, Pennsylvania. 314-323. British Columbia Cetacean Sightings Network. 2005. SPLASH: North Pacific humpback researchers working together. Sightings, the Newsletter of the British Columbia Cetacean Sightings Network. 17 (17). Baird R.W. and L.M. Dill. 1996. Ecological and social determinants of group size in transient killer whales. Behavioral Ecology. 7(4): 408-416. Baker, C. S., L. M. Herman, A. Perry, W. S. Lawton, J. M. Straley, A. A. Wolman, G. D. Kaufman, H. E. Winn, J. D. Hall, J. M. Reinke and J. Ostman. 1986. Migratory movement and population structure of humpback whales (Megaptera novaeangliae) in the central and eastern North Pacific. Marine Ecology Progress Series. 41: 103. Baker, C. S., A. Perry and L. M. Herman. 1987. Reproductive histories of female humpback whales (Megaptera novaeangliae) in the North Pacific. Marine Ecology Progress Series.41: 103. Bates, C. 2008. An Introduction to the Seaweeds of British Columbia. University of British Columbia.

Page 17-2 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Beckett, J. and K. Munro. 2010. Fish and Fish Habitat Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Beckett, J. and K. Munro. 2009. Fish and Fish Habitat Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Calgary, AB. Beissinger, S. R. 1995. Population trends of the Marbled murrelet projected from demographic analyses. In C.J. Ralph, G. L. Hunt, Jr., M. G. Raphael and J. F. Piatt (ed). Ecology and Conservation of the Marbled Murrelet General Technical Report PSW-GTR-152, Pacific Southwest Research Station, U.S.D.A. Forest Service, Albany, CA. 385-393. Bence, A.E., and W.A. Burns. 1995. Fingerprinting hydrocarbons in the biological resources of the Exxon Valdez spill area. In P.G. Wells, J.N. Butler and J.S. Hughes (ed.). Exxon Valdez Oil Spill: Fate and Effects in Alaskan Water. Philadelphia, PA. ASTM. 84-140. Ben-David M., G. M. Blundell and J. E. Blake. 2001. Post-release survival of river otters: effects of exposure to crude oil and captivity. Oiled Wildlife Care Network Symposium, May 2001, Sacramento, California. Berger, A.E. 1993. Effects of the Nestucca oil spill on seabirds along the coast of Vancouver Island. Canadian Wildlife Service Technical Report Series No. 179. Berkes, F. 1999. Sacred Ecology: Traditional Ecological Knowledge and Resource Management. Taylor & Francis. Philadelphia, PA. Bernatowicz, J.A., P.F. Schempf and T.D. Bowman. 1996. Bald eagle productivity in south-central Alaska in 1989 and 1990 after the Exxon Valdez oil spill. pp.785-797. In S.D. Rice, R.B. Spies, D.A. Wolfe and B.A. Wright (ed.). Proceedings of the Exxon Valdez Oil Spill Symposium. American Fisheries Society Symposium 18. Bethesda, MD. American Fisheries Society. Bigg, M. A. 1981. Harbour seal - Phoca vitulina and P. largha. In S. H. Ridgway and R. J. Harrison, eds. Handbook of marine mammals. Vol. 2. Academic Press, London. 359pp. Pages 1-27 Bigg, M.A. 1985. Status of the Steller sea lion (Eumetopias jubatus) and California sea lion (Zalophus californianus) in British Columbia. Canadian Special Publication on Fisheries and Aquatic Sciences. 77. Binderup, M., L.Duedahl-Olesen, S. Einarsson, B. Fabech, A.L. Haldorsen, H. Johnsson, H. K. Knutsen, A.K. Müller, P. J. Vuorinen, and M. L. Wiborg. 2004. The effect of oil spills on seafood safety: An example of the application of the Nordic risk analysis model. Nordic Council of Ministers. Copenhagen, DK. Birtwell, I.K. and M.K. McAllister. 2002. Hydrocarbons and their effects on aquatic organisms in relation to offshore oil and gas exploration and oil well blowout scenarios in British Columbia, 1985. Canadian Technical Report of Fisheries and Aquatic Sciences. 2391: 52. Bittner, J.E. and D.R. Reger. 1995. EVOS report 1994 spill area site and collection plan. Restoration report 94007-1. Exxon Valdez oil spill restoration project final report. Alaska Dept. of Natural Resources, Anchorage, AK. Parks and Outdoor Recreation Div.

April 2010 FINAL Page 17-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Bowman, T. D., P. F. Schempf, and J. A. Bernatowicz.. 1993. Effects of the Exxon Valdez oil spill on bald eagles. Exxon Valdez Oil Spill State and Federal Natural Resource Damage Assessment. Final Report. Bird Study 4. US Fish and Wildlife Service. Anchorage, AK. Cited in Bowman, T.D., P.F. Schempf and J.A. Bernatowicz. 1995. Bald eagle survival and population dynamics in Alaska after the Exxon Valdez oil spill. J. Wild. Mange. 59: 317-324. Bowyer, R. T., G. M. Blundell, M. Ben-David, S. C. Jewett,T. A. Dean, and L. K. Duffy. 2003. Effects of the ExxonValdez oil spill on river otters: injury and recovery of a sentinel species. Wildlife Monographs. 153: 1–53. Boyd, J.N., D. Scholz and A.H. Walker. 2001. Effects of oil and chemically dispersed oil in the environment. 2001 International Oil Spill Conference. British Columbia Forest Service. 1998. Eulachon: A significant fish for First Nations communities. Forest Sciences, Prince Rupert Forest Region. Smithers, BC. 6 pp. Brown, E.D., T.T. Baker, J.E. Hose, R.M. Kocan, G.D. Marty, M.D. McGurk, B.L. Norcross and J. Short. 1996. Injury to the early life history stages of Pacific Herring in Prince William Sound after the Exxon Valdez oil spill. In S. D. Rice, R. B. Spies, D. A. Wolfe & B. A. Wright (ed.). Proceedings of the Exxon Valdez Oil Spill Symposium. Anchorage, Alaska. American Fisheries Society. Vol. Symposium 18, 448-462. Browne, P., Laake, J.L. and R.L. DeLong. 2002. Improving pinniped diet analyses through identification of multiple skeletal structures in fecal samples. Fishery Bulletin. 100:423-433. Bue, B.G., S. Sharr, S.D. Moffitt and A.K. Craig. 1996. Effects of the Exxon Valdez Oil Spill on Pink Salmon Embryos and Preemergent Fry. Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium 18. Cairns, D.K., and R.D. Elliot. 1987. Oil spill impact assessment for seabirds: the role of refugia and growth centres. Biol. Conserv. 31:1-9. Calkins, D.G., E. Becker, T.R. Spraker and T.R. Loughlin. 1994. Impacts on Stellar sea lions. In T.R. Loughlin. (ed.). Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 119-139. Calkins, D.G. and K.W. Pitcher. 1982. Population assessment, ecology and trophic relationships of Steller sea lions in the Gulf of Alaska. Final Report: Research Unit 243 Contract #03-5-022-69. United States Department of the Interior. Calkins, D.G., E. Becker, T.R. Spraker and T.R. Loughlin. 1994. Impacts on Steller Sea Lions. In Loughin, T.R. (ed). 1994. Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 119-139. Cam, E., L. Lougheed, R. Bradley and F. Cooke. 2003. Demographic assessment of a Marbled Murrelet population from capture-recapture data. Conservation Biology. 17: 1-10. Campbell, R.W., N.K. Dawe, I. McTaggert-Cowan, J.M. Cooper, G.W. Kaiser and M.C.E. McNall. 1990. The Birds of British Columbia. Vol.1. Nonpasserines: Loons through Waterfowl. Royal BC Museum, Victoria, and Canadian Wildlife Service. Delta, BC.

Page 17-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Campbell, R.W., N.K. Dawe, I. McTaggert-Cowan, J.M. Cooper, G.W. Kaiser and M.C.E. McNall. 1990. The Birds of British Columbia, Vol. 2, Nonpasserines: Diurnal Birds of Prey through Woodpeckers. Royal BC Museum, Victoria and Canadian Wildlife Service. Delta, BC. Canadian Resourcecon Ltd. 1982. An Analysis of Three Sport Fisheries: Kitimat Arm/Douglas Channel. Prepared for Department of Fisheries and Oceans, Vancouver, British Columbia. Canadian Wildlife Service. 2007. British Columbia Marine Bird Area of Interest Database (March 2007). Digital datafiles. Pacific Wildlife Research Centre, Canadian Wildlife Service, Environment Canada. Canadian Council of Ministers of the Environment (CCME). 1996. A Framework for Ecological Risk Assessment: General Guidance. Canadian Council of Ministers of the Environment (CCME). 1997. A Framework for Ecological Risk Assessment: Technical Appendices. Carls, M.G. 1987. Effects of dietary and water-borne oil exposure on larval Pacific Herring (Clupea- Harengus-Pallasi). Marine Environmental Research. 22(4):253-270. Carls, M.G., A.C. Wertheimer, J.W. Short, R.M. Smolowitz and J. J. Stegeman.1996a. Contamination of Juvenile Pink and Chum Salmon by Hydrocarbons in Prince William Sound After the Exxon Valdez Oil Spill. In Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium 18. Carls, M.G., L. Holland, M. Larsen, J.L. Lum, D.G. Mortensen, S.Y. Wang and A.C. Wertheimer. 1996b. Growth, Feeding and Survival of Pink Salmon Fry Exposed to Food Contaminated with Crude Oil. Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium 18. Carls, M.G., J.E. Hose, R.E. Thomas and S.D. Rice. 2000. Exposure of Pacific herring to weathered crude oil: assessing effects on ova. Environmental Toxicology and Chemistry. 19(6):1649-1659. Challenger, G.E., and G.S. Mauseth. 1998. Closing and opening of fisheries following oil spills: A case study in Humboldt Bay, California. Proceedings of the Twenty first Arctic Marine Oil Spill Technical Program.167-180. Chambers, P.A., R.E. DeWreede, E.A. Irlandi and H. Vandermeulen. 1999. Management issues in aquatic macrophyte ecology: a Canadian perspective. Canadian Journal of Botany. 7:471-487. Chenelot, H., J. Matweyou and B. Konar. 2001. Investigation of the overwintering of the annual Macroalga Nereocystis luetkeana in Kachemak Bay, Alaska. In University of Alaska Sea Grant College Program Report. Cold water diving for science. University of Alaska Sea Grant College Program (a).19-24. Cheng, N.-S., A.W.-K. Law and A.N. Findikakis. 2000. Hydrocarbon transport in the surf zone. Journal of Hydraulic Engineering 126: 803-809. Chittleborough, R. G. 1965. Dynamics of two populations of the humpback whale Megaptera novaeangliae (Borowski). Australian Journal of Marine and Freshwater Research 16: 33-128.

April 2010 FINAL Page 17-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Clapham, P. J., L.S. Baraff, C.A. Carlson, M.A. Christian, D.K. Mattila, C.A. Mayo, M.A. Murphy and S. Pittman. 1993. Seasonal occurrence and annual return of humpback whales, Megaptera novaeangliae, in the southern Gulf of Maine. Canadian Journal of Zoology. 72(1):440-443. Clapham, P. J. and C.A. Mayo. 1987. Reproduction and recruitment of individually identified humpback whales, Megaptera novaeangliae, observed in Massachusetts Bay, 1979-1985. Canadian Journal of Zoology. 65: 2853-2863. Clarke, C.L., and G.S. Jamieson. 2006b. Identification of Ecologically and Biologically Significant Areas in the Pacific North Coast Integrated Management Area: Phase II – Final Report. Can. Tech. Rep. Fish. Aquat. Sci. 2686: 25 p. Clark, C.W. and J.H. Johnson.1984. The sounds of the bowhead whale, Balaena mysticetus, during the spring migrations of 1979 and 1980. Canadian Journal of Zoology 62:1436-1441. Connell, J.H. 1972. Community interaction on marine rocky intertidal shores. A. Rev. Ecol. Syst. 3: 169- 192. Conway, K.W., M. Krautter, J.V. Barrie and M. Neuweiler. 2001. Hexactinellid sponge reefs on the Canadian continental shelf: a unique "living fossil". Geoscience Canada 28(2):71-78. Costa, D.P. and G.L. Kooyman. 1982. Oxygen consumption, thermoregulation, and the effect of fur oiling and washing on the sea otter, Enhydra lutris. Canadian Journal of Zoology. 60: 2761-2767. Craig, A. S. and L.M. Herman. 1997. Sex differences in site fidelity and migration of humpback whales (Megaptera novaeangliae) to the Hawaiian Islands. Canadian Journal of Zoology. 75: 1923- 1933. Crawford, W., D. Johannessen, F. Whitney, R. Birch, K. Borg, D. Fissel, et al. 2007. Appendix C: Physical and chemical oceanography. In B. G. Lucas, S. Verrin & R. Brown (Eds.), Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA) (Can. Tech. Rep. Fish. Aquat. Sci. 2667. vii + 77. Crawford, W.R., J.Y. Cherniawsky and P.F. Cummins. 1999. Surface Currents in British Columbia Coastal Waters: Comparison of Observations and Model Predictions. Atmosphere-Ocean 37(3):255-280. Crawford, W.R., W.S. Huggett and M.J. Woodward. 1988. Water Transport through Hecate Strait, BC. Atmosphere-Ocean 26(3):301-320. Cretney, W.J., C.S. Wong, R.W. Macdonald, P.E. Erickson and B.R. Fowler. 1983. Polycyclic Aromatic Hydrocarbons in Surface Sediments and Age-Dated Cores from Kitimat Arm, Douglas Channel and Adjoining Waterways. In R.W. Macdonald (ed.). Proceedings of a Workshop on the Kitimat Marine Environment. Canadian Technical Report of Hydrography and Ocean Sciences No. 18. 163-190. Cucci, T.L. and C.E. Epifanio. 1979. Long-term effects of water-soluble fractions of Kuwait crude oil on the larval and juvenile development of the mud crab Eurypanopeus depressus. Marine Biology. 55(3):215-220.

Page 17-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Culbertson, J.B., Valiela, I., Peacock, E.E., Reddy, C.M., Carter, A. and R. VanderKruik. 2007. Long- term Biological Effects of Petroleum Residues on Fiddler Crabs in Salt Marshes. Marine Pollution Bulletin. 54: 955-962. Darling, J. D. and D.J. McSweeney. 1985. Observations on the migrations of North Pacific humpback whales (Megaptera novaeangliae). Canadian Journal of Zoology. 63: 308-314. Day, R. H., S. M. Murphy, J. A. Wiens, G. D. Hayward, E. J. Harner, and L. N. Smith. 1997. Effects of the Exxon Valdez oil spill on habitat use by birds in Prince William Sound, Alaska. Ecol. Appl 7:593–613. Day, R.H., S.M. Murphy, J.A. Wiens, G.D. Hayward, E.J. Harner and B.E. Lawhead. 1997a. Effects of the ExxonValdez oil spill on habitat use by birds along the Kenai Peninsula, Alaska. Condor 99: 728–742. Day, R. H., S. M. Murphy, J. A. Wiens, G. D. Hayward, E.J. Harner, and L. N. Smith. 1995. Use of oil- affected habitats by birds after the Exxon Valdez oil spill. Pages 727–761 in P. G. Wells, J. N. Butler, and J. S. Hughes, editors.Exxon Valdez oil spill: fate and effects in Alaskan waters.STP 1219, American Society for Testing and Materials,Philadelphia, Pennsylvania, USA Dahlheim, M. E., and C.O. Matkin. 1994. Assessment of injuries to Prince William Sound killer whales. In T.R. Loughlin. (ed.) Marine mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 163-171. Dean, T.A., M.S. Stekoll, S.C. Jewett, R.O. Smith and J.E. Hose. 1998. Eelgrass (Zostera marina L.) in Prince William Sound, Alaska: Effects of the Exxon Valdez Oil Spill. Marine Pollution Bulletin 36(3):201-210. Dean, T.A., S.C. Jewett, D.R. Laur and R.O. Smith. 1996. Injury to epibenthic invertebrates resulting from the Exxon Valdez oil spill. American Fisheries Society Symposium 18: 424-439. Dean, T.A. and S.C. Jewett. 2001. Habitat-specific recovery of shallow subtidal communities following the Exxon Valdez oil spill. Ecological Applications 11(5):1456-1471. d‘Entremont, M. 2010. Marine Birds Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. d‘Entremont, M. 2009. Marine Birds Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Burnaby BC. DeGange, A.R., A.M. Doroff, and D.H. Monson. 1994. Experimental recovery of sea otter carcasses at Kodiak Island, Alaska, following the Exxon Valdez oil spill. Marine Mammal Science.10:492- 496. Delvigne, G.A.L. and C. Sweeney. 1988. Natural dispersion of hydrocarbon. Hydrocarbon and Chemical Pollution. 4: 281-310. Det Norske Veritas (DNV). 2010. Quantitative Risk Analysis. Prepared for Northern Gateway Pipelines Inc. Calgary, AB.

April 2010 FINAL Page 17-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Det Norske Veritas (DNV). 2002. Optimized Escort Tug Operations At Fawley Terminal, DNV report no. 2002 – 0529 CONFIDENTIAL. DFO. 1999. Giant Red Sea Cucumber. DFO Science Stock Status Report, C6-10. DFO. 2000a. Hexactinellid sponge reefs on the British Columbia continental shelf: geological and biological structure. DFO Pacific Region Habitat Status Report 2000/02. DFO. 2000b. Red Sea Urchin:Underwater World. Department of Fisheries and Oceans DFO. 2001. Fish stocks of the Pacific Coast. Fisheries and Oceans Canada. Ottawa, ON. 1-162 pp. DFO. 2003. El Nino: Warm water events in the Pacific Ocean. Factsheet. Institute of Ocean Sciences. Fisheries and Oceans Canada, DFO. 2005. Canada‘s Oceans Action Plan For Present and Future Generations. DFO/2005-348. Ottawa, Canada. DFO. 2008a. Amended Integrated fisheries Management Plan: Crab by Trap January 1, 2008 to December 31, 2008. Department of Fisheries and Oceans. DFO. 2008b. Amended Integrated Fisheries Management Plan: Prawn and Shrimp by Trap, May 1, 2008 to April 30, 2009. Department of Fisheries and Oceans. DFO. 2008g. Amended Integrated Fisheries Management Plan: Prawn and Shrimp by Trap, May 1, 2008 to April 30, 2009. Department of Fisheries and Oceans. DFO. 2008m. Recovery Strategy for the Northern and Southern Resident Killer Whales (Orcinus orca) in Canada. Ottawa. ix + 81pp pp DFO (Fisheries and Oceans Canada) and Abalone Recovery Team. 2004. National Recovery Strategy for the Northern Abalone (Haliotis kamtschatkana) in Canada. 22 pp. Di Toro, D.M. and J.A. McGrath. 2000. Technical basis for narcotic chemicals and polycyclic aromatic hydrocarbon criteria. II. Mixtures and Sediments. Environ. Toxicol. Chem. 19: 1971-1982. Di Toro, D.M., J.A. McGrath and D.J. Hansen. 2000. Technical basis for narcotic chemicals and polycyclic aromatic hydrocarbon criteria. I. Water and Tissue. Environ. Toxicol. Chem. 19: 1951- 1970. DNV Maritime Solutions. 2010. Quantitative Risk Analysis and Intended Methods of Reducing Risks. Prepared for the Enbridge Northern Gateway Project. Norway. Driskell, W.B., J.L. Ruesink, D.C. Lees, J.P. Houghton and S.C. Lindstrom. 2001. Long-term signal of disturbance: Fucus gardneri after the Exxon Valdez oil spill. Ecological Applications 3: 815-827. Druehl, L. 2000. Pacific Seaweeds. Harbour Publishing, Madeira Park, B.C., Canada. Dumond, D.E. and D.G. Griffin. 2002. Measurements of the Marine Reservoir Effect on Radiocarbon Ages in the Eastern Bering Sea. Arctic. 55 (1): 77-86.

Page 17-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Edwards, R. and White, I. 1999. The Sea Empress Oil Spill: Environmental Impact and Recovery (1999). Proceedings of the International Oil Spill Conference 1999, 7-12 March 1999, Seattle, USA, 97- 102, American Petroleum Institute, Washington DC, USA Ellis, G.M., J.K.B. Ford and J.R. Towers. 2007. Northern resident killer whales in British Columbia photo-identification catalogue 2007. Pacific Biological Station. Nanaimo, BC. Etkin, D.S., J. Michel, D. French McCay, M. Boufadel and H. Li. 2008. Integrating state-of-the-art shoreline interaction knowledge into spill modelling. International Hydrocarbon Spill Conference. May 2008. Savannah, GA, USA. Evans, P. G. H. and J.A. Raga. 2001. Marine Mammals: Biology and Conservation. New York: Kluwer Academic/Plenum Publishers. Fargo, J., L. MacDougall, and I. Pearsall. 2007. Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA): Appendix G: Groundfish. Fedje, D.W., T. Christensen, H. Josenhans, J. MacSporran and J. Strang. 2005. Millenial Tides and Shifting Shores: Archaeology on a Dynamic Landscape. In D.W. Fedje and R.W. Mathewes. (ed.). Haida Gwaii: Human History and Environment from the Time of the Loon to the Time of the Iron People. University of British Columbia Press, Vancouver. Fingas, M. and B. Fieldhouse. 2004. Formation of water-in-hydrocarbon emulsions and application to hydrocarbon spill modelling. Journal of Hazardous Materials. 107: 37-50. Fingas, M. 2001. Basics of oil spill cleanup. 2nd ed. CRC Press LLC. Boca Raton, FL. Fisheries and Oceans Canada (DFO). 2001. Fish stocks of the Pacific Coast. Fisheries and Oceans Canada. Ottawa, ON. 1-162 pp. Fisheries and Oceans Canada (DFO). 2008. Amended Integrated fisheries Management Plan: Crab by Trap January 1, 2008 to December 31, 2008. Department of Fisheries and Oceans. Ford, R.G., M.L. Bonnel, D.H. Varoujean, G.W. Page, H.R. Carter, B.E. Sharp, D. H. Heinemann and J.L. Casey. 1990. Total direct mortality of seabirds resulting from the Exxon Valdez oil spill. In S.D. Rice, R.B. Spies, D.A. Wolfe and B.A. Wright (ed.). Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Symposium 18. Ford, J.K.B., G.M. Ellis, D.R. Matkin, K.C. Balcomb, D. Briggs and A.B. Morton. 2005. Killer whale attacks on minke whales: prey capture and antipredator tactics. Marine Mammal Science. 21(4): 603-618. Ford, R.G. G.K. Himes Boor and J.C. Ward. 2001. Seabird mortality resulting from the M/V New Carissa oil spill incident February and March 1999. R.G. Ford Consulting Company, Portland, Oregon. Ford, J.K.B. 2006. An Assessment of critical habitats of resident killer whales in waters off the Pacific coast of Canada. Canadian Science Advisory Secretariat, Research Document 2006/072. Ford, J. K. B., Ellis, G. M., Barrett-Lennard, L. G., Morton, A. B., Palm, R. S. and K.C. Balcomb III.1998. Dietary specialization in two sympatric populations of killer whales (Orcinus orca) in coastal British Columbia and adjacent waters. Canadian Journal of Zoology. 76(8): 1456-1471.

April 2010 FINAL Page 17-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

French-McCay, D.P, J.J. Rowe, N. Whittier, S. Sankaranarayanan and D.S. Etkin. 2004. Estimation of potential impacts and natural resource damages of oil. Journal of Hazardous Materials 107:11- 25. French-McCay, D. 2004. Hydrocarbon spill impact modeling: development and validation. Environmental Toxicology and Chemistry. 23: 2441-2456. French-McCay, D., C.J. Beegle-Krause, J. Rowe and W. Rodriguez. 2009. Oil Spill Risk Assessment: Relative Impact Indices by Oil Type and Location. In Proceedings of the 32nd AMOP Technical Seminar on Environmental Contamination and Response, Emergencies Science Division, Environment Canada, Ottawa, ON, Canada. 655-681. French-McCay, D.P, J.J. Rowe, N. Whittier, S. Sankaranarayanan and D.S. Etkin. 2004. Estimation of potential impacts and natural resource damages of oil. Journal of Hazardous Materials. 107:11- 25. Friesen, V.L., T.P. Birt, J.F. Piatt, R.T. Golightly, S.H. Newman, P.N. Hebert, B.C. Congdon and G. Gissing. 2005. Population genetic structure and conservation of marbled murrelets (Brachyramphus marmoratus). Conservation Genetics. 6:607-614. Friesen, V., T. Birt, M.Z. Peery and S. Beissinger. 2007. Conservation genetics of marbled murrelets throughout their range. Final report to Scotia Pacific Lumber and U.S. Fishand Wildlife Service. January 31, 2007. 21 pp. Frost, K.J., L.F. Lowrym, E.H. Sinclair, J. Ver Hoef and D.C. McAllister. 1994. Impacts on Distribution, Abundance and Productivity of Harbour Seals. In Loughlin, T.R. (ed). 1994. Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 97-118. Fry, D.M. 1995. Reproductive effects in birds exposed to pesticides and industrial chemicals. Environmental Health Perspectives. 103:165-171. Fry, D.M., J. Swenson, L.A. Addiego, C.R. Grau, and A. Kang. 1986. Reduced reproduction of wedge- tailed shearwaters exposed to single doses of weathered Santa Barbara crude oil. Archives of Environmental Contamination and Toxicology.15:453-463. Fry, D.M. and J. Lowenstein. 1985. Pathology of Common Murres and Cassin‘s Aukelets exposed to oil. Archives of Environmental Contamination and Toxicology. 14:725-737. Garrott, R.A., L.L. Eberhardt, and D.M. Burn. 1993. Mortality of sea otters in Prince William Sound following the Exxon Valdez oil spill. Marine Mammal Science. 9:343-359. Geiger, H.J., B.G. Bue, S. Sharr, A.C. Wertheimer and T.M. Willette. 1996. A Life History Approach to Estimating Damage to Prince William Sound Pink Salmon Caused by the Exxon Valdez Oil Spill. Proceedings of the Exxon Valdez Oil Spill Symposium. American Fisheries Society Symposium 18. Geraci, J. R. and D.J. St. Aubin. 1988. Synthesis of Effects of Oil on Marine Mammals.Department of Interior Minerals Management Services, Atlantic OCS Region.

Page 17-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Geraci, J. R. and D.J. St. Aubin. 1985. Expanded Studies of the Effects of Oil on Cetaceans. Final Report Part I, U.S. Department of the Interior, Bureau of Land Management, Washington, D.C. Geraci, J.R. 1990. Physiologic and Toxic Effects. In J.R. Geraci and D.J. St. Aubin (ed.). Sea Mammals and Oil: Confronting the Risks. Academic Press, Inc., San Diego, CA. 167-197. Gibson, M.J. and J. White. 1990. Results of the eagle capture, health assessment, and short-terrn rehabilitation program following the Valdez oil spill. Wildl. J. 13:49-57. Cited in White, C., M. Ritchie, J. Robert, and B.A. Cooper. 1995. Density and productivity of bald eagles in Prince William Sound, Alaska, after the Exxon Valdez oil spill. ASTM Special Technical Publication , no. 1219, pp. 762-779. Gillespie, G.E. 1999. Status of Olympia Oyster, Ostrea conchaphila, in Canada (No. 99/150). Canadian Stock Assessment Secretariat Research Document Gilroy, D.J. 2000. Derivation of shellfish harvest reopening criteria following the Hew Carissa oil spill in Coo Bay, Oregon. Journal of Toxicology and Environmental Health. 60A: 317-329. Gilsdorf, J.M., S.E. Hygnstrom and K.C. VerCauteren. 2002. Use of frightening devices in wildlife damage management. Integrated Pest Management Reviews. 7:29–45. Gonzalez, J., F.G. Figueirasb, M. Aranguren-Gassisa, B.G. Crespob, d, E. Fernándeza, X.A.G. Moránc and M. Nieto-Cidb. 2009. Effect of a simulated oil spill on natural assemblages of marine phytoplankton enclosed in microcosms. Estuarine, Coastal and Shelf Science. 83(3): 265-276. Gosling, E.M. 1992. Systematics and geographic distribution of Mytilus. In E. M. Gosling (ed.). The Mussel Mytilus: Ecology, Physiology, Genetics and Culture. Elsevier Science Publications, Amsterdam. Grahl-Nielsen, O., J.T. Staveland and S. Wilhelmsen. 1978. Aromatic hydrocarbons in benthic organisms from coastal areas polluted by Iranian crude oil. J. Fish. Res. Board Can. 35:615-623. Gregr, E.J. and A.W. Trites. 2001. Predictions of critical habitat for five whale species in the waters of coastal British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 58(7):1265-1285 Griffiths, D.J., Oritsland, N.A. and T. Oritsland. 1987. Marine Mammals and Petroleum Activities in Norwegian Waters. Fisken Havet. 1: 179. Groves, S. 1984. Chick growth, sibling rivalry, and chick production in American black oystercatchers. Auk. 101:525-531. Gundlach, E.R. 1987. Hydrocarbon-holding capacities and removal coefficients for different shoreline types to computer simulate spills in coastal waters. Proceedings of the International Hydrocarbon Spill Conference. 451-457. Washington, DC, USA. Haebler, R. 1994. Biological Effects: Marine Mammals and Sea Turtles. In J. Burger (ed.). Before and after an oil spill: the Arthur Kill. Rutgers University Press. 238-252. Haggett, P. 1983. Geography: A Modern Synthesis. Harper & Row. New York, NY.

April 2010 FINAL Page 17-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Hall, A. 2008. State of the Ocean in the Pacific North Coast Integrated Management Area (PNCIMA). David Suzuki Foundation. 159 pp. Hart J.L.1973. Pacific fishes of Canada. Fisheries Research Board of Canada (Bull. 180), Ottawa, Canada. 749 pp. 1973. Hartwick, E.B. and W. Blaylock. 1979. Winter ecology of a black oystercatcher population. Stud. Avian Biol. 2:207-215. Harwell, M.A. and J.H. Gentile. 2006. Ecological significance of residual exposures and effects from the Exxon Valdez oil spill. Integrated Environmental Assessment and Management. 2(3): 204-246. Hay & Company Consultants (Hay & Co). 2010. Wind Observations in Douglas Channel, Squally Channel and Caamaño Sound Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Hay, D. and P.B. McCarter. 2000. Status of the eulachon Thaleichthys pacificus in Canada (Research Document 2000/145). Canadian Stock Assessment Secretariat. Fisheries and Oceans Canada. Nanaimo, BC. 92 pp. Hay, D.E., P.B. McCarter, R. Kronlund and C. Roy. 1989. Spawning areas of British Columbia herring: a review, geographical analysis and classification. Volumes 1-6. Canadian Manuscript Report of Fisheries and Aquatic Sciences 2019. Health Canada. 1994. Human Health Risk Assessment for Priority Substances. Health Canada. 2007a. Federal Contaminated Site Risk Assessment in Canada. Part V. Guidance on Complex Site Specific Human Health Risk Assessment of Chemicals (SSRAchem). Health Canada. 2007b. Federal Contaminated Site Risk Assessment in Canada. Part I. Guidance on Screening Level Risk Assessment (SLRA). Health Canada. 2007c. Federal Contaminated Site Risk Assessment in Canada. Part II. Health Canada Toxicological Reference Values (TRVs). Health Canada. 2007d. Federal Contaminated Site Risk Assessment in Canada. Part III. Guidance on Peer Review of Human Health Risk Assessments for Federal Contaminated Sites in Canada. Heise, K., J.K.B. Ford and P.F. Olesiuk. 2007. Appendix J: Marine mammals and turtles. In: Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA): iv + 35p. B.G. Lucas, S. Verrin and R. Brown (Eds.). Canadian Technical Reports for Fisheries and Aquatic Sciences. Hobday, A.J. 2000. Persistence and transport of fauna on drifting kelp (Macrocystis pyrifera (L.) C. Agardh) rafts in the Southern California Bight. Journal of Experimental Marine Biology and Ecology. 253:75-96. Hoffman D.J. 1978. Embryotoxic effects of crude oil in mallard ducks and chicks. Toxicology and Applied Pharmacology. 46:183–190. Hoffman, D.J., B.A. Rattner, G.A. Burton Jr. and J. Cairns Jr. (ed.). 2003. Handbook of Ecotoxicology. CRC Press. Boca Raton FL USA.

Page 17-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Holmes, W, K. Cavanaugh and J. Cronshaw. 1978. Effects of Ingested Petroleum on Oviposition and Some Aspects of Reproduction in Experimental Colonies of Mallard Ducks (Anas Platyrhynchos). Journal of Reproduction and Fertility. 54 (2): 335. Houghton, J.P., R.H. Gilmour, D.C. Lees, W.B. Driskell, S.C. Lindstrom, A. Mearns. 1997. Prince William Sound intertidal biota seven years later: has it recovered? 1997 International Oil Spill Conference: 679-686. Hurst, R.J., P.D. Watts and N.A. Oritsland. 1991. Metabolic compensation in oil-exposed polar bears. Journal of Thermal Biology 16: 53–56, Hyatt, K., M.S. Johannes and M. Stockwell. 2007. Appendix I: Pacific Salmon. In B. G. Lucas, S. Verrin & R. Brown (Eds.), Ecosystem Overview: Pacific North Coast Integrated Management Area (PNCIMA) (Can. Tech. Rep. Fish Aquat. Sci. 2667. Vi + 55. International Chamber of Shipping. 2006. International Safety guide for Oil Tankers and Terminals, Fifth Edition. Oil Companies International Marine Forum. International Association of Ports and Harbours, Whitherby Publishing, London. 176-177. International Maritime Organization. 2002. Guidelines for formal safety assessment (FSA) for use in the IMO rule-making process. MSC/Circ.1023. MEPC/Circ.392 International Standards Organization. 1992. Sensory Analysis - Methodology - Vocabulary. Report ISO 5942. Geneva: ISO IPIECA (International Petroleum Industry Environmental Conservation Association). 1991. IPIECA Report Series, Volume One. Guidelines on Biological Impacts of Oil Pollution. Jacobs, R.P.W.M. 1980. Effects of the ' Amoco Cadiz ' Oil Spill on the Seagrass Community at Roscoff with Special Reference to the Benthic Infauna. Marine Ecology Progress Series. 2(3):207-212. Jamieson, G.S. and L. Chew. 2002. Hexactinellid sponge reefs: Areas of interest as Marine Protected Areas in the North and Central Coast areas (Research Document 2002/122). Canadian Science Advisory Secretariat. Jardine, I.D., K.A. Thomson, M.G.G. Foreman and P.H. LeBlond. 1993. Remote Sensing of Coastal Sea Surface Features off Northern British Columbia. . Remote Sensing Environment (45):73-84. Jehl, J. R. 1985. Hybridization and evolution of oystercatchers on the Pacific coast of Baja California. Ornithol. Monog. 36:484-504. Jenssen, B.M. 1994. Review article: effects of oil pollution, chemically treated oil, and cleaning on the thermal balance of birds. Environmental Pollution 86: 207–215. Jessup, D.A. 1998. Rehabilitation of Oiled Wildlife. Conservation Biology. 12:1153–1155. Jordan, T.G. and Rowntree, L. 1986. The Human Mosaic: A Thematic Introduction to Cultural Geography. Harper & Row. New York, NY. Kansanen, P.H. and J. Seppälä. 1992. Interpretation of mixed sediment profiles by means of a sediment mixing model and radioactive fallout. Hydrobiologia. 243/244: 371-379.

April 2010 FINAL Page 17-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Karki, R. and R. Hamelin. 2010. Non-traditional Land Use Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Karki and Hamelin, 2009. Non-Traditional Land Use Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Calgary, AB. Kennish, M.J. 1992. Ecology of Estuaries: Anthropogenic Effects. CRC Press Marine Science Series. 500 pp. King, J.G. and G.A. Sanger. 1979. Oil Vulnerability Index for marine oriented birds. In J.C. Bartonek and D.N. Nettleship (ed.). Conservation of marine birds of northern North America. U.S. Fish and Wildlife Service, Wildlife Research Report Number 11, Washington, D.C., USA. 227-239. Kingsford, M.J., 1992. Drift algae and small fish in coastal waters of northeastern New Zealand. Mar. Ecol. Prog. Ser. 80: 41– 55. Kokita, T. and M. Omori. 1998. Early life history traits of the gold-eye rockfish, Sebastes thompsoni, in relation to successful utilization of drifting seaweed. Marine Biology. 132(4): 1432-1793. Korotenko, K.A., R.M. Mamedov and C.N.K. Mooers. 2000. Prediction of the dispersal of hydrocarbon transport in the Caspian Sea resulting from a continuous release. Spill Science and Technology Bulletin. 6: 323-339. Koski, W. R., S. D. Kevan and W. J. Richardson. 1993. Bird dispersal and deterrent techniques for oilspills in the Beaufort Sea. Environmental Studies Research Funds Report No. 126. Calgary, AB. Krebs, C.T. and Burns, K.A. 1978. Long-term Effects of an Oil Spill on Populations of the Salt Marsh Crab Uca pugnax. Journal of Fisheries Research Board of Canada. 35:648-649. Law, R.J. and J. Hellou. 1999. Contamination of fish and shellfish following oil spill incidents. Environmental Geosciences 6(2): 90-98. Lehr, W.J., R.J. Fraga, M.S. Belen and H.M. Cekirge. 1984. A new technique to estimate initial spill size using a modified Fay-type spreading formula. Marine Pollution Bulletin. 15: 326-329. Leighton, F.A. 1995. The toxicity of petroleum oils to birds: an overview. In: Wildlife and Oil Spills. L. Frink, K. Ball-Weir and C. Smith (eds), Tri-State Bird Rescue and Research, Inc., Newark, Delaware. Leys, S.P., K. Wilson, C. Holeton, H.M. Reiswig, W.C. Austin and V. Tunnicliffe. 2004. Patterns of glass sponge (Porifiera, Hexactinellida) distribution in coastal waters of British Columbia, Canada. Marine Ecology Progress Series 283:133-149. LGL Limited Environmental Research Associates. 2004. A Review of the State of Knowledge of Marine and Shoreline Areas in the Queen Charlotte Basin. University of Northern British Columbia's Northern Land Use Institute. Prince George, BC. Lipscomb, T.P., R.K. Harris, A.H. Rebar, B.E. Ballachey and R.J. Haebler. 1996. Pathological studies of sea otters, Exxon Valdez Oil Spill State/Federal Natural Resource Damage Assessment Final Report (Marine Mammal Study 6-11), U.S. Fish and Wildlife Service, Anchorage, AK.

Page 17-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Lloyd‘s Register Fairplay (LRFP) 2007. Lloyd‘s Register Fairplay Incident database and World Fleet Statistics LRFP 2007. Lloyd‘s Register Fairplay Incident database and World Fleet Statistics Lockhart, W.L., D.A. Duncan, B.N. Billeck, R.A. Danell and M.J. Ryan. 1996. Chronic toxicity of the 'water-soluble fraction' of Norman Wells crude oil to juvenile fish. Spill Science & Technology Bulletin 3(4):259-262. Lougheed, C. 1999. Breeding Chronology, Breeding Success, Distribution and Movements of Marbled Murrelets (Brachyramphus marmoratus) in Desolation Sound, British Columbia. M.Sc. thesis. Simon Fraser University. Burnaby, BC. Loughlin, T. R., Ballachey, B. E. and B.A. Wright. 1996. Overview of studies to determine injury caused by the Exxon Valdez oil spill to marine mammals. Paper presented at the Am. Fish. Soc. Symp. Loughlin, T.R. 1994. Tissue hydrocarbon levels and the number of cetaceans found dead after the spill. In T.R. Loughlin. (ed.). Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 359-370. Love, M.S., M. Yoklavich and L. Thorsteinson. 2002. The Rockfishes of the Northeast Pacific. University of California Press. Berkley, CA. Lowry, L.F., K.J. Frost and K.W. Pitcher. 1994. Observations of oiling of harbor seals in Prince William Sound. In T.R. Loughlin (ed.) 1994. Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 209–225. Lucas, B.G., Verrin, S., and Brown, R. (ed.). 2007. Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA). Can. Tech. Rep. Fish. Aquat. Sci. 2667: 104 p. MacConnachie, S., J. Hillier and S. Butterfield. 2007b. Marine Use Analysis for the Pacific North Coast Integrated Management Area. Can. Tech. Rep. Fish. Aquat. Sci 2677: viii + 188 p. MacDonald, R.W. 1983. The distribution and dynamics of suspended particles in Kitimat fjord system. In R.W. MacDonald (ed.). Proceedings of a workshop on the Kitimat marine environment. Paper presented at the Proceedings of a workshop on the Kitimat marine environment, Institute of Ocean Sciences, DFO, Sidney, BC. 116-137. Mackay, D., W.Y Shiu and K.C. Ma. 2000. Physical-Chemical Properties and Environmental Fate Handbook on CD-ROM. CRC Press. Boca Raton, FL. Mackas, D., A. Peña, D. Johannessen, R. Birch, K. Borg and D. Fissel. 2007. Appendix D: Plankton. In Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA). Edited by Lucas, B.G. Verrin, S., and Brown, R. Can. Tech. Rep. Fish. Aquat. Sci. 2667: iv + 33 p. Maritime Safety Committee (MSC). 2006. FSA Study on ECDIS/ENCs: Details on Risk Assessment and Cost Benefit Assessments, Submitted by Denmark and Norway, International Maritime Organization MSC 81/INF.9. MARPOL. 1978. International Convention for the Prevention of Pollution from Ships, 1973, as modified by the Protocol of 1978 relating thereto (MARPOL). International Maritime Organization.

April 2010 FINAL Page 17-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Martin, A. R., Katona, S. K., Mattila, D., Hembree, D. and T.D. Waters.1984. Migration of humpback whales between the Caribbean and Iceland. Journal of Mammalogy. 65: 330-333. Marty, G.D., A. Hoffmann, M.S. Okihiro, K. Hepler and D. Hanes. 2003. Retrospective analysis: bile hydrocarbons and histopathology of demersal rockfish in Prince William Sound, Alaska, after the Exxon Valdez oil spill. Marine Environmental Research. 56(5):569-584. Matkin, C.O., Saulitis, E.L., Ellis, G.M., Olesiuk, P. and S.D. Rice. 2008. Ongoing Population-level Impacts on Killer Whales Orcinus orca Following the Exxon Valdez Oil Spill in Prince William Sound. Alaska. Mar Ecol Prog Ser. 356: 269-281. Matkin, C.O., G.M. Ellis, M.E. Dalheim and J. Zeh. Status of Killer whales in Prince William Sound, 1985-1992 . In T.R. Loughlin. (ed.). Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 141-162. Mauseth, G.S.,Martine, C.A. and K. Whittle. 1997. Closing and Reopening Fisheries Following Oil Spills; Three Different Cases with Similar Problems. Arctic and Marine Oilspill Program Technical Ministry of Supply and Services, Canada. Seminar. CONF 20; VOL 2, pages 1283- 1304. Mauseth, G.S. and G.E. Challenger 2001. Trends in Rescinding Seafood Harvest Closures Following Oil Spills. In Proceeding of the 2001 International Oil Spill Conference, Tampa, Florida, March 26- 29, 679-684. Mazet, J.A.K., S.H. Newman, K.V.K. Gilardi, F.S. Tseng, J.B. Holcomb, D.A. Jessup and M.H. Ziccardi. 2002. Advances in Oiled Bird Emergency Medicine and Management. Journal of Avian Medicine and Surgery 16: 146–149. McAlpine, D.F., M.C. James, J. Lien and S.A. Orchard. 2007. Status and conservation of marine turtles in Canadian waters. In C. N. L. Seburn & C. A. Bishop (Eds.), Ecology, Conservation and Status of Reptiles in Canada. Herpetological Conservation. (Salt Lake City, Utah: Society for the Study of Amphibians and Reptiles. Vol. 2, 85-112. McConnaughey, R.A., D.A. Armstrong, B.M. Hickey and D.R. Gunderson. 1992. Juvenile Dungeness Crab (Cancer-Magister) Recruitment Variability and Oceanic Transport During the Pelagic Larval Phase. Canadian Journal of Fisheries and Aquatic Sciences 49: 2028-2044. McEwan, E.H. N. Aitchison and P.E. Whitehead. 1974. Energy metabolism of oiled muskrats. Canadian Journal of Zoology 52: 1057–1062. McFarlane Tranquilla, L., K. Truman, D. Johannessen and T. Hooper. 2007. Appendix K: Marine Birds. In Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA). Edited by Lucas, B.G., Verrin, S., and Brown, R. Can. Tech. Rep. Fish. Aquat. Sci. 2667:vi + 68 p. McShane, C., T. Hamer, H. Carter, G. Swartzman, V. Friesen, D. Ainley, R. Tressler, and K. Nelson. 2004. Evaluation report for the 5-year status review of the Marbled Murrelet in Washington, Oregon, and California.U.S. Fish and Wildlife Service, Region 1, Portland, OR. MEPC 49/22/Add.2.

Page 17-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Michel, J. 2000. Appendix D: Adverse effects from oil. Research Planning Inc. Columbia, SC. Miller, L.A., M. Robert and W.R. Crawford. 2005. The large, westward-propagating Haida Eddies of the Pacific eastern boundary. Deep-Sea Research II 52:845-851. Milko, R., L. Dickson, R. Eliot and G. Donaldson. 2003. Wings Over Water. Canada‘s Water Bird Conservation Plan. Mizroch, S.A., L.M. Herman, J.M. Straley, D. Glockner-Ferrari, C. Jurasz, J. Darling, S. Cerchio, C. Gabriele, D. Salden and O. von Ziegesar. 2004. Estimating the adult survival rate of central North Pacific Humpback whales. J. Mammal. 85(5): 963-972. Moles, A., & Norcross, B. L. 1998. Effects of oil laden sediments on growth and health of juvenile flatfishes. Canadian Journal of Fisheries and Aquatic Science. 55: 605-610. Monson, D.H., D.F. Doak, B.E. Ballachey, A. Johnson, and J.L. Bodkin. 2000. Long-term impacts of the Exxon Valdez oil spill on sea otters, assessed through age-dependent mortality patterns. Proceedings of the National Academy of Science. 97: 6562-6567. Morgan, K.H., K. Vermeer and R.W. McKelvey. 1991. Atlas of pelagic birds of western Canada. Special publication. Delta, BC: Canadian Wildlife Service. 72 pp. Morrison, R. I. G., R. E. Gill, Jr., B. A. Harrington, S. Skagen, G. W. Page, C. L. Gratto-Trevor, and S. M. Haig. 2001. Estimates of shorebird populations in North America. Occasional Paper Number 104, Canadian Wildlife Service, Environment Canada, Ottawa, ON. 64 pp. Mulcahy, D.M. and B.E. Ballachey. 1994. Hydrocarbon residues in sea otter tissues. Pages 313-330 In T.R. Loughlin (ed). Marine mammals and the Exxon Valdez. Academic Press, San Diego. NRC (National Research Council). 2003. Oil in the Sea III: Inputs, Fates and Effects. The National Academies Press. National Research Council. Washington, D.C. 278 p. NRC (National Research Council). 1985. Oil in the Sea: Inputs, Fates and Effects. The National Academy Press. National Research Council. Washington, D.C. 601 p. Neff, J.M., E.H. Owens, S.W. Stoker and D.M. McCormick. 1995. Shoreline oiling conditions in Prince William Sound following the Exxon Valdez oil spill. In: P.G. Wells, J.N. Butler and J.S. Hughes (eds.). Exxon Valdez oil spill: fate and effects in Alaskan Waters. American Society for Testing and Materials. ASTM STP 1219, pp. 312-345. Nicholson, N.L. 1970. Field studies on the giant kelp Nereocystis. Journal of Phycology, 6:177-182. Nisbet, I. C. T. 1989. Organochlorines, reproductive impairment, and declines in Bald Eagle Haliaeetus leucocephalus populations: mechanisms and dose relationships. In B.U. Meyburg and R. D. Chancellor (ed.). Raptors in the modern world. World Working Group for Birds of Prey, Berlin, Germany. 483-489. Norcross, B.L., J.E. Hose, M. Frandsen and E.D. Brown. 1996. Distribution, abundance, morphological condition, and cytogenetic abnormalities of larval herring in Prince William Sound, Alaska, following the Exxon Valdez oil spill. Canadian Journal of Fisheries and Aquatic Science. 53:2376-2387

April 2010 FINAL Page 17-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Northern Maritime Research. 2002. Northern Shipwrecks Database CD, Northern Maritime Research, Bedford, Nova Scotia. Olesiuk, P.F. 1999. An assessment of the status of harbour seals (Phoca vitulina) in British Columbia. Canadian Stock Assessment Secretariat Research Document 99/33. Department of Fisheries and Oceans, Ottawa, ON, Canada. Olesiuk, P.F. 1993. Annual prey consumption by harbor seals (Phoca vitulina) in the Strait of Georgia, British Columbia. Fishery Bulletin. 91:491-515. Olesiuk, P.F., Bigg, M.A. and G.M. Ellis. 1990. Recent trends in the abundance of harbour seals, Phoca vitulina, in British Columbia. Canadian Journal of Fisheries and Aquatic Sciences. 47: 992–1003. Owens, E., E. Taylor and B. Humphrey. 2008. The persistence and character of stranded oil on coarse- sediment beaches. Marine Pollution Bulletin. 56(1): 14-26. Ott, R., C. Peterson and S. Rice. 2001. Exxon Valdez Oil Spill (EVOS) Legacy: Shifting Paradigms in Oil Ecotoxicology. Paci, C., Tobin, A. and P. Robb. 2002. Reconsidering the Canadian environmental impact assessment Act: A place for traditional environmental knowledge. Environmental Impact Assessment Review. 22: 111-127. Pacific Leatherback Turtle Recovery Team. 2006. Recovery Strategy for Leatherback Turtles (Dermochelys coriacea) in Pacific Canadian Waters. Fisheries and Oceans Canada. Vancouver. v + 41 pp pp. Paine, M.D., W.C. Leggett, J.K. McRuer and K.T. Frank. 1992. Effects of Hibernia Crude-Oil on Capelin (Mallotus-Villosus) Embryos and Larvae. Marine Environmental Research. 33(3):159-187. Paine, R.T. 1974. Intertidal community structure: experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia (Berl.) 15: 93-120. Patin, S. 1999. Environmental impact of the offshore oil and gas industry. East Northport, NY: EcoMonitor Publ. 425 pp. Pattee, O.H. and J.C. Franson. 1982. Short-term effects of oil ingestion on American kestrels (Falco sparverius). Journal of Wildlife Diseases.18(2):235-241. Pearson, W.H., E. Moksness and J.R. Skalski. 1995. A Field and Laboratory Assessment of Oil Spill Effects on Survival and Reproduction of Pacific Herring Following the Exxon Valdez Spill.In P. G. Wells, J. M. Butler and J. S. Hughes (ed.). Exxon Valdez Oil Spill: Fate and Effects in Alaskan Waters. Philadelphia. ASTM STP 1219. American Society for Testing and Materials. 626-661 Pellegrin, N., J. Boutillier, R. Lauzier, S. Verrin and D. Johannessen. 2007. Appendix F: Invertebrates, Ecosystem Overview: Pacific North Coast Integrated Management Area (PNCIMA). Canadian Technical Report of Fisheries and Aquatic Sciences 2667. Perry, R. I., M. Stocker and J. Fargo. 1994. Environmental effects on the distributions of Groundfish in Hecate Strait, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 51:1401.

Page 17-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Peterson, C.H. 2000. The Exxon Valdez Oil Spill in Alaska: Acute, Indirect and Chronic Effects on the Ecosystem. Advances in Marine Biology. 39: 103 p. Peterson, C.H., S.D. Rice, J.W. Short, D. Esler, J.L. Bodkin, B.E. Ballachey, et al. 2003. Long-term ecosystem response to the Exxon Valdez oil spill. Science. 302: 2082-2086. Peterson, C.H., M. C. Kennicutt II, R. H. Green, P. Montagna, D. E. Harper, Jr., E. N. Powell, and P. F. Roscigno. 1996. Ecological consequences of environmental perturbations associated with offshore hydrocarbon production: a perspective on long-term exposures in the Gulf of Mexico. Can. J. Fish. Aquat. Sci. 53: 2637–2654. Phillips, R.C. 1984. The ecology of eelgrass meadows in the Pacific Northwest: a community profile. US Fish and Wildlife Service. 85 pp. Piatt, J.F. C.J. Lensink, W. Butler, M. Kendziorek and D.R. Nysewander. 1990. Immediate impact of the Exxon Valdez oil spill on marine birds. The Auk. 107:387-397. Piatt, J.F., Kuletz, K.J., Burger, A.E., Hatch, S.A., Friesen, V.L., Birt, T.P., Arimitsu, M.L.,Drew, G.S., Harding, A.M.A., and K.S. Bixler. 2007. Status review of the Marbled Murrelet (Brachyramphus marmoratus) in Alaska and British Columbia. U.S. Geological Survey Open-File Report 2006- 1387, 258 p. Pinto, J.M., W.H. Pearson and J.W. Anderson. 1984. Sediment preferences and oil contamination in the Pacific sand lance Ammodytes hexapterus. Marine Biology. 83:193-204. Pojar, J. and MacKinnon, A. (ed.). 1994. Plants of the Pacific Northwest Coast. Lone Pine Publishing. Vancouver, BC. Polaris Applied Sciences Inc. 2009. Coastal Operations and Sensitivity Mapping for the CCAA. Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Polaris Applied Sciences Inc. 2010. Coastal Operations and Sensitivity Mapping for the OWA. Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Polaris Applied Sciences Inc. (Polaris). 2010a. Coastal Operations and Sensitivity Mapping of the Open Water Area Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Polaris Applied Sciences Inc. (Polaris). 2010b. Coastal Operations and Sensitivity Mapping for the Confined Channel Assessment Area. Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Polechla, P. 1990. Action plan for North American otters. In: P. Foster-Turley, S. Macdonald and C. Mason (compilers) and the IUCN/SSC Otter Specialist Group Otters: an action plan for their conservation, pp.74-79. IUCN, Gland, Switzerland. Ralls, K. and D.B. Siniff. 1990. Sea otters and oil: ecological perspectives. Pp. 199-210. In J.R. Geraci and D.J. St. Aubin (ed.). Sea Mammals and Oil: Confronting the Risks. Academic Press Inc. Rambeau, A.L. 2008. Determining abundance and stock structure for a widespread, migratory animal: The case of humpback whales (Megaptera novaeangliae) in British Columbia, Canada. MSc Thesis. University of British Columbia. Vancouver, BC.

April 2010 FINAL Page 17-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Raum-Suryan, K.L., Pitcher, K.W., Calkins, D.G., Sease, J.L., Loughlin, T.R. 2002. Dispersal, rookery fidelity, and metapopulation structure of Steller sea lions (Eumetopias jubatus) in an increasing and a decreasing population in Alaska. Marine Mammal Science. 18:746-764. Reed, M., E. Gundlach and T. Kana. 1989. A coastal zone hydrocarbon spill model: development and sensitivity studies. Hydrocarbon and Chemical Pollution 5: 411-449. Reed, M., N. Ekrol, P. Daling, Ø. Johansen, M.K. Ditlevsen, I. Swahn, J.L. Myrhaug Resby and K. Skognes. 2004. SINTEF Hydrocarbon Weathering Model User‘s Manual – Version 3.0. SINTEF Materials and Chemistry. Trondheim, Norway. Reger, D.R., J.D. McMahan and C.E. Holmes 1992. Effect of Crude Oil Contamination on Some Archaeological Sites in the Gulf of Alaska, 1991 Investigative Archaeological Study Number 1, Final Report. Alaska Department of Natural Resources Division of Parks and Outdoor Recreation, Office of History and Archaeology, Anchorage. Rice, S.D., S. J.W. and J.F. Karinen. 1976. Toxicity of Cook Inlet crude oil and No. 2 fuel oil to several Alaskan marine fishes and invertebrates. Sources, Effects and Sinks of Hydrocarbons in the Aquatic Environment. Proceedings of Symposium Paper presented at the Sources, Effects and Sinks of Hydrocarbons in the Aquatic Environment. Proceedings of Symposium, Washington, D.C. Cited in Birtwell, I.K. and C.D. McAllister. 2002. Hydrocarbons and their Effects on Aquatic Organisms in Relation to Offshore Oil and Gas Exploration and Oil Well Blowout Scenarios in British Columbia, 1985. Canadian Technical Report of Fisheries and Aquatic Sciences 2391. Fisheries and Oceans Canada. . 394-406. Am. Inst. Biol. Sc. Rocke, T.E., T.M. Yuill and R.D. Hinsdill. 1984. Oil and Related Toxicant Effects on Mallard Immune Defences. Environmental Research 33: 343–352. Rogers, F. 1992. More Shipwrecks of British Columbia. Douglas & McIntyre, Vancouver. Ronconi, R.A., C.C. St. Clair, P.D. O‘Hara and A.E. Burger. 2004. Waterbird deterrence at oil spills and other hazardous sites: potential applications of a radar-activated on-demand deterrence system. Marine Ornithology 32: 25–33. Savard, J.-P.L., D. Bordage and A. Reed. 1998. Surf Scoter (Melanitta perspicillata). In A. Poole & F. Gill (Eds.), The Birds of North America No. 363. Philadelphia, PA: The Birds of North America Inc. Cited in Fraser, D.F., W.L. Harper, S.G. Cannings, and J.M. Cooper. 1999. Rare birds of British Columbia. Wildlife Branch and Resource Inventory Branch, BC Ministry Environment, Lands and Parks, Victoria, BC. Savard, J-P.L. 1988. A summary of current knowledge on the distribution and abundance of moulting seaducks in the coastal waters of British Columbia. Technical Report Series No. 45. Canadian Wildlife Service, Delta, B.C. Scholz, D., Michel, J., Shigenaka, G., & Hoff, R. (1992). Introduction to Coastal Habitats and Biological Resources for Spill Response: Chapter 4 Biological Resources. In Hazardous Materials Response and Assessment Division. National Oceanic and Atmospheric Administration (Vol. HMRAD 92-4).

Page 17-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Schwiegert, J., B. McCarter, T. Therriault, L. Flostrand, C. Hrabok, P. Winchell and D. Johannessen. 2007. Appendix H: Pelagics. In Ecosystem overview: Pacific North Coast Integrated Management Area (PNCIMA). Can. Tech. Rep. Fish. Aquat. Sci. 2667:iv + 35 Schwarzenbach, R.P., P.M. Gschwend and D.M. Imboden. 2002. Environmental Organic Chemistry. 2nd edition. Wiley-Interscience. Hoboken, NJ. Sea Otter Recovery Team. 2007. Recovery Strategy for the Sea Otter (Enhydra lutris) in Canada. Vancouver. vii + 56pp. Sewell, A.T., J.G. Norris, S. Wyllie-Echeverria and J. Skalski. 2001. Eelgrass monitoring in Puget Sound: Overview of the Submerged Vegetation Monitoring Project. Prepared for the Washington State Department of Natural Resources. Olympia, WA. Short, J.W., M.R. Lindberg, P.M. Harris, J. Maselko, J.J. Pella and S.D. Rice. 2004. Estimate of oil persisting on the beaches of Prince William Sound 12 years after the Exxon Valdez oil spill. Environmental Science and Technology 38: 19-25. Slattery, S., B. Gray, L. Bogdan, K. Guyn, D. Buffet, J. Griffith, et al. 2000. A proposal for habitat conservation in the Fraser River Delta. Ducks Unlimited Canada British Columbia Coastal Field Office. Sloan, N.A. 1999. Oil impacts on cold-water resources: a review relevant to Parks Canada's evolving marine mandate. Parks Canada, National Parks, P. Canada SL Ross. 2009. Properties and Fate of Hydrocarbons from Hypothetical Spills in the Confined Channel Assessment Area and at the Marine Terminal Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Calgary, AB. SL Ross. 2010. Properties and Fate of Hydrocarbons from Hypothetical Spills in the Open Water Area Technical Data Report. Prepared for Enbridge Northern Gateway Pipelines Limited Partnership. Calgary, AB. SL Ross. 2010a. Properties and Fate of Hydrocarbons from Hypothetical Spills in the Confined Channel Assessment Area and at the Marine Terminal Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Smith, T. G., Geraci, J. R. and St. Aubin, D. J. 1983. Reaction of bottlenose dolphins, Tursiops truncatus, to a controlled oil spill. Can. J. Fish. Aquat. Sci. 40, 1522–1525. Smultea, M. A., & Wϋrsig, B. 1995. Behavioral reactions of bottlenose dolphins to the Mega Borg oil spill, Gulf of Mexico 1990. Aquatic Mammals, 21(3), 171-181. Speich, S.M., D.A. Manuwal and T.R. Wahl. 1991. The bird/habitat oil index: A habitat vulnerability index based on avian utilization. Wildlife Society Bulletin 19: 216-221. Springer, Y., Hays, C., Carr, M. and M. Mackey. 2007. Ecology and Management of the Bull Kelp, Nereocystis luetkeana: A Synthesis with Recommendations for Future Research. Lenfest Ocean Program. 53 p.

April 2010 FINAL Page 17-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Stephenson, M., A. St-Amand, P. Mazzocco and D. Yee. 2010a. Risk Assessment for Accidental Hydrocarbon Spills at the Kitimat Terminal and in the Confined Channel Assessment Area Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Stephenson, M., A. St-Amand, P. Mazzocco and J.-M. Devink. 2010b. Marine Ecological Risk Assessment for Kitimat Terminal Operations Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Steventon, J.D. and N.P. Holmes. 2002. A radar-based inventory of marbled murrelets (Brachyramphus marmoratus), northern mainland coast of British Columbia. Stephenson, M., A. St-Amand, P. Mazzocco and J.-M. Devink. 2009. Marine Ecological Risk Assessment for Kitimat Terminal Operations Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. Stephenson, R. 1997. Effects of oil and other surface-active organic pollutants on aquatic birds. Environmental Conservation 24: 121–129. Steventon, J.D. and N.P. Holmes. 2002. A radar-based inventory of marbled murrelets (Brachyramphus marmoratus), northern mainland coast of British Columbia. Suchanek, T.H. 1993. Oil Impacts on Marine Invertebrate Populations and Communities. Amer. Zool. 33: 510-523. Szaro, R. C. And P. H. Albers. 1977. Effects of external applications of No. 2 fuel on common eider eggs. In Fate and Effects of Petroleum Hydrocarbons in Marine Ecosystems and Organisms, D. A. Wolfe (ed.). New York: Pergamon Press. 164–167. Szaro, R. C. 1977. Effects of petroleum on birds. Transactions of the North American Wildlife and Natural Resource Conference 42: 374- 381. Teal, J.M. and R.W. Howarth. 1984. Oil spill studies: A review of ecological effects. Environmental Management. 8(1):27-43. Terres, J. K. 1980. The Audubon Society Encyclopedia of North American Birds. Alfred A. Knopf, New York. The Glosten Associates. 2004. Study of Tug Escorts in Puget Sound. Draft report prepared for State of Washington Department of Ecology, Lacey Washington. Thomson, R.E. 1989. The Queen Charlotte Islands: Physical Oceanography. In The Outer Shores. University of British Columbia, . Vancouver, BC. . Thorne, R. E, and G. L. Thomas. 2008. Herring and the "Exxon Valdez" oil spill: an investigation into historical data conflicts. ICES Journal of Marine Science. 65: 44-50. Triton Environmental Consultants Ltd. 1993. MTBE Trans-shipment Project Environmental Baseline and Sensitivity – Final Report. Prepared for Alberta Envirofuels Inc. Kitimat, BC United States Environmental Protection Agency (US EPA). 1998. Guidelines for Ecological Risk Assessment. EPA/630/R-95/002F.

Page 17-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

United States Environmental Protection Agency (US EPA). 2005. Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities, Final. EPA/530/R-05/006. Office of Solid Waste and Emergency Response, Washington, DC. Vadas, R.L. 1972. Ecological implications of culture studies on Nereocystis luetkeana. Journal of Phycology. 8:196-203 Vadas, R.L., S. Johnson and T.A. Norton. 1992. Recruitment and mortality of early post-settlement stages of benthic algae. British Phycological Journal. 27(3): 331-351. Varela, M., Bode, A., Lorenzo, J., Alvarez-Ossorio, M.T., Miranda, A., Patrocinio, T., Anadon, R., Viesca, L., Rodriguez, N., Valdes, L., Cabal, J., Urrutia, A., Garcia-Soto, C., Rodriguez, M., Alvarez-Salgado, X.A. and S. Groom. 2006. The Effect of the ―Prestige‖ Oil Spill on the Plankton of the N-NW Spanish Coast. Marine Pollution Bulletin. 53: 272-286. van Tamelen, P.G. and M.S. Stekoll. 1996. Population response of the brown alga Fucus gardneri and other algae in Herring Bay, Prince William Sound, to the Exxon Valdez oil spill. American Fisheries Society Symposium. 18: 193-211. von Ziegesar, O., E. Miller, E. and M.E. Dahleim. Impacts on humpback whales in Prince William Sound. In T.R. Loughlin. (ed.). Marine Mammals and the Exxon Valdez. Academic Press Inc., San Diego, CA. 173-191. Velando, A., Álvarez, D., Mouriño, J., Arcos, F., Barros, A. 2005. Population trends and reproductive success of European Shag following the Prestige oil spill in the Iberian Peninsula. Journal of Ornithology. 146: 116-120. Wang, J.Y., D.E. Gaskin and B.N. White. 1996. Mitochondrial DNA analysis of harbour porpoise, Phocoena phocoena, subpopulations in North American waters. Canadian Journal of Fisheries and Aquatic Science. 53: 1632-1645. Ware, D. and D. McQueen. 2006. Retrospective estimates of interannual and decadal variability in lower trophic level production in the Hecate Strait-Queen Charlotte Sound Region from 1958 to 1998. Canadian Technical Report of Fisheries and Aquatic Sciences. 2656. Weiner, A. H., C. Berg, T. Gerlach, J. Grunblatt, K. Holbrook, and M. Kuwada. 1997. The Exxon Valdez oil spill: habitat protection as a restoration strategy. Restoration Ecology. 5(1):45-55. Wertheimer, A.C. and A.G. Celewycz. 1996. Abundance and Growth of Juvenile Pink Salmon in Oiled and Non-Oiled Locations of Western Prince William Sound after the Exxon Valdez Oil Spill. Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium 18 Paper presented at Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium 18. Wertheimer, A.C., N.J. Bax, A.G. Celewycz, M.G. Carls and J.H. Landingham. 1996. Harpacticoid Copepod Abundance and Population Structure in Prince William Sound, One Year after the Exxon Valdez Oil Spill. Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium. 18.

April 2010 FINAL Page 17-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Wheeler, B., A. Rambeau and K. Zottenberg. 2009. Marine Mammals Technical Data Report. Prepared for Northern Gateway Pipelines Inc. Calgary, AB. White, C., M. Ritchie, J. Robert and B.A. Cooper. 1995. Density and productivity of bald eagles in Prince William Sound, Alaska, after the Exxon Valdez oil spill. ASTM Special Technical Publication. No. 1219. 762-779. Wiens, J.A., R.H. Day, S.M. Murphy and K.R. Parker. 2004. Changing habitat and habitat use by birds after the Exxon Valdez oil spill, 1989–2001. Ecological Applications 14, No. 6, pp. 1806-1825. Williams, G.L. 1989b. Coastal/Estuarine Fish Habitat Description and Manual, Part 1: Species/Habitats Outline. Prepared for Fisheries and Oceans Canada. Williams, T.M., R.A. Kastelein, R.W. Davis and J.A. Thomas. 1988. The effects of oil contamination and cleaning on sea otters (Enhydra lutris). I. Thermoregulatory implications based on pelt studies. Canadian Journal of Zoology 66: 2776–2781. Wϋrsig, B. 1990. Cetaceans and oil: Ecological perspectives. Pp. 129-165. In Geraci, J.R. and D.J. St. Aubin (eds.). Sea Mammals and Oil: Confronting the Risks. Academic Press, Inc.,San Diego, CA. 282 pp. Willette, M. 1996. Impacts of the Exxon Valdez Oil Spill on the Migration, Growth, and Survival of Juvenile Pink Salmon in Prince William Sound. Proceedings of the Exxon Valdez oil spill symposium. American Fisheries Society Symposium. 18. Williams, T.M., R.A. Kastelein, R.W. Davis, and J.A. Thomas. 1988. The effects of oil contamination and cleaning on sea otters (Enhydra lutris). I. Thermoregulatory implications based on pelt studies. Canadian Journal of Zoology 66: 2776-2781. Williams, G.L. 1989. Coastal/Estuarine Fish Habitat Description and Manual, Part 1: Species/Habitats Outline. Prepared for Fisheries and Oceans Canada. Wright, M. 2002. Eelgrass Conservation for the B.C. Coast: a Discussion Paper Wu, J. and X. Zhou. 1992. A study of the effects of petroleum on the early development of mussel Mytilus edulis. Transactions of oceanology and limnology/Haiyang Huzhao Tongbao 2:46-50. Würsig, B. 1990. Cetaceans and oil: Ecological perspectives. Pp. 129-165. In J.R. Geraci and D.J. St. Aubin (eds.). Sea Mammals and Oil: Confronting the Risks. Academic Press, Inc., San Diego, CA. 282 pp. Yender, R., J. Michel and C. Lord. 2002. Managing Seafood Safety after an Oil Spill. Hazardous Materials Response Division, Office of Response and Restoration, National Oceanic and Atmospheric Administration. Seattle. 72 p. Yelland, D. and W.R. Crawford. 2005. Currents in Haida Eddies. Deep-Sea Research II 52:875-892. Zieman, J.C., R.J. Orth, R.C. Phillips, G.W. Thayer and A. Thorhaug. 1984. The effects of oil on seagrass ecosystems. In J. Cairns & A. L. Buikema (ed.). Restoration of habitats impacted by oil spills. Boston, MA: Butterworth.

Page 17-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

17.2 Internet Sites Aboriginal Canada Portal. 2010a. Heiltsuk. http://www.aboriginalcanada.gc.ca/acp/community/site.nsf/en/fn538.html Aboriginal Canada Portal. 2010b. Quatsino. http://www.aboriginalcanada.gc.ca/acp/community/site.nsf/en/fn633.html Aboriginal Canada Portal. 2010c. Tlatlasikwala. http\p://www.aboriginalcanada.gc.ca/acp.community/site.nsf/en/fn632.html Andres, B.A. and G.A. Falxa. 1995. Black Oystercatcher (Haematopus bachmani), The Birds of North America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved from the Birds of North America Online: http://bna.birds.cornell.edu.bnaproxy.birds.cornell.edu/bna/species/155doi:10.2173/bna.155 BC CDC (BC Conservation Data Centre). 2009. Species Summary: Haematopus bachmani. B.C. Ministry of Environment. Available at: http://a100.gov.bc.ca/pub/eswp/ Accessed: Nov 25, 2009. BC Integrated Land Management Bureau, 2009. BCMEC Ecosections. Avalible at: http://www.ilmb.gov.bc.ca/cis/coastal/bcmec/ecosections.htm BC MOE (Ministry of Environment). 2009. BC Species Ecosystems Explorer. Northern resident killer whale. Available at: http://a100.gov.bc.ca/pub/eswp/reports.do?index=4 British Columbia Ministry of Environment (BC MoE). 2009b. BC Species Ecosystems Explorer. Northern resident killer whale. Available at: http://a100.gov.bc.ca/pub/eswp/ British Columbia Ministry of Environment (BC MoE). 2009c. BC Species Ecosystems Explorer. Humpback whale. Available at: http://a100.gov.bc.ca/pub/eswp/ BC Stats. 2008. British Columbia Tourism Room Revenue by Region. Annual 2007. Available at: http://www.bcstats.gov.bc.ca/data/bus_stat/busind/tourism/trra2007.pdf BC Stats. 2008a. Tourism Industry Monitor Annual 2007. Available at: http://www.bcstats.gov.bc.ca/data/bus_stat/busind/tourism/timcurr.pdf BC Stats. 2008b. British Columbia Tourism Room Revenue by Region. Annual 2007. Available at: http://www.bcstats.gov.bc.ca/data/bus_stat/busind/tourism/trra2007.pdf BC Stats. 2009a. 2006 Census of Canada: Census Profiles (A) Alphabetic List of Placenames. Available at: http://www.bcstats.gov.bc.ca/data/cen06/profiles/detailed/ch_alpha.asp BC Stats. 2009b. British Columbia Local Area Economic Dependencies: 2006. Available at: http://www.bcstats.gov.bc.ca/pubs/econ_dep.asp BC Stats. 2009c. Local Health Area 80 - Kitimat 2008 Statistical Profile. Available at: http://www.bcstats.gov.bc.ca/data/sep/lha/lha_80.pdf BirdLife International. 2008. Important Bird Areas. Available at: http://www.bsc-eoc.org/iba/spmaps.jsp. Accessed: November 2008.

April 2010 FINAL Page 17-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

BP Exploration (Alaska) Inc. 1996. Wildlife interaction and deterrence plan. GC-2 oil transit line. Greater Prudhoe Bay, Western Operating Area. 15 March 2006. Available at: http://www.dec.state.ak.us/SPAR/PERP/response/sum_fy06/060302301/plans/060302301_plan_ wildlife0315.pdf British Columbia First Nation Environmental Assessment Working Group. 2000. Workshop Report for the CEAA Five-Year Review. March 7th, 2000. Prepared by Praxis Pacific and submitted to the Canadian Environmental Assessment Agency. March 28th, 2000. Vancouver, BC. Available at: http://www.acee.gc.ca British Columbia Treaty Commission. 2010. Available at: http://www.bctreaty.net/nations British Columbia Wildlife Act. 1996. Available at: http://www.bclaws.ca/Recon/document/freeside/-- %20W%20--/Wildlife%20Act%20%20RSBC%201996%20%20c.%20488/00_96488_01.xml Brown, R.G.B. , A.R. Lock, and M. Kavanagh. 2003. Issues and Topics: Oil Pollution and Birds. HInterland's Who's Who. Environment Canada. Accessed: November 29, 2009.Available at: http://www.hww.ca/hww2.asp?id=229&cid=4#sid66 California Department of Fish and Game. 2005. Wildlife Response Plan for California. Office of Spill Prevention and Response. Available at: www.dfg.ca.gov/ospr/misc/WILDLIFE%20%20RESPONSE%20%20PLAN%206-30-2005.pdf. Appendices available at: http://www.dfg.ca.gov/ospr/report/wildlife_response_plan_6-30- 2005.pdf. Accessed: October 22, 2009. Canadian Coast Guard. 1995. Oil Handling Facilities Standards. TP 12402. Available at: http://www.tc.gc.ca/publications/EN/TP12402/PDF/HR/TP12402E.pdf Canadian Council of Ministers of the Environment (CCME). 2007. Canadian Water Quality Guidelines for the Protection of Aquatic Life. Summary Table. Update 7.1. Available at: http://ceqg- rcqe.ccme.ca/ Canadian Council of Ministers of the Environment (CCME). 2002. Canadian Sediment Quality Guidelines for the Protection of Aquatic Life. Summary Tables. Update 2002. Available at: http://ceqg-rcqe.ccme.ca/ CWS (Canadian Wildlife Service). 1999. Oil Spill Response Plan. Available at: http://www.atl.ec.gc.ca/reports/pdf/oil_spill_response.pdf Cetacealab. 2009. Cetacealab: Humpbacks. Accessed: January, 2009. Available at: http://cetacealab.org/humpbacks.htm CIT (Coast Information Team). 2003. Draft: An Ecosystem Spatial Analysis for Haida Gwaii, Central Coast, and North Coast British Columbia. Available at: http://www.for.gov.bc.ca/hfd/library/documents/bib92465.pdf

Page 17-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

COSEWIC. 2008. COSEWIC assessment and update status report on the Killer Whale Orcinus orca, Southern Resident population, Northern Resident population, West Coast Transient population, Offshore population and Northwest Atlantic / Eastern Arctic population, in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. viii + 65 pp. Available at: http://www.sararegistry.gc.ca/virtual_sara/files/cosewic/sr%5Fkiller%5Fwhale%5F0809%5Fe%2 Epdf COSEWIC. 2007. COSEWIC assessment and update status report on the sea otter Enhydra lutris in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. vii + 36 pp. http://www.sararegistry.gc.ca/document/dspDocument_e.cfm?documentID=1434 COSEWIC. 2003a. COSEWIC assessment and update status report on the humpback whale Megaptera novaeangliae in Canada. Committee on the Status of Endangered Wildlife in Canada, Ottawa, Canada. 25 pp. Available at: http://www.sararegistry.gc.ca/virtual_sara/files/cosewic/sr%5Fhumpback%5Fwhale%5Fe%2Epdf COSEWIC. 2003b. COSEWIC assessment and update status report on the Steller sea lion Eumetopias jubatus in Canada. Committee on the Status of Endangered Wildlife in Canada, Ottawa, ON. 47 pp. Available at: http://www.sararegistry.gc.ca/document/dspDocument_e.cfm?documentID=459 COSEWIC. 2003. COSEWIC assessment and update status report on the harbour porpoise Phocoena phocoena (Pacific ocean population) in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, ON. Available at: http://www.sararegistry.gc.ca/document/default_e.cfm?documentID=498 CRIMS (Coastal Resource Information System). 2009. Available at: http://www.ilmb.gov.bc.ca/cis/coastal/others/crimsindex.htm DFO (Fisheries and Oceans Canada). 2008a. Rockfish Conservation Areas – Pacific Region Area 5 – Goschen. Available at: http://www.pac.dfo-mpo.gc.ca/fm-gp/maps-cartes/rca-acs/rca-acs/north- nord/GoschenChart3927-eng.htm DFO (Fisheries and Oceans Canada). 2008b. Rockfish Conservation Areas – Pacific Region Area 4, 5 and 104 – Porcher Peninsula. Available at: http://www.pac.dfo-mpo.gc.ca/fm-gp/maps-cartes/rca- acs/rca-acs/north-nord/PorcherPeninsulaChart3927-eng.htm DFO (Fisheries and Oceans Canada). 2008c. Recovery strategy for the northern and southern resident killer whales (Orcinus orca) in Canada. Species at Risk Act Recovery Strategy Series. March 2008. Available at: http://www.dogwoodinitiative.org/in- depth/files/Final%20Recovery%20Strategy-Resident-Killer-Whale%20pdf.pdf DFO (Fisheries and Oceans Canada). 2008d. Dungeness, Red Rock Crab and Tanner Crab, Fact Sheet. Available at: http://www-comm.pac.dfo-mpo.gc.ca/publications/factsheets/species/crab_e.htm. Accessed: November 2008. DFO. 2009. Red Sea Urchin:Underwater World. Department of Fisheries and Oceans. Available at: http://www.dfo-mpo.gc.ca/science/publications/uww-msm/articles/urchin-oursin-eng.htm

April 2010 FINAL Page 17-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

DFO. 2008o. Dungeness, Red Rock Crab and Tanner Crab, Fact Sheet. Available at: http://www- comm.pac.dfo-mpo.gc.ca/publications/factsheets/species/crab_e.htm. Accessed: November 2008. DFO (Fisheries and Oceans Canada). 2007. 2005 Survey of recreational fishing in Canada. Available at: http://www.dfo-mpo.gc.ca/stats/rec/can/2005/index-eng.htm DFO (Fisheries and Oceans Canada). 2006a. The El Niño. Available at: http://www-sci.pac.dfo- mpo.gc.ca/osap/projects/default_e.htm Accessed: November, 2008. DFO (Fisheries and Oceans Canada). 2005. Amending the Marine Mammal Regulations. Consultation workbook. March 2005. Available at: http://www-comm.pac.dfo- mpo.gc.ca/pages/consultations/marinemammals/MMR%20Consultation%20Workbook%20Pacifi c%20Final%20II_TC_.pdf DFO (Fisheries and Oceans Canada). 2002. PACIFIC Marine Ecozone. Available at: http://www.eman- rese.ca/eman/network/pacific-ecozone.html Accessed: Novemeber, 2008. DFO (Fisheries and Oceans Canada). 2002a. Underwater world – Pacific Salmon. http://www.dfo- mpo.gc.ca/zone/underwater_sous-marin/salmon/salmon-saumon-eng.htm DFO (Fisheries and Oceans Canada). (Fisheries and Oceans Canada). 2002b. Underwater World. Dungeness crab. Available at: http://www.dfo-mpo.gc.ca/zone/underwater_sous-marin/crab- crabe/crab-eng.pdf DFO (Fisheries and Oceans Canada). 2002. Underwater World. Dungeness crab. Available at: http://www.dfo-mpo.gc.ca/zone/underwater_sous-marin/crab-crabe/crab-eng.pdf DFO (Fisheries and Oceans Canada). 2007. 2005 Survey of Recreational Fishing in Canada. Available at http://www.dfo-mpo.gc.ca/communic/statistics/recreational/Canada/2005/index_e.htm. Due West Charters. 2009. Kayak Mothership Cruises with Due West Charters on the B.C. Coast. Available at: http://www.duewestcharter.bc.ca/Home.htm Duen Sailing Adventures. 2009. The Natural Coast: Pacific Wildlife and Cultural Tours. Available at: http://www.thenaturalcoast.com/ Environment Canada. 2000. Bald Eagle: An indicator of wildlife sustainability in British Columbia. Available at: http://www.ecoinfo.gc.ca/env_ind/region/baldeagle/eagle_e.cfm. Accessed October 2009. Environment Canada. 2000. The Importance of Nature to Canadians: The Economic Significance of Nature-related Activities. Available at: http://www.ec.gc.ca/nature/pdf/nature_e.pdf Etkin, D.S. 2001. Comparative Methodologies for Estimating On-Water Response Costs for Marine Oil Spills in Proceedings of the 2001 International Oil Spill Conference pp 1281-1289. Available at: http://www.environmental-research.com/erc_papers/ERC_paper_3.pdf Etkin, D.S., D. French McCay, J. Jennings, N. Whittier, S. Subbayya, W. Saunders, and C. Dalton. 2003. Financial implications of hypothetical San Francisco bay oil spill scenarios: Response, socioeconomic, and natural resource damage costs. Proc. 2003 Int. Oil Spill Conf.: 1,317-1,325. Available at: http://www.environmental-research.com/erc_papers/ERC_paper_8.pdf

Page 17-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Government of Canada. 2009. Species Profile - Sea Otter. Species at Risk Public Registry. Available at: http://www.sararegistry.gc.ca Accessed November 4, 2009. Guilbault, J., M. Krautter, K.W. Conway and J.V. Barrie. 2006. Modern Foraminifera Attached to Hexactinellid Sponge Meshwork on the West Canadian Shelf: Comparison with Jurassic Counterparts from Europe. Paleontological Society. 9.1.3A. Available at: http://palaeo- electronica.org/2006_1/sponge/setting.htm. Accessed: November, 2008. Howes, D.E., M.A. Zacharias and J.R. Harper. 1997. British Columbia Marine Ecological Classification: Marine Ecosections and Ecounits. Available at: http://ilmbwww.gov.bc.ca/cis/coastal/mris/mec.htm Accessed: November, 2008. IBA Canada (Important Bird Areas Canada). 2009. Kitkatla Channel, Goschen Island North to Porcher Island. Available at: http://www.bsc-eoc.org/iba/site.jsp?siteID=BC119&seedet=Y Integrated Land Management Bureau (ILMB). 2009a. BCMEC Ecosections. Avalible at: http://www.ilmb.gov.bc.ca/cis/coastal/bcmec/ecosections.htm Integrated Land Management Bureau (ILMB). 2009b. Coastal Resource Information Management System (CRIMS). Available at: http://www.ilmb.gov.bc.ca/cis/coastal/others/crimsindex.htm International Convention on the Establishment of the International Fund for Compensation for Oil Pollution Damage. 1971. Information available at: http://www.iopcfund.org/ IMO (International Maritime Organization). 2002. International Safety Management (ISM) Code. Available at: http://www.imo.org/humanelement/mainframe.asp?topic_id=287 Indian and Northern Affairs Canada (INAC). 2009. Community Profile Information. Available at: http://www.ainc-inac.gc.ca/index-eng.asp. Indian and Northern Affairs Canada (INAC). 2006. Community Profile Information. Accessed: December 2006. Available at: http://www.ainc-inac.gc.ca/index-eng.asp. Indian and Northern Affairs Canada (INAC). 2008. Measuring Aboriginal Well-Being: The Human Development Index (HDI) and the Community Well-Being Index (CWB). Internet site: http://www.ainc-inac.gc.ca/ai/rs/pubs/rsh3-eng.asp INAC 2009, Internet site.Community Profile Information. Accessed: December 2009. Available at: http://www.aincinac.gc.ca/index-eng.asp Integrated Land Management Bureau (ILMB). 2009a. BCMEC Ecosections. Avalible at: http://www.ilmb.gov.bc.ca/cis/coastal/bcmec/ecosections.htm Integrated Land Management Bureau (ILMB). 2009b. Coastal Resource Information Management System (CRIMS). Available at: http://www.ilmb.gov.bc.ca/cis/coastal/others/crimsindex.htm International Maritime Organization (IMO). 2002. International Safety Management (ISM) Code. Available at: http://www.imo.org/humanelement/mainframe.asp?topic_id=287

April 2010 FINAL Page 17-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

International Oil Pollution Compensation Fund (IOPC).1971. International Convention on the Establishment of the International Fund for Compensation for Oil Pollution Damage. Available at: http://www.iopcfund.org/ International Tanker Owners Pollution Federation Limited. 2009. International Tanker Owners Pollution Federation Limited. Accessed: October, 2009. Available at: http://www.itopf.com Kauffman, K.E., F. Jaunes and M. McGinley. 2009. Alcids in Marine Ecosystems. In C.J. Cleveland (ed.). Encyclopedia of Earth. Environmental Information Coalition. National Council for Science and the Environment. Available at: http://www.eoearth.org/article/Alcids_in_marine_ecosystems Accessed: October 2009. King Pacific Lodge. 2009. A Place for Heart and Soul. Available at: http://www.kingpacificlodge.com/index.cfm Kitimat Chamber of Commerce. 2009. Available at: http://www.visitkitimat.com/ Lloyd‘s Register - Fairplay (LRFP). 2007. Marine Incident Database. Available at: http://www.lrfairplay.com/. Accessed: 2007. LRFP (Lloyd‘s Register - Fairplay). 2007. Marine Incident Database. 2007. Available at: http://www.lrfairplay.com/ Maple Leaf Adventures. 2009. Great Bear Rainforest. Available at: http://www.mapleleafadventures.com/elements/pdfs/brch/GreatBearRainforest-MapleLeaf.pdf Marine Liability Act. 2001. Available at: http://laws.justice.gc.ca/en/M-0.7/index.html MBCA (Migratory Birds Convention Act). 1994. Available at: http://laws.justice.gc.ca/en/M-7.01/ Ministry of Environment (BC MoE). 2009. BC Species Ecosystems Explorer. Northern resident killer whale. Available at: http://a100.gov.bc.ca/pub/eswp/ Ministry of Sustainable Resource Management. 2002. British Columbia Marine Ecological Classification. Marine Ecosections and Ecounits. Prepared for the Coastal Task Force, Resources Information Standards Committee. March 2002. Available at: http://www.ilmb.gov.bc.ca/risc/pubs/coastal/marine/version%202/bcmec_version_2.pdf. Accessed June 1, 2009. Norwegian Meteorological Institute. 2009. Available at: www.met.no. NMFS (National Marine Fisheries Service). 1991. Recovery Plan for the Humpback Whale (Megaptera novaeangliae). Prepared by the Humpback Whale Recovery Team for the National Marine Fisheries Service. Silver Spring, Maryland. 105 pp. Available at: http://www.nmfs.noaa.gov/pr/pdfs/recovery/whale_humpback.pdf NMFS (National Marine Fisheries Service). 1997. Investigation of scientific information on the impacts of California sea lions and Pacific harbor seals on salmonids and the coastal ecosystems of Washington, Oregon, and California. NOAA Tech. Mem. NMFS-NWFSC-28. 172 pp. Available at: http://www.nwfsc.noaa.gov/publications/techmemos/tm28/tm28.htm

Page 17-30 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

National Oceanic and Atmospheric Administration (NOAA). 2009a. Copepod. The global plankton database. Available at: http://www.st.nmfs.noaa.gov/plankton/content/region_npac.html National Oceanic and Atmospheric Administration (NOAA). 2009b. Oil spill response and killer whales. Available at: http://response.restoration.noaa.gov/type_topic_entry.php?RECORD_KEY(entry_topic_type)=en try_id,topic_id,type_id&entry_id(entry_topic_type)=799&topic_id(entry_topic_type)=1&type_id (entry_topic_type)=2 North King Lodge. 2009. Welcome to North King Lodge. Available at: http://www.northkinglodge.com/ Ocean Light II Adventures. 2009. Coastal Exploration Aboard the Ocean Light II. Available at: http://www.oceanlight2.bc.ca/index.html Olesiuk, P.F. 1999. An assessment of the status of harbour seals (Phoca vitulina) in British Columbia. Canadian Stock Assessment Secretariat Research Document 99/33. Department of Fisheries and Oceans, Ottawa, ON. Available at: http://www.dfo-mpo.gc.ca/csas/Csas/publications/ResDocs- DocRech/1999/1999_033_e.htm Ott, R., Peterson, C. and S. Rice. 2001. Exxon Valdez Oil Spill (EVOS) Legacy: Shifting Paradigms in Oil Ecotoxicology. Available at: http://www.alaskaforum.org Pacific States Marine Fisheries Commission. 1996. Dungeness Crab. Available at: http://www.psmfc.org/habitat/edu_crab_fact.html. Accessed: November 2009. Pacific Yellowfin Charters. 2009. Cruising the Great Bear Rain Forest. Available at; http://www.pacificyellowfin.com/site/our_cruises/great_bear_rainforest/cruising_the_great_bear_ rainforest.html Perry, R. I., M. Stocker and J. Fargo. 1994. Environmental effects on the distributions of Groundfish in Hecate Strait, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 51:1401. Picou, J.S. and D.A. Gill. 1996. ―The Exxon Valdez Oil Spill and Chronic Psychological Stress‖ in American Fisheries Society Symposium, 18: 879-893, 1996. Available at: http://jomiller.com/exxonvaldez/articles/picougill2.html Picou, J.S. and C.G. Martin. 2007. Long-Term Community Impacts of the Exxon Valdez Oil Spill: Patterns of Social Disruption and Psychological Stress Seventeen Years after the Disaster. Available at: http://www.arlis.org/docs/vol1/243478793.doc Pribilof Islands Wildlife Protection Subgroup. 2001. Wildlife Protection Guidelines: Pribilof Islands. Annex D – Attachment 4. Aleutian Islands Subarea Contingency Plan for Oil and Hazardous Substance Spills and Releases. Accessed July 5, 2006. Available at: http://www.akrrt.org/AIPlan/WPG_Pribs_Aug2005.pdf

April 2010 FINAL Page 17-31

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

Reed, M., N. Ekrol, P. Daling, Ø. Johansen, M.K. Ditlevsen, I. Swahn, J.L. Myrhaug Resby and K. Skognes. 2004. SINTEF Hydrocarbon Weathering Model User’s Manual – Version 3.0. SINTEF Materials and Chemistry. Trondheim, Norway. Available at: http://www.mms.gov/alaska/reports/2005rpts/2005_020/Final%20Report/Appendix%20D/User% 20Manual%20for%20SINTEF%20Hydrocarbon%20Weathering%20Model%203_0.pdf. Accessed: August 2009. SARA. 2007. Species at Risk Public Registry. Available at: http://www.sararegistry.gc.ca/default_e.cfm Accessed. Sea Duck Joint Venture. 2004. Sea Duck Information Series: Surf Scoter (Melanitta perspicillata). Available at: http://www.seaduckjv.org/infoseries/susc_sppfactsheet.pdf. Accessed: November 2009. Sefass, T. and P. Polechla. 2008. Lontra canadensis. In: IUCN 2009. IUCN Red List of Threatened Species. Version 2009.2. Available at: http://www.iucnredlist.org/apps/redlist/details/12302/0 Sellers, R.A., and S.D. Miller. 1999. Exxon Valdez Oil Spill. State/Federal Natural Resource Damage Assessment Final Report. Population Dynamics of Brown Bears After the Exxon Valdez Oil Spill. Alaska Department of Fish and Game. Division of Wildlife Conservation. Accessed October 22, 2009. Available at: http://www.evostc.state.ak.us/Files.cfm?doc=/Store/FinalReports/1991-TM04-Final.pdf& Sport Fishing Institute of British Columbia (SFIBC). 2008. Octopus. Available at: http://www.sportfishing.bc.ca/fish/octopus.htm Accessed: November, 2008. Statistics Canada. 2007. 2006 Community Profiles. 2006 Census Statistics Canada Catalogue no. 92-591- XWE. Ottawa. Released March 13 2007. Available at: http://www12.statcan.ca/census- recensement/2006/dp-pd/prof/92-591/index.cfm?Lang=E Statistics Canada. 2002. 2001 Community Profiles Released June 27, 2002. Last modified: 2005-11-30, Statistics Canada Catalogue no. 93F0053XIE. Available at: http://www12.statcan.ca/english/Profil01/CP01/Index.cfm?Lang=E Statistics Canada. 2007. 2006 Community Profiles. 2006 Census Statistics Canada Catalogue no. 92-591- XWE. Ottawa. Released March 13 2007. Available at: http://www12.statcan.ca/census- recensement/2006/dp-pd/prof/92-591/index.cfm?Lang=E Sea Duck Joint Venture. 2004. Sea Duck Information Series: Surf Scoter (Melanitta perspicillata). Accessed November, 2009. Available at: http://www.seaduckjv.org/infoseries/susc_sppfactsheet.pdf The Glosten Associates. 2004. Study of Tug Escorts in Puget Sound. Draft report prepared for State of Washington Department of Ecology, Lacey Washington United States Environmental Protection Agency (US EPA). 2009. Risk Assessment Portal. Available at: http://www.epa.gov/risk/basicinformation.htm#arisk

Page 17-32 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Section 17: References

U.S. Fish and Wildlife Service. 2004. Effects of Oil Spills on Wildlife and Habitat: Alaska Region. Accessed: September, 2009. Available at: http://alaska.fws.gov/media/unalaska/Oil%20Spill%20Fact%20Sheet.pdf Williams, G.L. 1989a. Coastal/Estuarine Fish Habitat Description and Assessment Manual. Part I. Available at: http://www.shim.bc.ca/species/species.htm Accessed: November, 2008.

17.3 Personal Communication Anderson, T. 2005. Project Biologist, Jacques Whitford Limited, Burnaby. BC. Verbal field report. August 15. Ford, J.K.B. 2005. Research Scientist, Fisheries and Oceans Canada, Conservation Biology Section, Nanaimo, British Columbia. Conference call November 1. Ford, J. 2008. Question regarding NR killer whale abundance. Head, Cetacean Research Program, Pacific Biological Station, Fisheries and Oceans Canada. Email, Nanaimo, BC. November 30, 2008. Wheeler, B. 2006. December recording of Humpback whales. Marine Science Practice Lead, Jacques Whitford Axys Ltd. Conversation, Burnaby, BC. Grant, L. 2009. Store manager. MK Bay Marina. Kitimat, BC. Personal communication. January 2009. McGee, G. 2009. Manager. Log Barging. Seaspan International. Vancouver, BC. Personal communication. February 2009. Thompson, R. 2009. Fishing Charter Owner-Outfitter. Kitimat, BC. Personal communication. January 2009.

April 2010 FINAL Page 17-33

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Appendix A Figure References

April 2010 FINAL Page A-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-1 Oceanographic Summary Figure References

Data layers Reference Legend title included Data source Number Comments Summer Current Digitized from hard Thomson, 1989 233 This layer represents the copy (Crawford et al summer oceanographic 2007) currents in BC that are a combination of tidal (gravitational pull) and non-tidal (wind driven, runoff/buoyancy driven etc.) currents Summer Digitized from hard Thomson, 1989 233 This layer represents the Variable Current copy (Crawford et al. summer oceanographic 2007) tidal and non-tidal currents that are variable throughout the season Winter Current Digitized from hard Thomson, 1989 233 This layer represents the copy (Crawford et al winter oceanographic 2007) tidal and non-tidal currents Winter Variable Digitized from hard Thomson, 1989 233 This layer represents the Current copy (Crawford et al. winter oceanographic 2007) tidal and non-tidal currents that are variable throughout the season Oceanographic 1: High aggregation DFO 235 Large identified IA that Areas of of macrozooplankton covers continental shelf Significance and consists of high aggregations of macrozooplankton 2: Learmouth Bank DFO 235 Learmouth Bank is an isolated bank that acts to trap a diversity of plankton in the surrounding water 3: Mc Intyre Bay DFO 235 McIntyre Bay eddies eddies have been shown to concentrate decapod larvae and support aggregations of plankton diversity 4: Dogfish Bank DFO 235 Dogfish Bank is the largest shallow bank in the region and serves as a larval rearing area for a large diversity of invertebrate species

April 2010 FINAL Page A-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-1 Oceanographic Summary Figure References (cont’d)

Data layers Reference Legend title included Data source Number Comments Oceanographic 5: Hecate Strait tidal DFO 235 The Hecate Strait tidal Areas of front front is effective from Significance spring through fall (cont’d) accumulating zooplankton as it moves. 6: Haida eddies DFO 235 & 238 The Haida Eddies form at the tip of Cape St James and act as a center of transport of concentrated plankton waters from PNCIMA to the Gulf of Alaska. The arrow represents the direction of water flow towards Alaska 7: High aggregation DFO 235 Canyons and troughs of of macrozooplankton Queen Charlotte Sound where there is high aggregations of macrozooplankton 8: High aggregation DFO 235 Canyons and troughs of of macrozooplankton Queen Charlotte Sound where there is high aggregations of macrozooplankton 9: Tidal mixing DFO 235 This area surrounds Scott Islands. The tidal mixing here drives high productivity 10: Brooks Peninsula DFO 235 At Brooks Peninsula north/south boundary there is often an offshore flow of nearshore waters and it is a significant north/south boundary for many eastern Pacific species 11 & 12: Tidal mixing DFO 235 These are 2 areas and high productivity identified because they have particularly high productivity as a result of tidal mixing.

Page A-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-2 Benthos Summary Figure References

Legend Data layers Reference title included Data source Number Comments Red and Important Areas DFO 27 The IA's represent areas identified Green sea as having high aggregations and urchins good fitness for Green urchins Marine resources- MAPSTER 62, 67 Represents Red and Green urchin local knowledge (DFO) distribution Red and Green CRIMS 160 Represents Red and Green urchin sea urchins distribution Sea Important Areas DFO 26 IA's were identified for Giant red Cucumber sea cucumber for concentrations of productivity and high densities Marine resources- MAPSTER 70, 74 Represents Sea cucumber local knowledge (DFO) distribution Olympia Important Areas DFO 25 Identified IA in the PNCIMA oyster Known SARA 228 Most of the distribution data came distributions from COSIEWIC report Shrimp Important Areas DFO 30 IA's were identified based on trawl industry log books and research trawl data, for aggregations and uniqueness Known CRIMS 158 Represents data derived from distributions fisheries records Prawn Known MAPSTER 93 Represents data derived from distributions (DFO) fisheries records Tanner Important Areas DFO 29 The areas identified are based on crab research surveys done on the continental shelf. However there is not much information at the moment and the area is subject to change Dungeness Important Areas DFO 28 Identified IA's in the PNCIMA were crab identified important because of uniqueness, aggregations and fitness. Corals and Important Areas DFO 56 IA's were identified from bycatch sponges from fisheries. Data was provided by Living Oceans Society and produced by DFO

April 2010 FINAL Page A-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-2 Benthos Summary Figure References (cont’d)

Legend Data layers Reference title included Data source Number Comments Sponge Important Areas DFO 55 Sponge reefs were identified for reefs their uniqueness, aggregations, naturalness and their low resiliance. Sponge reef Natural 219, 220 Layers of sponge reef for Hecate distribution Resources and Georgia Straits were used to Canada identify sponge reefs, and more specifically areas where the Hexatinellid glass sponges exist. Manila Important Areas DFO 22 IA's were identified for high clam densities of aggregations Razor clam Important Areas DFO 23 The IA was identified for high uniqueness and aggregations Geoduck Important Areas DFO 24 The IA was identified for high clam productivity and density beds.

Page A-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-3 Marine and Anadromous Fish Summary Figure References

Reference Legend title Data layers included Data source Number Comments Areas of Green Sturgeon Important DFO PNCIMA 2 Important habitat area as Important Areas Identified in PNCIMA Fish Habitat report Rockfish Important Areas DFO PNCIMA 18 Important habitat area as Identified in PNCIMA report Pollock Important Areas DFO PNCIMA 57 Important spawning and rearing habitat areas Lingcod Important Areas DFO PNCIMA 59 Important spawning and rearing habitat areas Pacific Halibut Important DFO PNCIMA 60 Important spawning and Areas rearing habitat areas Herring Important Areas DFO PNCIMA 21 Important spawning, rearing and migration habitat areas Major Herring Spawning CRIMS 168 Major Herring Spawning areas areas 2000-2007 Eulachon Important Areas DFO PNCIMA 3 Important spawning and summer feeding habitat areas Eulachon Spawning CRIMS 240 Represents the known Rivers spawning rivers for Eulachon Salmon Important Areas DFO PNCIMA 1 Important migratory, staging, rearing, and feeding habitat areas Sole and Flounder DFO PNCIMA 17 Important spawning and Important Areas rearing habitat areas Sablefish Important Areas DFO PNCIMA 16 Important habitat area as Identified in PNCIMA report Hake Important Areas DFO PNCIMA 20 Important migration habitat areas Pacific Cod Important DFO PNCIMA 59 Important spawning and Areas rearing habitat areas Rockfish Rockfish Conservation DFO PNCIMA 37 Areas of important rockfish Conservation Areas conservation closed to Areas commercial and recreational fishing

April 2010 FINAL Page A-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-4 Marine Birds Summary Figure References

Reference Legend title Data layers included Data source Number Comments Sea Bird Bald Eagles CRIMS 130 Habitat and distribution in BC Habitat Black Oystercatchers CRIMS 131 Habitat and distribution in BC Cormorants CRIMS 132 Habitat and distribution in BC Dabbling Ducks CRIMS 133 Habitat and distribution in BC Diving Ducks CRIMS 134 Habitat and distribution in BC Fulmars, Shearwaters and CRIMS 135 Habitat and distribution in BC Petrels Geese and Swans CRIMS 136 Habitat and distribution in BC Great Blue Heron CRIMS 137 Habitat and distribution in BC Gulls CRIMS 138 Habitat and distribution in BC Loons and Grebes CRIMS 139 Habitat and distribution in BC Shorebirds CRIMS 140 Habitat and distribution in BC Pelagic birds CRIMS 143 Habitat and distribution in BC Important CWS BC Marine Bird Area Environment 222 Marine areas with a heightened Bird Areas of Interest Canada ecological value for migratory of birds Significance Bird Important Areas PNCIMA 4 Important areas identified for marine birds; Layer does not include distribution beyond the continental shelf, south of Brooks Peninsula, or inlets. Globally identified important BirdLife 223 Polygon near Aristazabal Island bird area International was Digitized based on BirdLife International’s identification of IBAs of global or national significance Alcid Range Alcid Range CRIMS 129 Habitat and distribution in BC Marbled Marbled Murrelet Range CRIMS 142 Habitat and distribution in BC Murrelet Range Important Sea Bird Colonies CRIMS 144 Distribution of nesting areas for Bird bird colonies in coastal BC Colonies

Page A-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-5 Marine Mammals and Turtles Summary Figure References

Data layers Reference Legend title included Data source Number Comments Important Sperm Whale IA PNCIMA 6 Functional Habitat Use – deep Marine water area; Layer does not include Mammal distribution beyond the continental Habitat shelf or south of Brooks Peninsula. Grey Whale IA PNCIMA 8 Important migration and feeding areas; Layer does not include distribution south of Brooks Peninsula. Blue Whale IA PNCIMA 9 Functional Habitat Use– deep water area; Layer does not include distribution beyond the continental shelf or south of Brooks Peninsula. Northern Resident PNCIMA 5 Functional Habitat Use– Killer Whale IA socialization and migration area Humpback Whale PNCIMA 7 Functional Habitat Use– feeding IA area; Layer does not include distribution beyond the continental shelf or south of Brooks Peninsula. Sei Whale IA PNCIMA 10 Functional Habitat Use– deep water area; Layer does not include distribution beyond the continental shelf or south of Brooks Peninsula. Fin Whale IA PNCIMA 11 Functional Habitat Use based on sightings and historic whaling records; Layer does not include distribution beyond the continental shelf or south of Brooks Peninsula. Northern Fur Seal PNCIMA 13 Foraging and feeding areas; Layer IA does not include distribution beyond the continental shelf or south of Brooks Peninsula. Northern Fur Seal CRIMS 184 Extends range down SWVI –which was missing from PNCIMA layer Steller Sea Lion IA PNCIMA 12 Haulouts; Layer does not include distribution beyond the continental shelf or south of Brooks Peninsula. Steller Sea Lion CRIMS 189 Extends range down SWVI –which was missing from PNCIMA layer

April 2010 FINAL Page A-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-5 Marine Mammals and Turtles Summary Figure References (cont’d)

Data layers Reference Legend title included Data source Number Comments Important Leatherback Turtle PNCIMA 31 Feeding area; Layer does not Marine IA include distribution beyond the Mammal continental shelf or south of Brooks Habitat Peninsula. (cont’d) California Sea lion CRIMS 176 Range in BC Humpback Humpback Whale PNCIMA 7 Known Concentrations; Layer does Whale Known IA not include distribution beyond the Concentrations continental shelf or south of Brooks Peninsula. Northern N.R. Killer Whale PNCIMA 5 Important areas based on formally Resident Killer IA identified critical habitat, or areas Whale Critical listed as potential critical habitat in and Potential future Critical Habitat Sea Otter Sea Otter IA PNCIMA 14 Sea otter established range. Layer Known does not include distribution south Concentration of Brooks Peninsula. Sea Otter CRIMS 186 Extends range down SWVI –which was missing from PNCIMA layer Steller Sea Steller Sea Lion IA PNCIMA 12 Rookeries Lion Rookeries

Page A-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-6 Human Use Summary Figure References

Data layers Reference Legend title included Data Source Number Comments Oil / Gas Exploratory CRIMS 194 Represents areas Exploratory Well Wells explored for oil and gas potential. Oil / Gas Tenure Tenure- BC, CRIMS 197 Represents areas that will active likely be utilized for oil Tenure- CRIMS 198 and gas mining, if the Canada, active moratorium is lifted. This area encompasses some of the best prospective petroleum deposits in BC. High Wind wind energy Wind Energy Atlas 226 Represents areas of high Energy Potential potential wind energy potential that Area are likely to be explored for offshore wind farm licenses. Some investigative permits have already been issued within this area. Aquaculture Site Finfish farms CRIMS 123 Includes all finfish farms within the OWA. Shellfish farms, CRIMS 126, Includes all the shellfish hatcheries, and 127, farms within the OWA, aquaculture 128 some of which extend capability into the Confined Channel Area. These data points were small- scale and offered better visual data when merged with other aquaculture data. Processing Site Processors CRIMS 125 Represents locations of aquaculture processing which incorporate raw materials into marketable commodities. Rockfish Rockfish DFO / CRIMS 37 Represents locations of Conservation Conservation important rockfish habitat Area Areas and aims to protect areas with the most apparent rockfish declines.

April 2010 FINAL Page A-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-6 Human Use Summary Figure References (cont’d)

Data layers Reference Legend title included Data Source Number Comments Gwaii Haanas Gwaii Haanas Parks / CRIMS 227 Outline of the proposed National Marine NMCA NMCA, which Conservation encompasses the Gwaii Area / National Haanas National Park. Park Together these conservation initiatives aim to protect this important area of high biological, ecological, and physical characteristics. National Parks: CRIMS This data represents the Only Gwaii terrestrial National Park Haanas in OWA with boundaries that extend into the OWA to protect portions of the coastal marine environment. BC Park / Provincial Parks CRIMS 214 Includes terrestrial Parks Protected Area that have boundaries which extend into the OWA to protect portions of the coastal marine environment. Sponge Reef Sponge reefs PNCIMA / DFO / 55/ 219/ Represents important CRIMS 220 Sponge Reef habitat / protected areas. Sponge and Corals and PNCIMA 56 Represents Important Corals Sponges Sponge and Coral habitat / protected areas Recreational NC Sport Fish DFO 71 Sports fishing data was Fishing Area combined into one file CC Sport Fish DFO 75 because it represented data from different locations within the OWA. NC Rec. Crab DFO 66 Recreational crab fishing data was included with overall recreational fishing layer because the sites were often close-to or on-top-of the sport / finfish fishing areas.

Page A-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-6 Human Use Summary Figure References (cont’d)

Data layers Reference Legend title included Data Source Number Comments Recreational Sport lodge DFO 107 Point data for sport Fishing Area 2007 fishing lodges in BC (cont’d)

Ground Fish Groundfish CRIMS 163 Represents areas of Recreational important recreational Fishing Area groundfish fishery Dive Site - High Dive sites CRIMS / DFO 205 Represents areas of high Use use - DFO / CRIMS data show the same areas. DFO Staffed DFO Staffed DFO 120 Represents areas that Site sites are monitored by DFO staff and are thus, areas of easy accessibility / quick response by DFO staff in the event of an emergency situation (e.g. an oil spill). Coast Guard Coast Guard DFO 121 Represents areas that Site Staffed sites are monitored by Coast Guard staff and are therefore areas of easy accessibility / quick response by Coast Guard staff in the event of an emergency situation (e.g. an oil spill). Moorage / Small Craft DFO 115 Represent areas where Anchor / Marina Harbours people will be accessing Location Anchorages CRIMS 200 the OWA with their recreational / pleasure Marinas CRIMS 208 craft vessels. Moorage CRIMS 211 BC Ferry Route Ferry routes and CRIMS 206 Represents popular, terminals southern OWA, ferry routes - this data set on its own was incomplete.

April 2010 FINAL Page A-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix A: Figure References

Table A-6 Human Use Summary Figure References (cont’d)

Data layers Reference Legend title included Data Source Number Comments BC Ferry Route Additional MacConnachie et 225 Represents northern BC (cont’d) Northern BC al. 2007 MSP Ferry Routes that were Ferry Routes not included in the CRIMS database such as: Prince Rupert - Queen Charlottes and Prince Rupert through the inside channel. Industry Site Marine CRIMS 210 Represents various industries industries and all of the buildings associated with each industry. Additionally, it is not clear whether all industries are currently in operation. Thus, this data layer likely over-represents industrial operations in the OWA. Cruise Ship Cruise ship MacConnachie et Represents cruise ship Route routes to Alaska al. 2007 MSP routes during the peak season (June – early September). These routes are not heavily frequented outside of the summer window.

Page A-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Appendix B General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

April 2010 FINAL Page B-1

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Loons and Grebes Red-Throated Loon Widely distributed Kitimat Estuary Prefers shallow inshore Feeds in OWA year- throughout OWA. Minette Bay to Kitamaat and protected waters - round. Village bays, inlets, harbours, Collects food for Coste Rocks, Clio Bay lagoons, and estuaries. young in summer. and Walbran Point Princess Royal Island Pacific Loon Widely distributed along Kitimat Estuary Deeper and offshore Feeds in OWA year- coast. Minette Bay to Kitamaat waters, but also bays, round. Northwest part of Hecate Village estuaries, surge narrows, Most abundant in Straight is a Global IBA. channels, coves, inlets and spring and fall. lagoons. Common Loon Widely distributed Kitimat Estuary Inshore sea coasts and Feeds in OWA year- throughout OWA. Minette Bay to Kitamaat sheltered waters. round. Village Most abundant in Douglas Channel spring and fall. Yellow-Billed Loon Widely distributed Uncommon in CCAA Prefers more open water Rarer visitor. throughout OWA. than common. Least abundant in summer. Horned Grebe Widely distributed Throughout Douglas Inshore marine waters. Feeds in OWA year- throughout OWA. Channel round. Most abundant in spring and fall. Red-Necked Grebe Widely distributed Throughout Douglas Bays, inlets, estuaries and Feeds in OWA year- throughout OWA. Channel narrows. round. Most abundant in spring and fall.

April 2010 FINAL Page B-3

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Loons and Grebes (cont’d) Western Grebe In winter OWA has Throughout Douglas Winter on sheltered salt Feeds in OWA year- highest abundance in Channel and brackish water along round. North America. coast. Most abundant in Often congregate in large spring and fall. groups for extended periods of time. Albatrosses, Petrels and Shearwaters Black-Footed Albatross Widely distributed Rare Pelagic. Common from spring offshore along to fall, most abundant continental slope and in summer. Vancouver Island Shelf. Laysan Albatross Widely distributed Rare Pelagic. Regular visitor in low offshore along numbers, particularly continental slope and in summer, July– Vancouver Island Shelf. November. Short-Tailed Albatross Widely distributed Very Rare Pelagic. Rare, mostly summer offshore along Travels long distance visitor. continental slope and from Japanese breeding Vancouver Island Shelf. grounds to feed in OWA, particularly along continental slope.

Page B-4 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Albatrosses, Petrels and Shearwaters (cont’d) Northern Fulmar Widely distributed. Pelagic. Year-round, most Common over shelf break Breeding population abundant in winter. and northern waters. (critically imperiled in Southern Vancouver British Columbia) nests in Island Shelf is a National colonies primarily on sea IBA. cliffs, and less frequently on low flat rocky islands. Pink-Footed Common throughout Uncommon Pelagic. Mostly summer. Shearwater offshore and continental Common visitor to OWA. shelf waters. Would be sensitive to any offshore oil spills in summer. Buller's and Flesh- Common throughout Uncommon Pelagic. Mostly summer and Footed Shearwater offshore and continental fall. shelf waters. Buller’s common, flesh-footed rarer. Sooty and Common in Hecate Strait, Moore, Byers, Pelagic. Mostly summer. Short-Tailed Queen Charlotte Sound, McKenny, and Shearwater and along shelf break. Whitmore Islands

April 2010 FINAL Page B-5

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Albatrosses, Petrels and Shearwaters (cont’d) Leach's Storm QCI, north Vancouver Nest on Moore and Pelagic. Year-round. Petrel Island, Scott Islands, and Byers Islands Comes ashore only to Large numbers of Aristazabal Island are breed along west coast of summer breeders. Global IBAs. Vancouver Island and Small numbers winter West Vancouver Island is Haida Gwaii. on open sea of coast. also a National IBA. Estimated 1,400,000 breeding in British Columbia. Single birds in channels. Fork-Tailed Storm Global IBA nesting Nest on Moore and Nearshore protected Summer breeders Petrel colonies in British Byers Islands waters. and overwintering Columbia on QCI, north Seen at Coste Island Single birds in channels. birds. Vancouver Island and Also common over open west of Aristazabal Island. waters. Cormorants Double-Crested Coastal breeder in Strait Common in Douglas Nearshore: bays, harbours, Year-round resident. Cormorant of Georgia. Channel during Fall estuaries, inlets. Overwinters in OWA. Winters along coast. Roosts on islets. Nests on ground in dense colonies.

Page B-6 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Cormorants (cont’d) Pelagic Cormorant Widespread along inner Potential Breeding Site Rocky coasts. Year-round resident. (Pelagicus and outer coast. at Coste Island Rock Forages in bays, harbours, Overwinters in OWA. Subspecies) Nesting colonies on east lagoons, surge narrows Moresby Island, Scott and coves. Islands, and west Roosts on rocky, un- Vancouver Island. vegetated islets, reefs, cliffs. Breed in small colonies along coast. Brandt’s Cormorant Coastal breeder. Historic Breeding Prefers open-ocean. Year-round resident. West Vancouver Island is Activity in CCAA Often fishes in large flocks. Overwinters in OWA. a Global IBA. Roost on rock outcroppings just offshore. Waders American Bittern Sparsely distributed and Not Recorded in CCAA Estuarine marshes and Winter. not abundant. sloughs.

April 2010 FINAL Page B-7

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Waders (cont’d) Great Blue Heron One subspecies (fannini) Common in CCAA: Salt, brackish and Year-round. resident along coast. Emsley Cove, Bish freshwater environs: Nesting occurs in groups Cove, Princess Royal shallow bays, lagoons, of two or three pairs, Island, Dewdney and inlets, coves, tidal scattered across coast Glide Island Ecological mudflats, sloughs, and QCI. Reserve marshes, rivers, ditches. Subspecies herodias Forages at water edge. breeds inland, but winters coastally. Geese and Swans Trumpeter Swan Southern coastal OWA in Kitimat Estuary Sheltered, shallow, aquatic Year-round. winter. Emsley Cove habitat; estuaries, bays, More abundant in More northern during lagoons, tidal mud flats, winter and peaks in migration. beaches and inlets. spring. Snow Goose and Migrates offshore Estuarine marshes with Spring and fall Greater White- (Greater White-Fronted) bullrush rhizomes, sedge migrants. Fronted Goose and coastally (both) often rhizomes and shoots. in large flocks. Shallow waters of sloughs, and mud flats.

Page B-8 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Geese and Swans (cont’d) Brant Widely distributed in Eelgrass habitat. Primarily a spring OWA, particularly inner Estuaries, beaches, bays, migrant. coast. lagoons, mud flats, cobble Almost entire global beaches of outer coastline population migrates along only. the British Columbia coast Not in channels. each spring. WVI and EQCI are Global IBAs. Canada Goose Widely distributed in Migration through Anywhere with permanent Year-round. OWA. Kitimat Arm. Emsley water and grazing habitat. More abundant in Cove, and Princess Migrate offshore in fall. winter and peaks in Royal Island spring. Found throughout CCAA Dabbling Ducks Northern Pintail, Widely distributed in Kitimat River Estuary Tidal mudflats and Most abundant in American Wigeon, coastal nearshore areas. Emsley Cove marshes, estuaries, creek winter, and spring Green-Winged Teal Might be found in flocks Princess Royal Island mouths, shallow foreshore and fall migration. offshore during migration. and bays. Pintail and wigeon on Exposed eelgrass and QCI year-round. seaweed beds. Mallard, Northern Widely distributed in Migration North-South Prefers shallow open areas Present year-round. Shoveler, Gadwall coastal nearshore areas. Through and marshes, estuaries, Kitimat Arm sheltered bays, mudflats, Bish Cove and coastal marine waters.

April 2010 FINAL Page B-9

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Diving Ducks Ring-Necked Duck Throughout coastal OWA. Saltwater lagoons and Common winter sloughs, inlets, bays and resident and migrant. estuaries. Harlequin Duck Groups of two to six Kitimat Arm Turbulent waters adjacent Year-round resident usually associated with Bish Cove to rocky islets, shores and nearshore. freshwater along Might Breed in Creeks bays; saltwater lagoons, Further offshore in channels. flowing into Douglas inlets and harbours. winter. Breed in some small Channel Breed along coastal coastal creeks. creeks. Forage close to tideline. Long-Tailed Duck Throughout coastal OWA. Small Numbers in the Offshore and deeper Most common during Common in open-ocean. CCAA waters of straits, bays, winter (offshore) into Major concentration in harbours, channels and spring (inshore). northern QCI. fiords, also estuaries, mudflats. Form large rafts. Surf Scoter Mostly coastal. North End of Kitimat Open, shallower waters of Most abundant during During winter British Arm straits adjacent to winter and spring. Columbia OWA has Coste Island Rock beaches, spits and points. highest abundance in Bish Cove Also protected bays, North America. harbours and lagoons, Global IBA off SWVI and deep fiords. near Prince Rupert. Large concentrations near Large concentrations form herring spawning sites. big rafts in coastal areas.

Page B-10 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Diving Ducks (cont’d) White-Winged Throughout coastal and North End of Kitimat Marine brackish waters: Most common during Scoter and Black open waters of OWA. Arm bays, inlets, sounds, winter. Scoter Coste Island Rock channels, estuaries. Bish Cove White-winged prefers more open, deeper waters than surf scoter. Sandy gravelly bottoms with shellfish and mussel beds. Common and Throughout coastal OWA. Bish Cove Closely linked to intertidal Abundant during Barrow’s 60–90% of global Barrow’s Goldeneye habitats. winter, as well as Goldeneye, and population of Barrow’s abundant in the CCAA Estuaries, bays, harbours, during spring and fall Bufflehead Goldeneye breeds in utilizing bays, estuaries, lakes, lagoons, shallower migrations. British Columbia (on inlets, harbours and waters of straits adjacent Rare in summer. ponds, but uses coast in rocky shores to beaches and spits, winter). fiords with rocky shores and extensive mussel beds. Greater and Lesser Throughout coastal and Kitimat Arm Estuaries, bays, harbours Common winter Scaup open water bays of OWA. and saltwater lagoons. residents and Greater prefers large open migrants. waters of straits, near points, rocky islets or beaches, and brackish waters.

April 2010 FINAL Page B-11

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Diving Ducks (cont’d) Hooded Merganser Mostly seen in southern Bish Cove Estuaries, protected bays Most common during OWA. and inlets, coastal lakes, winter, year-round in marshes, sloughs. southern OWA. Common Merganser Widespread throughout Bish Cove Fresh and brackish waters Year round in OWA. coastal OWA. Breeding in Emsley of estuaries, protected Cove bays and inlets. Roosts in adjacent shoals, beaches, gravel/sand bars, logs, dry spits. Might breed near marine shores. Brood in later summer, otherwise mostly salmon stream outlets. Red-breasted Widespread in winter. Bish Cove Bays, estuaries and inlets. Most common during Merganser Breeds in summer in winter. Masset Inlet (QCI). Coastal Raptors Osprey Widespread throughout Kitimat Arm Protected coastal waters Mostly in summer coastal OWA. such as bays, lagoons, (breeds in OWA). inlets. Breeds wherever there are fish (almost exclusive diet).

Page B-12 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Coastal Raptors (cont’d) Bald Eagle Throughout coastal OWA. Kitimat Arm Seashores, sloughs, Year-round. 20–35% of global Princess Royal Island marshes. Largest influx from population breed along Nesting on Moore and Breed primarily in interior of whole coastline and rivers in Byers Islands coniferous forests near western continent as British Columbia. Dedney and Glide seashores, and marshes. salmon run in fall and Islands winter. Peale’s Peregrine Throughout coastal outer Kitimat Arm Typically nests on ledges Year-round. Falcon shores of OWA. British Columbia’s only of rocky island cliffs, Nests at scattered known Tree-Nesting usually near seabird locations along north Peregrine Falcons found colonies. coast. at the Byers, Conroy Occasionally, nests occur Northern and southern and Harvey Sinnett on mainland headland QCI, and Scott Islands Islands Ecological cliffs. are National IBAs. Reserve A few nests on grassy ledges on rock bluffs. More rarely, old nests of Pelagic Cormorants, Bald Eagles and Common Ravens have been used. Rails, Coots, and Cranes American Coot Throughout coastal OWA. Kitimat Area Marine and brackish Common in fall and waters, including marshes, winter. sloughs, estuaries, tidal mudflats.

April 2010 FINAL Page B-13

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Rails, Coots, and Cranes (cont’d) Sandhill Crane Not often in salt water but Nesting Habitat at Estuarine marshes, Primarily fall and might use low-lying Princess Royal Island, intertidal areas. spring migrants. patches of shoreline. Dewdney and Glide Breeds on bogs of coastal Breed along coast in Breeds on QCI and Island Ecological islands. summer. coastal mainland islands. Reserves Forages along cobble beaches and low-lying shoreline of outer islands. Shorebirds Black-Bellied Plover Coastal OWA. Tidal mudflats, sandy Spring and fall beaches, rocky islets and migrant. beaches. Migrant on outer islands. American Golden Coastal OWA. Lagoon shores, sandspits, Spring and fall Plover tidal mudflats, rocky migrant. beaches. Migrant on outer islands. Semipalmated Breeds sparingly in Tidal mudflats, Mostly spring and fall Plover coastal British Columbia. sandy/gravel beaches, migrant. small estuaries. and sometimes rocky beaches. Breeds on sandy beaches.

Page B-14 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Killdeer Breeds in coastal British Likely to breed in Kitimat Tidal mudflats and Year-round. Columbia. River Estuary sandspits. Roost on log booms and nearshore rocks. Breeds in open areas with minimal ground cover, including marine beaches. Black Oystercatcher Have been known to nest Breeding at Moore, Rocky islets, reefs and Year-round residents. along over 80% of British Byers, McKenny and spits, lagoons, gravel/mud Columbia’s shoreline Whitmore Islands flats, rocky beaches, sand (National IBA). bars and inlets. 30–35% of global Outer coast, cobble population breeds beaches. coastally in British Intimately tied to rocky Columbia. intertidal. Area around Princess Royal Island and WVI are Global IBA. Greater and Lesser Coastal OWA. Princess Royal Island Tidal mudflats in protected Spring and fall Yellowlegs bays and estuaries, edges migrant. of tidal channels, sandy beaches and spits; prefers areas with shallow waters over mud. Roosts in offshore rocks, reefs, rocky beaches.

April 2010 FINAL Page B-15

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Wandering Tattler Exposed outer coastal Surf-washed reefs and Spring and fall OWA. rocks of outer coast. migrant. Rare in protected waters. Exposed outer coast, cobble beaches, intertidal. Whimbrel Mostly coastal OWA. Offshore islets and rocks, Spring and fall mudflats, wind-swept migrant. sandy beaches and spits. Single birds, outer coast, cobble beaches, rocks. Ruddy Turnstone Mostly coastal OWA. Rocky shores, offshore Spring and fall islets and reefs or exposed migrant. peninsula. Occasionally pebble/sand beaches and mudflats. Outer coast, cobble beaches.

Page B-16 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Black Turnstone Mostly coastal OWA Emsley Cove Reefs, rocky beaches, Winter. during winter. jetties at mouths of rivers OWA has highest or along lagoons, adjacent abundance in North mud flats, wet sandy America. beaches, floating kelp National IBA near Prince beds. Rupert and SE Queen Outer coast, cobble Charlotte Sound. beaches flocks of several hundred on migration. Surfbird Mostly coastal OWA. Emsley Cove Rocky shorelines, including Winter. IBA of global significance islands, reefs, beaches, off SWVI. headlands, jetties and breakwaters. Occasionally on sandy beaches, tidal mudflats, log booms. Single birds, outer coast, cobble beaches. Red Knot Mostly coastal OWA. Mudflats, hard-packed Spring and fall (Two Subspecies sandy beaches, offshore migrant. rocks. COSEWIC Listed) Rarely single birds. Outer coast, cobble beaches. Prefers estuaries.

April 2010 FINAL Page B-17

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Sanderling Mostly coastal OWA. Ocean beaches, sandbars, Winter and coastal mudflats. migrant. Sandy beaches only. Western Sandpiper Most abundant shorebird Estuaries, mud/sand Spring and fall in British Columbia. beaches. migrant. Most of the world’s Flocks can exceed population migrates along 100,000 birds. coast in spring and fall. Global IBA on WVI. Spotted Sandpiper Breeds in coastal British Known to nest in Kitimat Shorelines, where small Summer. Columbia. River Estuary, Emsley streams drain into tidal Cove and Princess mud or boulder-strewn Royal Island beaches. Breeds on some coastal grassy beaches. Solitary, Semi- Coastlines throughout Emsley Cove Tidal mudflats and Spring and fall Palmated, OWA. Princess Royal Island estuaries, around tide migrant. Baird's, Least, and pools on rocky coastlines. Pectoral Sandpipers

Page B-18 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Rock Sandpiper Mostly coastal OWA. Offshore rocky islets, rocky Winter. headlands, peninsulas, beaches, occasionally mudflats. Rarely single birds. Outer coast, cobble beaches. Dunlin Throughout coastal OWA. Kitimat River Estuary Tidal mudflats. Spring and fall Roosts on spits, dykes, migrant. beached logs, log booms, Overwinters in breakwaters. southern OWA. Occasionally found on sandy beaches, rocky points. Flocks of several hundred. Short-billed Small breeding population Tidal mudflats, estuaries, Spring and fall Dowitcher in northern QCI. and offshore rocks. migrant. OWA estuaries.

April 2010 FINAL Page B-19

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Shorebirds (cont’d) Long-Billed Mostly coastal OWA. Prefers freshwater, Spring and fall Dowitcher although readily uses salt migrant. water during migration. Forages on tidal mudflats. Roosts on offshore rocks, islands and log booms. Estuaries only. Red-Neckeda Mostly offshore in OWA. Princess Royal Island Entirely pelagic in winter. Spring and fall Phalarope and Red Forage along tidelines and migrant. Phalarope edge of kelp beds. Small numbers in Flocks of several thousand winter. during migration. Possibly groups in wider channels in late summer. Jaegers Pomarine, Long- Mostly offshore. Very Rare in CCAA Highly pelagic offshore Seasonal migrant. Tailed, and Parasitic Long-tailed and Pomarine migrants. Jaeger most pelagic of three Small numbers in species. nearshore bays, coves, Parasitic somewhat more estuaries and surge coastal. narrows Single birds on outer coast cobble beaches.

Page B-20 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Gulls and Terns Bonaparte's Gull Common throughout Bish and Emsley Coves Bays, harbours, lagoons, Spring and fall and Sabine’s Gull OWA. estuaries, areas of tidal migrants. Sabine’s has an IBA of convergence and Bonaparte’s National Significance on upwelling. occasional winter SWVI. Passages and narrows. visitor. Roosts in kelp beds, offshore islets and log booms. Large flocks on migration. Sabine’s further offshore. Ring-Billed and Common throughout All marine habitats, bays, Spring and fall Western Gull OWA. lagoons, estuaries, migrants. beaches. Overwinter on Regular vagrant. Vancouver Island Shelf. Herring, Thayer’s, Throughout OWA. Bish and Emsley Coves Open ocean, beaches, Over winter. and Glaucous Gull bays, fish plants, harbours, lagoons, mudflats, inlets, estuaries, spits. Regular vagrants.

April 2010 FINAL Page B-21

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Gulls and Terns (cont’d) Mew Gull Occasional breeder in Nest in Alcan Settling Bays, estuaries, surge Year-round but OWA. Ponds narrows, beaches, mostly over winter. Common throughout mudflats, harbours. OWA. Open salt-water. Global IBA on SWVI. California Gull Throughout the OWA. Bish and Emsley Coves Beaches, bays, estuaries, Small numbers year SWVI is an IBA of global lagoons, fields, open round but mostly significance. ocean, brackish sloughs spring and fall. and freshwater lakes. Regular vagrant. Glaucous-Winged Only colonial breeding Nest at Coste Island Bays, harbours, estuaries, Year-round. Gull gull species on coast. Rock and at Moore, rivers. National IBA throughout Mckenny and Whitmore Roosts in sheltered waters. OWA, Global on WVI. Ecological Reserve Nests on small offshore islands, treeless, with patches of grass, herbs or shrubs.

Page B-22 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Gulls and Terns (cont’d) Black-Legged Throughout OWA. Rocky islets and Winter. Kittiwake Winters widely at sea and headlands, sandy beaches far offshore. and spits, bays, harbours, surge narrows, estuaries, open waters. Very large flocks of non- breeders on outer coast. Some small groups in channels in spring. Caspian Tern Rare throughout most of Very Rare Outside of Beaches, tidal mudflats, Primarily spring and OWA. the South Coast sheltered bays. fall migration. Roosts on sandbars, mudflats, beaches and rocks. Arctic Tern Offshore OWA during Migrates offshore. Migrant. migration. Pelagic. Alcids Common Murre Entire British Columbia Nests on steep cliffs. Year-round. population breeds on only Forages far offshore, but two sites in summer: also in protected marine Kerouard and Scott waters off straits, inlets, Islands. bays and channels. Rest of year widespread. Wintering birds in Hecate Strait.

April 2010 FINAL Page B-23

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Alcids (cont’d) Thick-Billed Murre British Columbia Not Found in CCAA Nests on steep cliffs. Rare migrant and population breeds Non-breeding population summer visitor along exclusively on Triangle seen mostly in northerly coast. Island in dense colonies. waters in winter. Year-round on Triangle Island. Pigeon Guillemot Nesting colonies mostly Historic Breeding at Nearshore zone of bays, Throughout OWA concentrated around QCI Moore and Byers inlets, rocky shorelines, year-round. (EQCI is Global IBA) and Islands channels, surge narrows, Scott Islands. Breeding at Coste sound, coves and Small widely scattered Island in Kitimat Arm harbours. breeding groups. Marbled Murrelet 20–30% of global Bish and Emsley Coves Bays, inlets, fiords, Year-round. breeding population occur Kitimat Estuary lagoons, harbours and and breed in OWA. coves, as well as exposed Global IBA on WVI and coastal waters. National IBA on NQCI. Shelves at mouths of inlets and shallow banks important for foraging. Ancient Murrelet Globally significant Might Nest in the Moore, Avoids protected coastal Winter throughout population. McKenney and waters unless blown in by OWA. 74% of global breeding Whitmore Ecological storm. Year-round in North population breeds in Reserve Forages in areas of and around QCI. British Columbia, almost upwelling and mixing - entirely on QCI, (entire surge narrows, channels QCI an IBA of global and areas with strong significance). eddies and tidal streams. Young reared at sea.

Page B-24 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Alcids (cont’d) Cassin's Auklet Globally significant Historic Breeding at Rarely inshore. Year round. population. Moore and Byers Deep water of Hecate 80% of global breeding Islands Strait. population breeds in Nesting at Glide Island Forms large flocks well British Columbia. and Byers, Conroy and offshore. Global IBA on Scott Harvey Sinnett Nests widely on remote Islands, and NWVI, all Ecological Reserves offshore islands. around QCI, and Aristazabal Island. Rhinoceros Auklet Globally significant Historic Breeding at Prefers open marine Year round. population. Moore and Byers waters; mouths of bays 56% of global breeding Islands and inlets and outer limits population breeds in of estuaries. British Columbia. Rarely in inlets, fiords and Moore Islands, Prince estuaries. Rupert, Scott Island and Breeds at Moore Islands. QCI IBA. Forages at channel mouths. Tufted Puffin Breed in British Columbia. Historic Breeding at Prefers outer coastal Winters offshore and Roughly 90% of British Moore and Byers waters. in Hecate Strait. Columbia’s tufted puffins Islands Rarely seen in harbours Year-round at occur on Scott Islands and bays. Triangle Island and (Global IBA). Breeding colonies on QCI. Others on SQCI (National southern tip of QCI and IBA). Scott Island.

April 2010 FINAL Page B-25

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-1 Marine Birds Associated with Ecosections in the OWA (cont’d)

Sighting Locations within Species1 Habitat Use in OWA the CCAA Marine Habitat Seasonal Occurrence Alcids (cont’d) Horned Puffin Tiny British Columbia Historic Breeding at Pelagic. Offshore waters in population breeds only on Moore and Byers Breeds colonially on cliffs. winter. Triangle Island and Islands Winters offshore. Year-round at southern QCI. Triangle Island and QCI. Kingfisher Belted Kingfisher Breeds and winters in Marshes, coastal Coastal areas OWA. shorelines, lagoons, throughout OWA. tidepools, estuaries, beaches, sloughs. Nests in cut banks. Single birds in channels and estuaries close to shore. NOTE: 1 The areas assessed for this volume are not the same as those assessed for the environmental and socio-economic assessment (ESA); therefore, the species listed here are not identical to those assessed in the ESA.

Page B-26 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Baleen Whales Blue Whale Not well understood. 2,000–3,000 in Endangered Continental Slope Historically generally offshore, but Eastern North Pacific Dixon Entrance seen occasionally in Hecate Strait. Subarctic Pacific Use area for foraging and migration. Transitional Pacific Fin Whale Not well understood. 5,700 in Threatened Continental Slope Historically present in many areas. Eastern North Pacific Dixon Entrance Likely use area for foraging and Transitional Pacific migration. Queen Charlotte Sound Sei Whale Not well understood. 35 in Endangered Continental Slope Historically generally offshore but Eastern North Pacific Dixon Entrance occasionally in Hecate Strait and Transitional Pacific Queen Charlotte Sound. Minke Whale Not well understood. Unknown but Not at Risk Dixon Entrance Likely year round resident. Not Abundant; North Coast Fjords Most frequently found in nearshore <600 in California, Queen Charlotte Strait waters. Washington, and Oregon Humpback Frequently seen. 1,500 in British Threatened Continental Slope Whale Use area for foraging and migration. Columbia; Dixon Entrance 6,500 in Eastern Hecate Strait North Pacific Queen Charlotte Sound Vancouver Island Shelf North Coast Fjords Queen Charlotte Strait

April 2010 FINAL Page B-27

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Baleen Whales (cont’d) Grey Whale Frequently seen. 18,800 in Eastern Special Concern Continental Slope Use area for foraging and migration, North Pacific Vancouver Island Shelf especially from April through June Hecate Strait and December through January. Queen Charlotte Strait Some whales remain resident throughout summer to forage. North Pacific Very little is known. <300 in Eastern Endangered Nearshore Waters Right Whale Historically present. North Pacific Now extremely rare due to over- exploitation. Might return to area if North Pacific population recovers in other areas. Toothed Whales Sperm Whale Not well understood. Unknown but Not at Risk Continental Slope Generally offshore. 885 in California, Transitional Pacific Males move further inshore in Washington, and summer to feed. Oregon Calving might occur offshore. Killer Whale - Found year round. 205 Threatened Dixon Entrance Northern Forage and breed in the area. North Coast Fjords Resident Queen Charlotte Strait Queen Charlotte Sound Killer Whale - Occasionally in area, especially NW 85 Endangered Vancouver Island Shelf Southern Vancouver Island. Resident Likely for foraging.

Page B-28 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Toothed Whales (cont’d) Killer Whale - Found year round. 220 Threatened Transient Forage and breed throughout area. Killer Whale - Not well understood. 250 Identified to Date Special Concern Continental Slope Offshore Likely use the area for foraging and Transitional Pacific breeding. Subarctic Pacific Pacific White- Not well understood. Unknown but Not at Risk Continental Slope Sided Dolphin Likely year round residents. 26,860 in Central Use entire area for foraging and North Pacific; breeding. 39,822 in California, Washington, and Oregon Dall’s Not well understood. Unknown but Not at Risk Dixon Entrance Porpoise Year round residents. 83,400 in Alaska; Use area for foraging and calving. 75,915 in California, Washington, and Oregon Harbour Not well understood. Unknown but Special Concern Dixon Entrance Porpoise Year round residents. 28,967 in Washington Most commonly found in shallow and Oregon; (<200 m) nearshore areas. 37,450 in Southeast Alaska Striped Rare visitor to area. Unknown but Not at Risk Transitional Pacific Dolphin 9,165 in California, Subarctic Pacific Washington, and Continental Slope Oregon Vancouver Island Shelf

April 2010 FINAL Page B-29

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Toothed Whales (cont’d) Common Rare visitor to area. Unknown but Not at Risk Transitional Pacific Dolphin Only been sighted in OWA on a few 25,000 in California, Subarctic Pacific occasions. Washington, and Continental Slope Oregon Vancouver Island Shelf Risso’s Uncommon visitor to area. Unknown but Not at Risk Transitional Pacific Dolphin Recently seen in western Hecate 12,748 in California, Subarctic Pacific Strait with some regularity. Washington, Oregon Continental Slope Hecate Strait Northern Right Uncommon. Unknown but Not at Risk Transitional Pacific Whale Dolphin Regularly found in large groups in 16,417 in California, Subarctic Pacific offshore and continental slope Washington, and Continental Slope waters often with Pacific white-sided Oregon dolphins. Short-Finned Rare visitor to area. Unknown but Not at Risk Transitional Pacific Pilot Whale 149 in California, Subarctic Pacific Washington, and Continental Slope Oregon Vancouver Island Shelf False Killer Rare visitor to area. Unknown Not at Risk Transitional Pacific Whale Subarctic Pacific Continental Slope Vancouver Island Shelf

Page B-30 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Toothed Whales (cont’d) Baird’s Uncommon. Unknown but Not at Risk Transitional Pacific Beaked Whale Generally found in waters >1,000 m 152 in California, Subarctic Pacific deep. Washington, Oregon Continental Slope Recent sightings along Continental Vancouver Island Shelf slope. Stejneger’s Likely rare. Unknown Not at Risk Transitional Pacific Beaked Whale Subarctic Pacific Continental Slope Vancouver Island Shelf Hubbs’ Likely rare. Unknown Not at Risk Transitional Pacific Beaked Whale Known only from stranding records Subarctic Pacific near northern Vancouver Island. Continental Slope Vancouver Island Shelf Cuvier’s Likely rare. Unknown but Not at Risk Transitional Pacific Beaked Whale Have been seen in Hecate Strait in 1,121 in California, Subarctic Pacific recent years. Washington, and Continental Slope Oregon Hecate Strait Vancouver Island Shelf Pygmy Sperm Rare in area. Not at Risk Transitional Pacific Whale Subarctic Pacific Continental Slope Vancouver Island Shelf Dwarf Sperm Rare in area. Unknown, but Data Deficient Transitional Pacific Whale estimates 11,000 in Subarctic Pacific Eastern Pacific Continental Slope Vancouver Island Shelf

April 2010 FINAL Page B-31

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Pinnipeds Steller Sea All three breeding rookeries in 18,400–19,700 in Special Concern Dixon Entrance Lion Canada are within OWA. British Columbia Continental Slope Hecate Same areas are important haul-out Strait sites and foraging grounds for Queen Charlotte Sound population. Vancouver Island Shelf North Coast Fjords California Sea Males and sub-adults migrate north 1,500–3,000 in British Not at Risk Vancouver Island Shelf Lion from California in winter. Columbia Small numbers occur in OWA from September through May. Harbour Seal Year-round resident. 108,000 in British Not at Risk Coastal Area Ubiquitous throughout British Columbia Haul-Out Sites Columbia. Widely distributed in OWA. Northern Not well understood. 101,000 in North Pacific Not at Risk Continental Slope Elephant Seal Seals disperse widely to continental shelf and slope waters to forage. Winter breeding rookeries and moulting sites in Mexico and California.

Page B-32 FINAL April 2010

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Northern Fur Pelagic. 888,000 in North Pacific Threatened Dixon Entrance Seal Winter off Continental Shelf and Hecate Strait shelf break. North Coast Fjords In spring migrate north along the Queen Charlotte Sound west side of Haida Gwaii and farther Continental Slope offshore. Vancouver Island Shelf Some overwinter in Hecate Strait and up inlets. Table B-2 Marine Mammals and Turtles with Ecosections in the OWA (cont’d)

Species Name Use of Assessment Area Estimated Population COSEWIC Status Important Areas Other Sea Otter Year round residents of central 500 on Central British Special Concern Coastal Areas of coast. Columbia Coast; Northern and Western Often found in small groups off 2,700 off west coast of Vancouver Island northern Vancouver Island. Vancouver Island Coastal Areas of Central Use area for feeding and breeding. Mainland Leatherback Rare visitor July through September. Unknown Endangered Continental Slope Turtle Likely feed in area. Vancouver Island Shelf Sightings have increased in recent Transitional Pacific years. Coastal area of Haida Gwaii Green Sea Rare. Unknown Not Assessed East Side of Graham Turtle Potentially accidental visitor from Island September through December. Vancouver Island Shelf SOURCE: Modified from Heise et al. 2007; additional sources of information include COSEWIC Status Reports, the BCCSN, and Expert Knowledge.

April 2010 FINAL Page B-33

Northern Gateway Pipelines Inc. TERMPOL STUDY NO. 3.15: General Risk Analysis and Intended Methods of Reducing Risk Appendix B: General Marine Mammmal and Marine Bird Species Lists for the Open Water Area

Page B-34 FINAL April 2010