ECONOMIC AND ENVIRONMENTAL IMPACTS OF TRANSITIONING TO A CLEANER ELECTRICITY GRID

Transcription

1 Study No. 174 CANADIAN ENERGY RESEARCH INSTITUTE ECONOMIC AND ENVIRONMENTAL IMPACTS OF TRANSITIONING TO A CLEANER ELECTRICITY GRID IN WESTERN CANADA Canadian Energy Research Institute Relevant Independent Objective

2

3 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada i ECONOMIC AND ENVIRONMENTAL IMPACTS OF TRANSITIONING TO A CLEANER ELECTRICITY GRID IN WESTERN CANADA

4 ii Canadian Energy Research Institute Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada Authors: With contributions from: Ganesh Doluweera Hossein Hosseini Evar Umeozor Duncan Lucas Ammar Hyder ISBN Copyright Canadian Energy Research Institute, 2018 Sections of this study may be reproduced in magazines and newspapers with acknowledgement to the Canadian Energy Research Institute Printed in Canada Front cover photo courtesy of Google images Acknowledgements: The authors of this report would like to extend their 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 and Megan Murphy. The authors would also like to thank the following individuals and institutions for providing data and helpful insights for this study. Responsibility for any errors, interpretations, or omissions lies solely with CERI. Dr. Matthew Ayres, University of Calgary Mr. Nick Martin, Canada West Foundation Mr. Amir Motamedi, Alberta Electric System Operator Mr. Kevin Gawne, Manitoba Hydro Mr. Leonard Olien, Solas Energy Consulting Inc. Mr. Sanjay De Soyza, BC Hydro Dr. Joule Bergerson, University of Calgary ABOUT THE CANADIAN ENERGY RESEARCH INSTITUTE CANADA S VOICE ON ENERGY Founded in 1975, the Canadian Energy Research Institute (CERI) is an independent, registered charitable organization specializing in the analysis of energy economics and related environmental policy issues in the energy production, transportation and consumption sectors. Our mission is to provide relevant, independent, and objective economic research of energy and environmental issues to benefit business, government, academia and the public. For more information about CERI, visit CANADIAN ENERGY RESEARCH INSTITUTE 150, Street NW Calgary, Alberta T2L 2A6 info@ceri.ca Phone:

5 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada Table of Contents iii LIST OF FIGURES... LIST OF TABLES... ACRONYMS AND ABBREVIATIONS... EXECUTIVE SUMMARY... CHAPTER 1 INTRODUCTION... 1 Study Scope and Objectives... 1 Scenario Descriptions... 3 Carbon Pricing... 3 Emissions Reduction Target... 4 Renewable Electricity Targets... 5 Phase-out of Coal-fired Power Plants... 5 Interprovincial Coordination... 5 Provincial Electric Power Systems... 6 CHAPTER 2 METHODOLOGY Electric Power System Planning and Operations Simulation Model Model Objective Model Constraints Representation of Generating Units Model Outputs Main Assumptions and Parameters Demand Data Renewable Resource Data Electricity Trade with the United States Capital Costs and Technology Learning Cycling Costs Installed Capacity Fuel Prices Energy Policy and Climate Change Policy-related Parameters Current Interprovincial Transmission Intertie Capacities CHAPTER 3 RESULTS AND DISCUSSION Electricity Supply and GHG Emissions Cost of Power System Operations The Value of Interprovincial Coordination Operational Challenges Value of Flexible Generation Technologies CHAPTER 4 CONCLUSIONS BIBLIOGRAPHY APPENDIX A HYBRID CARBON PRICING: OUTPUT-BASED ALLOCATION v vii ix xi

6 iv Canadian Energy Research Institute

7 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada List of Figures v E.1 Electricity Generation in Western Canada in 2030 and xii 1.1 Alberta Installed Capacity Alberta Electricity Generation in British Columbia Installed Capacity British Columbia Electricity Generation in Manitoba Installed Capacity Total Energy Generation in Manitoba Saskatchewan Generation Capacity, Annual Electricity Demand Forecast Hourly Demand Forecast Developed for the Study Average Wind Power Availability by Season in the Four Western Provinces Average Solar Power Availability by Season in the Four Western Provinces Natural Gas Price Forecast Used for the Analysis Installed Generation Capacity under Different Scenarios by Technology Annual Electricity Generation under Different Scenarios by Technology in Different Operating Periods Total GHG Emissions from Electricity Generation under Different Scenarios GHG Intensity of Electricity Generation under Different Scenarios Power System Costs System Dispatch in Alberta System Dispatch in Saskatchewan... 47

8 vi Canadian Energy Research Institute

9 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada List of Tables vii E.1 Summary of Scenario Results... xiii 1.1 Study Scenarios British Columbia s International Trade with the United States British Columbia s Interprovincial Trade Manitoba s International Trade with the United States Manitoba s Trade with Ontario Learning Rates for the Technology Sources Costs and Heat Rates Typical Lower Bound Costs of Cycling and Impacts on Equivalent Forced Outage Rates Assumed Lifetime of Different Technologies Fuel Prices Used for the Analysis Carbon Pricing System Parameters GHG Emissions Cap Current Intertie Capacities Average Residential Cost of Electricity The Value of Interprovincial Coordination in Western Canada A.1 Assumed Benchmark Emissions Intensities and Carbon Prices A.2 Carbon Cost/Credits for Different Technologies Over the Period

10 viii Canadian Energy Research Institute

11 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada Acronyms and Abbreviations ix AB AESO BC BEI CAD CCIR CCMP-NC CCMP-WC CCS CERI CO2e COE CTG DGHG-NC DGHG-WC ECCC EFOR FOM gco2e/tco2e GHG GW/GWh IPPs kw/kwh MB MW/MWh NEB NGCC NGCC-CCS NGCogen NGSC O&M OBA PV SK VOM Alberta Alberta Electric System Operator British Columbia Benchmark emissions intensity Canadian dollar Carbon Competitiveness Incentive Regulation Current carbon management plan with no coordination Current carbon management plan with higher coordination among provinces Carbon capture and storage Canadian Energy Research Institute Carbon dioxide equivalent Cost of electricity Coal-to-gas converted unit Deep GHG reduction with no coordination Deep GHG reduction with higher coordination among provinces Environment and Climate Change Canada Forced Outage Rates Fixed operating and maintenance cost Grams/tonnes of carbon dioxide equivalent Greenhouse Gas Emissions Gigawatt/Gigawatt-hour Independent Power Producers kilowatt/kilowatt-hour Manitoba Megawatt/Megawatt-hours National Energy Board Natural Gas Combined Cycle Natural Gas Combined Cycle with Carbon Capture and Storage Natural Gas Cogeneration Natural Gas Simple Cycle Operations and maintenance Output-based allocation Photovoltaic Saskatchewan Variable operating and maintenance cost

12 x Canadian Energy Research Institute

13 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada Executive Summary xi With the Pan-Canadian Framework on Clean Growth and Climate Change, Canada has set an ambitious goal to reduce greenhouse gas (GHG) emissions to mitigate global climate change while sustaining economic growth. Guided by the Pan-Canadian Framework, Canadian provinces are acting through the implementation of programs and regulations to reduce GHG emissions in respective economies. Central to the GHG emissions reduction plan is reducing GHG emissions from the electric power sectors of Canadian provinces. A multitude of policies and programs are being implemented to reduce GHG emissions from electricity generation. The purpose of this CERI study is to assess economic impacts and GHG emissions under alternative pathways of transforming the electricity grid in the four western provinces: British Columbia, Alberta, Saskatchewan, and Manitoba. The alternative pathways are primarily characterized by energy and climate change policies that are already adopted or could potentially be adopted by the provincial and federal governments. In this study, the pathways are developed into five scenarios. These scenarios consist of a reference scenario (Reference), two carbon pricing scenarios (CCMP-NC and CCMP-WC), and two carbon emissions cap scenarios (DGHG-NC and DGHG-WC). Two scenarios in each of the latter two categories have the same policy targets but differ by whether provinces coordinate with each other by increasing electricity trade (CCMP- WC and DGHG-WC) or no higher level of coordination among provinces is sought (CCMP-NC and DGHG-NC). Higher coordination scenarios assume doubling of current intertie capacities. This study focuses on the period CERI has developed an electric power system investment planning and economic dispatch simulation model to conduct the underlying analysis in this report. Figure E.1 shows the combined electric power generation mix of the four western provinces under each scenario in five-year periods starting in 2030 and As can be seen from the figure, in most cases the dominant power generation options in the western grid are hydropower (in British Columbia and Manitoba) and the natural gas-fired technologies (in Alberta and Saskatchewan). Under the Reference scenario, the share of electricity produced by natural gasfired technologies in Alberta rises to 99% by This is due to the variable nature of renewable electricity generation and low natural gas prices. Despite the dominance of natural gas, both the total GHG emissions and emissions intensities will be lower than 2005 levels in Alberta and Saskatchewan due to the retirement of coal-fired generation. Under all scenarios, power generation in Manitoba and BC will be dominated by hydropower. Some natural gas-fired generating units get added to these two provinces to compensate for variations in seasonal hydropower supply.

14 xii Canadian Energy Research Institute Figure E.1: Electricity Generation in Western Canada in 2030 and 2050 Under the CCMP-NC & CCMP-WC scenarios (carbon pricing scenarios) the generation mix of Alberta and Saskatchewan notably changes. The share of renewable energy in Alberta s generation mix increases from 16% in 2020 to 30% in 2030 and after. In Saskatchewan, the renewable share increases from 17% in 2020 to 26-35% after The increase in renewable share is driven mainly by renewable energy targets as opposed to the credits provided by the carbon pricing systems. Nonetheless, renewables receive incentives under the carbon pricing system. The dominant renewable energy technology in Alberta is wind power. By , solar PV and geothermal energy start competing with wind for the 30% renewable energy target set by Alberta. The two emissions cap scenarios (DGHG-NC and DGHG-WC) lead to transformative changes in the electricity generation mix in Alberta and Saskatchewan after The deep GHG emissions cap of 80% drives conventional natural gas-fired generation away from the generation mix. Up to 2040, the capacity and generation mix in the two provinces remains close to those under the carbon pricing scenarios but changes significantly after that. In both Alberta and Saskatchewan, a large amount of renewable energy gets added to the generation mix under the carbon cap scenarios. By 2050, as the carbon cap becomes more stringent, the power system in Alberta will require a low carbon flexible generation source. Consequently, natural gas combined cycle units with carbon capture and storage and geothermal

15 Economic and Environmental Impacts of Transitioning to a Cleaner Electricity Grid in Western Canada xiii energy enter the generation mix. Under these scenarios, the share of electricity from renewable sources and CCS units reached over 75% in both Alberta and Saskatchewan by Table E.1 presents some key metrics pertaining to the four provincial electric power systems under each scenario. As can be seen from Table E.1, compared to the Reference scenario, the average cost of electricity is higher when different climate change policies are enforced (i.e., carbon pricing, renewable targets, and carbon caps). There are few exceptions. Under carbon pricing scenarios, credits received by hydropower in BC and Manitoba lowers the average cost of electricity in those two provinces. The highest cost increase compared to the Reference scenario is observed under the carbon cap scenarios. Table E.1: Summary of Scenario Results Emissions intensity (gco2e/kwh) Relative GHG emissions (% of 2005 GHG emissions from electricity generation) Average cost of electricity for residential customers (cents/kwh) Reference CCMP-NC CCMP-WC DGHG-NC DGHG-WC Province AB BC MB SK AB 57% 73% 46% 52% 43% 52% 47% 20% 44% 21% BC 38% 72% 38% 98% 53% 55% 38% 20% 53% 6% MB 93% 221% 91% 211% 82% 128% 62% 20% 5% 26% SK 59% 63% 42% 52% 41% 52% 50% 20% 50% 19% AB BC MB SK In the higher coordination scenarios, provinces increase the intertie capacities to increase electricity trade. The intended benefit is reducing the overall cost by sharing resources and providing system balancing services. In the carbon cap scenarios, when provinces have higher coordination, a combined carbon cap is set for the whole western region. The intention here is to achieve higher emissions reductions where it is cheaper to do so. In scenarios where provinces have higher coordination, the total cost is lower compared to the scenario with the same policy goals. As such, the value of coordination for carbon pricing scenarios is CAD$1,691 million. For the carbon cap scenarios, the value of coordination is estimated to be CAD$1,812 million. The effectiveness of coordination to reduce overall cost is constrained by the amount of investment needed to enhance the intertie capacities. Both sets of policy scenarios (carbon pricing scenarios and carbon cap scenarios) can achieve the intended goal of reducing GHG emissions while not significantly increasing the average cost of

16 xiv Canadian Energy Research Institute electricity. In addition to the main results summarized here, the report also provides insights into renewable energy regimes and impacts of operating a power system with the higher amount of variable generating sources.

17 Economic and Environmental Impacts of Transitioning to a 1 Cleaner Electricity Grid in Western Canada Chapter 1: Introduction Electricity has a pivotal role in achieving economy-wide deep emissions reductions. It is a highly versatile form of energy and converting electricity into end-use energy services can be done at high efficiencies. In addition, commercially proven technologies exist for example, renewable energies, nuclear power to produce electricity with low or zero GHG emissions. More centralized ownership, control, and prior experience with regulations have made the emissions management in the electricity sector more manageable. As such the electricity sector is a primary sector of focus for carbon management in all countries, and Canada is no exception. In fact, in 2015, global investment in coal and gas-fired power generation plants fell to less than half that in renewable electricity generation technologies, in a record year for clean energy. It was the first time that renewable energy made up a majority of all the new electricity generation capacity under construction around the world, and the first year in which the financial investment by developing countries in renewables outstripped that of the developed world (The Guardian 2016). Examples of current and proposed GHG emissions reduction policies in Canadian jurisdictions include carbon pricing, phase-out of coal-fired generation plants, and a mandated minimum level of renewable energy integration. These policies will have diverse impacts on electricity systems in Canadian provinces regarding overall cost, system reliability, and level of emissions reductions achievable. Transitioning to an electricity system where renewable and cleaner generation technologies dominate production would inevitably have economic impacts resulting from new investments, stranded assets, and changes to energy markets. On the other hand, some Canadian provinces already have very low emissions-intensive electricity systems. Therefore, higher levels of coordination and electricity trade among provinces can potentially contribute to overall emissions reductions. It is important that these factors are investigated at a systems level to inform decision makers. The cost associated with transitioning to a cleaner generation system stem from several factors. New investments in power generation systems required to transition into an electricity system where renewable and cleaner generation technologies dominate energy supply may not be the least-cost option. Additional infrastructure, such as new transmission systems needed to connect site-constrained generation, may add to the cost. There may be operational level impacts that would lead to higher costs and emissions. For example, variable renewable energy sources may lead to cyclical operation of other generating units in the generation fleet, deviating them from the economically and technically optimal operating conditions. Additional standby and ramping reserves may also be required to meet reliability standards. Assessment of these factors requires power systems-level analysis, focusing on both investments and operational constraints. Study Scope and Objectives The main objective of this study is to assess the economic impacts and GHG emissions under alternative pathways of transforming the electricity grid in the four western provinces: British

18 2 Canadian Energy Research Institute Columbia, Alberta, Saskatchewan, and Manitoba. The alternative pathways are primarily characterized by energy and climate change policies that are already adopted or could potentially be adopted by the provincial and federal governments. In this study, the pathways are developed into five scenarios that are listed in Table 1.1. We provide insights into the implications of each of these scenarios answering the following questions: What are the total costs of electric system operations and system-level average costs of electricity including generation investments, operations, systems support services, transmission costs, and distribution costs? What are the GHG emission levels (both total emissions and emissions per unit of electricity generated)? What is the value of a higher level of inter-provincial coordination regarding electricity trade and aligned climate change policy? All scenarios assume the retirement of the coal-fired generation fleet in Alberta by Saskatchewan s coal units are assumed to operate until the end of service life (i.e., 50 years). Only Alberta (30% of the energy by 2030) and Saskatchewan (50% of the installed capacity by 2030) have renewable energy targets. Table 1.1: Study Scenarios Scenario Name Scenario Description Carbon Pricing GHG Reduction Target (reference year = 2005) Renewable Energy Targets Interprovincial Transmission Reference Reference case None None None Current levels CCMP -NC Current carbon management plan OBA system with a carbon levy None Provincial Targets Current levels CCMP-WC Current carbon management plan with higher coordination among provinces OBA system with a carbon levy None Provincial Targets Doubling of current intertie capacity DGHG-NC Deep GHG reduction None 30% by 2030, 80% by 2050 Provincial Targets Current levels DGHG-WC Deep GHG reduction with higher coordination among provinces None 30% by 2030, 80% by Shared emissions target Provincial Targets Doubling of current intertie capacity.

19 Economic and Environmental Impacts of Transitioning to a 3 Cleaner Electricity Grid in Western Canada The study focuses on the period CERI has developed an electric power system expansion and operation planning model to conduct the underlying analysis in this report. The project answers the above questions by simulating the power system operations in the four provinces using this model. This report is organized as follows. The remainder of this chapter provides a description of the five scenarios and a concise description of the power systems of the four provinces that are assessed in the study. Chapter 2 describes the CERI electric power system expansion and operation planning model. The general model framework and underlying assumptions are also described in Chapter 2. The results of the analysis are presented and discussed in Chapter 3. Chapter 4 concludes. Scenario Descriptions Table 1.1 shows the five scenarios that this study evaluates. These scenarios are characterized by energy policy and climate policy goals. The ambitions of Canada s current climate change mitigation plan is set by the Government of Canada s Pan-Canadian Framework on Clean Growth and Climate Change (ECCC 2016). The framework sets the stage to meet Canada s commitments for the Paris Agreement. While the Pan-Canadian Framework sets the end policy goals, each province is given the flexibility to set the regulations and policies to reach those goals. In the framework, there are several action plans that target the electricity sector. These include carbon pricing, phasing-out coal-fired electricity, an adaptation of renewable electricity technologies, and improving energy efficiency. Some provinces have already implemented policies and regulations (CERI 2018). Actions by the four western provinces are described in the latter part of this chapter. It should be noted that the provinces of Alberta and British Columbia have carbon pricing systems that predate the Pan-Canadian Framework. Here, we discuss the main policy objectives that are used to develop the five scenarios assessed in this study. Carbon Pricing Carbon pricing is a key regulatory tool stipulated in the Pan-Canadian Framework. While the provinces are given the flexibility to set their carbon pricing system, the federal government has announced the implementation of a carbon backstop pricing system that will be enforced on the provinces that would not implement a carbon pricing system (ECCC 2017a). Under the federal carbon backstop, carbon pricing on electricity is expected to take a hybrid approach composed of a carbon levy and an output-based allocation (OBA) pricing system (this pricing system is henceforth referred to as an OBA system). The output-based standard will be set at a level that represents best-in-class performance to drive reduced emissions intensity in the electricity sector. In provinces and territories where the federal carbon pricing system has been implemented, the regulated facilities would be assigned an annual limit on the total quantity of greenhouse gases they can emit before the carbon price is applied. The allowed emissions would be set regarding greenhouse gas emissions per unit of product for various types of industrial activities (i.e., benchmark emissions intensity (BEI)). If a facility emits more than the limit, it would have to pay the carbon levy for the excess emissions

20 4 Canadian Energy Research Institute (i.e., the carbon levy is compliance obligation times the carbon price). If a facility reduces its emissions below its limit, it could earn revenues by selling credits to other facilities that exceed their limits (ECCC 2018b). It is evident that for a given generation unit, the compliance obligation will be determined by the benchmark intensity and the carbon price. The provinces are provided with the flexibility to design and implement their own carbon pricing system. However, it is reasonable to assume that a pricing system set by a province will not deviate significantly from the minimum levels required by the federal regulations. The province of Alberta has announced the implementation of a hybrid carbon pricing system that is closely aligned with the proposed federal system. The province has set the emissions accounting and compliance options under the Carbon Competitiveness Incentive Regulation (CCIR) (Government of Alberta 2017b; Province of Alberta 2017). Under the CCIR, electricity generation is subjected to carbon pricing. However, only emissions beyond the established baseline, which reduces annually, are subjected to carbon pricing. In two scenarios (referred to as current climate management plan (CCMP) scenarios), we assume the carbon pricing system is implemented in all four provinces. In developing those scenarios, we assume the carbon pricing system under the CCIR for Alberta. For the other three provinces, we assume that a pricing system similar to the federally-announced OBA system is implemented in the outlook period. However, the benchmark emission intensity is assumed to remain constant, unlike Alberta s benchmark that becomes more stringent over time. Emissions Reduction Targets The most recent major international climate change mitigation agreement that Canada has ratified is the Paris Agreement whose main objective is to strengthen the global response against climate change by keeping a global temperature rise in the current century below 2 degrees Celsius above pre-industrial levels. As announced by a joint federal-provincial declaration (the Vancouver Declaration), Canada is to undertake joint efforts to reduce GHG emissions by 30 percent below 2005 levels by 2030 to meet or exceed the Paris agreement commitments. Canada s 2050 emissions reduction targets are set at 80% below 2005 levels. However, the federal action plan on climate change mitigation does not explicitly require any of the economic sectors including the electric power sector to reduce GHG emissions by the levels above. Carbon pricing and other complementary measures are expected to drive emissions down to achieve those emissions reduction levels. In two of the scenarios (referred to as deep GHG reduction (DGHG) scenarios), we gain insights into the implications of requiring the electricity power sectors of all four provinces to reduce GHG emissions by 30% below 2005 levels by 2030 and 80% below 2005 levels by 2050.

21 Economic and Environmental Impacts of Transitioning to a 5 Cleaner Electricity Grid in Western Canada Renewable Electricity Targets Large-scale adaptation of renewable electricity generation technologies is a policy that has been implemented by many jurisdictions to reduce GHG emissions from the electricity sector. Of the four western provinces, Alberta and Saskatchewan have set explicit renewable energy targets. Alberta requires a minimum of 30% of electricity produced in the province be sourced from renewable sources by 2030 (AESO 2018c). It should be noted that this is a minimum energy target. In contrast, Saskatchewan has set an installed capacity target by requiring a minimum of 50% of installed electricity generation capacity be from renewable electricity technologies by 2030 (Saskpower 2018). In both provinces, existing renewable electricity generating units including hydropower can be counted towards meeting the target. These respective targets are enforced in our four scenarios. It is worth mentioning that Canada s electricity grid is relatively clean; 65% of its electricity is procured from renewable energy sources, and over 80% of all electricity comes from non-ghg emitting sources (NRCan 2018). Total hydropower installed capacity is 78 gigawatts (GW) (NRCan 2016). Two of the western provinces, British Columbia and Manitoba, produce more than 90% of electricity through hydropower. British Columbia s Clean Energy Act requires at least 93% of the electricity in British Columbia to come from clean or renewable resources. This requirement was maintained in the carbon pricing and carbon cap scenarios. Manitoba has not announced any explicit renewable energy targets. As such, we do not enforce renewable energy targets in Manitoba in any of the scenarios. Phase-out of Coal-fired Power Plants Phasing-out of coal-fired electricity by 2030 is a main regulatory plan enforced by the federal government to reduce GHG emissions from the electricity sector (Government of Canada 2015). Coal-fired electricity is the dominant power generation technology (regarding energy generation) in Alberta and Saskatchewan. Under the Climate Leadership Plan, Alberta has already set the course to retire its coal-fired electricity fleet by Therefore, under all five scenarios, we assume that Alberta retires its coal-fired generation fleet. However, the coal units can convert to operate with natural gas as it is allowed under provincial and federal regulations. On the other hand, Saskatchewan has made arrangements with the federal government to operate the provinces coal fleet beyond 2030 (Saskpower 2018). The provinces have offered to implement alternative plans to reduce an equivalent amount of GHG emissions reductions. As such, we allow the coal-fired generating units in Saskatchewan to operate until they reach maximum service life, which is assumed to be 50 years. Interprovincial Coordination A higher level of interprovincial coordination is potentially a way to reduce overall electric power system operation cost as well as the cost of GHG emissions reductions. By a higher level of coordination, we refer to increased electricity trade among provinces, and shared climate change mitigation goals. In the case of the electricity sector, the ability to increase electricity trade is

22 6 Canadian Energy Research Institute limited by the electricity transmission intertie capacity among provinces. In three scenarios (Reference, CCMP-NC and DGHG-NC), the capacity of the interprovincial electricity interconnections is assumed to remain at current levels. In two scenarios (referred to as with higher coordination (WC) scenarios ) we assume doubling of electricity intertie capacities. The five scenarios whose implications are assessed in this study are summarized below and in Table 1.1. The Reference scenario refers to a no climate or renewable energy policy future. In this case, we assume that there are no GHG emissions reduction targets or carbon pricing system is adopted. Similarly, there are no renewable electricity targets. The capacity of interprovincial electricity interconnections is assumed to remain at current levels. The CCMP-NC scenario assumes implementation of carbon pricing and where applicable, provincial renewable energy targets. However, no higher level of interprovincial coordination and therefore, the capacity of interprovincial electricity interconnections are assumed to remain at current levels. The CCMP-WC scenario is like CCMP-NC, but a higher level of coordination is achieved through a doubling of intertie capacities. The GHG-NC scenario assumes implementation of emissions reduction targets in each of the four electric power systems. However, to isolate the impact of reduction targets, no carbon pricing is enforced. DGHG-WC is like DGHG-NC, but it assumes doubling of intertie capacities. Furthermore, DGHG-NC sets a combined GHG emissions target for the western provinces. In this case, there is no emission reduction target for an individual province, but the combined emissions should be 30% below by 2030 and 80% below by 2050 compared to combined GHG emissions from power generation in 2005 in the four provinces. A reader may question the validity of the Reference scenario as the current provincial and federal policies are such that carbon pricing and some renewable energy targets are enforced. Our intention of using a no climate policy reference scenario is to test whether cleaner technologies with no GHG emissions would be competitive without policy intervention. Provincial Electric Power Systems The electric power systems of the four provinces assessed in this study vary regarding ownership structure, capacity, current generation system, and resource availability to produce electricity, and applicable climate change regulations and goals. This section provides a concise introduction to the four power systems. Alberta Alberta began the process of restructuring its electricity industry to an energy-only market design in 1996, with full retail and wholesale competition established by 2001 (Olmstead and Ayres 2014). In an energy-only market, electricity generators are compensated for the energy services they provide in hourly markets. In this setting, generators rely on high prices that occur when supply conditions are tight to recover their large fixed costs of investing in a generation facility. This scarcity pricing is the primary instrument to motivate capacity investment to facilitate competition in hourly energy markets and ensure a reliable supply of electricity (Brown 2018).

23 Economic and Environmental Impacts of Transitioning to a 7 Cleaner Electricity Grid in Western Canada On November 23, 2016, the provincial government announced that Alberta would transition to a capacity market model instead of requiring generators to rely solely on the energy-only market for their revenue streams. Rather than paying generators only for the energy they produce, a capacity market pays companies both for the capacity they could offer the market, even when their facilities are not operating, plus the price they receive for the electricity they generate when operational (Saric, Carson, and Bachmann 2017). Currently, Alberta s generation fleet (16,390 MW capacity) is dominated by natural gas-fired generation (7,599 MW or 46 percent of the total installed capacity) and coal-fired generation (6,003 MW or 37 percent of capacity), as shown in Figure 1.1 and Table 1.2. In recent years, net capacity additions have been dominated by natural gas-fired facilities as well as notable growth in wind-fueled capacity (Olmstead and Ayres 2014). Figure 1.1: Alberta Installed Capacity 46% 37% 9% 5% 3% Coal Hydro Wind Others Natural Gas Source: (AESO 2018b). Total installed capacity at the beginning of 2018 was 16,390 MWh.

24 8 Canadian Energy Research Institute Figure 1.2: Alberta Electricity Generation in % 36% 4% 3% 1% Coal Hydro Wind Others Natural Gas Source: (AESO 2018b). Total generation in 2017 was 82,572 GWh. Alberta s electricity industry has faced substantial changes in government policies and economic forces in recent years. Alberta GHG emissions totalled 274 Mt in 2014, with 16.2% of emissions coming from the electricity sector (Government of Alberta 2017c). On January 15, 2018, the Government of Canada released legislative and regulatory proposals for the new Greenhouse Gas Pollution Pricing Act (Government of Canada 2018). The proposals consider a federal carbon pricing system that includes two key elements: 1) a charge on fossil fuels that would be paid by fuel producers or distributors, and 2) an output-based pricing system (OBPS) for industrial facilities with high levels of emissions (ECCC 2017b). Approximately 96% of GHG emissions from the electricity generation (i.e., 15.5% of total GHG emissions in Alberta) were regulated under the Specified Gas Emitters Regulation (SGER) (in effect in 2007, expired December 31, 2017) (Government of Alberta 2017d). Output-based Allocation (OBA) is a part of the Carbon Competitiveness Incentive Regulation (AB Regulation 255/2017) under the Climate Change and Emissions Management Act (Government of Alberta 2017d, 2017c) that replaces the SGER. The OBA applies to: facilities that are large emitters over 100,000 tonnes sectors and facilities that qualify to opt-in (sectors that are emissions-intensive and trade exposed or like for like (the facility competes directly with another facility that is regulated under the CCIR))

25 Economic and Environmental Impacts of Transitioning to a 9 Cleaner Electricity Grid in Western Canada In 2016, the Government of Alberta announced several policies that increase the stringency of environmental regulation on carbon emissions, mandated a phase-out of over 6,250 MW of coal generation capacity by 2030, and also announced a target to integrate 5,000 MW of renewable generation capacity by 2030 via the Renewable Electricity Program (Government of Alberta 2018a, 2018b). The Renewable Electricity Program (REP) is intended to encourage the development of large-scale renewable electricity generation to support the Government of Alberta s target of 30 percent renewable electricity by As outlined in the Renewable Electricity Act, the REP will encourage the development of 5,000 MW of renewable electricity generation by These policies, alongside the anticipated retirement of Alberta s entire coal fleet by 2030, which represents 38 percent of generation capacity and over 50 percent of electricity generation in Alberta, compound the reliability concerns of the electricity system in Alberta (AUC 2017). British Columbia British Columbia has 17,302 MW of installed capacity (see Figure 1.3). Most of the electricity generated in BC comes from hydroelectric sources, a mix of large and small, run-of-river and storage. Most hydroelectric installations are owned and operated by BC Hydro, including all the largest generating stations like Revelstoke, Mica, and GM Shrum. In total, BC Hydro owns 32 hydro facilities as well as two natural gas plants (BC Hydro 2017). Combined, BC Hydro has a total installed capacity of about 12,052 MW, generating on average 46,000 GWh of energy each year and services about 1.8 million residential accounts and 200,000 commercial and industrial accounts in (BC Hydro 2017, 2018c). About 99% of BC Hydro s capacity by nameplate rating is hydro. Alongside BC Hydro, there are 124 Independent Power Producers (IPPs) currently operating in BC. They sell some or all their output to BC Hydro for distribution. They have a combined capacity of 5,250 MW, about 30% of all installed capacity in the province, and sell 21,425 GWh per year to BC Hydro for distribution (BC Hydro 2018b). Many are storage or non-storage hydro, ranging from less than 0.1 MW to the 900 MW Kemano plant in Kitimat. IPPs also include biomass, wind, biogas, ERG, and fossil fuel generators. Figures 1.3 and 1.4 show the installed capacity of all generating stations in BC by fuel, as well as the energy produced for the grid (it does not include energy produced purely for industrial users). Because of the dominance of hydro in British Columbia, there are relatively little greenhouse gas emissions resulting from electricity generation. BC is on track to continue investing in renewable generation through hydropower. The largest generating project in development is Site C on the Peace River in northeastern BC, near the town of Fort St. John. Expected to be completed in 2024, this 1,100 MW storage hydro facility will generate approximately 5,100 GWh of electricity per year (BC Hydro 2018d). Currently, the total cost is estimated at $8.8 billion. Another major project, an upgrade of the Revelstoke generating station, will add another 500 MW of capacity from storage hydro. This is expected to be completed in 2026.

26 10 Canadian Energy Research Institute Figure 1.3: British Columbia Installed Capacity 8% 0% 5% 4% 0% Biomass Wind Biogas Hydro 83% Thermal Other Source: (BC Hydro 2018b, 2018c), Total installed capacity in 2017 was 17,302 MW Figure 1.4: British Columbia Electricity Generation in 2017 (67,113 GWh) 5% 0% 6% 3% 0% Biomass Wind Biogas 86% Hydro Thermal Other Source: (BC Hydro 2018b 2018c) Independent Power Producers (IPPs) are expanding their own hydro and wind operations. Wind operations have seen strong recent growth in BC. In the last ten years, 670 MW of wind generation has been installed in 5 major plants and 30 MW from various smaller sites (all are IPPs). Another 30 MW of wind and 116 MW of storage and non-storage hydro is currently under development by IPPs (BC Hydro 2018a).

27 Economic and Environmental Impacts of Transitioning to a 11 Cleaner Electricity Grid in Western Canada Currently, there are no geothermal generating stations in BC. While some parties have expressed interest in geothermal, and some sites have been identified with the potential for hundreds of MW of capacity, no plants are being developed right now. The most developed site so far is the South Meager Geothermal Project, north of Pemberton. Some exploratory drilling was carried out in the mid-2000s, but no generating station is in place. Regarding electricity trade, BC is one of the most active provinces, both importing and exporting. Most trade is with the United States and the rest with Alberta. BC s more recent international trade (all with the US) is shown in Tables 1.2 and 1.3. Table 1.2: British Columbia s International Trade with the United States Year Total Exports (GWh) Total Exports (millions $CAD) Total Imports (GWh) Total Imports (millions $CAD) Source: (National Energy Board 2018a) Table 1.3: British Columbia s Interprovincial Trade Source: (CERI 2017) Year Total Exports (GWh) Total Imports (GWh) BC s carbon management regulations include both direct pricing of GHG emissions (Carbon Tax Act) and non-pricing regulations and technology-specific policies, such as the Greenhouse Gas Reduction Targets Act (2007), Greenhouse Gas Emission Reporting Regulation (2016), and Climate Leadership Plan (2016). The Greenhouse Gas Reduction Targets Act (passed in November 2007; came into force January 2008) sets aggressive legislated targets for reducing greenhouse gases. Under the Act, BC's GHG emissions are to be reduced by at least 33% below 2007 levels by A further emission reduction target of 80% below 2007 levels is set for (Province of British Columbia 2007). The Carbon Tax Act (passed in May 2008, came into force July 2008). BC s revenue-neutral carbon tax puts a price on greenhouse gas emissions, providing an incentive for choices that produce fewer emissions. The escalating tax was phased-in on July 1, The tax covers around 70% of BC s total GHG emissions. Carbon tax rates started at $10/tonne of CO2e emissions in 2008, increasing by $5/tonne each year until reaching the current rate of $30/tonne of CO2e emissions

28 12 Canadian Energy Research Institute in The BC Government is currently taking a new approach to carbon pricing and climate action, increasing carbon tax rates annually by $5 per tonne of CO2e emissions beginning April 1, 2018 ($35), until rates reach $50 per tonne in 2021 (Government of BC n.d.). The Greenhouse Gas Emission Reporting Regulation (enacted on January 1, 2016) requires that industrial operations that emit 10,000 tco2e/year (and those that have emitted more than 10,000 tonnes in any of the previous three years) report their GHG pollution each year. Operations emitting over 25,000 tco2e/year are required to have their emission reports independently verified. Compliance reporting requirements for regulated operations are prescribed. Types of operations that may have an obligation to report include electricity generation and co-generation (thermal), electricity transmission (SF6 emissions) and electricity import operation. Specifically, related to renewable electricity generation, the Clean Energy Act (came into force June 2010) sets provincial energy objectives and mechanisms related to electricity selfsufficiency, clean and renewable energy, energy efficiency, greenhouse gas emission reductions and fuel switching to lower carbon-intensity energy (Government of BC 2017). The Act requires at least 93% of the province s electricity be generated by clean or renewable resources. The Clean Energy Act also requires that by 2020, at least 66% of BC Hydro s incremental power demand must be met through conservation and efficiency improvements (Government of BC 2016) Manitoba Manitoba Hydro owns and operates all 15 hydro generating stations in Manitoba, one natural gas, one combined natural gas/coal plant, and four diesel generating stations for remote communities with overall transmission system comprising 18,500 kilometres of lines (Manitoba Hydro 2017b, 2018c). Total installed capacity in Manitoba is about 6,000 MW of which around 90% is hydro (see Figure 1.5). Therefore, very little of the capacity of the natural gas or coal generating stations are used, and about 97% of energy is produced from hydro as shown in Figure 1.6 (National Energy Board 2018b; Natural Resources Canada 2018). The two large wind farms in the province are owned by private entities, who sell all their output to Manitoba Hydro for distribution. Also, there are some small-scale biomass operations that are privately owned.

29 Economic and Environmental Impacts of Transitioning to a 13 Cleaner Electricity Grid in Western Canada Figure 1.5: Manitoba Installed Capacity (Total: 5,962 MW) 5% 1% 0% 6% Hydro Wind 88% Natural Gas Coal Diesel Source: (Manitoba Hydro 2017) Figure 1.6: Total Energy Generation in Manitoba Source: (National Energy Board 2018b)

30 14 Canadian Energy Research Institute Regarding trade, since Manitoba generates much more electricity than it consumes, it can export large amounts of energy to the United States. Sales through trade are one of the major justifications for new generation and transmission projects. They have trade connections with Saskatchewan and Ontario as well. Tables 1.4 and 1.5 show Manitoba s electricity trade volumes with the United States and with Ontario. Table 1.4: Manitoba s International Trade with the United States Year Total Exports (GWh) Total Exports (CAD$ millions) Total Imports (GWh) Total Imports (CAD$ millions) Source: (National Energy Board 2018a) Source: (IESO 2018) Table 1.5: Manitoba s Trade with Ontario Year Total Exports (GWh) Total Imports (GWh) Despite the overwhelming use of hydroelectricity, Manitoba Hydro looks to be continuing its investment in renewable energy. They converted the Selkirk coal generator to natural gas in 2002, and plan to retire their last coal generator in No new thermal generating stations are planned. The next major generating station will be Keeyask, a hydro storage station on the Lower Nelson River. Expected to be completed in 2021, it will have a capacity of 695 MW, generating on average 4,400 GWh per year (Manitoba Hydro 2018b). Another large project, the Conawapa generating station, would be a 1,485 MW storage hydro facility. If developed, the project has the potential to generate 7,000 GWh of electricity per year. The project is on hold until changes in the electricity market to justify its construction (CBC News 2014). Also, the Bipole III 500 kv HVDC transmission line is being constructed to add redundancy to the transmission system and will add an extra 2,000 MW of line capacity to accommodate development like Keeyask and Conawapa (Manitoba Hydro 2018a). It is expected to be completed in 2018, at the cost of $4.9 billion. To accommodate extra trade with the US, Manitoba Hydro has also proposed a new transmission line to the Minnesota border. Manitoba unveiled its Made-in-Manitoba Climate and Green Plan on October 27, A central component of the plan is to introduce a $25 per tonne carbon levy in 2018, frozen at that rate until 2022 an amount that the government notes are half that mandated by the federal

31 Economic and Environmental Impacts of Transitioning to a 15 Cleaner Electricity Grid in Western Canada government and will eventually be the second-lowest in Canada. Manitoba claims that cumulative reductions under this approach will be higher than outcomes achieved under the Pan- Canadian benchmark. The proposed approach will be more than the benchmark price until 2020; however, Manitoba s approach will fall short of the federal benchmark between 2020 and 2022 (Government of Manitoba 2017; Flanagan et al. 2017). Saskatchewan Saskatchewan Power Corporation (SaskPower) has been the leading energy supplier for the province since 1929 and today serves 528,000 local customers and manages $11 billion in generation, transmission, and distribution. With a generation capacity of 3,542 MW of electricity, SaskPower accounts for 80% of the electricity produced in the province. IPPs are responsible for the additional 949 MW. Saskatchewan s record power generation sits at 24,374 GWh. Distribution of electricity in Saskatchewan is maintained by 159,000 kilometres of power lines over the course of 195 distribution stations. SaskPower currently owns and operates a total of 18 power generating stations across the province. By fuel type, these can be broken down into five hydropower stations, three coal-fired stations, six natural-gas power stations, and two wind facilities. SaskPower also has two wholly-owned subsidiaries NorthPoint Energy Solutions and SaskPower International. NorthPoint regularly trades electricity in markets outside of Saskatchewan. SaskPower International is involved as joint operator of Cory Cogeneration Station and Cory Cogeneration Funding Corporation, alongside its investment in MRM Cogeneration Station in Alberta (Saskpower 2017). In the pursuit of supplying increasingly higher volumes of electricity generated from renewable sources, SaskPower buys power from a variety of IPPs. SaskPower also trades with three naturalgas fired power stations. Between the Sunbridge, Morse, and Red Lily Wind Energy facilities, SaskPower has begun the process of acquiring an additional 200 MW of wind generation. All in all, SaskPower s dealings with IPPs amount to a total net generating capacity of 808 MW. Amongst the numerous long-term power purchase agreements that SaskPower is involved with, one of the more notable ones is the Manitoba Hydro Northern Power Purchase Agreement. The Manitoba Hydro Northern Power Purchase Agreement is a 20-year agreement set to commence in 2020 involving 100 MW of renewable hydroelectricity to be sold to SaskPower from Manitoba Hydro. To ensure uninterrupted energy deliveries, a new 80 km 230 kv transmission line will have to be constructed between Birtle (MB) and Tantallon (SK). SaskPower has also previously entered in a purchase agreement involving $100M worth of electricity, starting in November 2015 and running through May With regards to international trade, NorthPoint Energy Solutions exported a total of 197,835 MWh of energy through the period January-December A further breakdown from the National Energy Board reveal the recipients of SaskPower s exports; 41,542 MWh to Minnesota

32 16 Canadian Energy Research Institute and 156,293 MWh to North Dakota. Additionally, SaskPower also exports 184,358 MWh of nonrevenue energy (Saskpower 2017). With 24 power stations 17 of which are fully owned by SaskPower Saskatchewan s electricity mix is primarily sourced through natural gas, coal, and hydropower (4,491 MW). As indicated in Figure 1.7, natural gas and coal account for nearly three-quarters of the electricity generated in the province. Figure 1.7: Saskatchewan Generation Capacity, % 1% 20% 40% 34% Natural Gas Coal Hydro Wind Other Source: (Saskpower 2017) Wind energy is growing in importance and is continuing to contribute higher proportions in the generation mix. Furthermore, a competitive process for Saskatchewan s first 10 MW utility-scale solar project has been initiated, which is expected to add 60 MW to the electricity system by 2021, in alignment with a renewables strategy implemented in the province. With regards to coal-fired electricity generation, by 2030, SaskPower will be required to retire Units #4, #5, and #6 at Boundary Dam Power Station and Units #1 and #2 at Poplar River Power Station to adhere to regulatory requirements. The regulations state that units commissioned before 1975 will reach the end of their useful life on December 31 st of the 50 th year after their commissioning date. Alternatively, SaskPower has the option to invest in cleaner technologies that will allow the units above to remain within 420 tonnes of CO2 per GWh generated, as specified by the regulations. This will allow them to continue operation of all existing units (Saskpower 2017). SaskPower has agreed to the re-siting of a 177 MW wind project from an area near Chaplin to a location at Blue Hills in southwestern Saskatchewan. SaskPower aims to reduce its greenhouse gas emissions to approximately 40% below 2005 levels. A continued plan to increase renewable

33 Economic and Environmental Impacts of Transitioning to a 17 Cleaner Electricity Grid in Western Canada electricity generation up to 50% of total capacity has been implemented to achieve this goal. Reflecting the nature of gas-fired generation and renewable integration, it is anticipated that the province s emissions profile will slightly rise until 2020, after which it is expected to fall dramatically. Recently, SaskPower had reached the milestone of capturing nearly 1.5 million tonnes of CO2 at the Boundary Dam Integrated Carbon Capture and Storage (CCS) Demonstration Project since its startup. In 2014, the power station became the first in the world to successfully use CCS technology. At a glance, the Boundary Dam Unit #3 can power 100,000 Saskatchewan homes, and successfully reduces SO2 emissions by up to 100%, and CO2 emissions by 90%. SaskPower had invested $47M at Boundary Dam in refurbishing its coal fleet (Saskpower 2017). Overall, the province is considering increasing electricity generation capacity from 4,500 MW (2017) to 7,000 MW (2030) to respond to the province s growing demand for electricity (Saskpower 2018). It is planning to double the percentage of renewable electricity from 25% of overall capacity (2015) to 50% by 2030, with 30% coming from wind power (The Globe and Mail 2016) and 60 MW of solar generation to be added by 2021 (Saskpower 2018). Wind power capacity will increase from 221 MW in 2017 (5% of total electricity generating capacity) to approximately 2,100 MW by 2030 (Saskpower 2018). SaskPower plans to add as much as 1,800 MW of wind generation in 12 years ( ) to meet its 2030 goals (EnergiNews 2018). Pursuant to the Management and Reduction of Greenhouse Gases and Adaptation to Climate Change Act (Bill 126, implemented May 2010), regulated emitters (facilities that emit 50,000 tonnes or more of CO2e annually) are required to reduce their annual GHG emissions by a prescribed amount relative to a baseline in order to collectively meet the provincial emissions reduction target (Harper et al. 2016). In addition to the proposed amendments to Canada s coalfired electricity regulations, the governments of Canada and Saskatchewan have an agreement in principle to develop an equivalency agreement for the existing coal-fired electricity regulations that would apply to coal-fired electricity units, until the end of 2029 (EnergiNews 2018). In fact, unlike provinces such as Alberta and Ontario, Saskatchewan does not have plans to phase-out its use of coal. It has, instead, focused on the development of CCS technology, and the use of that technology to retrofit coal-fired generation facilities in the province (Government of Saskatchewan 2017).

34 18 Canadian Energy Research Institute

35 Economic and Environmental Impacts of Transitioning to a 19 Cleaner Electricity Grid in Western Canada Chapter 2: Methodology Electric Power System Planning and Operations Simulation Model In this study, we assess the impacts of transforming the electric power systems of western Canadian provinces to reduce GHG emissions and the impact of interprovincial electricity trade. An electric power system is a complex system that consists of hundreds of generating units, demand points, and transmission and distribution lines that link electricity supply with demand. In a typical power system, at any given instance, the electricity demand is almost instantaneously matched by electricity supplied from a set of generating units. The generating units that would supply the required electricity are usually selected by minimizing the total operating cost. In power system operations, this is known as least-cost economic dispatch. Economic dispatch is constrained by existing generating units, transmission capacity, fuel and resource (e.g., water flows for hydropower, blowing wind for wind power plants, etc.) availability, and generating unit specific operational constraints. The latter point refers to technical limitations of generators such as response time, planned and unplanned outages, and environmental constraints. To make reasonable energy, economic, and environmental impact assessments of power system operations, the generating units (including their technical constraints), resource supply, demand dynamics (including current and future demand), applicable regulations, and typical decision making should be simulated as a system. Furthermore, as the demand for electricity grows over time, new generating units need to be added. The analysis framework should include a systematic way to determine the most likely generating unit additions. For this analysis, we have developed an electric power system planning and dispatch simulation model. This model simulates operations of interconnected power systems that are being assessed by minimizing the total investment and operating cost. The model is developed as a linear mathematical optimization problem that minimizes the total economic cost (the objective function) subjected to a set of constraints that represent the aforementioned constraining parameters. Electricity system planning and operations models based on mathematical programming are widely used as integrated resource planning (IRP) models by utility decision makers. Due to the recent restructuring of power systems in some jurisdictions (e.g., in Alberta) the applicability of IRP type models has lessened. However, this type of model can still be used to gain insights into the impacts of public policy and energy policy decisions. This type of model has been widely used in recent analyses to assess impacts of electric power system transitions in North America and elsewhere (Johnston et al. 2013; van der Weijde and Hobbs 2012; Eurek et al. 2016). Here we provide a brief description of the model structure and underlying assumptions. Interested readers are invited to contact CERI for full model details.

36 20 Canadian Energy Research Institute Model Objective The model objective is to minimize the present value of capital investment in and operation costs of the power systems over the period In the current analysis, we focus only on minimizing the generation and investment cost. Furthermore, in the current analysis, each provincial electric power system is modelled as a single node, where all generating units and electricity loads are assumed to be connected to a single point. As such, intra-province bulk transmission system is not modelled. We assume sufficient bulk transmission is available within the province. We, however, include the cost of spur transmission that is needed to connect generating units to the bulk transmission system. Different provinces are connected to neighbouring provinces through electricity interties. Intertie capacities are assumed to remain at current levels or exogenously increased depending on the scenario. The simulation period is divided into seven operating periods of five years. Each year in a given operating period is assumed to be identical to the others in the same period regarding demand and cost parameters. Investment decisions on generation are made at the beginning of the operating period to ensure there is generation available to satisfy demand. As such the model minimizes the following cost components aggregated across seven investment points and seven operating periods: Cost of adding new generation and connecting them to the bulk transmission system Variable operating expenses (including expenditures for fuel, operation and maintenance and the carbon cost) and fixed operation and maintenance costs for all installed capacity Variable operating expenses and carbon cost incurred in providing spinning reserves to maintain system reliability In the model, generation dispatch to satisfy the demand is resolved at an hourly time resolution. The model needs to simulate only seven representative years with 8,760 hours in a typical year (24 hrs/day x 365 days/year). To reduce the size of the optimization, we model a year by a sample of representative hours. This is a typical modelling practice used in this type of modelling (van der Weijde and Hobbs 2012; Johnston et al. 2013).

37 Economic and Environmental Impacts of Transitioning to a 21 Cleaner Electricity Grid in Western Canada Here we represent each month in each year by two days one representing a weekday and the other a weekend day. As such we have a sample of 576 hours (24 hrs/day x 2 days/month x 12 months/year) representing each year. In this way, we can reduce computing needs while retaining details about seasonal variations of demand and variable resource supply. Model Constraints The main model constraints that represent economic, technical and resource availability parameters are discussed below. The demand-meeting constraints require generation and transmission (i.e., interties) infrastructure be dispatched in such a manner as to meet demand in every simulated hour in every province. Also, hourly production cannot be larger than installed capacity and generation availability is constrained by planned and forced outage levels. Electricity production from variable renewable electricity sources such as wind and solar cannot be more than the amount permitted by hourly resource availability. Cogeneration units must maintain a minimum production level to satisfy the host facility heat demand, and the biomass units must run with a minimum capacity factor to be consistent with the historical production. Furthermore, for each intertie path in every hour, the total amount of electricity dispatched along the transmission path between two provinces in each hour cannot exceed the sum of the thermal transmission capacity allowed in a scenario. Hydropower production is constrained by water availability. In most cases, CERI could find data on water availability in monthly resolution. However, where applicable, it is possible to keep water in reservoirs and release it to produce electricity as needed. This is a valuable attribute of hydropower for system operations as this energy storage capability of hydropower can be used to match variable demand as well as undispatchable variable electricity supply sources. In the current model, we set an energy constraint where hourly hydropower production can vary depending on the demand and variable supply. However, aggregated daily production cannot be greater than the amount sustained by daily water availability. Similar to Johnston et al. (2013), we also require the hourly production of at least 50% of the average historical production level to maintain the river flows. The model requires the system contain sufficient planning and operating reserve capacity to maintain system reliability. The planning reserve capacity requirement is set at 12% of the hourly demand (Martinez et al. 2013). Operating reserves consist of contingency, quick start, and spinning reserves. Contingency and quick start reserves are required to be 3.5% of hourly demand. Spinning reserves are critical to maintaining system supply and balance under intrahour variabilities and any demand forecast errors. Furthermore, additional spinning reserves are required to maintain reliability when the system is operated with undispatchable generating sources such as variable renewables.

38 22 Canadian Energy Research Institute In this analysis, we require the total spinning reserve requirement to be 3.5% of the hourly demand and 5% of the hourly output of dispatchable generating units (in this analysis, wind and solar PV). When required by a scenario, policy goals such as renewable energy targets and emissions caps are enforced as another set of constraints. Representation of Generating Units In this model, individual generating units are aggregated by technology and dispatched as a single unit. For example, all natural gas combined cycle units are aggregated and represented as a single large unit. By doing so, we reduce the data and computing requirements. However, we lose the ability to model some cost and operational details such as minimum operating limits of generators, additional fuel requirements for thermal unit cycling (off/on cycles) and additional fuel requirements due to load following (varying levels of output) of thermal units. In addition to higher fuel requirements, cycling and load following can lead to increases in operation and maintenance costs due to fatigue on system components. Integration of higher levels of renewables can potentially exacerbate these impacts. As such these factors can potentially increase the cost of renewable energy integration (Kumar et al. 2012b; Jordan and Venkataraman 2012). Reliable estimations of these impacts require individual unit level simulations, which we are unable to do with the current state of the model. We address this further in the discussion section. Model Outputs The main model outputs derived through simulations are as follows: Electricity generation capacity available at each province at different operating periods by technology Electricity production in each hour by technology and by province Investment costs and FOM costs by period, by technology, and by province Hourly operating costs that include VOM, fuel cost, and carbon cost by period and by province These outputs are aggregated across time and space and further processed to estimate different metrics that are necessary to assess the main research questions. For example, the average cost of electricity (measured in cents per kwh) is calculated by dividing the total annual cost by total production. Main Assumptions and Parameters The simulation model requires data on demand, costs, and resources. This section describes and summarizes the main parameters that we used for this analysis.

39 Economic and Environmental Impacts of Transitioning to a 23 Cleaner Electricity Grid in Western Canada Demand Data This analysis simulates system operations at an hourly resolution to study the impacts on power system operations due to time-varying demand and renewable electricity sources. Therefore, we require demand data from each of the future representative years at hourly resolution. To develop an hourly load forecast, we first extracted the typical electricity demand profiles of the four power systems by following a method proposed by MacCormack et al. (2008). We used historical hourly demand datasets obtained from the operators of the four power systems (AESO 2018a; Manitoba Hydro 2017a; SaskPower 2018b; BC Hydro 2018e). Up to five-year-long datasets were used to extract the demand profile. The demand profile is a zero-mean hourly dataset that represents one year. By doing so, we retain the seasonal and diurnal demand dynamics. We then use an annual electricity demand projection developed by the NEB for the Canada s Energy Future 2017 report (NEB 2017). This demand outlook is depicted in Figure 2.1. Figure 2.1: Annual Electricity Demand Forecast Source: Forecast by NEB (2017); Figure by CERI

40 24 Canadian Energy Research Institute The annual demand data is then used to estimate the average hourly demand of each province. The average demand data is added to the load profile to produce an hourly demand forecast for each of the seven years simulated in the study. As discussed earlier, an 8,760 hour per year long dataset is then reduced to a 576 hour per year sample by selecting a representative weekday and a weekend day from each month in a given year. Figure 2.2 shows the hourly demand data for a weekday in each of the four seasons in all four provinces in the representative year of the period starting in Figure 2.2: Hourly Demand Forecast Developed for the Study Source: CERI. The figure shows the hourly demand in a weekday in January (Winter), May (Spring), July (Summer) and October (Fall) in 2020 in all four provinces. A total of 24 days (1 weekday and one weekend day per month) are simulated for each of the seven representative years.

41 Economic and Environmental Impacts of Transitioning to a 25 Cleaner Electricity Grid in Western Canada Renewable Resource Data The main source of renewable resource data (wind, solar PV biomass, and geothermal) is CERI Study 168, A Comprehensive Guide to Electricity Generation Options in Canada (CERI 2018). The report produced detailed renewable energy resource datasets for all Canadian provinces. The dataset used in this study indicates the hourly availability factor of renewable electricity sources in the four provinces assessed. Figures 2.3 and 2.4 depict the average wind power and solar power resources in western Canada. Figure 2.4 depicts the electricity production of a utility-scale solar PV system with single-axis tracking. Figure 2.3: Average Wind Power Availability by Season in the Four Western Provinces Source: CERI

42 26 Canadian Energy Research Institute Figure 2.4: Average Solar Power Availability by Season in the Four Western Provinces Source: CERI Hydropower production potential in the four provinces is estimated using historical production data obtained from different sources (AESO 2018a; BC Hydro and Kurschner 2014; Manitoba Hydro 2017a; Martinez et al. 2013). Hydropower data was estimated at the monthly resolution and all hours in a given month is assumed to have the same hydropower production potential. Given the seasonal water flow patterns, in our opinion, this is a reasonable assumption. Electricity Trade with the United States In this analysis, exports to the US and imports from the US are added as exogenous demand and supply, respectively. Trade data is assumed to be equal to average hourly trade flows in the last three years. Historic hourly trade datasets for Alberta and BC are estimated using the data obtained from the AESO and BC Hydro (AESO 2018a; BC Hydro 2018e). For Manitoba, we were unable to find hourly data, and therefore, we used monthly average data obtained from the NEB commodity tracking system (NEB 2018). Saskatchewan is assumed to have no trade with the US. Capital Costs and Technology Learning Cost data for different technologies are gathered from different sources such as Fu et al. (Martinez et al. 2013), Black & Veatch (2012), EIA (2017), and CERI (2018). The changes in the capital costs over time depends on the technology learning. The most common model used in the energy literature to forecast changes in technology cost is the one-factor learning curve (or

43 Economic and Environmental Impacts of Transitioning to a 27 Cleaner Electricity Grid in Western Canada experience curve ). This widely used formulation is derived from empirical observations across a variety of energy technologies that frequently indicate a log-linear relationship between the unit cost of the technology and its cumulative output (production) or installed capacity. The reduction in cost associated with a doubling of experience is referred to as the learning rate, reported in Table 2.1 (Rubin et al. 2015a). Capital costs are projected until 2050 using the total installed capacity projection from the International Energy Agency (IEA) (2015) and the learning rates for different technologies from Rubin et al. (2015b). Capital and other generation-specific costs assumed for this study are listed in Table 2.2. Source: (Rubin et al. 2015a) Table 2.1: Learning Rates for the Technology Sources Technology and Energy Source Learning Rates Coal 8.3% NGCC 14% NGSC 15% NG-Cogen 4.5% Nuclear 6% Hydro 1.4% Biomass 11% Wind 12% Solar- Residential 23% Solar- Commercial 23% Solar- Utility- Tracking 23% CCS-gas 14% CCS-coal 5.5% Geothermal 30% The data used for costs and heat rates are shown in Table 2.2. Although listed in the table, nuclear power was not made available for investment in the model. This exclusion was made considering that no nuclear power project is proposed or considered in any of the western provinces.

44 28 Canadian Energy Research Institute Technology Table 2.2: Costs and Heat Rates Variable O&M (CAD$/MWh) Fixed O&M (CAD/kW year) Total Capital Requirement (CAD$/kW) Heat Rate (GJ/MWh) Coal NGCC NGSC NG-Cogen Nuclear Hydro Biomass Wind Solar-Residential Solar-Commercial Solar-Utility (tracking) Natural Gas with CCS* Coal with CCS Geothermal Coal to Gas (CTG) * Variable cost of CCS units includes CO2 transport and storage cost that was assumed to be $12/tCO2. CCS capture rate was assumed to be 90%. Another restriction we placed on the analysis dealt with natural gas-fired CCS technology. Given this is an emerging power generation technology with very limited operational experience, natural gas CCS was not made available for investment until Cycling Costs In a power system characterized by increasing shares of renewable power generation, the flexibility requirements placed on existing conventional capacities rise significantly (Agora Energiewende 2017). Penetration of renewable generation into the electricity grid results in higher operations and maintenance (O&M) expenditures of the conventional fossil fuel plants. A major cause of this increase in O&M cost for many fossil units is unit cycling. Cycling refers to the operation of electric generating units at varying load levels, including on/off, load following, and minimum load operation, in response to changes in system load requirements. While cyclingrelated increases in failure rates may not be noted immediately, critical components will eventually start to fail. Shorter component life expectancies will result in higher plant equivalent forced outage rates (EFOR) and higher capital and maintenance costs to replace components at or near the end of their service lives (Kumar et al. 2012). Table 2.3 shows the lower bounds for cycling costs and EFORs when the share of renewables increases.

45 Economic and Environmental Impacts of Transitioning to a 29 Cleaner Electricity Grid in Western Canada Table 2.3: Typical Lower Bound Costs of Cycling and Impacts on Equivalent Forced Outage Rates Unit Type Large Coal-fired Sub-Critical Steam ( MW) Large Coal-fired Super-Critical Steam (500-1,300 MW) Gas-fired Combined Cycle (CT-ST and HRSG) Gas-fired Simple Cycle Large Frame CT Gas-fired Steam ( MW) Cost Item Typical Hot Start Data O&M cost ($/MW cap.) EFOR Impact % % % % % Typical Warm Start Data O&M cost ($/MW cap.) EFOR Impact % % % % % Typical Cold Start Data O&M cost ($/MW cap.) EFOR Impact % % % % % Source: (Kumar et al. 2012b) The cycling costs in the table are provided as a reference. Since we are modelling thermal units at an aggregated level, this data is not used in the simulation model. One exception is the inclusion of increased fuel requirements due to load following operation when thermal units provide spinning reserves. In this case, we assume the fuel requirements of a thermal power plant would be 15% higher than rated values due to part load operations (Jordan and Venkataraman 2012). Installed Capacity Changes in the existing generation capacity over time is projected until 2050 using the current installed generation capacity, the provinces long-term plans for new generations, and lifetime assumption for different technologies. We assume the lifetime of different technologies; this is shown in Table 2.4.

46 Canadian Energy Research Institute Source: CERI Table 2.4: Assumed Lifetime of Different Technologies Technology Lifetime Coal 50 Natural Gas 30 Hydro 60 Biomass 35 Wind 25 Solar 25 Other assumptions for capacity projection until 2050 are as follows: Fuel Prices 1. There is no retirement for hydro units in all provinces until 2050, 2. Some of the coal units in Alberta are retired before 2030 because of either the specified date of forced retirement or they will reach the maximum lifetime before 2030, but all of them are retired after 2030, and 3. Some of the coal units in Saskatchewan are retired before 2050 because of either the specified date of forced retirement or they will reach the maximum lifetime before Otherwise, they will operate until Figure 2.5 and Table 2.5 show the fuel prices used for the analysis Figure 2.5: Natural Gas Price Forecast Used for the Analysis (CAD$/GJ) AB BC MB SK Data Source: Natural gas price forecast is developed by CERI using data obtained from Statistics Canada (Statistics Canada 2018). Figure by CERI.

47 Economic and Environmental Impacts of Transitioning to a 31 Cleaner Electricity Grid in Western Canada Fuel Table 2.5: Fuel Prices Used for the Analysis Price (CAD$/GJ) Sources Biomass 6 Keller et al. (2018) Coal 0.14 Bank of Canada (2018) Prices shown in Table 2.5 are assumed for the entire analysis period. Energy Policy and Climate Change Policy-related Parameters As discussed in Chapter 1, in two scenarios, we assume that the carbon pricing system based on OBA is enforced in all four provinces. Our assumptions about the pricing system are also described there. Table 2.6 lists the carbon pricing and benchmark emissions intensities (BEI) assumed for the analysis. These prices and BEI values are used for the scenarios CCMP-NC and CCMP-WC. Operating Period Table 2.6: Carbon Pricing System Parameters Carbon Price ($/tco2e) Benchmark Emissions Intensity (tco2e/mwh) AB BC MB SK [30-50] Two other scenarios enforce GHG emissions reduction limits that require 30% GHG emissions reduction by 2030 and 80% GHG reductions by 2050 compared to 2005 emissions. Table 2.7 lists the emissions caps enforced under these two scenarios. GHG emissions levels from electricity generation in 2005 in the four provinces are obtained from Environment Canada s National GHG Inventory Report (ECCC 2018a Part 3, Annex 13).

48 32 Canadian Energy Research Institute Table 2.7: GHG Emissions Cap Scenario Province Emissions from Electricity Generation in 2005 (ktco2e)* Emissions Cap (ktco2e) 2030 (30% below 2005 levels) 2050 (80% below 2005 levels) DGHG-NC AB 51,900 36,330 10,380 DGHG-NC BC 1, DGHG-NC MB DGHG-NC SK 15,200 10,640 3,040 DGHG-WC Combined target for all western provinces 68,798 48,159 13,760 *Source of 2005 emissions levels: Environment Canada (ECCC 2018a Part 3, Annex 13). All except the Reference scenario requires Alberta to produce at least 30% of its electricity from renewable sources by Similarly, Saskatchewan is required to have at least 50% of its electricity generation capacity from renewable sources. No renewable energy targets are enforced in Manitoba and BC. Current Interprovincial Transmission Intertie Capacities Current intertie capacities assumed for the analysis are listed in Table 2.8. These values are determined by reviewing different transmission system expansion plans and status reports published by the provincial electricity system operators (AESO 2015; Manitoba Hydro 2013; SaskPower 2018b). Origin Intertie Destination Table 2.8: Current Intertie Capacities Export Limit (origin to destination flow) (MW) Import Limit (destination to origin flow) (MW) BC AB 1,200 1,000 AB SK MB SK For the BC-AB intertie, we assume that interties restoration plans that have been announced by AESO are implemented by 2020, and the intertie can be loaded to its full line rating. For the MB- SK intertie, we assume that the planned Birtle, MB to Tantallon, SK 230 kv transmission line project is completed by 2020 (SaskPower 2018a).

49 Economic and Environmental Impacts of Transitioning to a 33 Cleaner Electricity Grid in Western Canada Chapter 3: Results and Discussion We used the power system investment and dispatch simulation model along with the data and assumptions presented in Chapter 2 to assess the impacts of a cleaner generation fleet under five scenarios. The model selects and operates the optimal set of power generation technologies that minimize the total cost while satisfying future power demand. These scenarios represent alternative pathways that the electric power systems of the four western Canadian provinces could take under different energy and climate policies. The five scenarios are described in detail in Chapter 1. In summary, they consist of a reference scenario (Reference), two carbon pricing scenarios (CCMP-NC and CCMP-WC), and two carbon emissions cap scenarios (DGHG-NC and DGHG-WC). The latter four scenarios are aligned with the course set by the Pan-Canadian Framework to reduce GHG emissions while sustaining economic growth. Under each scenario, we estimate the power generation capacity additions, electricity generation, different cost metrics, and GHG emissions metrics. Electricity Supply and GHG Emissions Figure 3.1 shows the installed generation capacity by technology in the four western provinces at different time periods. Figure 3.2 shows the electricity produced by different units. From these results, several observations can be made about the impact of climate policies on power generation investments. Figure 3.3 shows the total GHG emissions from electricity generation. Figure 3.4 shows the GHG emissions intensity of power generation. Under the Reference scenario, in the absence of climate change policies, natural gas becomes the dominant electricity generation fuel in Alberta and Saskatchewan. With the growing electricity demand, the share of natural gas-fired generation (NGCC, NGSC, NGCogen, CTG) in Alberta s electricity supply rises from 58% in 2020 to 99% in This is also due to the retirement of the coal-fired power plants. Coal units retire either because they reach the end of their life or due to phasing-out of coal from the generation fleet in Alberta. In Saskatchewan, the same share rises from 27% in 2020 to 83% in Despite the dominance of natural gas, both total GHG emissions and GHG intensity of electricity will be lower than 2005 levels in the two provinces due to the retirement of coal-fired generation (Figure 3.3). Despite not surpassing 2005 emissions levels, total emissions and emission intensity gradually rise due to the growing electricity production to satisfy the forecasted demand. Under the Reference scenario, British Columbia (BC) and Manitoba already have a large fleet of hydropower generators. The two provinces also have some hydropower units that are currently under construction. As such, hydropower generators dominate the electricity supply. A marginal amount of power is produced by natural gas units (2% in BC and 3% in Manitoba) to compensate for variations in seasonal hydropower supply variations.

50 34 Canadian Energy Research Institute GHG emissions from power generation in Manitoba would be higher and continue to grow throughout the simulated outlook period. In BC, emissions will also grow but would not pass 2005 levels within the analysis period. The absolute emissions amount and emissions intensity will remain low in both provinces due to hydropower dominance. Figure 3.1: Installed Generation Capacity under Different Scenarios by Technology Source: CERI

51 Economic and Environmental Impacts of Transitioning to a 35 Cleaner Electricity Grid in Western Canada Each figure set in a row (for Figures 3.1 to 3.4) shows the key parameter in the four provinces in the periods starting in 2020, 2030, 2040, and 2050 by technology under a given scenario (Reference, CCMP-NC, CCMP-WC, DGHG-NC, DGHG-WC). Some operating periods (2025, 2035, and 2045) are not shown to improve clarity. Figure 3.2: Annual Electricity Generation under Different Scenarios by Technology in Different Operating Periods Source: CERI

52 36 Canadian Energy Research Institute Figure 3.3: Total GHG Emissions from Electricity Generation under Different Scenarios Source: CERI Under the two carbon pricing scenarios (CCMP-NC and CCMP-WC), the electricity generation systems of BC and Manitoba do not undergo significant change compared to the Reference scenario. Hydropower remains as the dominant power generation technology. Neither province sees investments in other renewable sources under the carbon pricing scenarios. Manitoba adds approximately 1,000 MW of new gas-fired generation capacity which, mostly (63%) consists of NGSC. These units are mainly used for load balancing and manage hydropower supply variations. Actual electricity generation by gas units is in the range of 1-3%. Under THE carbon pricing scenarios, both BC and Manitoba face rising GHG emissions, although the amount and the emission intensity remain at very low levels compared to the total provincial emissions.

53 Economic and Environmental Impacts of Transitioning to a 37 Cleaner Electricity Grid in Western Canada Figure 3.4: GHG Intensity of Electricity Generation under Different Scenarios Source: CERI Under the CCMP-NC and CCMP-WC scenarios, the generation mix of Alberta and Saskatchewan see notable changes. The share of renewable energy in Alberta s generation mix increases from 16% in 2020 to 30% in 2030 and afterward. In Saskatchewan, the renewable share increases from 17% in 2020 to 26-35% after The increase in renewable share is driven mainly by renewable energy targets as opposed to the credits provided by the carbon pricing systems. Nonetheless, renewables receive incentives under the carbon pricing system. Under the hybrid carbon pricing system that is designed as an OBA system, generators whose emission intensity is lower than the benchmark intensity receive credits. In Alberta, as the benchmark intensity gets more stringent over time, credits received by renewables increase by 5-8% per year (see Appendix A for carbon costs/credits received by different generation technologies under the carbon pricing system). The opposite happens to fossil fuel power generating sources, where they would have to pay carbon costs that increase the total operating

54 38 Canadian Energy Research Institute cost. Therefore, the carbon credits help renewable energy sources to compete with thermal energy sources. In Alberta, the dominant renewable energy technology is wind power. Alberta has excellent wind resources, and the cost of wind power is lower than other renewable energy sources. Alberta has stable wind regimes, especially in winter months. Demand in Alberta and all other provinces is higher in winter. The correlation of wind with demand in high demand periods also helps to manage the variability. These factors lead to wind becoming the lower-cost, dominant generation option to satisfy the 30% renewable energy target in 2030 and afterward. Interestingly, by the operating period, solar PV and geothermal energy start competing with wind for the 30% energy share. Several factors lead to this result. By 2050, due to technology learning, the capital cost of both technologies declines significantly (34% and 55% decline for solar PV and geothermal, respectively, from 2020 to 2050). Wind is already a mature technology with less room for capital cost reduction. The cost of wind would decline only by about 14% by As the cost of renewables is dominated mainly by capital cost, this is the major driving factor for solar PV and geothermal to be able to start competing with wind. Other factors also influence this technology competition. Some of the wind power that was added earlier would have retired upon reaching their maximum service life by As demand rises, the capacity of renewables required to meet the 30% energy target increases. Balancing the supply and demand under a high amount of variable generating sources becomes challenging. Therefore, generating technologies or a combination of technologies that introduce lower variability will be favoured by the cost minimization algorithm. Geothermal is inherently a dispatchable generation technology and therefore would not introduce a notable amount of supply-side variability. Solar PV has negative temporal correlations with wind. Consequently, combining solar PV with wind will reduce the net variability. These factors lead to geothermal and solar PV entering the least-cost generation mix by 2050 under climate and energy policy constraints. In Saskatchewan, where a minimum of 50% renewable generation capacity is required after 2030, the choice of selected renewable generation technologies under carbon pricing scenarios seem to be influenced by available intertie capacity. Under the CCMP-NC scenario, where intertie capacities remain at current levels, the 50% renewable electricity capacity requirement is shared among existing hydro, wind, and solar PV. Towards the later operating periods (2045 and after), solar PV dominates other renewables. However, under CCMP-WC, where intertie capacities are doubled, wind energy becomes the dominant technology. Solar PV enters the generation mix by a considerable amount only after The main reason for the discrepancy is, having a higher amount of intertie capacity provides more technical options to manage variability introduced by a single dominant technology.

55 Economic and Environmental Impacts of Transitioning to a 39 Cleaner Electricity Grid in Western Canada In this case, renewable electricity that gets produced in low demand periods can be exported to neighbouring jurisdictions without uneconomic curtailment. The export capacity to Manitoba from Saskatchewan helps to manage wind variability as hydropower in Manitoba can act as a balancing storage option. With limited intertie capacity under CCMP-NC, the system favours a low volatile generation mix provided by a combination of wind and solar PV. Despite renewable energy entering the generation mix in larger volumes under the CCMP-NC and CCMP-WC scenarios, natural gas remains as the dominant power generation fuel in Alberta and Saskatchewan (natural gas share is 70% in Alberta and around 65% in Saskatchewan after 2030). The total GHG emissions continue to grow, albeit at a slower rate (see Figure 3.3). The average GHG intensity in Alberta and Saskatchewan on average remains around 0.3 tco2/mwh. Compared to the Reference scenario, this is approximately a 25% reduction in Alberta and a 31% reduction in Saskatchewan. The total GHG emissions reductions achievable are around 50% below 2005 levels by both 2030 and 2050 in the two provinces. The two emissions cap scenarios (DGHG-NC and DGHG-WC) lead to transformative changes in the electricity generation mix in Alberta and Saskatchewan after The deep GHG emissions cap of 80% drives conventional natural gas-fired generation away from the generation mix. Until 2040, the capacity and generation mix in the two provinces remains close to those under the carbon pricing scenarios but changes significantly after that. Under the carbon cap scenarios, the share of renewable energy in Saskatchewan s generation approaches 68% by In Alberta, the renewable share in the generation mix reaches 50% by In both provinces, wind is the dominant renewable electricity source. The total amount of installed wind capacity in Alberta reaches 12,700 MW, which is approximately 60% of the total generation capacity. In 2050, natural gas combined cycle units with carbon capture and storage (NGCC-CCS) also enters the generation mix in Alberta (3,590 MW) and, to a lesser extent, in Saskatchewan (150 MW) under the carbon cap scenarios. Both Alberta and Saskatchewan install geothermal energy as well. Under this scenario, NGCC- CCS and geothermal provide the much-needed low-carbon baseload generation service that the two power systems require. Here, by baseload we refer to a minimum demand that needs to be satisfied at all times. Serving this demand with high GHG emissive technologies will have greater carbon management challenges (e.g., having to pay a higher carbon levy or not being able to meet the emissions cap). NGCC units become an operating reserve and load following service provider that helps to balance the electricity supply and demand. Under the DGHG-WC scenario, BC also installs 1,000 MW of geothermal capacity. Most of the energy produced by these units get exported to Alberta through additional interties added under this scenario. This is mainly because we constrain the total geothermal energy potential in Alberta and BC to be 1,000 MW per province as geothermal is still an emerging technology with limited experience. Furthermore, for geothermal to be competitive, capital costs need to decline. This is expected to occur over time due to technology learning. The estimated geothermal potentials in Alberta and BC are much higher than 1,000 MW (Grasby et al. 2011).

56 40 Canadian Energy Research Institute Under the carbon cap scenarios, total emissions and average grid emission intensities decline steadily (Figures 3. 3 and 3.4). This is because the policy targets require emissions reduction of 80% below 2005 levels by Under these scenarios, the share of electricity from renewable sources and CCS units reached over 75% in both Alberta and Saskatchewan by It should be noted that, although it was made available in 2025 (considering the construction lead time), new coal CCS units did not get built anywhere in western Canada. This is due to the higher capital cost as well as the relatively high GHG intensity even after capturing 90% of GHG produced in combustion. It should be noted that the Boundary Dam coal CCS unit in Saskatchewan was operating in all operating periods. However, no new coal CCS units were built by the model. Similarly, no new hydropower was added anywhere in western Canada due to higher capital cost and seasonal variation of hydro resources. Under the DGHG-WC scenario, where intertie capacities are doubled and a common carbon cap was set for the western region as opposed to individual provincial caps, the total operating cost was lower than GDHG-NC (the scenario with the same policy constraints with no higher level of coordination). This shows the value of coordination and inter-provincial trade. In the case of Alberta, one observation made across all scenarios is that about 3,000 MW of coal is converted to gas (CTG). CTG conversion can be done at a very low capital cost (approximately $240/kW), and these units can provide the operating reserves and balance services that the system needs. Since the amount of energy provided by CTG is low, the impact on emissions is not significant. However, under a very stringent carbon cap (i.e., DGHG scenarios), CTG stops providing reserves due to higher GHG emissions. Cost of Power System Operations In this section, we estimate the total operating cost and the average cost of electricity under each of the scenarios assessed in this study. Figure 3.5 shows the main cost components that are incurred in operating the four power systems under different scenarios. Here, Investment cost refers to capital investments made on new generation units at different time periods. It only includes the investments made by the power system planning and dispatch model developed for this study. FOM is the fixed operating and maintenance costs of new, as well as existing, generating units. VOM (energy) is the variable operating and maintenance cost in producing electricity. Similarly, VOM (spinning) is the variable operating and maintenance cost in providing spinning reserves by appropriate units. For thermal units, fuel cost dominates the VOM cost. Carbon cost refers to payments made, or credits received (negative cost) by electricity generating units under the carbon pricing scenarios. As can be seen, in the case of the carbon pricing scenarios, all power systems incur net negative carbon costs. In the case of Alberta and Saskatchewan, the carbon payments made by GHG

57 Economic and Environmental Impacts of Transitioning to a 41 Cleaner Electricity Grid in Western Canada emissive technologies is negated by credits received by cleaner power sources. In the case of BC and Manitoba, significant carbon credits are received by hydropower units, reducing the overall power system operating cost. If implemented this way (i.e., hybrid carbon pricing with OBA), this should lead to lower electricity costs to the consumers. We then use these costs to calculate the average cost of electricity (COE). For COE calculations we annualized the investment costs using a capital recovery factor of 10%/year. We used publicly available data and estimates for the capital cost of existing units and estimated the capital costs of generating units that are currently under construction (e.g., the Site C hydropower project in BC, numerous wind power projects in Alberta, Keeyask hydropower project in Manitoba). We used publicly available information and our best estimates to quantify the value of existing generating units (SaskPower 2018b; Manitoba Hydro 2017a; CERI 2018). These costs are exogenously added to the cost components in Figure 3.5 for COE calculations. Table 3.1 shows the estimated COE values under different scenarios.

58 42 Canadian Energy Research Institute Figure 3.5: Power System Costs Source: CERI In addition to the generation cost that we estimated, the cost of electricity seen by consumers (e.g., residential customers) includes an electricity delivery charge that covers bulk electricity transmission, distribution and administrative charges. Since our model is unable to estimate these, we exogenously added a delivery charge of 7.5 cents/kwh (2.5 cents/kwh transmission and 5 cents/kwh for distribution and administrative charges). This value is estimated by observing recent price statistics and estimates made by system operators (AESO 2018d; Goulding and Atanasov 2014; Utilities Consumer Advocate 2018). Also note that the same delivery charge

59 Economic and Environmental Impacts of Transitioning to a 43 Cleaner Electricity Grid in Western Canada is used under all five scenarios. Therefore, this does not impact the comparative analysis of individual scenario results. The average cost of electricity for residential customers is shown in Table 3.1. Compared to the Reference scenario, the average COE values are higher when different climate change policies are enforced (i.e., carbon pricing, renewable targets, and carbon caps). There are a few exceptions. Under the carbon pricing scenarios, credits received by hydropower in BC and Manitoba lowers the COE in those two provinces. Table 3.1: Average Residential Cost of Electricity (cents/kwh) Province Period Reference Scenario CCMP-NC CCMP-WC DGHG-NC DGHG-WC AB AB AB AB BC BC BC BC MB MB MB MB SK SK SK SK One may rightly question the notion of providing carbon credits for large hydropower. A counterargument is, in those provinces, credits received by the electricity sector can be used to cover the cost of reducing emissions elsewhere in the economy. To test the impact of providing carbon credits for large hydro on system operations and generation mix, we simulated the CCMP-NC and CCMP-WC scenarios without carbon credits for large hydro. This did not change the generation mix or installed capacity mix in any of the provinces. The Value of Interprovincial Coordination In two scenarios we tested the value of higher coordination among provinces in reducing GHG emissions through decarbonizing the electricity grid. In higher coordination scenarios, provinces increase the intertie capacities to increase electricity trade. The intended benefit is reducing the overall cost by sharing resources and providing system balancing services. Under the carbon cap

60 44 Canadian Energy Research Institute scenarios, when provinces have higher coordination, a combined carbon cap is set for the whole western region. The intention here is to achieve higher emissions reductions where it is cheaper to do so. In this analysis, we define the value of coordination as the difference between the present values of investment and operating cost of scenarios with the same policy goals but different levels of coordination. Under this definition, we can estimate the value of coordination for carbon pricing and carbon cap scenarios. Table 3.2 shows the present value of the investment and operating cost of the four power systems in western Canada. In scenarios where provinces have higher coordination, the total cost is lower for the same policy goals. As such, the value of coordination for the carbon pricing scenarios is CAD$1,691 million. For the carbon cap scenarios, the value of coordination is estimated at CAD$1,812 million. Scenario Type Table 3.2: The Value of Interprovincial Coordination in Western Canada Scenario Present Value of Total Power System Investment and Operating Cost (CAD$ million) Value of Coordination (CAD$ million) Gross Value Net Value* Carbon pricing scenarios CCMP-NC 57,166 CCMP-WC 55,475 1, Carbon cap scenarios DGGHG- NC 78,127 DGHG- 332 WC 76,315 1,812 *Net value of coordination is calculated by subtracting the estimated cost to double the intertie capacity from the gross value of coordination. The minimum infrastructure needed to facilitate higher coordination in the scenarios we assessed in this study is the doubling of existing intertie capacities among neighbouring provinces (i.e., BC-AB, AB-SK, and MB-SK). Investments on these interties are justified if the value of coordination is higher than the transmission investment cost. It is difficult to estimate the transmission investment cost without selecting a specific transmissions corridor. Using high-level transmission cost estimates obtained from the Pan-Canadian Wind Integration Study (GE Energy Consulting 2016 Section 7, p19-20), we estimate the cost of additional interties as follows. All new lines are assumed to follow the existing intertie corridors closely. Additional substations are added as needed.

61 Economic and Environmental Impacts of Transitioning to a 45 Cleaner Electricity Grid in Western Canada BC-AB intertie (1,000 MW increase; add a new 500 kv line, ~350km): CAD$750 million AB-SK intertie (150 MW increase; add a new 230 kv line, ~ 225km): CAD$380 million MB-SK intertie (300 MW increase; add a new 230kV line, ~200km): CAD$340 million Under these assumptions, the minimum cost to double the intertie capacity is CAD$1,480 million. This makes the net value of coordination CAD$211 million and CAD$332 million for carbon pricing and carbon cap scenarios. With this estimate, investment in transmission to facilitate coordination is marginally justified. It is also possible to increase only some of the interties. Using a system-level assessment such as the one developed in this study, the intertie expansions with optimal value can be identified. In order to gain insights into the value of reinforcement of individual interties, we ran several other sensitivity analysis simulations with carbon pricing and higher coordination scenario (CCMP-WC). In these simulations, we doubled the capacity of only two interties and kept the other at current capacity. This analysis showed that doubling only the Alberta/BC intertie and Saskatchewan/Manitoba intertie but keeping the Alberta/Saskatchewan intertie at current capacity will increase the net benefit of coordination for the carbon pricing scenarios by CAD$156 million (70% increase). This is due to not having to invest in an intertie that brings in the least value. The sensitivity analysis also showed that not increasing the intertie capacity between Saskatchewan and Manitoba reduces the value of coordination by 78%. Operational Challenges Results of the dispatch simulations provide some interesting insights about demand and resource dynamics as well as some technical challenges resulting from operations under the modelled policy constraints. Extraction of these insights is enabled by our use of time-synchronized hourly profiles of resources and demand over the simulation period. This shows the strength of economic dispatch simulations under operating constraints and allows the user to obtain information that is not possible through comparative assessments of individual technology modelling such as the levelized cost of energy screening assessments. Figure 3.6 shows the generation dispatch simulation of Alberta power systems on a winter day and a summer day in 2030 and 2050 under different scenarios. Figure 3.7 shows the same information for Saskatchewan. As can be seen from Figure 3.6, wind resources in winter months show baseload-like behaviour where availability is consistent and steady. Demand in winter is higher in Alberta, and steady wind regimes in winter increase the value of wind as a clean energy resource. Similar results are observed for Saskatchewan. The differences between the dispatch profiles in different scenarios are due to the use of interties between the jurisdictions.

62 46 Canadian Energy Research Institute Figure 3.6: System Dispatch in Alberta Source: CERI.

63 Economic and Environmental Impacts of Transitioning to a 47 Cleaner Electricity Grid in Western Canada Figure 3.7: System Dispatch in Saskatchewan Source: CERI.