Study No. 155 January 2016

CANADIAN AN ASSESSMENT OF HYDROELECTRIC OPTIONS TO SATISFY OIL SANDS RESEARCH LECTRICITY EMAND INSTITUTE E D

Canadian Energy Research Institute | Relevant • Independent • Objective

AN ASSESSMENT OF HYDROELECTRIC POWER OPTIONS TO SATISFY OIL SANDS ELECTRICITY DEMAND

An Assessment of Hydroelectric Power Options to Satisfy Oil Sands Electricity Demand

Author: Ganesh Doluweera

ISBN 1-927037-39-3

Copyright © Canadian Energy Research Institute, 2016 Sections of this study may be reproduced in magazines and newspapers with acknowledgement to the Canadian Energy Research Institute

January 2016

Printed in Front Photo Courtesy of istockphoto.com

Acknowledgements: The author of this report would like to extend his thanks and sincere gratitude to all CERI staff that provided insightful comments and essential data inputs required for the completion of this report, as well as those involved in the production, reviewing, and editing of the material, including but not limited to Allan Fogwill, Dinara Millington and Megan Murphy. The author would also like to personally thank Dr. David Layzell of the University of Calgary for providing helpful insights for this study.

ABOUT THE CANADIAN ENERGY RESEARCH INSTITUTE The Canadian Energy Research Institute is an independent, not-for-profit research establishment created through a partnership of industry, academia, and government in 1975. Our mission is to provide relevant, independent, objective economic research in energy and environmental issues to benefit business, government, academia and the public. We strive to build bridges between scholarship and policy, combining the insights of scientific research, economic analysis, and practical experience.

For more information about CERI, visit www.ceri.ca

CANADIAN ENERGY RESEARCH INSTITUTE 150, 3512 – 33 Street NW Calgary, Alberta T2L 2A6 Email: [email protected] Phone: 403-282-1231

January 2016 An Assessment of Hydroelectric Power Options to Satisfy iii Oil Sands Electricity Demand Table of Contents

LIST OF FIGURES ...... v LIST OF TABLES ...... vii EXECUTIVE SUMMARY ...... ix CHAPTER 1 INTRODUCTION ...... 1 Oil Sands Electricity Demand and the Opportunity to Reduce GHG Emissions ...... 3 Hydroelectric Power Generation Options ...... 7 Options to Move Hydropower ...... 11 Scope and Objectives ...... 13 CHAPTER 2 METHODOLOGY ...... 15 Hydropower Generation Options ...... 15 Hydropower Generation Options in Alberta ...... 16 Hydropower Generation Options in ...... 17 Hydropower Generation Options in Manitoba ...... 18 Transmission Options...... 19 Selection of Transmission Line Corridors ...... 20 Levelized Cost of Delivered Electricity and GHG Emissions Abatement Cost ...... 20 Uncertainty Assessment ...... 22 Estimation of Environmental and Social Impacts ...... 23 Residential and Property Value Impacts ...... 24 Agricultural Impacts ...... 24 Impacts on Indigenous Populations ...... 24 Environmental Impacts ...... 25 CHAPTER 3 RESULTS ...... 27 Hydropower Generation and Transmission Options ...... 27 Alberta-BC Options ...... 27 Alberta-Manitoba Option ...... 28 Alberta Slave River Options ...... 29 Levelized Cost of Electricity and Cost of Avoided CO2 ...... 30 Sensitivity Analysis against Discount Rate ...... 37 Economic Assessment of Electrical Extraction ...... 39 Social and Environmental Impacts of Hydropower Options ...... 41 Employment and Other Economic Development Benefits ...... 48

January 2016 iv Canadian Energy Research Institute

CHAPTER 4 DISCUSSION AND CONCLUDING REMARKS ...... 49 Implications of the Alberta Electricity Market Structure on Hydropower Project Financing ...... 49 Alberta-British Columbia Hydropower Options ...... 50 Alberta Slave River Hydropower Options ...... 50 Alberta-Manitoba Hydropower Options...... 51 Electrical Extraction Technologies as a Carbon Management Option ...... 52 Long-term Planning ...... 52 APPENDIX A SOCIAL AND ENVIRONMENTAL IMPACTS WITHIN THE STUDY AREA ...... 55 APPENDIX B CALCULATION OF ENERGY COST AND EMISSIONS OF BITUMEN EXTRACTION THROUGH SAGD...... 59

APPENDIX C UNCERTAINTY ANALYSIS OF LCOE AND CACO2 ...... 63 BIBLIOGRAPHY ...... 67

January 2016 An Assessment of Hydroelectric Power Options to Satisfy v Oil Sands Electricity Demand List of Figures

E.1 Average Cost of Delivered Electricity of Different Generation and Transmission Options...... xi E.2 Cost of Avoided GHG Emissions ...... xi E.3 Land Cover within the Direct Impact Area of the Hydropower Generation and Transmission Options...... xiii 1.1 Historic and Forecasted Bitumen Extraction and Upgrading Capacity in Alberta by Operation ...... 2 1.2 Historic and Forecasted Electricity Demand in Alberta by Consumption Sector ...... 3 1.3 Electricity Intensity of Oil Sands Operations by Type of Operation ...... 5 1.4 GHG Emissions Intensity of Average Electricity Supply Mix in Alberta and the Oil Sands Sector ...... 6 1.5 GHG Emissions Intensity of the Oil Sands Industry and Individual Operations ...... 7 1.6 Electric Power Demand of Oil Sands Operations by Type of Operation ...... 8 1.7 Currently Installed Hydropower Generation Capacity and Technical Potential to Develop New Capacity in Canadian Provinces and Territories ...... 9 2.1 Overview of Hydropower Generation and Transmission Options ...... 15 3.1 Alberta-BC Hydropower Generation and Transmission Options...... 28 3.2 Alberta-Manitoba Hydropower Generation and Transmission Option ...... 29 3.3 Alberta Slave River Hydropower Generation and Transmission Options ...... 30 3.4 Levelized Cost of Electricity Delivered to Oil Sands Operations in Alberta under Different Generation and Transmission Options ...... 32 3.5 Cost of Avoided GHG Emissions ...... 33 3.6 Capital Cost Contribution of Different Transmission System Components ...... 34 3.7 Electricity Supply Curve ...... 35 3.8 GHG Emissions Abatement Supply Curve ...... 36 3.9 Electricity Supply Curve Sensitivity Analysis against Discount Rate ...... 38 3.10 GHG Emissions Abatement Supply Curve Sensitivity Analysis against Discount Rate ...... 38 3.11 Sensitivity Analysis of CACO2 for Electrical Extraction Technologies ...... 41 3.12 Land Cover within the Direct Impact Area...... 42 3.13 Population and Number of Dwellings in the Direct Impact Area ...... 43 3.14 Number and Area of Reserve Lands in the Direct Impact Area ...... 44 3.15 Protected Areas and Water Bodies in the Direct Impact Area ...... 45 3.16 Woodland Caribou Population in the Study Area ...... 47 A.1 Land Cover of Respective Study Areas ...... 55 A.2 Residential and Property Value Impacts ...... 56 A.3 Impacts on Indigenous Populations ...... 56 A.4 Protected Areas and Water Bodies within the Respective Study Areas ...... 57 B.1 Energy System of the SAGD Facility ...... 59

January 2016 vi Canadian Energy Research Institute

C.1 Monte Carlo Simulation Results: Distributions of LCOE ...... 64 C.2 Monte Carlo Simulation Results: Cumulative Distributions of LCOE ...... 64 C.3 Monte Carlo Simulation Results: Cumulative Distributions of CACO2 ...... 65

January 2016 An Assessment of Hydroelectric Power Options to Satisfy vii Oil Sands Electricity Demand List of Tables

E.1 Summary of Hydropower Generation and Transmission Options ...... x E.2 LCOE and CACO2 Estimates Compared to Different Reference Cases ...... xiii 2.1 Assumptions Made in Calculating Capital Charge Factor ...... 21 3.1 Summary of Hydropower Generation and Transmission Options ...... 27 3.2 LCOE and CACO2 Estimates ...... 31 3.3 Minimum CACO2 and Other Metrics of Electrical Extraction Scenarios ...... 40 B.1 Main Parameters and Results of the Illustrative Case Example ...... 61 C.1 Selected Metrics of Uncertainty Assessment Results ...... 65

January 2016 viii Canadian Energy Research Institute

January 2016 An Assessment of Hydroelectric Power Options to Satisfy ix Oil Sands Electricity Demand Executive Summary

The oil sands sector in Alberta is an important player in the global petroleum supply chain and a major contributor to the provincial and Canadian economy. Extracting and upgrading bitumen is an energy-intensive process where large amounts of thermal energy and electricity are utilized. The energy-intensity of its operations, in addition to heightening the marginal cost of production, have made the oil sands sector a dominant greenhouse gas (GHG) emitter in Alberta. With growing concerns about climate change, GHG-intensive operations have created a challenging environment for the oil sands sector. Consequently, oil sands operators and the provincial government are exploring options to reduce GHG emissions. Decarbonizing the electricity consumed by oil sands operations is one option to reduce GHG emissions.

Hydropower is a proven option to deliver a reliable supply of low carbon electricity. It is a major source of in Canada. A high potential to develop new hydropower plants is available within Alberta and neighbouring jurisdictions. However, development of new hydropower plants requires long distance transmission lines to connect the oil sands region to sites with high hydropower generation potential. Furthermore, development of hydropower plants and transmission lines can potentially have land use impacts with greater environmental and social implications. This study identifies six options to generate and transmit hydropower to the oil sands region in Alberta and provides a multi-attribute evaluation of those options. This study also provides comprehensive economic assessments and high-level land use impact evaluations. The reference electricity generation option used in this analysis is natural gas-fired cogeneration units.

This study assesses new hydropower development options available in Alberta, British Columbia (BC) and Manitoba. Two long distance electricity transmission technologies – high (HVDC) and high voltage (HVAC) – were assessed as options to transmit hydropower to the oil sands region. The six hydropower generation and transmission options assessed in this report are summarized in Table E.1.

January 2016 x Canadian Energy Research Institute

Table E.1: Summary of Hydropower Generation and Transmission Options Hydropower Generation Plant Transmission System Option Site and River Province Rated No. of Length System Capacity Lines (km/line) (MW) Site C-DC Site C on the British 1100 ±500 kV HVDC 1 600 Columbia bipole Site C-AC Site C on the British 1100 Single circuit 2 600 Peace River Columbia 500kV HVAC BC British Increase Single circuit Intertiei Columbia by 500 500kV HVAC Slave Alternative 4 Alberta 1100 ±500 kV HVDC 1 400 River-DC site on the bipole Slave River Slave Alternative 4 Alberta 1100 Single circuit 2 400 River-AC site on the 500kV HVAC Slave River Manitoba Conawapa site Manitoba 1485 ±500 kV HVDC 1 1100 DC on the Nelson bipole River iThe BC Intertie option assumes a case where the existing BC-AB intertie is reinforced to import higher amounts of hydropower purchased from the BC Hydro system. Therefore, no new hydropower plants or new transmission lines are attributed to this option. Source: CERI

Figure E.1 depicts the average cost of delivered electricity (taking into account both generation and transmission cost; measured in $/megawatt hour [MWh]) under the six hydropower options. The reference case is also shown.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy xi Oil Sands Electricity Demand

Figure E.1: Average Cost of Delivered Electricity of Different Generation and Transmission Options

Note: All costs are in 2014 Canadian dollars. Source: CERI

Figure E.2 shows the GHG emissions abatement cost (measured in $/tonne of carbon dioxide equivalent [tCO2e]) of the hydropower options. GHG emissions abatement costs are calculated in comparison to the cogeneration reference case.

Figure E.2: Cost of Avoided GHG Emissions (CACO2)

Source: CERI

January 2016 xii Canadian Energy Research Institute

Each of these six hydropower options can deliver sufficient electricity to satisfy the demand of in-situ bitumen extraction operations with production capacity of 0.5 million bbl/day to 1.1 million bbl/day. The average cost of delivered electricity is in the range of $81-$162/MWh.

In contrast, natural gas-fired cogeneration would cost about $57/MWh. Hence, without a price on GHG emissions, the likelihood of hydropower options reducing the marginal cost of oil sands operations is low. As a carbon emissions mitigation option, utilizing hydropower can potentially reduce the GHG emissions of oil sands operations by 13-16 percent at a cost of $75-$332/tCO2e.

The lowest average cost of delivered electricity and GHG emissions abatement cost results from purchasing hydropower from the BC Hydro system and delivering it by utilizing the existing transmission intertie between Alberta and BC (BC Intertie option). This option requires implementation of mitigation measures to enable the full capacity utilization of the Alberta-BC intertie. The BC Intertie option also has the advantage of being able to deliver low GHG-intensive electricity in the near term (within 2-5 years). Furthermore, as the BC Intertie option would utilize existing electricity infrastructure, it would lead to zero to minimal new environmental and social impacts.

Compared to the other new hydropower options assessed in this study, the Manitoba DC option has a number of advantages. The Manitoba DC option has the lower average cost of delivered electricity compared to the two new BC hydropower options. The average cost of delivered electricity is very close to the Slave River-DC option and lower than the Slave River-AC option. The Conawapa hydropower project, which is the generation option pertaining to Manitoba DC, is in the advanced planning stage and Manitoba Hydro has already completed feasibility assessments.

Hydropower generation and transmission options assessed in this study have the ability to reduce GHG emissions of oil sands operations by decarbonizing the electricity consumed for bitumen extraction and upgrading. However, GHG emissions from bitumen extraction and upgrading are dominated by the emissions associated with the thermal energy portion. Therefore, with a larger supply of hydroelectric power, it is possible to achieve deeper emissions in the oil sands sector by deploying electrical extraction technologies for in-situ recovery of bitumen.

The reference case used in this analysis is onsite cogeneration. Adding onsite cogeneration to an oil sands operation requires additional investments and increases operational complexity. Therefore, it is also plausible that the oil sands operators may choose to purchase electricity from the Alberta electricity market, instead of onsite generation. In that situation, the reference case could be the grid average cost and emissions of an average price of $66/MWh and higher emissions of 710 KgCO2/MWh. The reference case could also be the highest emissions source – coal-fired generation – using a coal-based electricity price of $83/MWh and emissions of 820 KgCO2/MWh. This latter option is being phased out in Alberta but still forms a plausible reference case to compare displacement costs. If these six options were considered, the carbon emissions abatement cost changes as shown in Table E.2.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy xiii Oil Sands Electricity Demand

Table E.2: LCOE and CACO2 ($/tCO2e) Estimates Compared to Different Reference Cases HVDC Options HVAC Options Intertie Option Slave Site C-DC Manitoba DC Site C-AC Slave River-AC BC Intertie River-DC LCOE 141 110 124 162 121 81 CACO2 – Cogen 266 165 207 332 198 75 CACO2 – Grid Avg. 107 62 82 137 77 21 CACO2 – Coal 72 33 50 98 47 -3 Source: CERI

Abatement costs are sensitive to the reference case. For the situation replacing new coal-fired generation with the BC Intertie, the abatement cost is negative because the cost of BC-intertie electricity is less than coal-based generation.

Figure E.3 depicts the land cover within the direct impact area of the hydropower generation and transmission options. In this study, the direct impact area is defined as the area formed by a combination of a 1 km wide buffer that encloses the selected transmission line corridor and a circular buffer with a 10 km radius that encloses the hydropower plant.

Figure E.3: Land Cover within the Direct Impact Area of the Hydropower Generation and Transmission Options

Source: CERI

As depicted in Figure E.3, due to the greater transmission distance, the Alberta-Manitoba option (Manitoba-DC) has the highest land use impact while the shorter transmission distance makes the Alberta hydropower options (Slave River-DC/AC) the ones with the lowest land use impacts.

January 2016 xiv Canadian Energy Research Institute

However, careful assessment of the land cover reveals some interesting findings. Of the new hydropower options, the Alberta-BC options (Site C-DC/AC) have the highest agricultural and residential (in terms of populated areas within the direct impact area) impacts. Despite the longer transmission distance, the Alberta-Manitoba option has the lowest residential impacts. In all cases, the majority of the populated areas that would be impacted by new hydropower options are within Alberta. Alberta Slave River options would likely have the highest amount of environmental impacts in terms of the environmentally sensitive areas within the direct impact area. Environmental impacts could be exacerbated by potential impacts on the Wood Buffalo National Park and the Peace-Athabasca Delta, a wetland ecosystem with global significance. Moreover, these two options would have the highest impacts on aboriginal populations in terms of the number of First Nations reserves within the direct impact area.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 1 Oil Sands Electricity Demand Chapter 1 Introduction

Oil sands deposits in northern Alberta is a nationally important hydrocarbon reserve. At a total reserve estimate of 173 billion barrels of oil equivalent (BOE) at the end of 2014, it is the third largest proven hydrocarbon reserve in the world, behind Venezuela (298 BOE) and Saudi Arabia (267 BOE) (BP, 2015). In 2014, total capital investments in the oil sands industry amounted to CDN$24.3 billion, which was 26 percent of total capital investments in Alberta and 7 percent that of Canada (Alberta Government, 2015; CAPP, 2015b). The enormous amount of capital investments as well as technological developments has led to steady growth in bitumen extraction from oil sands and upgrading it into synthetic crude oil.

According to the Alberta Energy Regulator (AER), at the end of 2014, production capacity of oil sands crude1 was 2.2 million barrels per day (bbl/d) (AER, 2015). Oil sands crude production capacity is expected to rise to 5.8 million bbl/d by 2030 (Murillo, 2015). The capital investments and operations of the oil sands sector contributes to economic growth in Canada, creates jobs, and drives the growth in other sectors of the economy such as manufacturing, transportation, financial, professional services, etc. (CERI, 2011).

The main hydrocarbon product that is being extracted from the oil sands is bitumen, a heavy, highly viscous form of petroleum that does not flow at normal reservoir conditions (Charpentier et al., 2011). Consequently, unconventional and advanced extraction techniques are being applied to recover and extract bitumen from oil sands deposits and upgrade bitumen into synthetic crude oil (SCO).2 Bitumen is extracted from the oil sands deposits through surface mining or through in-situ extraction techniques (Murillo, 2015). The two main in-situ techniques currently being employed are steam assisted gravity drainage (SAGD) and cyclic steam stimulation (CSS).

Figure 1.1 depicts a long-term forecast of bitumen production and upgrading levels in Alberta by type of operation as estimated in a recent CERI study (Murillo, 2015; Business as Usual scenario). Recent economic conditions, particularly lower oil prices, may potentially lead to a lower production growth rate than the ones depicted in Figure 1.1. For example, a recent outlook by the Canadian Association of Petroleum Producers (CAPP) estimates the total bitumen extraction level in 2030 to be approximately 800,000 bbl lower than that of the level depicted in Figure 1.1 (CAPP, 2015a). The production level, nonetheless, is forecasted to grow steadily from current levels.

Due to intense processing requirements, bitumen extraction and upgrading are energy-intensive operations (Bergerson et al., 2012; Englander et al., 2015; Nduagu & Gates, 2015). In general,

1 This includes both upgraded and non-upgraded bitumen 2 As of 2014, approximately 31% of bitumen is upgraded into synthetic crude oil (AER, 2015).

January 2016 2 Canadian Energy Research Institute

energy demands for oil sands projects can be divided into three categories: thermal energy demand, which is primarily satisfied through combustion of natural gas; electricity demand; and demand for transportation (such as diesel ) (Murillo, 2015). Natural gas is the primary feedstock to produce hydrogen (H2) used for bitumen upgrading.

Figure 1.1: Historic and Forecasted Bitumen Extraction and Upgrading Capacity in Alberta by Operation

Source: Business as Usual (BAU) scenario, CERI Study 151 (Murillo, 2015)

With the rapid growth of oil sands operations, the oil sands sector has become a dominant consumer of primary energy in Alberta. For example, in 2014 the oil sands sector accounted for 30 percent of natural gas use, 20 percent of diesel fuel use, and 21 percent of electricity use in Alberta (Murillo, 2015). The fossil fuel dependence of the oil sands sector3 has also led to a significant amount of GHG emissions as well. In 2013, GHG emissions resulting from oil sands operations was 61.4 metric tons of carbon dioxide equivalent (MtCO2e), which was 23 percent of Alberta’s emissions and 8 percent of Canadian emissions (Environment Canada, 2015a).

3 Natural gas for thermal energy; natural gas, coal, and petroleum coke for electricity production within the oil sands sector and Alberta grid; and diesel for transportation

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 3 Oil Sands Electricity Demand

With growing worldwide concerns about GHG emissions, the oil sands sector has been put under scrutiny. Current and future provincial (Government of Alberta, 2015) and national climate change mitigation policies as well as policies or regulations in other jurisdictions, for example, California’s Low Carbon Fuel Standard (CARB, 2015), can potentially influence oil demand.

Mounting concerns about the sector’s GHG emissions, and other associated environmental (Gosselin et al., 2010) and public health impacts (Kelly et al., 2009; McLachlan, 2014) has created a challenging environment for the oil sands sector both locally (ACFN, 2015; Droitsch & Simieritsch, 2010) as well as internationally (McCarthy, 2014). Consequently, options are being explored to reduce greenhouse gas emissions and other environmental impacts through fuel switching, energy efficiency and process improvements.

Oil Sands Electricity Demand and the Opportunity to Reduce GHG Emissions Extracting and upgrading bitumen requires large amounts of electricity. As such, oil sands operations require a reliable supply of electricity. According to the Alberta Electric System Operator (AESO) the total share of electricity consumed by the oil sands sector is expected to rise to 30 percent of total provincial electricity demand by 2022 (see Figure 1.2) (AESO, 2014a). Consequently, decarbonizing the electricity consumed by oil sands operations is one option to reduce GHG emissions.

Figure 1.2: Historic and Forecasted Electricity Demand in Alberta by Consumption Sector

Source: Data from AESO (2014); figure by CERI

January 2016 4 Canadian Energy Research Institute

Advancements in electricity generation, transmission, distribution, and storage have provided many technically and commercially feasible options to produce and deliver low carbon electricity (IEA, 2014). These options include generating electricity using low or zero carbon-intensive technologies such as renewable power sources, , and use of carbon capture and storage (CCS) in fossil fuel-fired power plants, reducing transmission and distribution losses, and the use of storage systems for optimal electricity supply and demand matching.

Electricity is a very versatile form of energy that can provide virtually any end use energy service. Unlike many other energy carriers, conversion of electricity into end use energy services can be done at very high efficiencies. Furthermore, the marginal cost of reducing carbon emissions in the electric power sector is reported to be lower than other sectors such as the industrial and transportation sectors (Apt, Keith, & Morgan, 2007). All of these factors have made widespread electrification of energy end use services a driving across the global energy system (IEA, 2014).

Hydropower is a proven option to deliver a reliable supply of low carbon electricity. It is a major source of electricity generation in Canada. More than 60 percent of current electricity supply in Canada is produced using hydropower plants (Statistics Canada, 2015). The total installed hydropower capacity in Canada amounts to 76 GW (approximately 58 percent of total Canadian power generation capacity). In addition to providing a clean and reliable supply of electricity, the hydropower industry is a significant contributor to the Canadian economy. In 2013, the hydropower industry contributed CDN$37 billion to Canada’s gross domestic product (GDP) and supported 135,000 full equivalent (FTE) jobs4 (PRISM Economics and Analysis, 2015).

Decarbonizing electricity supply by utilizing hydroelectric energy can potentially reduce the overall GHG emissions from oil sands operations. Figure 1.3 depicts the electricity demand of different oil sands operations (measured in kWh per barrel of output, kWh/bbl). Currently, these electricity demands are being satisfied by a mixture of electricity sources. A vast majority of bitumen and SCO are being produced by projects that have onsite electricity generation. The main source of onsite generation is natural gas-fired cogeneration, where high fuel efficiencies are achieved by co-producing electricity and thermal energy (Doluweera et al., 2011). While natural gas is the primary fuel, some projects use petroleum coke as a fuel for cogeneration.

4 As estimated by PRISM Economic Analysis (2015), job contribution includes direct, indirect, and induced effects.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 5 Oil Sands Electricity Demand

Figure 1.3: Electricity Intensity of Oil Sands Operations by Type of Operation

Source: CERI Study 151 (Murillo, 2015); Note the logarithmic scale in the y-axis

Projects that do not have cogeneration, purchase electricity from the Alberta grid. A large fraction of electricity in Alberta is produced using coal-fired generating units5 resulting in a higher average GHG emissions intensity. However, the Alberta provincial government has recently announced coal plant retirements so it is uncertain if any existing plants will be operating post- 2030 (AESO, 2014a; Murillo, 2015).

Figure 1.4 depicts the average GHG emissions intensity of the average Alberta grid mix and that of the electricity mix of the oil sands sector.6

5 In 2014, coal-fired generating units produced 65% of electricity production. This excludes behind the fence power generation that served onsite demands (AESO, 2015a). 6 Alberta grid mix intensity is estimated by taking into account the current generation mix, unit retirements, and most likely future additions. Oil sands electricity emissions intensity is estimated by taking into account the onsite generation fuel mix and electricity import volumes from the grid.

January 2016 6 Canadian Energy Research Institute

Figure 1.4: GHG Emissions Intensity of Average Electricity Supply Mix in Alberta and the Oil Sands Sector

Source: CERI Study 151 (Murillo, 2015).

Figure 1.5 shows GHG emissions intensity of the oil sands industry and individual operations. By synthesizing the information presented in Figures 1.3-1.5, it can be seen that the use of low carbon hydropower7 to satisfy electricity demands leads to GHG emissions reductions. For example, in 2025, the use of hydropower can reduce GHG emissions of mining operations by 16 percent and for in-situ operations by 13 percent (for SAGD)8 to 15 percent (for CSS).

7 Although no direct GHG emissions result from hydropower generation, a very low amount of life cycle GHG emissions is associated with hydropower due to construction phase emissions and emissions from lacustrine conditions created by altering riverine systems by hydropower dams (PNNL, 2013). Those indirect leads to an emissions intensity that is in the order of 3-10 kgCO2 eq/MWh. 8 For example, SAGD operations require about 16.5 kWh of electricity per barrel of bitumen. The estimated (Murillo, 2015) GHG emissions intensity of in-situ operations (including both SAGD and CSS) in 2025 is 72 kgCO2 eq/bbl. Oil sands electricity emissions intensity in 2025 is 0.55 kgCO2/kWh. Therefore, use of hydroelectric power results in an emissions reduction of 9 kgCO2/bbl (= 12.5% of 72 kgCO2 eq).

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 7 Oil Sands Electricity Demand

Figure 1.5: GHG Emissions Intensity of the Oil Sands Industry and Individual Operations

Source: CERI Study 151 (Murillo, 2015).

In addition to prevailing bitumen extraction technologies, it is also possible to extract bitumen using electrical extraction technologies. In these cases, electricity is utilized as the primary energy source, excluding the use of steam produced through fuel combustion. To date, two innovative electrical extraction processes have been demonstrated in Alberta. One demonstration project is based on a process called Electro-Thermal Dynamic Stripping Process (E-T DSP™) where electrical current passes between electrodes through a water envelop that acts to heat the bitumen (McGee, 2012).

A second experimental project is based on a process called Effective Solvent Extraction Incorporating Electromagnetic Heating (ESEIEH). The ESEIEH process eliminates the use of water. Bitumen is concurrently heated with electrical energy and further diluted with the injection of a solvent in a gravity drainage recovery process (Patterson, 2015). If a low/zero GHG emissive electricity supply, such as hydropower is available, it is possible to extract bitumen with negligible amounts of GHG emissions.

Hydroelectric Power Generation Options Hydroelectricity is produced by harnessing the energy in flowing water through turbines. There are two types of hydroelectric power plants: 1) plants with a storage reservoir that use a dam to store water and discharge to produce electricity; and 2) run-of-river (ROR) plants that do not use a dam but divert the river to flow it through turbines. In the case of ROR plants, the operators have limited control over the level of electricity production, as the plant relies on the natural flow of the river it has tapped. There are modified run-of-river plants that rely on the flow of water

January 2016 8 Canadian Energy Research Institute

to produce electricity and store water using a smaller reservoir to gain more control over the level of production.9

Hydroelectric plants generally have a long operational life that is in the order of 60 to 100 years. Therefore, once developed, a hydroelectric plant can provide a clean reliable supply of electricity for a long period.

As depicted in Figure 1.6, within 10 years, the electric power demand of oil sands operations could rise to 3,500 MW, which is approximately 1,400 MW beyond current demand. This only includes the direct electricity requirements of oil sands operations and does not take into account the electricity demand growth due to higher induced economic activities. Even under the lower oil sands growth as forecasted by CAPP, the direct electricity demand of oil sands operations will likely grow over 1,000 MW (Murillo, 2015).

Only about 3 percent of Alberta’s electricity is currently produced using hydropower and the total installed hydropower production capacity is only 890 MW (AESO, 2015a).

Figure 1.6: Electric Power Demand of Oil Sands Operations by Type of Operation

Source: CERI Study 151 (Murillo, 2015).

9 For example, maintaining reservoir will allow the operator to produce more electricity than the amount allowed by the water flow. This controllability can potentially be a valuable attribute of the plant when satisfying time varying electricity demand.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 9 Oil Sands Electricity Demand

Figure 1.7 depicts the currently developed hydroelectric capacity and technical potential to develop new capacity (Irving, 2013). As can be seen from the figure, there is significant hydropower potential available in Alberta and neighbouring jurisdictions. As estimated by the Canadian Hydropower Association, new hydropower development potential in Alberta is about 11,800 MW (Irving, 2013).

Prince Edward Island has a technical potential of 3 MW, which is not shown in the figure. The values indicated in the figure correspond to hydropower production capacity (measured in MW). The energy production capacity (measured in MWh in a given time period) depends on river flow rates and precipitation levels in the period of interest.

Figure 1.7: Currently Installed Hydropower Generation Capacity and Technical Potential to Develop New Capacity in Canadian Provinces and Territories

Source: Canadian Hydropower Association (Irving, 2013).

Another assessment made by HATCH (2010) for the Alberta Utilities Commission (AUC) estimated Alberta’s new hydroelectric energy potential to be 53,000 GWh/year. The HATCH assessment identified 36 sites10 suitable for hydroelectric power development in three northern river basins. The single site on the Slave River, near the border between Alberta and NWT has the highest capacity potential (up to 1,500 MW) and energy potential (8,500 GWh/year). The feasibility to develop a hydropower plant on the Slave River has been investigated by both the Alberta

10 Seventeen on the Athabasca River, eighteen on the Peace River, and one on the Slave River.

January 2016 10 Canadian Energy Research Institute

government and private investors since 1980 (ATCO Power & TransCanada, 2009; Legislative Assembly of Alberta, 2013; SRSC, 1980).

The neighbouring jurisdictions of British Columbia (BC), Saskatchewan (SK), the Northwest Territories (NWT), and Manitoba (MB) have a combined potential of 57,400 MW to develop hydropower plants.

Although a large amount of hydroelectric potential has been reported within Alberta and neighbouring jurisdictions, the feasibility to tap into that potential depends on economic, social, and environmental factors.

Development of hydropower generation capacity is very capital-intensive and the development cost is site sensitive. According to the International Energy Agency (IEA), the capital cost of large (larger than 10 MW) hydropower plants ranges from US$1,750/kW to US$6,250/kW11 (IEA ETSAP, 2010). However, the operational costs are very low (in the order of 1.5 percent to 2.5 percent of the investment cost per year). On a per kW basis, the capital cost of plants with smaller capacities can be several higher than that of larger plants. As such, the cost of energy produced by hydroelectric plants is dominated by its capital cost. Therefore, for a hydroelectric plant to be economically feasible, it must have the ability to operate at a high capacity factor (cf).12

A larger fraction of the capital cost of a hydroelectric power plant has to be spent on items that are disproportionate to the nameplate capacity of the power; for example, on planning, feasibility studies, permitting, environmental impact assessments and access roads. Furthermore, the magnitude of some environmental and social impacts such as impacts on natural ecosystems, fishery, and involuntary relocation of people tend to be disproportionate to the capacity of the plant. Therefore, developing the full capacity of a given site leads to more favourable economics than partial development.

Sites with good hydroelectric potential tend to be in remote areas away from major demand centers, requiring new transmission lines to connect them. Transmission development is also capital-intensive. Therefore, it is important that high capacity factors be maintained to keep transmission costs low. Furthermore, due to the amount of transmission requirements, if sufficient demand is available, developing larger sites with high capacity factors leads to lower overall costs than developing several smaller sites.

In the case of identifying hydroelectric power supply options for oil sands operations, it is important that they have the ability to supply large amounts of economically priced electric energy with high reliability. As such, although higher amounts of hydroelectric potential have

11 Values are in 2008 US$ 12 The capacity factor is the ratio between a power plants actual output in a given time period (usually a year) to its poetical output if the plant was continuously operated at its full nameplate capacity over the same period, i.e., cf=(power generation in MWh)/(nameplate capacity in MW x no. of hours in the period in h)

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 11 Oil Sands Electricity Demand been reported, only a limited number of sites possess the necessary attributes. This is further discussed in Chapter 2.

Transmission Options to Move Hydropower Sites with good hydroelectric potential tend to be in remote locations. For example, sites that are suitable to develop a hydropower plant on the Slave River are approximately 400 km north of the region where oil sands operations are concentrated. Therefore, new long distance transmission lines are required to connect them.

The Alberta electric power system currently has limited transmission interconnections with neighbouring jurisdictions. Therefore, in order to import larger amounts of hydropower into Alberta, new transmission interconnections need to be developed.

Electricity is generated, transmitted, and most of the time utilized in the form of alternating current (AC). The other form is direct current (DC) electricity, where the flow of electric charge is in one direction. In AC systems the (or electric charge) periodically changes direction. The period at which the current changes its direction is known as the system frequency.13 The way electricity is generated and transmitted has led to the dependence on AC electricity. The majority of large-scale power generating plants utilize rotating machines that are driven by a prime mover,14 taking advantage of electromagnetic induction. These rotating machines essentially produce electricity in AC form.

The most important reason for the dominance of AC is the fact that it enables the use of to increase the voltage at which electricity is transmitted and then decreases it back to distribution/utilization .15 Use of higher voltages minimize transmission losses that incur when electric current passes through a conductor.16 As a result, the most prevalent form of electricity transmission is high-voltage alternating current (HVAC) that utilizes voltages higher than 115 kilovolts (kV).

In some situations, however, use of a high voltage direct current transmission system (HVDC) is beneficial and the preferred choice.17 The primary reason for this is that HVDC eliminates the problem of a stability limit of a transmission system.

13 In North America, the AC system frequency is 60 cycles per second or 60 Hertz (Hz). 14 For example, water turbines in the case of hydropower plants, steam or gas turbines in the case of thermal power plants, wind turbines in the case of wind farms, etc. 15 Transformers also rely on electromagnetic induction and only off AC electricity 16 Amount of power transmitted (Pt) through a conductor is equal to the product of the voltage difference (V) at which power is transmitted and the electric current (I). i.e., Pt=VI. Consequently, for a given amount of power, increasing the voltage leads to lower current. Power loss in the form of heat in the transmission wire (Pl) is equal 2 to the product of the resistance of the wire (R) and the square of the current (i.e., Pl=I R). For a given transmission line, the cable resistance is proportional to the length of the conductor. Hence, lowering the current reduces transmission losses. 17 Detailed technical and other relevant descriptions about different transmission technologies can be found in the following reports that are in the public domain: (Simens, 2013; Stantec, 2009).

January 2016 12 Canadian Energy Research Institute

AC current changes its direction at the system frequency. The frequency is directly linked to the rotational of the generators. It is important that all generators in a given electric power system are in sync with each other and producing AC electricity within a very narrow margin around the system frequency. Speeding up or slowing down of a certain generator impacts other generators. Stability of an electric power system then refers to its ability to maintain the synchronism among the generators.

In addition to the resistance, long HVAC transmission lines have a significant reactance.18 Beyond a certain length, the system’s ability to maintain the synchronism among the generators that are at either end of the lines becomes challenging due to the high reactance. This limit is known as the stability limit of an HVAC line, and it can be significantly lower than the thermal limit (determined by the resistance of the line). In the case of HVDC lines, since reactance is irrelevant, the line can be utilized up to the thermal limit of the conductors, enabling more power transmission within a comparably sized transmission corridor.

Transmission loss in HVDC lines and electrical cable cost can be lower than HVAC lines with comparable length and capacity. An HVAC transmission line generally transmits power in three phases and therefore requires three conductor bundles.19 In the case of HVDC, the same amount of power can be transmitted with only one single or two conductor bundles. Reduction in the number of conductor bundles lowers the cost and losses.

Despite these benefits, a few challenges limit the widespread use of HVDC for long distance transmission. As discussed, generation and utilization of electric power is done in AC form. Furthermore, electrical transformers that work only off AC are required to elevate the voltages to levels suitable for transmission. Therefore, a HVDC transmission system requires two additional components to convert AC to DC (this converter is called a rectifier) at the sending end and then back to AC (using an inverter) at the receiving end. These converters add a significant amount to the capital cost of the system.20 Due to the high cost of converters at the sending and receiving ends, intermediate taps to a HVDC line may not be economically feasible, essentially making it a point-to-point transmission option. Furthermore, these additional system components add more losses (in the order of 1 percent) and reduces the reliability.

As the transmission length increases, the cost impact of the converter to the overall project cost is reduced. This makes the HVDC option more economic for longer transmission lines (Gavriovic, 2003; Simens, 2013; Stantec, 2009). For example, at a distance of 600 km, the transmission cost (measured in $/MW-km) of a single circuit 500 kV HVDC line is 19 percent lower than that of a

18 Reactance is proportional to the inductance of the line and the frequency at which power is flown through the conductor. For DC, since there is no change in the direction of current (i.e., frequency is zero), reactance is irrelevant (von Meier, 2006). 19 Generally a bundle of three conductors. 20 In the order of $300-500 million or 25% or more of the total project cost.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 13 Oil Sands Electricity Demand

500 kV HVAC line. In Canada, Manitoba Hydro and Hydro Quebec have significant experience in utilizing HVDC lines to move hydropower.21

Due to the geographic separation between the oil sands region and good sites to develop new hydropower plants, HVDC can potentially be the more suitable transmission option. However, both AC and DC options are considered in this paper.

Scope and Objectives The energy-intensity of oil sands production contributes to the higher than average marginal cost of production and GHG emissions intensity. With the climate change imperative, GHG-intensive activities have drawn criticism, including bitumen production. Using hydroelectric power to meet oil sands electricity demand can potentially provide an option to reduce GHG emissions. Furthermore, despite the high initial cost, the lower operating cost and long operational life of hydropower plants and associated transmission lines can potentially reduce the cost of electricity.

The main objective of this report is to identify hydroelectricity options that would reduce GHG emissions in oil sands production.

Despite the economic and societal benefits provided by large hydroelectric infrastructure projects, there is opposition to these projects due to stakeholder concerns of land use and water impacts and as a result permitting and siting them has become increasingly difficult (Schively, 2007; Vajjhala & Fischbeck, 2007). Project attributes that lead to such difficulties include complex social and environmental implications. In general, public opposition arises from concerns related to unwanted interference with private property rights, environment degradation, and Indigenous peoples traditional uses.

This report, therefore, provides a multi-attribute evaluation of the hydropower and transmission options by quantifying the economic, social, environmental, and land use impacts. Such multi- attribute evaluations provide investors, Indigenous Peoples, non-governmental organizations, regulators, and policy-makers useful insights into costs and benefits of different options. Having a common understanding of the attributes of a project facilitates stakeholder discussions and can contribute to effective decision making.

The specific contributions of this report are as follows:

1. Identify six options to produce and deliver hydroelectric power to satisfy electricity demands of oil sands operations; 2. Provide economic assessments of those six options;

21 In Manitoba, two ±500 HVDC lines known as Bipole I & Bipole II transmit power from hydropower plants on the Nelson River in northern Manitoba to Winnipeg, the main demand center. In Quebec, a ±450 HVDC line (known as Radisson - Nicolet - Des Cantons circuit) moves hydropower from Northern Quebec to southern parts of the province as well as to New England (US).

January 2016 14 Canadian Energy Research Institute

3. Estimate GHG emissions reduction potential and GHG emissions abatement costs of those six options; 4. Provide a preliminary economic assessment and GHG abatement cost of electrical extraction of bitumen with hydroelectric power; and 5. Provide an assessment of land use, social, and environmental impacts of those options using publicly available data sets.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 15 Oil Sands Electricity Demand Chapter 2 Methodology

Hydropower Generation Options Figure 2.1 provides an overview of the generation and transmission options for hydropower generation.

Figure 2.1: Overview of Hydropower Generation and Transmission Options

Source: CERI

As discussed in Chapter 1 (and depicted in Figure 1.7), hydropower potential is available in Alberta and neighbouring jurisdictions (Irving, 2013). Specific information about those sites are available in the public domain (BC Hydro, 2013a; Government of the Northwest Territories, 2011; HATCH, 2010; Manitoba Hydro, 2013; SaskPower, 2010). In order to narrow down the sites to be analyzed in this study, the following factors are taken into account.

 Oil sands operations require large quantities of electricity that is reliably supplied with minimal variation. Therefore, the potential hydroelectric sites need to have both higher production capabilities and capacity factors. This study considered only sites with higher than 500 MW of technical potential that can be operated at an average capacity factor of 50 percent or more.1

1 The equivalent energy supply potential is 2,190 GWh/year or more. Based on (Murillo, 2015), 2,190 GWh of electricity is sufficient to satisfy the electricity demand of a SAGD facility with a bitumen production rate of approximately 360,000 bbl/day.

January 2016 16 Canadian Energy Research Institute

 As new transmission lines need to be constructed to connect potential hydropower plants, this study examines only hydropower potential available in Alberta, British Columbia and Manitoba.  Developing several geographically dispersed hydropower plants with relatively smaller capacities was not considered as an option because that would lead to higher transmission costs and cumulative environmental impacts. Developing a single hydropower plant also enables the use of a single HVDC transmission line, which is essentially a point-to-point transmission option.  Significant hydropower potential is reported to be available in the Northwest Territories (NWT). However, that was not considered for this study due to the unavailability of reliable resource data and potentially longer transmission distances. Furthermore, the hydropower potential available on the NWT side of the Slave River is shared with Alberta.  Saskatchewan is reported to have relatively smaller hydropower potential (SaskPower, 2010). The reported sites are geographically dispersed and have smaller technical potential (250 MW or lower). Therefore, hydropower potential in Saskatchewan is excluded from this study.

Hydropower Generation Options in Alberta The main data source used to assess Alberta’s hydropower potential is an assessment conducted by HATCH (2010) for the Alberta Utilities Commission. The HATCH report provides an assessment of hydropower potential in five main river basins of Alberta. Of the five basins, only the Peace, Athabasca, and Slave River basins appear to have the energy potential that meets the selection criteria. A limited amount of information – such as geological viability of sites, cost estimates – is available for the sites in the Peace and Athabasca River basins.

TransAlta, a major utility company in Alberta, has been allowed to develop a hydropower plant at the Dunvegan site on the Peace River. However, that was excluded as the available capacity is reported to be only 100 MW. Furthermore, new hydropower developments in the BC section of the Peace River – mainly the Site C project – may constrain the capacity available on the Alberta section. A number of sites that are suitable for hydropower plants have been identified in the Athabasca River basin, of which some are closer to Fort McMurray. All have relatively low annual hydroelectric generation potential and were excluded from the study based on the selection criteria.

After taking into account those factors, only the hydropower potential available on the Slave River was selected for this study. The Slave River is formed where the Peace River meets the Riviere des Rochers and the river terminates upon flowing into Great Slave Lake in the NWT. Just south of the Alberta-NWT border, the Slave River enters a series of rapids known as Smith Rapids; a 30 km reach caused by the outcrop of the Canadian Shield. The hydropower potential within these rapids has been recognized for many years. The average flow rate of the Slave River, within the rapids near Fort Fitzgerald, is about 3,440 m3/s, which is by far the highest flow level of any river in Alberta (HATCH, 2010). Annual flow variation of the river is relatively low due to the natural regulation provided by Lake Athabasca and reservoir regulation by the W.A.C. Bennett

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 17 Oil Sands Electricity Demand

Dam in BC. The high and steady flow rates enable development of a hydropower plant with limited or no storage capacity.2

An extensive study of the hydropower potential including exact site identification, geological assessments, detailed cost assessments, and transmission options (SRSC, 1980) was conducted by the Alberta Government in the early 1980s. Upon exploring many options, that assessment identified three sites to develop a hydropower plant: Mountain Rapids site and Alternative 4 Site (A4S) in Alberta and Rapids of the Drowned in the NWT (see Figure 2.1). A4S has been identified as the most suitable site. ATCO Power and TransCanada, two energy companies in Alberta, are currently exploring the potential to develop a hydropower plant, likely at the A4S (ATCO Power & TransCanada, 2009). The project did not proceed to the feasibility assessment stage due to the opposition of the Smith’s Landing First Nations (Bell, 2010). AESO has included transmission lines to connect a hydropower plant on the Slave River in its long term transmission plan (AESO, 2014b), indicating the continued interest in this site.

The hydropower potential at the A4S is estimated to be as high as 1 500 MW. Annual flow rates and energy potential at A4S were obtained from (HATCH, 2010; SRSC, 1980) and also cross validated by using recent flow rate data (Environment Canada, 2015c). Upon assessing this data, the hydropower plant capacity at A4S is set to be 1,100 MW. With observed historic annual flow rates, the average capacity factor is estimated at 68 percent.

Additional hydropower opportunities are available within Alberta in the form of expanding the capacity of existing hydropower plants. This option is likely more economically preferable than building new hydropower plants. However, based on available information (HATCH, 2010), the incremental hydropower potential is lower than 500 MW and therefore did not satisfy the screening criteria used in this study.

Hydropower Generation Options in British Columbia Remaining hydropower resource potential in BC amounts to a staggering 32,000 MW. The vast majority of this capacity, however, is suitable only for ROR hydropower plants with limited dependable capacity (BC Hydro, 2013a; Appendix 8-A).3 Options to develop hydropower plants with high dependable capacity in BC are limited to Site C on the Peace River and adding additional turbines to existing but unused bays of some existing hydropower plants. Two main options in the latter category are the Mica units 5 and 6 and Revelstoke unit 6. Currently BC Hydro is installing Mica units 5 and 6 that have a combined capacity of 1,000 MW. Revelstoke unit 6 is in the active planning stage and has a potential capacity of 500 MW (BC Hydro, 2008).

The Site C dam and the associated generating station is a 1,100 MW hydropower plant that is now under construction. The power plant is estimated to have an annual energy generation

2 i.e., either as a ROR plant or modified ROR plant. 3 According to BC Hydro’s 2013 integrated resource plan, due to limited energy potential, the estimated average cost of generation at a larger number of sites, with a cumulative capacity of approximately 12,000 MW, is more than $200/MWh.

January 2016 18 Canadian Energy Research Institute

potential of 5,100 GWh. Construction of the project started in summer 2015 and is expected to be completed in 2024. It is the third hydropower plant on the BC segment of the Peace River. The proposed site is approximately 7 km southwest of the town of Fort St. John, BC. The length of a direct transmission interconnection to the oil sands region is approximately 600 km.

Two BC hydropower options are assessed in this study:

1) Site C generating station with a direct transmission link to the oil sands region. 2) Buying hydropower from the BC grid and transmitting it using the Alberta-BC interconnection. In this case, the source of hydropower could potentially be BC’s legacy hydropower system and new turbine additions to unused bays of existing hydropower plants. In order to import higher amounts of hydropower from BC, the existing transmission intertie between Alberta and BC needs to be reinforced to utilize its full capacity (AESO, 2015b).

Hydropower Generation Options in Manitoba Manitoba is a major hydropower producer and exporter. Approximately 98 percent of Manitoba’s 5,485 MW power generating capacity is composed of hydropower plants. About 30 percent of electricity produced in Manitoba is exported to other provinces and to the United States, making electricity export revenues an important contributor to the province’s economy.

Remaining hydropower potential in Manitoba is estimated to be 8,800 MW (Irving, 2013). Due to greater transmission developments required to connect to the oil sands region in Alberta (more than 1,000 km), only potential sites in Manitoba with high dependable hydropower capacity are considered for this study.

In an assessment of resource options to develop new power generation capacity in the province, Manitoba Hydro identified 16 new hydropower options (Manitoba Hydro, 2013; Appendix 7.2). Of those options, the proposed Keeyask generating station (695 MW) and Conawapa generating station (1,485 MW) are reported to have the lowest average cost of production. Both sites are on the Nelson River in northern Manitoba and are reported to have the ability to operate at relatively high capacity factors.4 The Manitoba Public Utilities Board has approved the Keeyask generating station for construction. Construction of Keeyask generating is required to satisfy future demand growth of Manitoba and Manitoba Hydro’s electricity export contractual obligations. Consequently, it is uncertain whether capacity of Keeyask would be available to export to Alberta. Therefore, the proposed Conawapa generating station is assessed in this study as the hydropower import option from Manitoba.

The Conawapa power generation project is in the advanced planning stage and therefore reliable resource and cost estimates are available from Manitoba Hydro. Straight-line transmission distance between the proposed site for the Conawapa project and the oil sands region is 1,100 km. It should be noted that this distance is in the same order as the length of HVDC transmission

4 80 percent for Keeyask and 57 percent for Conawapa.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 19 Oil Sands Electricity Demand lines that transmit power from hydropower generating stations on the Nelson River in northern Manitoba to demand centres and export terminals in Winnipeg.5

Transmission Options Transmission options that are assessed in this report are selected taking into account technical constraints (such as transmission losses, reliability, frequency variation, and system stability) and cost. Both HVDC and HVAC transmission options are assessed for Slave River (Alberta) and Site C (BC) hydropower options. For Conawapa (MB), only HVDC transmission options are considered. One main reason for that is Manitoba and Alberta are in different interconnections6 of the North American electric power system. An interconnection is a regional entity where the individual power systems in given interconnections coordinate their operations to ensure system reliability. Any intertie between two separate interconnections must be asynchronous (i.e., must utilize DC links) so that any frequency variation and contingency in one interconnection will not affect the stability of the other. Therefore, HVAC options are not considered for the Conawapa option.

All transmission options are assumed to operate at 500 kV. HVDC options are assumed to operate ±500 kV bipole mode with return. Bipole mode has higher reliability as the HVDC line can be operated at half the capacity even if a contingency occurred in one pole (Gavriovic, 2003; Simens, 2013).

Due to the large amount of power being transferred through the transmission corridor, to increase system reliability, a given HVAC interconnection is assumed to be composed of two single circuit 500 kV HVAC lines.7 This is consistent with transmission options explored by AESO to interconnect a potential hydropower plant on the Slaver River (AESO, 2014b). The amount of power transmitted through an HVAC line can be increased by adding series capacitors (also known as series compensators) to increase the stability limit. However, no series compensation is assumed for HVAC options in this study as sufficient capacity is available between two 500 kV single circuit lines.

All transmission systems are assumed to have static VAR (SVAR) compensators at the receiving end for reactive power compensations.

Capital costs of transmission lines, substations, compensating devices, land, and construction are calculated by utilizing a tool developed by Black & Veatch (2014) for the Western Electricity Coordinating Council (WECC). The tool has been reviewed by transmission line siting professionals from various electric power systems within the Western interconnection. Transmission losses are also estimated using this tool.

5 The two HVDC transmission systems are 895 km and 937 km in length. 6 Alberta is in the Western Interconnection and Manitoba in the Eastern Interconnection 7 Two lines would provide the system operator added reliability as a contingency on a single corridor but would not completely disconnect the hydropower plant from the system

January 2016 20 Canadian Energy Research Institute

Selection of Transmission Line Corridors For all transmission options, Fort McMurray, Alberta has been selected as the receiving end terminus point. Sending ends are the respective hydropower generating sites. Selection of exact transmission right of way (ROW) is a complex process that requires extensive surveying, land appraisal, and stakeholder consultation. No prior study that assessed transmission options to connect hydropower plants to Fort McMurray has been found in the public domain. Therefore, the ROW of a given transmission line is selected by using the following criteria.

 First, a 50 km wide buffer around the straight line connecting the sending and receiving ends is selected as the study area  The straight line connecting the sending and receiving ends is selected as the initial ROW  Within the study area, where possible, obstacles such as large water bodies and protected areas such as National Parks are avoided  If a transmission corridor is available within the study area, the ROW is adjusted to follow existing lines

Transmission ROW formulation was done by utilizing a geographic information system (GIS). The QGIS8 software tool and publicly available GIS data obtained from Natural Resources Canada (Natural Resources Canada, 2014, 2015) is used for this task.

ROW width of a single circuit 500 kV HVAC and that of a ±500 kV HVDC bipole is about 60m. Those ROW widths are used for ROW land cost calculation. However, environmental, economic, and social impacts of the transmission line extend beyond its ROW. For example, visual impacts may be apparent as far as 500m from the center of the ROW; fragmentation of forested areas by transmission lines would affect wildlife movements; transmission towers and lines on agricultural land would impede the ability to utilize farming machinery, reducing productivity.

Therefore, a 1 km wide corridor is used for ecological and social impact estimations. Impacts within the full study area (i.e., 50 km wide buffer that encompasses the ROW) are also estimated and presented in Appendix A.

Levelized Cost of Delivered Electricity and GHG Emissions Abatement Cost Generation and transmission options described in the preceding sections are combined to formulate six hydropower generation and transmission options.

In addition to the six hydropower options, this study also assesses a baseline or reference power generation option. The reference option assumes that electricity demand will be satisfied by using natural gas-fired cogeneration plants. This assumption is consistent with current operations and announcements made by investors (Murillo, 2015; OSDG, 2013). Furthermore, cogeneration is an efficient and economical option to satisfy both thermal and electricity

8 QGIS is an open source GIS tool. It is available from: www.qgis.org

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 21 Oil Sands Electricity Demand demands of oil sands operations at a high reliability (Doluweera et al., 2011). Since cogeneration units are onsite, as far as the electricity consumed by a given oil sands operation is concerned, no transmission costs or losses are incurred. The cogeneration reference plant in this study is assumed to be a gas turbine-based cogeneration plant with duct firing ability in the heat recovery steam generator to have higher control over thermal energy production. Since a cogeneration plant produces both electricity and thermal energy, costs and emissions that are chargeable for the electricity portion is considered for this analysis (AESO, 2012; Doluweera et al., 2011; Suncor Energy & Jacobs Consultancy, 2012).

In this study, the identified electricity generation and delivery options are screened using a levelized cost of electricity (LCOE) metric (Kammen & Pacca, 2004; NETL, 2008). LCOE, measured in $/MWh, is a metric that represents the cost of constructing and operating an electric power generating plant. It is a standard metric used for screening and comparing different power generating options. LCOE is calculated using the information available at the point of decision making, based on discounted project cash flow analysis (Kammen & Pacca, 2004; Rubin, 2012). In this study we calculate the LCOE of delivered electricity as follows, taking into account both the generation and transmission systems.

(퐶퐶푔 ∙ 푐푐푓푔 + 퐹퐶푔 + 푉퐶푔 + 퐶퐶푡 ∙ 푐푐푓푡 + 퐹퐶푡 + 푉퐶푡) 퐿퐶푂퐸 = 퐸푔푒푛 + 퐸푙표푠푠 Where,

CCx = capital cost ($) FCx = annual fixed operating and maintenance cost ($/year) VCx = annual variable cost ($/year) Egen = annual electricity generation (MWh/year) Egen = Rated capacity of the power plant (MW) · 8760 (h) · capacity factor (%) Eloss = transmission losses (MWh/year) x = g for generating plant x = t for transmission system ccfx = annual capital charge factor (/year) and it is calculated using the following equation 푟(1 + 푟)푁 푐푐푓 = (1 + 푟)푁 − 1 Where, r = discount rate; N = Project economic life.

Table 2.1: Assumptions Made in Calculating Capital Charge Factor Option Discount rate Discount Rate Project ccf (base value) (range used for Economic Life (base value) sensitivity analysis) (years) Cogeneration plant 10% 8% - 12% 30 10.6%/year Hydropower plant 6% 4% - 8% 60 6.19%/year Transmission lines 6% 4% - 8% 60 6.19%/year Source: CERI

January 2016 22 Canadian Energy Research Institute

Real discount rates (i.e., discount rate adjusted for inflation) are used for this analysis. Furthermore, different discount rates are used for the cogeneration reference case (baseline option), hydropower options, and transmission system. It should be noted that in contrast to the cogeneration reference case, hydropower plants and transmission systems are very long-lived assets (60-100 years for a hydropower plant and 40-60 years for a transmission system).

Discount rates for this analysis are selected by taking into account the operational lives of respective assets and weighted average cost of capital (WACC) of relevant investors (AltaLink, 2012; BC Hydro, 2013a; Manitoba Hydro, 2013). For high capital-intensive projects such as hydropower plants and transmission lines, ccf (specifically the discount rate r) has a high impact on LCOE. Therefore, a sensitivity analysis is conducted against the discount rate. Discount rate and project life assumptions used for this analysis are given in Table 2.1.

Cost of GHG emissions abatement (henceforth referred to as cost of avoided carbon dioxide, or CACO2) of a given hydropower generation and transmission option against the cogeneration baseline option is calculated as follows:

퐿퐶푂퐸ℎ푝 표푝푡𝑖표푛 − 퐿퐶푂퐸푏푎푠푒푙𝑖푛푒 표푝푡𝑖표푛 퐶퐴퐶푂2 = 퐸푀퐼푏푎푠푒푙𝑖푛푒 표푝푡𝑖표푛 − 퐸푀퐼ℎ푝 표푝푡𝑖표푛 Where,

LCOEx = Levelized cost of delivered electricity (in $/MWh) of the hydropower option (x=hp option) / baseline option (x=baseline option) EMIx = GHG emissions intensity (in tCO2e/MWh)

CACO2 is a standard metric used for screening GHG emissions abatement options. It calculates the cost of avoiding a tonne of atmospheric CO2 emissions while providing a unit of useful energy (i.e., MWh of electricity) (Rubin, 2012). It should be noted that CACO2 is measured against a reference option, which in general, has a higher GHG emissions intensity and lower LCOE. CACO2 can also be interpreted as the price of CO2 emissions that would make the LCOE of the low GHG emissive option the same as that of the reference plant.

Uncertainty Assessment Parameters such as capital costs, fuel prices, and capacity factors that are used for point estimates of LCOE and CACO2 have considerable degrees of uncertainty associated with them. For example, the hydropower options assessed in this study are in different planning stages and the accuracy of the capital cost estimates, as indicated by the project developers, are in the range of -15 percent to +30 percent of the reported value; capacity factors depend on river flow levels and precipitation patterns.

In order to ensure the robustness of the information available for decision makers with respect to large infrastructure projects, it is important that uncertainty associated with the results be explicitly assessed. In order to assess uncertainty associated with the LCOE and CAOC2 of the

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 23 Oil Sands Electricity Demand electricity generation and delivery options explored in this study, a Monte Carlo simulation (MCS) is developed.

To carry out a MCS, the probability distributions of the main parameters are identified. Each parameter is then sampled from their probability distributions. These samples are then used to calculate the relevant metrics (in this study LCOE and CACO2) and important attributes by which those metrics are identified by utilizing statistical techniques (e.g., mean, variance and percentiles of the metrics).

In this study, the impact of the following parameters are assessed by developing a MCS with 100,000 samples. Each parameter is assumed to be uniformly distributed.

 Capital costs of generation and transmission assets  Capacity factors  Natural gas price  Electricity price at the Mid-Columbia (Mid-C) electricity hub

Decision making under uncertainty is sometimes based on point source information instead of uncertainty ranges. For clarity, this analysis focuses on point source estimates; however, Appendix C contains a detailed probability assessment of the different options showing how the selection order may change based on the MCS.

Estimation of Environmental and Social Impacts Large energy infrastructure projects such as development of hydropower plants and long distance transmission lines inevitably lead to undesirable social and environmental impacts. Stakeholder perception on those impacts can potentially delay or halt permitting, siting, and construction of energy infrastructure. Stakeholder opposition of hydropower and transmission development, particularly against three new hydropower plants assessed in this study, was observed across Canada (Anderson, 2013; CBC News, 2010; Henton, 2013; Hunter, 2015; Larkins, 2015). Therefore, it is important that these impacts are quantified and stakeholders are consulted early in the screening phase of a project.

Social and environmental impacts are estimated by identifying a set of relevant project attributes and quantifying them (Keeney & Gregory, 2005). For example, when quantifying the impact on sensitive ecosystems by an electricity transmission line project, one attribute that can be utilized is the ecosystem area (say, in hectares) within the transmission ROW. Social and environmental impacts quantified in this study and the associated attributes are described in the following sections.

Quantification is not monetization. Monetization can be situation-specific and based on the judgement of the stakeholder. Physical quantification provides greater transparency and allows decision makers to assess their own value of the environmental and social impacts.

January 2016 24 Canadian Energy Research Institute

The area considered for impact estimates includes a 1 km wide buffer that encloses the selected transmission line corridor and a circular buffer with a 5 km radius that encloses the hydropower plant. Combination of these two buffers are henceforth referred to as the direct impact area. All impacts are calculated using publicly available GIS data.

Residential and Property Value Impacts Residential and property value impacts encompass a multitude of public concerns against energy infrastructure. These include, but are not limited to visual impacts, perceived health impacts due to exposure to electromagnetic fields (EMF) caused by the lines, impacts on future development, and potential reduction of property value due to a transmission line being within close proximity of respective properties. Furthermore, development of hydropower plants could potentially lead to involuntary displacement of a population due to reservoir flooding. This study estimates the population and number of dwellings in the direct impact area as two high-level metrics of residential and property value impacts. Population and number of dwellings are estimated by using 2011 Canada Census data (Statistics Canada, 2012) as follows:

푃표푝푢푙푎푡𝑖표푛 푑푒푛푠𝑖푡푦 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛 푎푟푒푎 푃표푝푢푙푎푡𝑖표푛 = ∑ ( ) × ( ) 표푓 푎 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛 푤𝑖푡ℎ𝑖푛 푡ℎ푒 푑𝑖푟푒푐푐푡 𝑖푚푝푎푐푡 푎푟푒푎 푎푙푙 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛

푁표. 표푓 푑푤푒푙푙𝑖푛푔푠 퐷푤푒푙푙𝑖푛푔푠 푑푒푛푠𝑖푡푦 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛 푎푟푒푎 = ∑ ( ) × ( ) 표푓 푎 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛 푤𝑖푡ℎ𝑖푛 푡ℎ푒 푑𝑖푟푒푐푐푡 𝑖푚푝푎푐푡 푎푟푒푎 푎푙푙 푐푒푛푠푢푠 푠푢푏푑𝑖푣𝑖푠𝑖표푛

Agricultural Impacts Agricultural impacts stem from loss of agricultural land due to reservoir flooding and siting transmission lines on agricultural lands. Areas of agricultural land within the direct impact area are estimated by using land cover data set obtained from Natural Resources Canada (Natural Resources Canada, 2014).

Impacts on Indigenous Populations Development of energy infrastructure in areas inhabited by Indigenous people affects their lives in multiple ways. Direct impacts stem from developing hydropower plants and transmission lines on Indigenous lands. Other adverse impacts of hydropower and transmission line development can potentially impede fishing, hunting, and trapping rights as well as cultural heritage resources of the affected population. In this study, as a proxy measure of impacts on Indigenous populations, reserve areas within the direct impact area is estimated by utilizing GIS tools. No other designated Indigenous land (e.g., Metis settlements) is observed within the study area. Indigenous lands geographic datasets were obtained from Natural Resources Canada (2015). Both number (partially or fully enclosed by the direct impact area) and First Nations reserves are estimated.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 25 Oil Sands Electricity Demand

Environmental Impacts Altering the natural habitat for hydropower and transmission developments will directly and indirectly impact the biodiversity of the habitat and ecosystem services attributable to the area. These ecosystem services include water regulation, soil formation and erosion control, food production, provision of raw material, climate regulation and recreational services. Quantifying these impacts is important at the screening stage of options to satisfy energy demands to ensure responsible resource developments and to identify the necessary regulatory requirements that need to be complied with.

The main environmental impacts associated with options assessed in this study include potential impacts on environmentally sensitive areas (e.g., wetlands, national parks, etc.), methylmercury formation due to reservoir flooding, avian mortality due to collision with power lines, soil erosion, and wildlife habitat defragmentation due to new linear disturbances. This study estimated two metrics as proxies of environmental impacts. One metric is the area of protected areas (national parks, provincial parks, migratory bird sanctuaries, etc.) within the direct impact area. The other metric is area of water bodies (rivers, lakes, and wetlands) within the direct impact area. GIS data from the CanVec+ catalogue is primarily used to estimate environmental impacts (Natural Resources Canada, 2014).

It is plausible that environmentally significant land may lie outside the boundaries of protected areas (for example, boreal forest, grasslands, private land). Therefore, the full land cover of the direct impact area is estimated and classified using GIS tools.

January 2016 26 Canadian Energy Research Institute

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 27 Oil Sands Electricity Demand Chapter 3 Results

Hydropower Generation and Transmission Options Specific details of the six hydropower generation and transmission options assessed in this report are described in this section. Key details of these six options are summarized in Table 3.1.

Table 3.1: Summary of Hydropower Generation and Transmission Options Hydropower Generation Plant Transmission System Site and River Province Rated Technology No. of Length Option System Capacity Lines (km/line) (MW) Site C-DC Site C on the British 1100 ±500 kV HVDC 1 600 Peace Rive Columbia bipole Site C-AC Site C on the British 1100 Single circuit 2 600 Peace Rive Columbia 500kV HVAC BC British Increase Single circuit Intertiei Columbia by 500 500kV HVAC Slave Alternative 4 Alberta 1100 ±500 kV HVDC 1 400 River-DC site on the bipole Slave River Slave Alternative 4 Alberta 1100 Single circuit 2 400 River-AC site on the 500kV HVAC Slave River Manitoba Conawapa site Manitoba 1485 ±500 kV HVDC 1 1100 DC on the Nelson bipole River iThe BC Intertie option assumes a case where the existing BC-AB intertie is reinforced to import higher amounts of hydropower purchased from the BC Hydro system. Therefore, no new hydropower plants or new transmission lines are attributed to this option. The capital cost of this option is assumed to be the reinforcement cost of the BC-AB transmission intertie. Source: CERI

Alberta-BC Options Two hydropower generation and transmission options assume that electricity produced by the Site C dam and associated hydropower plant is transported to Alberta via a new transmission line. The Site C dam is being constructed approximately 7 km southwest of the town of Fort St. John. The Site C-DC option assumes that a single ±500 kV HVDC bipole transmission system is used to transmit electricity from the Site C hydropower plant to Fort McMurray, AB.

January 2016 28 Canadian Energy Research Institute

The Site C-AC option assumes that two 500 kV single circuit HVAC transmission systems are used to move electricity from the Site C plant to Fort McMurray, AB. As discussed in Chapter 2, in the case of HVAC options, two transmission circuits are used to ensure system reliability. Figure 3.1 depicts the power plant locations and the transmission corridor. As can be seen in Figure 3.1, the transmission corridor would span across a mixture of agricultural land, populated areas, forests, and protected areas.

Figure 3.1: Alberta-BC Hydropower Generation and Transmission Options (Site C-DC and Site C-AC Options)

A third BC-Alberta option, the BC Intertie, assumes reinforcement of the existing AB-BC intertie (a 500 kV HVAC line between Cranbrook, BC and Langdon, AB) to utilize the full 1,200 MW capacity of the line and deliver hydroelectric power purchased from the BC electric power system. Currently only about 50 percent of the existing BC-AB intertie is being utilized.

Alberta-Manitoba Option The Manitoba DC option assumes that the proposed Conawapa hydropower plant on the Nelson River in Manitoba is developed and a single ±500 kV HVDC bipole transmission system is used to transmit electricity to Fort McMurray, AB. If developed, the Conawapa plant would be the eighth hydropower plant on the Nelson River. The Nelson River hydroelectric system currently has six

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 29 Oil Sands Electricity Demand plants in operation and another one under construction. Figure 3.2 depicts the location of the hydropower plants and the transmission line corridor.

Figure 3.2: Alberta-Manitoba Hydropower Generation and Transmission Option (Manitoba DC Option)

Alberta Slave River Options The Slave River-DC and Slave River-AC options assume that a hydropower plant would be developed on the Alberta Slave River. Based on previous proposals, the most likely location of the plant would be a site called Alternative 4 Site, just south of the border between Alberta and the Northwest Territories. Figure 3.3 depicts the most likely location of the power plant and transmission lines that would connect it to Fort McMurray.

The Slave River-DC option assumes that a single ±500 kV HVDC bipole transmission system is used to transmit electricity from a hydropower plant at the Alternative 4 Site to Fort McMurray, AB. The Slave River-AC option assumes that two 500 kV single circuit HVAC transmission systems are used to transmit power to Fort McMurray from the same hydropower plant.

January 2016 30 Canadian Energy Research Institute

Figure 3.3: Alberta Slave River Hydropower Generation and Transmission Options

Levelized Cost of Electricity (LCOE) and Cost of Avoided CO2 The LCOE of the cogeneration reference case and six hydropower options are estimated using the data listed in Table 3.2. CACO2 of the hydropower options are also estimated. Estimated values of LCOE and CACO2 are presented in Table 3.2, Figure 3.4 (LCOE), and Figure 3.5 (CACO2).

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 31 Oil Sands Electricity Demand

Table 3.2: LCOE and CACO2 Estimates (base case estimates) Intertie Cogen HVDC Options HVAC Options Option Reference Site Slave Manitoba Slave BC Case Site C-AC C-DC River-DC DC River-AC Intertiei Generation Rated capacity (MW) 1000 1100 1100 1485 1100 1100 500 Capacity factor (%) 85 53 68 54 53 68 75 Annual power generation 7446 5100 6552 7000 5100 6552 3285 (GWh/year) Capital cost (million $) 2070 8444 8800 9541 8444 8800 Overnight cost 1448 4830 4830 5610 4830 4830 Contingency & 435 1564 1835 1704 1564 1835 escalation IDCii 188 1668 1739 1908 1668 1739 Capital cost ($/kW) 2071 7677 8003 6425 7677 8003 Capital charge factor (%) 10.61 6.19 6.19 6.19 6.19 6.19 Fixed O&M cost (million 14 16 17 16 16 17 $/year) Fuel cost ($/MWh)iii 26.0 5.5 3.8 1.8 5.5 3.8 42.5 GHG intensity 325 11 5 5 11 5 11 (kgCO2e/MWh) LCOE (Generation only) 57.4 111 89.4 88.4 111 89.4 42.5 Transmission Trans. circuit length (km) 0 600 400 1100 1200 800 318 Capital cost (million $)iv 0 1984 1789 3141 3079 2560 280 Capital charge factor (%) 6.34 6.34 6.34 6.34 6.34 6.34 6.34 O&M cost (million 0 10 9 16 15 11 93 $/year) Transmission losses 0 122 127 253 292 250 196 (GWh/year) LCOE 57 141 110 124 162 121 81 v CACO2 Cogen ref. option 266 165 207 332 198 75 vi CACO2 AB Grid 107 62 82 137 77 21 vii CACO2 Coal 72 33 50 98 47 -3 Notes: All costs are in 2014 Canadian dollars. LCOE excludes taxes charged by respective jurisdictions iPower generation capital and operating costs are irrelevant for the BC Intertie option as no dedicated hydropower plant is assumed. Electricity imports are assumed to be purchased at the average peak electricity price at the Mid- Columbia (Mid C) gate. iiIDC = Interest accumulated during construction (also known as capitalized interest) iiiFuel cost of the cogeneration reference case is calculated assuming a natural gas price of $4/GJ (hhv) and heating value of 6.5 GJ/MWh; fuel cost of hydropower options is the water rental fees; non-fuel variable O&M are assumed to be negligible ivInterest accumulated during construction and contingencies are included in the capital cost v CACO2 is estimated against the cogeneration reference case vi CACO2 is estimated against the Alberta grid average with an average price of $66/MWh and average grid emissions intensity of 710 KgCO2/MWh for the 2010-2014 period. These parameters are estimated using data published by AESO (2015a)

January 2016 32 Canadian Energy Research Institute

vii CACO2 is estimated against coal-fired generation displacement reference option with an average cost of generation of $82/MWh and emissions intensity of 820 kgCO2/MWh. Data sources: Generation overnight capital cost, capacity factors: Site C-DC, Site C-AC: BC Hydro (2013a); Manitoba DC: Manitoba Hydro (2013); cogeneration: AESO (2012) Slave River-DC/Slave River-AC overnight capital cost is assumed the same as Site C. Slave River-DC/Slave River-AC capacity factor is estimated using flow rate data from Environment Canada (2015a) and SRSC (1980) New transmission costs and losses: Black & Veatch (2014) AB – BC intertie reinforcement costs: AESO (2015b)

The abatement costs of CO2 are sensitive to the reference case used. Costs are lowest when compared to higher emitting sources such as coal generation and lower compared to the more efficient generation of electricity through natural gas cogeneration.

Figure 3.4: Levelized Cost of Electricity (LCOE) Delivered to Oil Sands Operations in Alberta under Different Generation and Transmission Options

Source: CERI

LCOE represents the average cost of constructing and operating an electric power generation or transmission system. The reference option (base case) assumes that natural gas-fired cogeneration units satisfy electricity demands. All hydropower options have higher LCOE than the cogeneration reference option when both generation and transmission costs are taken into account.

The BC Intertie option, which assumes hydropower is bought from the BC Hydro system and delivered to Alberta using the existing intertie, is the hydropower option with the lowest LCOE. Generation costs of Slave River hydropower options (Slave River-DC and Slave River-AC) and the Conawapa hydropower option (Manitoba DC) appear to be the same, although the greater

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 33 Oil Sands Electricity Demand transmission distance makes the LCOE of Manitoba DC higher than the Slave River options. When both HVDC and HVAC transmission options are considered for a given hydropower plant, the HVDC option leads to lower transmission costs due to shorter effective circuit length (HVAC options have two independent transmission circuits) and lower losses.

Figure 3.5: Cost of Avoided GHG Emissions (CACO2)

Source: CERI

Among the options where new hydropower plants are developed, the Alberta Slave River options (Slave River-DC/Slave River-AC) have the lowest LCOE and CACO2. Both generation and transmission costs of the Slave River options are lower than other new hydropower options. This is mainly due to high capacity factors that lowered the generation cost and lowered transmission distance. For the two hydropower options where two transmission options are assessed, HVDC options have lower costs than HVAC options. This is because two HVAC lines are required to ensure sufficient transmission capacity and system reliability (see Figure 3.6).

January 2016 34 Canadian Energy Research Institute

Figure 3.6: Capital Cost Contribution of Different Transmission System Components

Source: Cost data from Black & Veatch (2014); figure by CERI

In this assessment, transmission line investment costs are estimated by utilizing a model developed by Black & Veatch (2014) for the western region of the North American power system. In terms of investment cost per transmission distance basis, the estimated values that are depicted in Figure 3.6 are comparable to recently completed or planned major transmission line projects in Western Canada. For example, the investment cost of HVDC lines of Site C-DC, Slave River-DC, and Manitoba DC options are in the order of $3.3-$4.4 million/km. Capital costs of two HVDC monopole lines are in the order of $4-$4.4 million/km. The capital cost of the Bipole III project, which would build a 1,400 km HVDC bipole line in Manitoba, is estimated to be $3.3 million/km.

Figure 3.7 indicates the amount of electricity that can be supplied by different hydropower options along with the respective LCOE values. It also indicates the equivalent in-situ bitumen extraction amount, of which electricity demand can be satisfied by power delivered under a given option. Site C-AC and Slave River-AC options are not depicted in this figure as they connect the same hydropower plant as Site C-DC and Slave River-DC. Similarly, Figure 3.8 depicts the amount of GHG emissions that can be avoided by the different options.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 35 Oil Sands Electricity Demand

Figure 3.7: Electricity Supply Curve

Source: CERI

Figure 3.7 depicts the amount of electricity that can potentially be delivered to Fort McMurray under different hydropower options along with respective LCOE estimates. Equivalent bitumen extraction estimates assume in-situ bitumen extraction. As can be seen from the figure, although the BC Intertie has the lowest LCOE, the estimated amount of delivered energy is less than half of other hydropower options.

Figure 3.8 depicts the emissions abatement potential of hydropower options along with their CACO2 estimates. CACO2 estimates of new hydropower options are double or more compared to the BC Intertie option.

January 2016 36 Canadian Energy Research Institute

Figure 3.8: GHG Emissions Abatement Supply Curve

Source: CERI

Estimates of CACO2 are sensitive to the reference case electricity supply that is to be displaced by hydropower options. When the cogeneration reference is considered, the estimated CACO2 values are relatively high for most hydropower options due to lower average costs of cogenerated electricity and lower emissions intensity.

Adding onsite cogeneration to an oil sands operation requires additional investments and increases operational complexity. Therefore, it is also plausible that the oil sands operators may choose to purchase electricity from the Alberta electricity market, instead of onsite generation. Therefore, CACO2 is also estimated against a case where Alberta average grid electricity is displaced by hydropower. The average price of grid electricity and average emissions intensity are assumed to be $66/MWh and 710 kg CO2eq/MWh, respectively. Those values represent the average conditions in the Alberta electricity market over the period 2010-2014.1 The reference case could also be the coal-fired generation that currently accounts for the largest share of electricity generation in Alberta. Coal-fired generation is being phased out in Alberta but still forms a plausible reference case to compare GHG abatement costs of displacing high emissive electricity. For the coal-fired electricity reference case, the reference plant is assumed to be the Keephills 3 (Capital Power, 2011). This is the last coal power plant that was added to the Alberta generation fleet and the unit that is most likely to remain in the fleet over the next few decades in the absence of GHG emissions constraints (AESO, 2012). For this reference case, a LCOE of

1 Calculated using the data published in (AESO, 2014a)

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 37 Oil Sands Electricity Demand

2 $83/MWh and a GHG emissions intensity of 820 kg CO2eq/MWh are assumed. CACO2 estimates based on these two alternative reference options are presented in Table 3.2.

As can be seen from Table 3.2, the CACO2 values correspond to the Alberta grid replacement reference case and are 60-70 percent lower than those of the cogeneration reference case. When replacing new coal-fired generation with the BC Intertie, the abatement cost is negative because the assumed cost of BC Intertie electricity is less than the reference coal power plant. The relative ranking of hydropower options in terms of CACO2 remains unchanged with the choice of reference option.

Sensitivity Analysis against Discount Rate Sensitivity analysis against the discount rate used for the LCOE calculation was carried out and the resulting supply curves are depicted in Figures 3.9 and 3.10. The analysis used discount rates of Base = 6 percent; Low = 4 percent; High = 8 percent. As can be seen from those results, the relative placements of different options did not change with the discount rate. The impact of a 3 discount rate on LCOE and CACO2 is significant for options with high capital expenditure. Although not depicted here, the impact of discount rate variation on LCOE of the cogeneration reference option is relatively low.

Impacts of variations in other important parameters (e.g., capital costs, capacity factors, etc.) are examined in Appendix C, through a probabilistic uncertainty assessment based on a Monte Carlo simulation. This assessment provides insights into the robustness of the results and how the selection order may change with variations in different parameters.

2 Calculated using the information available at: Capital Power, “Keephills 3 Power Plant Begins Commercial Operation,” 2011, http://www.capitalpower.com/MediaRoom/newsreleases/2011-news- releases/Pages/090111.aspx. A discount rate of 10% over 30 years is assumed. 3 1 percent change in discount rate leads to 13-14 percent change in LCOE and 23-24 percent change in CACO2

January 2016 38 Canadian Energy Research Institute

Figure 3.9: Electricity Supply Curve Sensitivity Analysis against Discount Rate

Source: CERI

Figure 3.10: GHG Emissions Abatement Supply Curve Sensitivity Analysis against Discount Rate

Source: CERI

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 39 Oil Sands Electricity Demand

Economic Assessment of Electrical Extraction Technologies If a low/zero emission supply of electricity such as hydroelectric power is available, deeper GHG emissions reductions can be achieved in the oil sands sector through adaptation of electrical extraction technologies. As such, this section provides estimates of CO2 abatement costs of using six hydropower options assessed in the report for electrical extraction of bitumen. The main challenge in estimating the CO2 abatement cost of electrical extraction technologies is that those technologies are still in demonstration or experimental stage and data on their costs, energy intensities, and other relevant performance parameters (e.g., full list of energy and material inputs) are not available in the public domain. Therefore, this assessment provides preliminary estimates of abatement costs and technology costs in the form of a scenario analysis.

In contrast to current in-situ extraction methods, electrical extraction techniques such as Effective Solvent Extraction Incorporating Electromagnetic Heating (ESEIEH) and Electro-Thermal Dynamic Stripping Process (E-T DSP™) employ alternative extraction techniques where all energy inputs are replaced with electricity. Therefore, the abatement cost is the cost of CO2 emissions that would make the average production cost4 per barrel of an electrical extraction technology the same as that of a reference bitumen extraction process.

For this assessment, the reference (or the baseline) extraction process is:

 SAGD bitumen extraction facility  production capacity of 30,000 barrels per day  thermal demand is 1.3 GJ/bbl, which represents the sector wide average SAGD thermal energy demand (Nduagu & Gates, 2015)  electricity demand is 16.5 kWh/bbl (Murillo, 2015)  natural gas-fired 85 MWe cogeneration system and a natural gas-fired supplementary boiler

Excess electricity produced by the cogeneration system is assumed to be exported to the Alberta grid and costs and emissions associated with exported electricity are excluded from the facility’s accounts. Full details of the reference SAGD facility along with its energy costs and emissions are presented in Appendix B. According to the estimates presented in Appendix B, the energy cost of the reference facility is $5.9/bbl5 and associated GHG emissions are 67 kgCO2e/bbl.

In the 2014 long term outlook, AESO (2014a) assumes that the electricity demand for electricity based extraction technologies to be 180 kWh/bbl. A recent assessment by CERI assumes the same energy intensity. An estimate based on process simulations reports the electricity demand of ESEIEH technology to be 44 kWh/bbl (= 0.16 GJ/bbl or approximately 15 percent of SAGD energy demand) (Patterson, 2015). Due to the data limitations, this assessment takes into account four different energy intensities, where the electricity energy demand for bitumen

4 Average production cost takes into account both capital and operating costs 5 Natural gas price is assumed to be $4/GJ

January 2016 40 Canadian Energy Research Institute

extraction is assumed to be 15 percent, 25 percent, 50 percent and 100 percent of the total energy demand of the reference SAGD facility (= 1.36GJ/bbl or 378 kWh/bbl).

For each combination of the four electricity demand cases and six hydropower generation and transmission options (a total of 24 scenarios), the bitumen production potential, electrical energy cost, avoided GHG emissions, and minimum abatement cost (i.e., CACO2) are calculated. For this assessment, it is assumed that geological costs (site preparation, drilling, etc.) are the same for the reference SAGD facility and the electrical extraction facility.

Minimum CACO2 of a given scenario is the price of CO2 that would make its energy cost the same as the average energy cost of conventional SAGD when the non-fuel energy cost6 is set to be zero. Impacts of taxes and royalties are excluded from this assessment. Impacts of emissions associated with non-energy material inputs (e.g., solvents for ESEIEH process) are also excluded.

Table 3.3: Minimum CACO2 and Other Metrics of Electrical Extraction Scenarios

i Production Potential Electrical Energy Cost Avoided Emissions Minimum CACO2 (1,000 bbl/day) ($/bbl) (million tCO2e/year) ($/tCO2e) 15% 25% 50% 100% 15% 25% 50% 100% 15% 25% 50% 100% 15% 25% 50% 100% Site C-DC 247 148 74 37 8 13 27 53 6.0 3.6 1.8 0.9 31 112 319 752 Slave River- 317 190 95 48 6 10 21 42 7.7 4.6 2.3 1.1 5 67 225 547 DC Manitoba DC 339 203 102 51 7 12 23 47 8.2 4.9 2.4 1.2 17 87 265 629 Site C-AC 247 148 74 37 9 15 31 61 6.0 3.6 1.8 0.8 49 142 380 880 Slave River- 317 190 95 48 7 11 23 46 7.7 4.6 2.3 1.1 14 83 257 612 AC BC Intertie 159 95 48 24 5 8 15 31 3.8 2.3 1.1 0.5 - 26 145 393 Source: CERI

The four metrics estimated for each scenario are presented in Table 3.3. As can be seen, compared to the reference SAGD facility, higher amounts of avoided emissions could be achieved by deploying electrical extraction techniques (95-99 percent reduction). However, as more electricity is needed for bitumen extraction, the bitumen production potential of a given hydropower generation and transmission option is relatively lower (24,000-339,000 bbl/day) resulting in a higher energy cost per barrel ($5-61/bbl).7 In the case of the BC Intertie option, when the non-energy cost is set to zero, the electrical extraction option is preferred to conventional SAGD even without a price on CO2. The bitumen extraction potential of all scenarios that involve the BC Intertie electricity supply option are the lowest.

Minimum CACO2 estimates of electrical extraction scenarios presented in Table 3.3 assume the non-fuel energy costs are zero. However, it is inevitable that the extraction process would incur additional costs. Therefore, minimum CACO2 can be interpreted as the lower limit of the CACO2

6 Non-fuel energy costs include all fixed and operating costs associated with implementing the extraction process such as all additional equipment, non-energy inputs (e.g., solvents used for ESEIEH process), etc. 7 Note that this only includes the cost of electricity

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 41 Oil Sands Electricity Demand that would make electrical extraction preferred to conventional SAGD.8 In order to gain insights into the magnitude of the non-fuel energy costs, a sensitivity analysis against the CACO2 is conducted in Figure 3.11.

Figure 3.11: Sensitivity Analysis of CACO2 for Electrical Extraction Technologies

Source: CERI

Figure 3.11 provides useful information to assist investment and research and development (R&D) decisions pertaining to electrical extraction technologies. At the assumed natural gas price and hydropower development cost, the non-fuel extraction cost represents the upper limit of the metric that could be afforded at a given CACO2 value.

For a given electricity intensity (e.g., 15 percent, 25 percent, etc.) the area below the corresponding line represents the values of CACO2 and non-fuel extraction costs that would make the electrical extraction method preferred to SAGD extraction. None of the scenarios with electricity intensity of 100 percent appear in this figure as their minimum CACO2 is more than $300/tCO2e.

Social and Environmental Impacts of Hydropower Options The social and environmental impacts are estimated by quantifying the land cover within the direct impact area.9 Only the impacts pertaining to new hydropower options are quantified,

8 At the assumed natural gas price and hydropower development costs 9 Direct impact area is defined as the area formed by a combination of a 1 km wide buffer that encloses the selected transmission line corridor and a circular buffer with a 10 km radius that encloses the hydropower plant)

January 2016 42 Canadian Energy Research Institute

excluding the cogeneration reference case. It is assumed that cogeneration units would be sited within an oil sands facility, resulting in no additional impacts attributable to electricity. The main exception is GHG emissions, which have been quantified for all options in CACO2 calculations. Similarly, the BC Intertie option that utilizes existing energy infrastructure would not have additional impacts. HVDC and HVAC transmission options are not distinguished, as the 1 km wide buffer is sufficient to site either of the transmission systems. Consequently, the five new hydropower options are re-categorized as follows for impact calculations:

 AB Slave River: Corresponds to Slave River-DC or Slave River-AC  AB-BC: Corresponds to Site C-DC or Site C-AC  AB-MB: Corresponds to Manitoba DC

Land cover within the respective direct impact areas is quantified in Figure 3.12. Land categories and quantification is carried out by utilizing the CanVec+ dataset (Natural Resources Canada, 2014). It can be seen that the direct impact areas mainly encompass boreal forest, wetlands, and water bodies (i.e., rivers and lakes). In the case of the AB Slave River options, those three land categories cover 96 percent of the direct impact area. For the AB-BC and AB-MB option the coverage of those three land categories are 66 percent and 76 percent, respectively.

Figure 3.12: Land Cover within the Direct Impact Area

Source: CERI

In Figure 3.13, the population and number of dwellings within the direct impact area are quantified as a proxy metric of the residential and property value impacts. According to the results, the vast majority of residential and property value impacts are concentrated in Alberta.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 43 Oil Sands Electricity Demand

New hydropower options from BC (Site C-DC and Site C-AC) appears to have the highest residential and property value impacts and they are mainly concentrated to areas surrounding the towns of Fort St. John, BC and Peace River, AB.

Figure 3.13: Population and Number of Dwellings in the Direct Impact Area

Source: CERI

One notable caveat of this study is that the land use impact assessment may not fully capture the impacts of widening the Peace River by Site C dam construction (Site C-DC and Site C-AC options). Construction of the Site C dam would widen the Peace River by up to three times the current levels and flood an approximate 83 km stretch along the river. Furthermore, under all options, construction of hydropower plants and transmission lines would inevitably require building of construction roads and camps. Impacts that would result from these temporary structures are not fully captured in this analysis. Assessment of those impacts require detailed, temporally explicit land use assessments that are beyond the scope of this study.

Only the AB-BC options appear to cover notable areas of agricultural land. Based on the dataset utilized (Natural Resources Canada, 2015), other new hydropower options do not have notable agricultural impacts. Furthermore, the AB-BC options have the highest residential and property value impacts. As can be seen from Figure 3.13, the population and number of dwellings in the direct impact area of the AB-BC options are several times greater than the other options.10

It should be noted that these impacts could be minimized by careful route selection. Furthermore, population and number of dwellings are estimated assuming uniform distribution

10 Based on 2011 Canadian census data

January 2016 44 Canadian Energy Research Institute

within a given census sub-division. The actual distribution may be contrary to that assumption. Nonetheless, the estimates depicted in Figure 3.13 provide general metrics of residential and property value impacts.

Figure 3.14: Number and Area of First Nations Reserve Lands in the Direct Impact Area

Source: CERI

The number and area of First Nations reserves within the direct impact area are shown in Figure 3.14. According to these estimates, as far as direct impact areas are concerned, only the AB Slave River and AB-MB options appear to have impacts on Indigenous populations. Furthermore, Indigenous population impacts pertaining to the AB Slave River options occur only within Alberta and those pertaining to the AB-MB option occur only within Manitoba. Impacts that would potentially stem from factors such as water regulation by hydropower dams and construction activities are not estimated in this study. Furthermore, quantification of impacts on Indigenous peoples within the full study area (i.e., 50 km wide buffer enclosing the transmission ROW; Figures 3.1-3.3) result in a much higher number of reserve areas affected (see Figures 3.1-3.3 and Figure A.3 of Appendix A). When the full study area is taken into account, impacts on Indigenous peoples extend to the AB-BC options as well as to five First Nations reserves in Saskatchewan.

The potential land impacts that would stem from widening the Peace River and some of the tributaries by the construction of the Site C dam are not captured by this assessment. All of these

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 45 Oil Sands Electricity Demand factors emphasize the importance in consulting the relevant Indigenous communities in any attempt to develop hydropower and transmission options assessed in this study.

Land cover within respective direct impact areas that are designated as protected areas at federal or provincial levels due to their ecological value are depicted in Figure 3.15. Figure 3.15 also depicts the area of water bodies within respective direct impact areas. These two proxy indicators provide metrics of environmental impacts of the new hydropower options. The amount of protected areas and water bodies impacted by new transmission options follow the same order as the transmission ROW length (for example, AB-MB, with its longest ROW, impacts the highest amount of protected areas).

Figure 3.15: Protected Areas and Water Bodies in the Direct Impact Area

Source: CERI

Protected areas impacted by new hydropower options are mostly provincial parks in respective jurisdictions. The Alberta Slave River options, however, would impact federally protected Wood Buffalo National Park (WBNP) and Richardson Lake Migratory Bird Sanctuary. WBNP, a UNESCO World Heritage Site, is the second largest national park in the world. The park is the nesting grounds of the last remaining wild flock of endangered Whooping Cranes in the world (Environment Canada, 2015b).

All four North American migratory bird flyways cross the WBNP. Furthermore, development of a hydropower plant and associated transmission infrastructure on the Slave River would inevitably impact the Peace-Athabasca Delta (PAD), a globally important ecologically sensitive

January 2016 46 Canadian Energy Research Institute

area. Due to its ecological importance, the PAD is protected under the Ramsar Convention (SRSC, 1980; Struzik, 2013).

The PAD is formed at the confluence of the Peace, Athabasca and Birch rivers at the western end of Lake Athabasca. Since the landscape is relatively flat, many of its waterways can flow in two directions depending on relative water levels, forming the delta (PADEMP, 2015). It is one of the world’s largest freshwater deltas and has a profound impact on flora and fauna – which includes 214 species of birds, 42 species of mammals, and 20 fish species – as well as the traditional land use practices of the surrounding Indigenous peoples communities.

The delta contains some of the largest undisturbed grass and sedge meadows in North America (PADEMP, 2015). Ecosystem health of the PAD is directly linked to the hydrology of the northward flows of the Peace and Slave rivers. Consequently, water level regulations by a potential hydropower plant would inevitably impact the PAD. Moreover, construction activities11 may disturb wildlife movement and increase the risk of spreading invasive species.

A hydropower plant development on the Slave River would impact a waterway that has not yet been impounded. In contrast, Site C and Conawapa would be developed on rivers that have already been impounded by existing hydroelectric dams.12 Due to these factors, of all the new hydropower options assessed in this study, the Slave River options likely have the highest amount of environmental impacts.

In this study, the extent of protected areas impacted by new hydropower options is estimated as a proxy measure of environmental impacts. However, ecological impacts of hydropower and transmission developments would extend well beyond the protected areas. As evident on Figure 3.12, a greater proportion of land within the respective direct impact areas contains Canada’s boreal forest and wetlands intermingled with it.

Canada’s boreal forest region is a complex ecosystem that supports thousands of species of flora and fauna and contains a significant proportion of the world’s surface freshwater (Pew Environment Group, 2012). Only a fraction of the boreal region is designated as a protected area – although all Canadian jurisdictions have enacted regulations to ensure the integrity of the natural environment that may be impacted by anthropogenic activities – and therefore, the metrics depicted in Figure 3.12 may not fully quantify the ecological impacts of the assessed new hydropower generation and transmission options. For example, Figure 3.16 depicts the habitats of the woodland caribou within the jurisdictions pertaining to this study. Woodland caribou – an iconic species whose population is often seen as an indicator of the ecosystem health of the boreal region – is an endangered species whose population is at risk of declining due to habitat loss (Environment Canada, 2015b).

11 Construction period of a large hydropower plant is in the order of 10-12 years and that of transmission lines is in the order of 1-4 years. 12 W.A.C. Bennett dam is upstream of the proposed Site C dam site on the Peace River; currently there are six hydroelectric dams on the Nelson River upstream of the proposed Conawapa site.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 47 Oil Sands Electricity Demand

Figure 3.16 also depicts the viability of self-sustaining the boreal woodland caribou population without conservation intervention as reported in a scientific study by Environment Canada (2008). It can be seen that transmission ROW associated with all new hydropower options, if developed, would impact the woodland caribou habitat. Moreover, a greater proportion of that habitat is unable to support a self-sustaining woodland caribou population. Therefore, new transmission development could exacerbate the risk of a declining woodland caribou population.

Figure 3.16: Woodland Caribou Population in the Study Area

Source: Environment Canada (2008); Figure by CERI

Another direct impact of hydropower development is the impact on fish habitat, populations, migrations, and movements in general. Adverse impacts on fish population may stem from loss of spawning sites. A hydropower plant could also impact the movement of fish within their riparian habitat. Establishment of fish ways, allowing fish to circumvent the hydroelectric development, is one mitigation option, although it would not enable the movement of 100 percent of the fish (Legislative Assembly of Alberta, 2013). Fish sluiceways are an adaptive means that would allow the fish to move from the reservoir to downstream.

Development of hydropower plants and transmission lines would lead to complex social and ecological impacts and this study provides some high-level metrics of those impacts. Quantification of those impacts and identifying parties that would be affected is important in developing mitigation measures, facilitating constructive stakeholder deliberations, and ensuring fairness and equity.

January 2016 48 Canadian Energy Research Institute

Employment and Other Economic Development Benefits Major infrastructure projects such as the development of hydropower plants and transmission systems contribute to economic development. Construction and operational phases of such infrastructure projects creates a large number of direct and indirect employment. Significant capital spending and electricity sales revenues lead to GDP growth. Contributions to different levels of government (municipal, provincial, etc.) revenues are made in the form of taxes and grants-in-lieu. While this study did not directly estimate those benefits, project proponents and stakeholder organizations have provided estimates of those benefits.

For example, Manitoba Hydro estimates that development of the Conawapa hydropower plant would create 10,700 person-years of direct and indirect employment during the construction and operation phases. Taxes, water rental, and other fees that would be paid to the province of Manitoba are estimated at $87.2 million per year (Manitoba Hydro, 2013).

Similarly, the Site C hydropower project is estimated to create 33,000 person-years of direct and indirect employment during the construction phase and 160 person years of jobs per year during the operations phase. The project is estimated to increase the provincial GDP by $3,230 million during the construction phase and by $7 million per year in the operations phase (BC Hydro, 2013b). Water rental fees that would be paid to the province are estimated at $40.2 million/year. School taxes and grants-in-lieu paid to local governments are estimated at $2.6 million/year.

Transmission system developments also support economic developments. An assessment by Pfeifenberger and Hou (2011) surveyed a number of new transmission projects across North America and it is estimated that US$1 million invested on transmission developments in Alberta would create 10 full-time equivalent jobs during the construction phase across Canada (7 in Alberta and 3 in the rest of Canada).

The Bipole III project in Manitoba, which would construct a 1,400 km long ±500 kV HVDC transmission line, is estimated to create 14,392 person years of direct and indirect employment across Canada (8,782 person years of employment in Manitoba) during the construction phase. Operations phase employment is estimated to be 150.1 person years of direct and indirect employment per year (Manitoba Bureau of Statistics, 2011). The Bipole III construction phase is estimated to increase Manitoba’s GDP by $688.7 million and Canada’s GDP by $1,185 million. Provincial tax revenue attributable to the Bipole III project is estimated to be $352.4 million during the construction phase and $9.4 million/year during the operations phase.

Based on these facts, it can be concluded that hydropower options assessed in this study, if developed, have the potential to contribute to local, provincial, territorial, and national economic development.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 49 Oil Sands Electricity Demand Chapter 4 Discussion and Concluding Remarks

The oil sands sector in Alberta is an important economic contributor to the province and Canada. Energy-intensive extracting and upgrading of bitumen in the oil sands sector has led to high marginal costs of production and GHG emissions. With the climate change imperative, initiatives by various jurisdictions have exerted pressure on Alberta′s oil sands sector to reduce its GHG emissions. This study assessed the viability of reducing GHG emissions of oil sands operations by utilizing hydropower to satisfy its electricity demands. The study identified six options to generate and transmit hydropower to the oil sands region, taking into account hydropower developmental potential available in Alberta and neighbouring Canadian jurisdictions.

Each of these six options can deliver sufficient electricity to satisfy the demand of in-situ bitumen extraction operations with production capacity of 0.5 million bbl/day to 1.1 million bbl/day. The average cost of delivered electricity (i.e., LCOE) is in the range of $81-$162/MWh. In contrast, utilizing natural gas-fired cogeneration units to satisfy the electricity demand would cost about $57/MWh. To put these numbers in perspective, the average electricity pool price in Alberta over the last 10 years has been in the range of $48-$90/MWh (AESO, 2015a). Hence, without a price on GHG emissions, the likelihood of hydropower options reducing the marginal cost of oil sands operations is low.

As a carbon emissions mitigation option, utilizing hydropower can potentially reduce the GHG emissions of oil sands operations by 13-16 percent. The associated GHG emissions abatement cost (i.e., CACO2) is in the range of $75-$332/tCO2e. The baseline assumed for the CACO2 calculation is the aforementioned cogeneration reference case. One reason for high CACO2 estimates of hydropower options is the lower LCOE and GHG emissions intensity of cogenerated electricity.

When viewed from a different reference point such as the average Alberta grid or coal-based generation, the abatement costs change. If the hydro project were replacing grid-based electricity, the abatement costs range from $21/tCO2e to $137/tCO2e. If the hydropower options are targeted at replacing coal-based generation, the costs range from minus $3/tCO2e to $98/tCO2e. Careful consideration of the reference point is needed to properly assess the economic cost of reducing carbon dioxide emissions.

Implications of the Alberta Electricity Market Structure on Hydropower Project Financing The structure of Alberta’s electricity market has implications on investments on hydropower generation and transmission options assessed in this study. Unlike many other Canadian jurisdictions, Alberta’s electric power system is deregulated with a competitive market in operation for electric power generation. Within this structure, electricity that is not self-supplied

January 2016 50 Canadian Energy Research Institute

must be exchanged through the competitive market. The AESO coordinates and operates an energy-only market for electricity, where generators submit offers to supply electricity for each hourly period and consumers submit bids to purchase electricity. Electricity producers receive the hourly pool price (measured in $/MWh) for the electricity supplied. Electricity transmission and distribution remain regulated where remunerations are set through cost of service reviews by the Alberta Utilities Commission.

The pool price of a given hour depends on the electricity demand and generators available to satisfy it. The minimum pool price is $0/MWh and the maximum is $999/MWh. The average pool price in Alberta over the last 10 years has been in the range of $48-$90/MWh.

Within this market structure, there is no organized program for long-term firm electricity contracts, although generators and consumers can negotiate power purchase agreements. As indicated in Table 3.2, the hydropower generation transmission options assessed require massive amounts of irreversible capital investments. Some form of investment guarantee may be needed to assist with the development of a hydropower project for Alberta. Given the magnitude of capital costs and operations life of hydropower plants, it is unlikely a proponent would make investment decisions solely based on observed and forecasted pool prices in Alberta.

Alberta-British Columbia Hydropower Options The lowest LCOE and CACO2 result from purchasing hydropower from the BC Hydro system and delivering it by utilizing the existing transmission intertie between Alberta and BC (BC Intertie option). This option requires implementation of mitigation measures to enable the full capacity utilization of an Alberta-BC intertie. This option also has the advantage of being able to deliver electricity in the short- to medium-term.

The highest LCOE and CACO2 are associated with two new BC hydropower options that assume electricity produced by the Site C generating unit is delivered to the oils sands region utilizing a HVDC (Site C-DC option) or HVAC (Site C-AC option) transmission system. Of the two options, the HVAC option has the higher LCOE and CACO2.

One main factor that leads to uncertainty in the viability of importing hydroelectric power from BC is the high forecasted electricity demand growth in BC; the BC Hydro system appears to run into electrical energy and capacity shortages in the early 2020s. Therefore, it is uncertain whether the BC hydropower options would be able to export large volumes of baseload electricity into oil sands operations in Alberta.

Alberta Slave River Hydropower Options Of the new hydropower development options assessed in this study, developing a hydropower plant on the Alberta Slave River and connecting it to Fort McMurray (Slave River-DC option) through a HVDC transmission system is the most attractive option in terms of LCOE and CACO2. In contrast, the Slave River-AC option that assumes a hydropower plant on the Slave River would be connected using two single circuit HVAC lines has higher LCOE and CACO2 estimates. However, the difference between LCOE values is 9 percent and the Slave River-AC option has the

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 51 Oil Sands Electricity Demand added reliability result from maintaining two independent transmission corridors. The transmission system in the latter option, however, has higher environmental impacts due to the need for two transmission ROW.

Of all the hydropower options assessed, the Slave River-DC and Slave River-AC options would likely have the highest amount of environmental impacts. Environmental impacts would be exacerbated by the impacts on Wood Buffalo National Park and the Peace-Athabasca Delta, a wetland ecosystem with global importance. Moreover, these two options would likely have higher impacts on Indigenous populations. It should be noted that a previous attempt to develop a hydropower plant on the Slave River did not proceed as the project proponents were unable to reach an agreement with the Smith’s Landing First Nations, who opposed the project citing environmental concerns (Bell, 2010).

The process to obtain the necessary regulatory approval for the Slave River-DC and Slave River- AC options appears to be less complex than the other options as both the hydropower plant and transmission system would be sited within the Alberta provincial boundary. However, the permitting process would be complicated by potential impacts on federally protected environmentally sensitive areas and the fact that hydrological changes induced by water regulation can potentially impact the Peace River, Lake Athabasca, and the system, would concern jurisdictions both upstream (e.g., BC and Saskatchewan) and downstream (e.g., Northwest Territories).

Alberta-Manitoba Hydropower Options The Manitoba DC option that assumes the development of the proposed Conawapa hydropower plant in Manitoba and connecting it to Fort McMurray by a HVDC transmission system is the most attractive new hydropower option for Alberta hydropower developments. The Conawapa project is in the advanced planning stage and Manitoba Hydro has already completed feasibility assessments. The Conawapa project site is on the Nelson River, which has already been impounded by six upstream hydroelectric dams, minimizing additional environmental impacts.

Manitoba Hydro has extensive experience with hydropower and HVDC systems that can facilitate efficient planning, developing, and operating of the hydropower generating plant and transmission system. The Manitoba Public Utilities Commission put the project on hold until Manitoba Hydro secures electricity export opportunities. A long-term contract to export electricity to the oil sands operations in Alberta can provide the level of certainty the province needs, justifying the investment. An additional large-scale hydropower option is available on the Nelson River downstream of the Conawapa site at the Gillam Island site. Therefore, the HVDC system pertaining to the Manitoba DC option can be utilized to deliver additional electricity to Alberta from the current and future hydropower plants on the Nelson River, lowering the transmission cost as well as potentially supporting larger amounts of oil sands operations.

Siting the transmission system of the Manitoba DC option would likely be very complex in terms of obtaining the necessary regulatory approvals. The transmission system would span across three provinces and have higher impacts on the boreal forest than other transmission options

January 2016 52 Canadian Energy Research Institute

assessed in this study. Another factor that would complicate the permitting is the uncertainty in benefits to Saskatchewan. Approximately 47 percent of the transmission ROW of Manitoba DC would lie in Saskatchewan, but the province would not receive any power sales revenues or electricity to satisfy local demand. The transmission system, however, would create employment in Saskatchewan during the construction and operation phases of the transmission system. It would also contribute to Saskatchewan GDP growth as well as government revenues through taxes (e.g., linear property taxes).

Another uncertainty about the viability of the Manitoba DC option stems from other potential export markets Manitoba Hydro is pursuing. Historically, Manitoba exports about 30 percent of its electricity production; more than 85 percent of that total is exported to electricity utilities in the US Midwest region. Manitoba’s electricity export capability – determined by transmission interconnections – is almost four times that of export capability to neighbouring Canadian provinces. Manitoba Hydro coordinates its operations with the Midwestern US electricity market operated by the Midcontinent Independent Systems Operator (MISO). The MISO market size, in terms of peak demand, is 10 times that of Alberta.

As indicated in its long-term development plan, those factors have led to Manitoba Hydro pursuing greater electricity export opportunities in the MISO region, instead of Canadian markets such as Alberta (Manitoba Hydro, 2013). Therefore, potential development of the Manitoba DC would require additional financial or policy considerations for this project to move forward. An arrangement between Manitoba and Alberta would demonstrate interprovincial cooperation as noted in the Canadian Energy Strategy (2015).

Electrical Extraction Technologies as a Carbon Management Option Deploying electrical extraction technologies and utilizing hydroelectric power as the electricity supply would lead to much larger emissions reductions for bitumen production. However, higher electricity demand for electrical extraction technologies would lower the bitumen extraction potential of the hydropower options assessed in this report. A preliminary economic assessment conducted through scenario analysis showed that the economic feasibility of this option is tied to the electricity intensity of the extraction process. Further research is required to gain insights into the conditions that would make this option economically viable.

Long-term Planning Hydropower generation and transmission options assessed in this study have the ability to reduce GHG emissions of oil sands operations by decarbonizing the electricity consumed for bitumen extraction and upgrading. Furthermore, those hydropower options have the ability to contribute to provincial, territorial, and national economic development by creating employment and increasing GDP.

Development of the hydropower plants and transmission systems require massive amounts of irreversible capital investments. Therefore, the viability of implementation of any of the hydropower options assessed in this study depends on providing the project developers the certainty of capital cost recovery. Moreover, development of these hydropower plants would

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 53 Oil Sands Electricity Demand lead to environmental and socioeconomic impacts that extend beyond the jurisdictions in which they would be sited. Therefore, greater levels of inter-jurisdiction coordination and stakeholder consultation is vital for the successful implementation of any of these hydropower generation and transmission options.

January 2016 54 Canadian Energy Research Institute

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 55 Oil Sands Electricity Demand Appendix A Social and Environmental Impacts within the Study Area

Social and environmental impacts of the respective hydropower options within the full study area (i.e., the area formed by a combination of a 50 km wide buffer that encloses the selected transmission line corridor and a circular buffer with a 5 km radius that encloses the hydropower plant) are presented in Figures A.1-A.4.

Figure A.1: Land Cover of Respective Study Areas

Source: CERI

January 2016 56 Canadian Energy Research Institute

Figure A.2: Residential and Property Value Impacts

Source: CERI

Figure A.3: Impacts on Indigenous Populations

Source: CERI

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 57 Oil Sands Electricity Demand

Figure A.4: Protected Areas and Water Bodies within the Respective Study Areas

Source: CERI

January 2016 58 Canadian Energy Research Institute

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 59 Oil Sands Electricity Demand Appendix B Calculation of Energy Cost and Emissions of Bitumen Extraction through SAGD

In order to estimate the cost of energy and CO2 emissions associated with bitumen extraction through SAGD, an illustrative case example is developed. In this example, a SAGD facility with a bitumen production capacity of 30,000 bbl/day is modeled. Electricity and thermal energy demands of the facility are satisfied by a natural gas-fired cogeneration system and a supplemental boiler (see Figure B.1). Excess electricity produced by the cogeneration system is assumed to be exported to the Alberta electricity market and costs and emissions associated with the exported electricity are subtracted from the bitumen production cost and emissions.

Figure B.1: Energy System of the SAGD Facility

Source: CERI

January 2016 60 Canadian Energy Research Institute

Where,

FT = Fuel input to gas turbine FG = Fuel input to HRSG FB = Fuel input to supplementary boiler Pc = Electricity production PE = Exported electricity P = Onsite electricity consumption for bitumen extraction (16.5 kWh/bbl) HC = Enthalpy of the steam produced by the cogeneration system (HRSG) Hcfw = Enthalpy of the HRSG feed water HB = Enthalpy of the steam produced by the supplementary boiler Hbfw = Enthalpy of the supplementary boiler feed water H = Onsite thermal energy consumption for bitumen extraction (1.3 GJ/bbl) T = Gas turbine electricity generation efficiency (30%) R = HRSG heat recovery efficiency (50%) G = HRSG supplemental firing efficiency (95%) B = Supplementary boiler efficiency (85%) (All efficiencies are higher heating value basis)

Fuel and emissions attributable to electricity are calculated as follows (Doluweera et al., 2011):

퐹 +퐹 −(퐻 −퐻 )⁄휂 퐹푢푒푙 푎푡푡푟𝑖푏푢푡푎푏푙푒 푡표 푒푙푒푐푡푟𝑖푐𝑖푡푦, 퐹퐶퐸 = 푇 퐺 퐶 푐푓푤 퐵 푃퐶 퐸푚𝑖푠푠𝑖표푛푠 푎푡푡푟𝑖푏푢푡푎푏푙푒 푡표 푒푙푒푐푡푟𝑖푐𝑖푡푦, 퐼푒푙 = 퐹퐶퐸 × 퐶푂2 푒푚𝑖푠푠𝑖표푛푠 𝑖푛푡푒푛푠𝑖푡푦 표푓 푛푎푡푢푟푎푙 푔푎푠

The average cost of thermal energy is assumed to be that of a natural gas-fired boiler with the same attributes as the supplemental boiler and are calculated as follows:

퐵표𝑖푙푒푟 퐶퐶 × 푐푐푓+푓𝑖푥푒푑 표푝푒푟푎푡𝑖푛푔 푐표푠푡+푎푛푛푢푎푙 푓푢푒푙 푐표푠푡 퐴푣푒푟푎푔푒 푐표푠푡 표푓 푡ℎ푒푟푚푎푙 푒푛푒푟푔푦, 퐶 = 푡ℎ 퐴푛푛푢푎푙 푡ℎ푒푟푚푎푙 푒푛푒푟푔푦 푝푟표푑푢푐푡𝑖표푛

Average cost of electricity of the cogeneration system is calculated after adjusting for thermal energy production as follows:

퐴푣푒푟푎푔푒 푐표푠푡 표푓 푒푙푒푐푡푟𝑖푐𝑖푡푦, 퐶푒푙 = 퐶표푔푒푛 퐶퐶 × 푐푐푓+푓𝑖푥푒푑 표푝.푐표푠푡+푎푛푛푢푎푙 푓푢푒푙 푐표푠푡−퐶 × 푎푛푛푢푎푙 푡ℎ푒푟푚푎푙 푒푛푒푟푔푦 푝푟표푑푢푐푡𝑖표푛 푡ℎ 퐴푛푛푢푎푙 푒푙푒푐푡푟𝑖푐𝑖푡푦 푝푟표푑푢푐푡𝑖표푛

Average energy cost and emissions per barrel of bitumen are calculated as follows:

퐴푣푒푟푎푔푒 푒푛푒푟푔푦 푐표푠푡 푝푒푟 푏푏푙 = (퐶표푔푒푛 퐶퐶+푆푢푝.푏표𝑖푙푒푟 퐶퐶)×푐푐푓+푐표푔푒푛 푓𝑖푥푒푑 표푝.푐표푠푡+푠푢푝.푏표𝑖푙푒푟 푓𝑖푥푒푑 표푝.푐표푠푡+ ( ) 푎푛푛푢푎푙 푐표푔푒푛 푓푢푒푙 푐표푠푡+푎푛푛푢푎푙 푠푢푝.푏표𝑖푙푒푟 푓푢푒푙 푐표푠푡− 퐶푒푙×푎푛푛푢푎푙 푒푙푒푐푡.푒푥푝표푟푡푠 퐴푛푛푢푎푙 푏𝑖푡푢푚푒푛 푝푟표푑푢푐푡𝑖표푛 (퐹 +퐹 +퐹 −퐹퐶퐸×푃 ) ×퐶푂2 푒푚𝑖푠푠𝑖표푛푠 𝑖푛푡푒푛푠𝑖푡푦 표푓 푛푎푡푢푟푎푙 푔푎푠 퐴푣푒푟푎푔푒 퐶푂2 푒푚𝑖푠푠𝑖표푛푠 푝푒푟 푏푏푙 = 푇 퐺 퐵 퐸 퐵𝑖푡푢푚푒푛 푝푟표푑푢푐푡𝑖표푛

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 61 Oil Sands Electricity Demand

The main parameter assumptions and results of the illustrative example are listed in Table B.1.

Table B.1: Main Parameters and Results of the Illustrative Case Example

Parameter Value Assumptions Bitumen production capacity 30,000 bbl/day Steam to oil ratio 3 Thermal energy demand of bitumen extraction 1.3 GJ/bbl Electricity demand of bitumen extraction 16.5 kWh/bbl Annual production capacity utilization 85% Cogeneration system Electricity production capacity 85MW Steam production capacity 918 GJ/h Heat to power ratio 3 Electricity generation efficiency, T 30% HRSG heat recovery rate, R 50% HRSG supplemental firing efficiency, G 95% Total capital requirement $2071/kW Fixed operating costs $14/kW/y Supplementary boiler Steam production capacity 707 GJ/h Efficiency 85% Total capital requirement $17,240/(GJ/h) Fixed operating cost $0.05/GJ Natural gas price $4/GJ Natural gas carbon emissions intensity 53 kgCO2e/GJ Capital charge factor 10.61% Results Average cost of electricity $55/MWh Average cost of thermal energy $3.9/GJ Average energy cost per bbl of bitumen $5.9/bbl Emissions attributable to electricity 325 kgCO2/MWh Average emissions per bbl 67 kgCO2/bbl Source: CERI

January 2016 62 Canadian Energy Research Institute

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 63 Oil Sands Electricity Demand Appendix C Uncertainty Analysis of LCOE and CACO2

An uncertainty analysis of LCOE and CACO2 estimates is conducted by developing a Monte Carlo simulation (MCS). For the MCS, capital costs (both generation and transmission systems), capacity factors, natural gas price, and electricity price at the Mid-C electricity hub are sampled from their distributions. Each parameter is assumed to be uniformly distributed within the identified uncertainty range as follows:

 Capital cost of power plants o Cogeneration, Site C (Site C-DC/Site C-AC), Conawapa (Manitoba DC): -15 percent to +30 percent of the base case estimates (Table 3.2) o Slave River (Slave River-DC/Slave River-AC): -20 percent to +50 percent of the base case estimates (Table 3.2)1  Capital cost of transmission system: -15 percent to +30 percent of the base case estimates  Capacity factor: -30 percent to +30 percent of the base case estimates  Natural gas price: $3/GJ to $8/GJ  Mid-C electricity price: $26/MWh to $83/MWh2

Each parameter is sampled 100,000 times and the LCOE and CACO2 are calculated. Cumulative probability distributions of LCOE of each option are depicted in Figure C.1. Cumulative probability distributions of LCOE are depicted in Figure C.2. Similarly, cumulative probability distributions of CACO2 are depicted in Figure C.3. The following metrics are calculated using MCS results and are listed in Table C.1.

 LCOE and CACO2 estimates that correspond to cumulative probability of 50 percent (henceforth referred to as P50 estimates). P50 estimates of a given metric is the value that the relevant metric, based on 50 percent probability, will not exceed  Probability of CACO2 being less than $100/tCO2e and $200/tCO2e (i.e., P(CACO2<100) and P(CACO2<200)

1 Slave River capital cost is varied over a wider range (-15% to 50%) as the project is in the exploratory stage and reliable capital cost estimates are not available. 2 The range used for the MCS is the monthly average peak period electricity prices observed in the Mid-C exchange over the period 2009-2014

January 2016 64 Canadian Energy Research Institute

Figure C.1: Monte Carlo Simulation Results: Distributions of LCOE

Notes: The boxes depict the plausible range of LCOE of different electricity generation and transmission options when the uncertainty associated with the main parameters are taken into account. In a given box, the bottom and top edges of the boxes correspond to the 25th and 75th percentile, respectively. The solid horizontal line inside the box corresponds to the median. The whiskers extend to the maximum/minimum plausible value that is within 1.5 times the interquartile range from the top/bottom edge of the box. Outlier data points are marked as ‘o’. Source: CERI

Figure C.2: Monte Carlo Simulation Results: Cumulative Distributions of LCOE

Notes: Cumulative distribution of a given option indicates the likelihood of LCOE estimates when the parameter uncertainty is taken into account. The dotted horizontal line corresponds to P50 estimate of LCOE. MCS results reiterate the fact that, without a price on carbon, the likelihood of the hydropower options yielding a lower LCOE than the cogeneration reference option is low. Source: CERI

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 65 Oil Sands Electricity Demand

Figure C.3: Monte Carlo Simulation Results: Cumulative Distributions of CACO2

Notes: The dotted horizontal line corresponds to P50 estimate of CACO2. When the uncertainty associated with major parameters is taken into account, the CACO2 of the Slave River-DC option (Slave River hydropower plant with HVDC transmission) and Manitoba DC option (Conawapa hydropower plant with HVDC transmission) are indiscernible. A similar result was observed for LCOE for those two options. Source: CERI

Table C.1: Selected Metrics of Uncertainty Assessment Results Metric Reference Site C- Slave Manitoba Site C- Slave BC Case DC River-DC DC AC River-AC Intertie P50 estimates LCOE ($/MWh) 63 146 127 128 168 139 100

CACO2 ($/tCO2e) 264 201 204 334 237 118 P(CACO2<100) (%) 0 3 0 0 0 17 P(CACO2<200) (%) 15 49 48 0 30 100 Source: CERI

The MCS results show that, under the assumed conditions, without a price on carbon emissions, the likelihood of LCOE of the cogeneration option being higher than the hydropower option is negligible. Site C-DC and Site C-AC options continue to be the most expensive options and the likelihood of either of the options having lower LCOE or CACO2 than other options is negligible.

Under point estimates (Table 3.2), of all hydropower options, the BC Intertie option had the lowest LCOE and CACO2. It is the only hydropower option that has a non-negligible probability (17 percent) of having a CACO2 less than $100/tCO2e. However, when the uncertainty associated with the parameters is taken into account, there is about a 10 percent probability that the Slave River-DC option would have lower LCOE and CACO2 than the BC Intertie.

Another observation is that as far as the distributions of LCOE and CACO2 are concerned, the Slave River-DC and Manitoba DC options are indiscernible. The probability of one being preferred

January 2016 66 Canadian Energy Research Institute

over the other is approximately 50 percent. Furthermore, the difference between P50 estimates of LCOE and CACO2 of the two options is only about 1 percent and 1.5 percent, respectively.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 67 Oil Sands Electricity Demand Bibliography

ACFN. (2015). Athabasca Chipewyan First Nation pulls out of Grand Rapids Hearings: Citing Prejudiced Process that favors Industry. Retrieved from https://acfnchallenge.wordpress.com/2014/07/15/athabasca-chipewyan-first-nation- pulls-out-of-grand-rapids-hearings-citing-prejudiced-process-that-favors-industry/ AER. (2015). Alberta’s Energy Reserves 2014 and Supply/Demand Outlook 2015–2024 (No. ST- 98). Alberta Energy Regulator. Retrieved from http://www.aer.ca/documents/sts/ST98/ST98-2015.pdf AESO. (2012). Long-term Transmission Plan. Alberta Electric System Operator. Retrieved from http://www.aeso.ca/downloads/AESO_2012_LTP_Sections_1.0_to_5.0.pdf AESO. (2014a). 2014 Long-term Outlook. Alberta Electric System Operator. Retrieved from http://www.aeso.ca/downloads/AESO_2014_Long-term_Outlook.pdf AESO. (2014b). Northeast Region Plan. Retrieved from http://www.aeso.ca/transmission/30376.html AESO. (2015a). 2014 Annual Market Statistics. Alberta Electric System Operator. Retrieved from http://www.aeso.ca/downloads/2014_Annual_Market_Stats_WEB.pdf AESO. (2015b). Intertie Restoration Project: AESO - BCH Joint Planning Study. Alberta Electric System Operator. Retrieved from http://www.aeso.ca/market/20238.html Alberta Government. (2015). Alberta Economic Dashboard. Retrieved from http://economicdashboard.albertacanada.com/ AltaLink. (2012). 2013-2014 General Tariff Application (Volume 1). Retrieved from http://www.altalink.ca/files/pdf/about/Volume-1-Application.pdf Anderson, N. (2013, September). Not-so-Clean Hydropower is Damming Us All. Retrieved June 5, 2015, from http://voices.nationalgeographic.com/2013/09/25/not-so-clean- hydropower-is-damming-us-all/ Apt, J., Keith, D. W., & Morgan, G. (2007). Promoting Low-Carbon Electricity Production. Issues in Science and Technology, 23(3). Retrieved from http://search.proquest.com/openview/c1411cfcb4b6766cc65db8282c2924e2/1?pq- origsite=gscholar ATCO Power, & TransCanada. (2009). Slave River Hydro Development: Overview and Current Status. Presented at the NWT Chamber of Commerce AGM. Retrieved from http://www.nwtchamber.com/docs/SRHD__Presentation_-_NWT_Chamber_AGM.pdf BC Hydro. (2008). Revelstoke/Mica Sequence Identification Project. Retrieved from https://www.bchydro.com/content/dam/hydro/medialib/internet/documents/info/pdf /info_revelstoke_mica_sequence_identification_project_com.pdf BC Hydro. (2013a). Integrated Resource Plan: 2013 Resource Options Report Update. Retrieved from https://www.bchydro.com/energy-in- bc/meeting_demand_growth/irp/document_centre/reports/2013-ror-update.html BC Hydro. (2013b). Site C Clean Energy Project Environmental Impact Statement: Executive Summary. Retrieved from https://www.sitecproject.com/sites/default/files/site_c_eis_executive_summary_amen ded_%28july_2013%29.pdf

January 2016 68 Canadian Energy Research Institute

Bell, S. (2010, November). No dam on Slave River. Slave River Journal. Retrieved from http://www.printfriendly.com/print/n?url=http%3A%2F%2Fsrj.ca%2Fno-dam-on-slave- river-p5411-207.htm Bergerson, J. A., Kofoworola, O., Charpentier, A. D., Sleep, S., & MacLean, H. L. (2012). Life Cycle Greenhouse Gas Emissions of Current Oil Sands Technologies: Surface Mining and In Situ Applications. Environmental Science & Technology, 46(14), 7865–7874. http://doi.org/10.1021/es300718h Black & Veatch. (2014). Capital Costs for Transmission and Substations: Updated Recommendations for WECC Transmission and Expansion Planning. Western Electricity Coordinating Council. Retrieved from https://www.wecc.biz/Reliability/2014_TEPPC_Transmission_CapCost_Report_B+V.pdf BP. (2015). BP Statistical Review of World Energy. Retrieved from http://www.bp.com/en/global/corporate/about-bp/energy-economics/statistical- review-of-world-energy/statistical-review-downloads.html Capital Power. (2011). Keephills 3 power plant begins commercial operation. Retrieved from http://www.capitalpower.com/MediaRoom/newsreleases/2011-news- releases/Pages/090111.aspx CAPP. (2015a). 2015 CAPP Crude Oil Forecast, Markets & Transportation. Canadian Association of Petroleum Producers. Retrieved from http://capp.ca/publications and statistics/publications/264673 CAPP. (2015b). CAPP’s Statistical Handbook is a historical summary of the petroleum industry’s progress. Canadian Association of Petroleum Producers. Retrieved from http://www.capp.ca/publications-and-statistics/statistics/statistical-handbook CARB. (2015). Low Carbon Fuel Standard Program. California Air Resources Board. Retrieved from http://www.arb.ca.gov/fuels/lcfs/lcfs.htm CBC News. (2010). Slave River hydro project nixed. Retrieved June 5, 2015, from http://www.cbc.ca/1.962503 CERI. (2011). Economic Impacts of Staged Development of Oil Sands Projects in Alberta (No. 125). Canadian Energy Research Institute. Retrieved from http://www.ceri.ca/images/stories/2011-08-24_CERI_Study_125_Section_1.pdf Charpentier, A. D., Kofoworola, O., Bergerson, J. A., & MacLean, H. L. (2011). Life Cycle Greenhouse Gas Emissions of Current Oil Sands Technologies: GHOST Model Development and Illustrative Application. Environmental Science & Technology, 45(21), 9393–9404. http://doi.org/10.1021/es103912m Doluweera, G. H., Jordaan, S. M., Moore, M. C., Keith, D. W., & Bergerson, J. A. (2011). Evaluating the role of cogeneration for carbon management in Alberta. Energy Policy, 39(12), 7963–7974. http://doi.org/10.1016/j.enpol.2011.09.051 Droitsch, D., & Simieritsch, T. (2010). Canadian Aboriginal Concerns With Oil Sands. Pembina Institute. Retrieved from https://www.pembina.org/reports/briefingnoteosfntoursep10.pdf Englander, J. G., Brandt, A. R., Elgowainy, A., Cai, H., Han, J., Yeh, S., & Wang, M. Q. (2015). Oil Sands Energy Intensity Assessment Using Facility-Level Data. Energy & Fuels, 29(8), 5204–5212. http://doi.org/10.1021/acs.energyfuels.5b00175

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 69 Oil Sands Electricity Demand

Environment Canada. (2008). Scientific Review for the Identification of Critical Habitat for Woodland Caribou, Boreal Population (Rangifer tarandus caribou) in Canada. Ottawa: Environment Canada. Retrieved from http://www.sararegistry.gc.ca/document/default_e.cfm?documentID=1761 Environment Canada. (2015a). National Inventory Report: 1990 - 2013 Greenhouse Gas Sources and Sinks in Canada. Environment Canada. (2015b). Species at Risk Public Registry. Retrieved October 30, 2015, from http://www.sararegistry.gc.ca/default.asp?lang=En&n=24F7211B-1 Environment Canada. (2015c). Water Survey of Canada. Retrieved from https://www.ec.gc.ca/rhc-wsc/default.asp?lang=En&n=4EED50F1-1 Gavriovic, A. (2003). HVDC. In Electrical Engineers Reference Book (16th ed., pp. 32/1–32/46). Elsevier. Retrieved from http://www.helitavia.com/books.htm Gosselin, P., Naeth, A., Plourde, A., Therrien, R., Van Der Kraak, G., & Xu, Z. (2010). Environmental and Health Impacts of Canada’s Oil Sands Industry. Royal Society of Canada. Retrieved from http://www.rsc-src.ca/en/expert-panels/rsc- reports/environmental-and-health-impacts--oil-sands-industry Government of Alberta. (2015). Climate Leadership Plan. Retrieved from http://alberta.ca/climate/leadership-plan.cfm Government of the Northwest Territories. (2011). Northwest Territories Energy Report. Retrieved from http://www.assembly.gov.nt.ca/sites/default/files/11-05-20td36- 166.pdf HATCH. (2010). Update on Alberta’s Hydroelectric Energy Resources. Alberta Utilities Commission. Henton, D. (2013). Hydro power not so cheap anymore, experts say. Retrieved August 31, 2015, from http://www.calgaryherald.com/technology/Hydro+power+cheap+anymore+experts/79 29301/story.html Hunter, J. (2015). B.C.’s Site C dam still waiting on permits delayed by group. Retrieved June 1, 2015, from http://www.theglobeandmail.com/news/british-columbia/bcs-site-c- dam-still-waiting-on-permits-delayed-by-treaty-8-group/article24717382/ IEA. (2014). Energy Technology Perspectives 2014 - Harnessing Electricity’s Potential. International Energy Agency. Retrieved from http://www.iea.org/publications/freepublications/publication/energy-technology- perspectives-2014---executive-summary---english-version.html IEA ETSAP. (2010). Hydropower. IEA Energy Technology Systems Analysis Program. Retrieved from http://www.etsap.org/E-techDS/PDF/E07-hydropower-GS-gct.pdf Irving, J. (2013). The Canadian Hydropower Renaissance. Presented at the 44th Annual H.G. Acres Seminar, Niagra Falls. Kammen, D. M., & Pacca, S. (2004). Assessing the Costs of Electricity. Annual Review of Environment and Resources, 29(1), 301–344. http://doi.org/10.1146/annurev.energy.28.050302.105630 Keeney, R. L., & Gregory, R. S. (2005). Selecting Attributes to Measure the Achievement of Objectives. Operations Research, 53(1), 1–11. http://doi.org/10.1287/opre.1040.0158

January 2016 70 Canadian Energy Research Institute

Kelly, E. N., Short, J. W., Schindler, D. W., Hodson, P. V., Ma, M., Kwan, A. K., & Fortin, B. L. (2009). Oil sands development contributes polycyclic aromatic compounds to the Athabasca River and its tributaries. Proceedings of the National Academy of Sciences, 106(52), 22346–22351. http://doi.org/10.1073/pnas.0912050106 Larkins, D. (2015, April). Hydro kept building Conawapa: Tories. Retrieved October 29, 2015, from http://www.winnipegsun.com/2015/04/01/hydro-kept-building-conawapa- tories?utm_source=facebook&utm_medium=recommend- button&utm_campaign=Hydro kept building Conawapa: Tories Legislative Assembly of Alberta. (2013). Review of the Potential for Expanded Hydroelectric Energy Production in Northern Alberta. Standing Committee on Resource Stewardship,. Manitoba Bureau of Statistics. (2011). Bipole III Transmission Project Economic Impact Assessment Technical Report. Retrieved from https://www.hydro.mb.ca/projects/bipoleIII/eis/BPIII_Economic_Impact_Assessment_T echnical_Report_November%202011.pdf Manitoba Hydro. (2013). Needs for and Alternatives To. Retrieved from https://www.hydro.mb.ca/projects/development_plan/nfat_business_case.shtml McCarthy, S. (2014, April 18). U.S. once again delays decision on Keystone XL pipeline. The Globe and Mail. Retrieved from http://www.theglobeandmail.com/report-on- business/industry-news/energy-and-resources/us-once-again-delays-decision-on- keystone-xl-pipeline/article18066497/ McGee, B. (2012). E-T DSP Technical Presentation. Presented at the Calgary Technical Event: Canadian Prairies Group pf Chartered Engineers. Retrieved from http://www.e- tenergy.com/media/files/upload/CPGCE-2012_cqn_khf.pdf McLachlan, S. (2014). Environmental and Human Health Implications of the Athabasca Oil Sands for the Mikisew Cree First Nation and Athabasca Chipewyan First Nation in Northern Alberta. Retrieved from http://onerivernews.ca/health-study-press-release-2014/ Murillo, C. (2015). Oil Sands Industry Energy Requirements and Greenhouse Gas (GHG) Emissions Outlook (2015-2050). Canadian Energy Research Institute. Retrieved from http://www.ceri.ca/images/stories/Study_151_-_Oil_Sands_Energy_Requirements_- _August_2015.pdf Natural Resources Canada. (2014). CanVec+ Feature Catalogue. Retrieved from http://ftp2.cits.rncan.gc.ca/pub/canvec+/ Natural Resources Canada. (2015). GeoGratis (API) - Product Search. Retrieved from http://geogratis.gc.ca/api/en Nduagu, E. I., & Gates, I. D. (2015). Unconventional Heavy Oil Growth and Global Greenhouse Gas Emissions. Environmental Science & Technology, 49(14), 8824–8832. http://doi.org/10.1021/acs.est.5b01913 NETL. (2008). Recommended Project Finance Structures for the Economic Analysis of Fossil- Based Energy Projects (No. DOE/NETL-401/090808). US National Energy Technology Laboratory. Retrieved from http://www.netl.doe.gov/energy- analyses/pubs/Project%20Finance%20Parameters%20-%20Final%20Report%20- %20Sept%202008_1.pdf OSDG. (2013). 2013 Oil Sands Co‐generation and Connection Report. The Oil Sands developers Group.

January 2016 An Assessment of Hydroelectric Power Options to Satisfy 71 Oil Sands Electricity Demand

PADEMP. (2015). Peace-Athabasca Delta Ecological Monitoring Program. Retrieved from http://pademp.com/research-and-monitoring/ Patterson, C. (2015). An Overview of Oil Recovery Using Radio Frequency Heating Technology: The Tools, Techniques and Processes Behind the ESEIEH Hydrocarbon Extraction Process. Presented at the Canadian Heavy Oil Conference. Pew Environment Group. (2012). A Forest of Blue: Canada’s Boreal. Retrieved from http://borealscience.org/wp-content/uploads/2012/06/report-forestofblue.pdf Pfeifenberger, J., & Hou, D. (2011). Employment and Economic Benefits of Transmission Infrastructure Investment in the U.S. and Canada. WIRES and The Brattle Group. Retrieved from http://www.wiresgroup.com/docs/reports/Brattle_WIRES_JobsStudy_May2011.pdf PNNL. (2013). Evaluating greenhouse gas emissions from hydropower complexes on large rivers in Eastern Washington. Pacific Northwest National Laboratory. Retrieved from http://www.pnl.gov/main/publications/external/technical_reports/PNNL-22297.pdf PRISM Economics and Analysis. (2015). Hydropower and the Canadian Economy: Jobs and Investments in Canada’s Largest Electricity Source. Canadian Hydropower Association. Rubin, E. S. (2012). Understanding the pitfalls of CCS cost estimates. International Journal of Greenhouse Gas Control, 10, 181–190. http://doi.org/10.1016/j.ijggc.2012.06.004 SaskPower. (2010). Powering A Sustainable Energy Future: Wind Power and Hydroelectric Opportunities in Saskatchewan. Retrieved from http://www.emtfsask.ca/presentations/saskatoon/10-06-02-Mohr.pdf Schively, C. (2007). Understanding the NIMBY and LULU Phenomena: Reassessing Our Knowledge Base and Informing Future Research. Journal of Planning Literature, 21(3), 255–266. http://doi.org/10.1177/0885412206295845 Simens. (2013). High Voltage Direct Current Transmission – Proven Technology for Power Exchange. Retrieved from http://www.siemens.com/about/sustainability/pool/en/environmental- portfolio/products-solutions/power-transmission- distribution/hvdc_proven_technology.pdf SRSC. (1980). Slave River Hydro feasibility Study - Final Report. Slave River Streering Committee (SRSC) for the Government of Alberta. Stantec. (2009). Assessment and Analysis of the State Of-the-Art Electric Transmission Systems with Specific Focus on High-Voltage Direct Current (HVDC), Underground or Other New or Developing Technologies. Alberta Energy. Retrieved from http://www.energy.alberta.ca/electricity/pdfs/transmissionsystemsstudy.pdf Statistics Canada. (2012). 2011 Census - Boundary files. Retrieved from https://www12.statcan.gc.ca/census-recensement/2011/geo/bound-limit/bound-limit- 2011-eng.cfm Statistics Canada. (2015). CANSIM Table 127-0007 Electric power generation, by class of electricity producer. Retrieved from http://www5.statcan.gc.ca/cansim/a26 Struzik, E. (2013). Canada’s Great Inland Delta: A Precarious Future Looms (Yale Environment 360). Retrieved from http://e360.yale.edu/feature/canadas_great_inland_delta_precarious_future_looms/27 09/

January 2016 72 Canadian Energy Research Institute

Suncor Energy, & Jacobs Consultancy. (2012). A Greenhouse Gas Reduction Roadmap for Oil Sands. Retrieved from http://ccemc.ca/wp-content/uploads/2012/12/C101221-CCEMC- GHG-Reduction-Roadmap-Final-Report.pdf Vajjhala, S. P., & Fischbeck, P. S. (2007). Quantifying siting difficulty: A case study of US transmission line siting. Energy Policy, 35(1), 650–671. http://doi.org/10.1016/j.enpol.2005.12.026 von Meier, A. (2006). Electric Power Systems: A Conceptual Introduction. Wiley.

January 2016