Citation
Quantitative evaluation of sustainable energy pathways for Colorado's power sector

Material Information

Title:
Quantitative evaluation of sustainable energy pathways for Colorado's power sector focus on greenhouse gas mitigations
Creator:
Barhaghi, Saeed G
Publication Date:
Language:
English
Physical Description:
xiii, 127 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Greenhouse gas mitigation -- Colorado ( lcsh )
Electric utilities -- Colorado ( lcsh )
Electric utilities ( fast )
Greenhouse gas mitigation ( fast )
Colorado ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Colorado Denver, 2008.
Bibliography:
Includes bibliographical references (leaves 122-127).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Saeed G. Barhagi.

Record Information

Source Institution:
University of Colorado Denver
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
259704484 ( OCLC )
ocn259704484

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

QUANTITATIVE EVALUATION OF SUSTAINABLE ENERGY PATHWAYS FOR COLORADO'S POWER SECTOR: FOCUS ON GREENHOUSE GAS MITIGATIONS by Saeed G. Barhaghi B.S., Civil Engineering, University of Kansas, 1977 M.S., Aeronautical Engineering, Wichita State University, 1983 A thesis submitted to the University of Colorado Denver in partial fulfillment of the requirements for the degree of Doctor of Philosophy Civil/Environmental Engineering May 2008

PAGE 2

This thesis for the Doctor of Philosophy degree by Saeed G. Barhaghi has been approved by Anu Ramaswami Bruce Janson Stephen Lawrence Paul Komor Date

PAGE 3

Barhaghi, Saeed, G. (Doctor of Philosophy, Civil Engineering) Quantitative Evaluation of Sustainable Energy Pathways for Colorado's Power Sector: Focus on Greenhouse Gas Mitigation Thesis directed by Professor Anu Ramaswami ABSTRACT This thesis develops a Colorado specific model to depict the evolution of statewide electric power sector, evaluating various scenarios for environmental and economic sustainability, with focus on greenhouse gas mitigation. This model simultaneously evaluates interactions between the various climate action regulations, current power infrastructures, future advanced technologies, limits of renewable energy and demand-side-management (DSM). For this purpose, a MARKAL optimization model and database is developed that minimizes total system costs while maximizing environmental benefits within transmission and other system constraints. The objectives of the thesis were: 1) Database Development & Calibration: To develop a Colorado-specific (CO MARKAL) database for power sector representing Business-as-Usual (BAU) for 30 years planning horizon (2005-2035}, calibrated to 2005 Energy Information Administration State Electricity Profiles data; 2) Development of Model: Including model refinements for 1) Finer Time-Slices (peak hour representation) suitable for Colorado power generation landscape with sizeable amount of natural gas combustion turbines; 2) Renewable Portfolio Standards Rule-Based constraints; 3) Aggregated DSM and Energy Efficiency measures to represent citywide actions and evaluate statewide Energy Efficiency impacts. 3) Future Scenario Evaluation: The four scenarios studied were: Advanced emerging technologies incorporating carbon-capture and sequestration (CCS) technology: Pulverized Coal with 50% CCS, Coal IGCC with 50% CCS, Gas IGCC with 90% CCS, Advanced Combustion Turbine & Combined Cycle, and Advanced Nuclear Technology.

PAGE 4

Energy Efficiency scenario, including linkage of citywide actions with statewide system to achieve 300 GWh per year and 1% per year reduction in energy consumption Regulatory Policy scenario evaluating both C02 caps consistent with the Colorado Climate Action Plan, and C02 taxes applied both upstream as Btu tax ( cents per million Btu) and downstream (dollar per ton of C02 in production of electricity) Sensitivity scenarios, including energy demand forecasts, natural gas price volatility, and demand elasticity were also evaluated for the various scenarios. Model output showed the following results relevant for future planning and policy making decision for Colorado power sector: Aggressive DSM and Energy Efficiency (DSM/EE) scenario was the most favorable scenario with societal gain of achieving over 7% C02 reduction from BAU, with economic savings from avoided infrastructure investments of at least $9 Billion (2005$) over a 30 year planning horizon. However, the carbon emissions from DSM/EE scenario alone did not meet any of the C02 caps goals. Regulations that cap C02 at 1990 levels in 2035 resulted in a system cost of around $7 Billion (2005$) above BAU costs, and were slightly lower (by 11 %) than costs incurred by capping C02 at 2005 levels in 2020. Btu taxes (upstream) did not have much impact on C02 mitigation because they were uniformly applied to all fossil-fueled generation. C02 taxes (downstream) did not make much of impact in C02 mitigation because of constrained wind power in the system (at 30% of production by 2035). Removing the constraints, the system became sensitive to the downstream C02 tax at $60/mt. Carbon taxes were the most expensive options costing the system from $8 to $21 Billion (2005$), above BAU costs, over the planning horizon depending on the amount of tax. Sensitivity analysis indicated that Colorado Power Sector is at high risk to future carbon regulations, possible increases in natural gas prices, and future growth, suggesting that mandates for considering sizeable renewable energy and energy efficiency are needed immediately. This abstract accurately represents the content of the candidate's thesis. recommend its publication. Anu Ramaswami

PAGE 5

DEDICATION I dedicate this thesis to my family who encouraged and inspired me to continue my goal of completing this work. To my wife Shahnaz, an exceptionally talented architect and artist with great passion for arts and sciences and love for the environment, who has devoted much concentration on zero-emission building design; to my daughter Michelle, who as a young physician has a great appreciation for the environment and the ecosystem and climate change issues; and last but not least to my youngest daughter Krystle, who as a young physicist has already gotten involved with ground breaking research that will soon help the mankind to cope with dreadful diseases. iv

PAGE 6

ACKNOWLEDGMENT I would like to thank my adviser, Dr. Anu Ramaswami, Professor of Environmental Engineering and Director of Sustainable Urban Infrastructure at the University of Colorado Denver for giving me the opportunity to work on this project and her continuous support and valuable feedback while working on this research. I also would like to thank Dr. Bruce Janson, the Chairman of Civil Engineering Department for providing financial support and technical support as my committee member. I'm thankful to all my committee members; Drs John Trapp, Stephen Lawrence, and Paul Komor for their valuable feedback and comments throughout this work. I am also thankful to Gary Goldstein, for fruitful discussions and feedback on MARKAL model during the execution of this project. v

PAGE 7

TABLE OF CONTENTS LIST OF FIGURES ......................................... ............................................................... ix LIST OF TABLES ........................................................................................................... xi PREFACE ..................................................................................................................... xiii 1 INTRODUCTION ..................................................................................................... 1 1.1 Global Greenhouse Gas initiatives ............................................................... 1 1.2 Greenhouse Gas Initiatives in the United States ........................................... 2 1.3 Regional Greenhouse Gas Initiatives (RGGI) .............................................. .4 1.4 California GHG initiatives .............................................................................. 5 1.4.1 West Coast Governors' Global Warming Initiative .............................. 5 1.5 Research Motivation ..................................................................................... 6 1.6 Research Objective ...................................................................................... 6 1.7 Research Impact .......................................................................................... 7 2 BACKGROUND ON ENERGY OUTLOOK .............................................................. 8 2.1 The U.S. Energy Outlook .............................................................................. 8 2.1.1 The U.S. Carbon Dioxide Emissions .................................................. 8 2.1.2 The U.S. Renewable Energy ............................................... .............. 8 2.2 Colorado Energy Outlook ............................................................................. 8 2.2.1 Colorado Sectoral Carbon Dioxide Emissions .................................. 11 2.2.2 Xcel Energy (PSCo) Carbon Dioxide (C02) Intensity ....................... 14 2.2.3 The Colorado Renewable Energy Standard ..................................... 16 2.3 Air Quality Impact of Electricity Generation ................................................. 16 2.3.1 State's Electric Power Sector Air Emission Regulations ................... 17 2.3.2 Colorado Electric Power Sector Air Emission Regulations ................ 17 3 MODELING METHODOLOGY ............................................................... .............. 21 3.1 MARKAL Model .......................................................................................... 21 3.2 MARKAL Model Applications ...................................................................... 23 3.3 EPA National MARKAL model .................................................................... 23 3.4 EPA Regional MARKAL model ................................................................... 24 3.5 MARKAL Model Application to Colorado .................................................... 24 4 MODELING METHODOLOGY FOR COLORAD0 ................................................. 26 4.1 Colorado Energy Model and Database Development ................................. 26 4.2 Energy Demand Forecast ........................................................................... 26 4.3 Existing Generation Resources Data Sources ............................................ 26 4.4 Renewable Portfolio Standards (RPS) requirements ......... ................. ........ 27 5 SCENARIO ANALYSIS ......................................................................................... 28 5.1 Approach .................................................................................................... 28 5.2 Reference Scenario .................................................................................... 29 5.3 Demand Forecast ....................................................................................... 30 vi

PAGE 8

5.3.1 RMATS Study .................................................................................. 30 5.3.2 CEF Report ...................................................................................... 30 5.4 Demand Load Shape .................................................................................. 34 5.5 Supply Options Load Management. ............................................................ 36 5.6 Fuel Supply Prices ...................................................................................... 36 5. 7 Existing Installed Capacity .......................................................................... 38 5.8 New Regulatory Landscape for Renewable and DSM/EE ........................... 39 5.8.1 Renewable Portfolio Standards ........................................................ 40 5.8.2 DSM and Energy Efficiency (DSM/EE) ............................................ .41 5.8.3 Regulated utilities DSM in Colorado ................................................. 44 5.8.4 Impact of New Legislation on DSM in Colorado ................................ 45 5.9 Integrated Gasification Combined Cycle (IGCC) ......................................... 48 5.1 0 Renewable Technologies ..................................................................... 49 5.1 0.1 Biomass ........................................................................................... 50 5.1 0.2 Geothermal ...................................................................................... 51 5.10.3 Solar ................................................................................................ 52 5.10.4 Wind ................................................................................................. 53 5.11 Near-Term Power Plants Retirements .................................................. 53 5.12 Approved and Proposed Future Power Plants ...................................... 54 5.13 New Power Plants ................................................................................ 54 5.14 Discount Rate and Inflation Rate .......................................................... 56 5.15 Transmission Constraints and Infrastructure Improvement Costs ......... 57 5.16 Resource Bounds ................................................................................ 59 5.16.1 Power Imports .................................................................................. 59 5.16.2 Biomass Limits ................................................................................. 60 5.16.3 Geothermal Limits ............................................................................ 60 5.16.4 Solar Limits ...................... ............................................................... 60 5.16. 5 Wind Limits ...................................................................................... 61 6 SCENARIO DESIGN AND ANALYSIS RESULTS ................................................. 62 6.1 Energy Supply ............................................................................................ 63 6.2 Installed and New Capacity Addition ........................................................... 66 6.3 Reference Scenario (BAU): Projected Emissions Profile ............................ 68 6.4 Advanced Technology Scenario ................................................................. 69 6.4.1 Energy Supply .................................................................................. 70 6.4.2 New Capacity Additions ................................................................... 71 6.4.3 System Costs and C02 Emissions Profile ........................................ 72 6.5 Energy Efficiency and C02 Emissions Reduction ....................................... 72 6.5.1 Statewide DSM/EE Plans ................................................................. 75 6.5.2 Cities and DSM/EE Projects and Costs ............................................ 75 6.6 Aggressive DSM/EE Scenario .................................................................... 75 6.6.1 DSM/EE Scenario Electricity Generation ..................... .................... 77 6.6.2 New Capacity Additions ................................................................... 78 6.6.3 DSM/EE System Cost-Benefits ........................................................ 79 6.6.4 DSM/EE C02 Emissions Profile ....................................................... 80 vii

PAGE 9

6. 7 Carbon Policy Scenarios .................. ......................................................... 80 6.7.1 Carbon Cap ...................................................................................... 81 6. 7.2 Co-Benefit of Carbon Cap ................................................................ 85 6.7.3 Carbon Tax ...................................................................................... 85 6.7.4 Biomass Co-Fire .............................................................................. 89 6.8 Sensitivity Analysis ..................................................................................... 90 6.8.1 Gas Prices Sensitivity Analysis ........................................................ 91 6.8.2 Load Forecast Sensitivity Analysis ................................................... 93 6.8.3 Sensitivity Analysis using Monte Carlo Simulations .......................... 95 6.8.4 Uncertainty of Model's lnpuUOutput ................................................. 96 6.9 Elastic Demand .......................................................................................... 97 6.9.1 Elastic Demand Fuel Consumptions ................................................. 99 6.10 Scenario Comparison and Conclusion ............................................... 102 6.11 Future Research Work ....................................................................... 104 APPENDIX A: MODEL INPUT FOR BASE-YEAR (2005) POWER PLANTS CAPACITY AND HEAT RATE ..................................................................................... 105 APPENDIX B: EXISTING POWER PLANTS EMISSION FACTORS .......................... 109 APPENDIX C: MODEL INPUT ASSUMPTIONS AND RESOURCE BOUNDS ........... 111 REFERENCES ............................................................................................................ 122 viii

PAGE 10

LIST OF FIGURES Figure 1: Colorado Electric Power Generation Capability in 2004 ............................. 9 Figure 2: Colorado Net Generation in 2005 ............................................................ 11 Figure 3: Colorado Fossil Fuel C02 Emissions ...................................................... 12 Figure 4: Xcel Energy (PSCo) Energy Resource Mix C02 Intensity ....................... 15 Figure 5: Metro Denver Voluntary Air Pollution Reduction Program ....................... 18 Figure 6: Colorado Power Plants Air Emissions ..................................................... 19 Figure 7: Price/Demand Trade-Off Curve ............................................................... 22 Figure 8: Generic Power Sector RES ................................. ................................... 22 Figure 9: Colorado Peak Demand Trend (2005 -2035) ........................................... 31 Figure 10: Colorado Energy Trend (2005-2035) ..................................................... 32 Figure 11: Colorado Energy Demand Projection (Low, Base & High) ..................... 34 Figure 12: Colorado Demand Projection (Low, Base & High) ................................. 34 Figure 13: Colorado Nine Time-Slice Proxy Load Distribution ................................ 36 Figure 14: Natural Gas Prices Forecast. ................................................................. 37 Figure 15: Coal Prices Forecast ............................................................................. 37 Figure 16: Nuclear Prices Forecast ........................................................... ............ 38 Figure 17: DSM RIM Test Example ....................................................................... .43 Figure 18: DSM TRC Test Example ....................................................................... 44 Figure 19: Xcel Energy DSM Programs .................................................................. 47 Figure 20: Xcel Energy DSM Avoided Marginal Energy Prices ...... ................. ....... .47 Figure 21: Xcel Energy DSM Avoided Emissions Costs ........................................ .48 Figure 22: Colorado Transmission Constrained Diagram ....................................... 57 Figure 23 Market Prices Forecast for Power Imports to Colorado ........................... 60 Figure 24: Colorado Sectoral Energy Demand .................................. ....... ... ........ 63 Figure 25: Reference Scenario (BAU) Primary Fuel Consumption ......................... 63 Figure 26: Reference Scenario (BAU) Electricity Generation .................................. 64 Figure 27: Reference Scenario (BAU) Installed and New Capacity Additions ......... 66 Figure 28: Reference Scenario (BAU) Projected C02 Emissions Profile ................ 69 Figure 29: Advanced Technology Scenario Electricity Generation ..................... .... 70 Figure 30: Advanced Technology Scenario New Capacity Additions ...................... 71 Figure 31: Advanced Technology Scenario Costs and C02 Emissions Profile ....... 72 Figure 32: DSM/EE Scenarios Total Electricity Generation .................................... 77 Figure 33: Energy Efficiency Scenarios Generation by Fuel Type .......................... 78 Figure 34: Aggressive DSM/EE Scenarios New Capacity Additions .............. ........ 79 Figure 35: DSM/EE Cost Benefits (2005M$) .......................................................... 79 Figure 36: DSM/EE C02 Emissions Profile ............................................................ 80 Figure 37: Carbon Cap Policy Electricity Generation by Fuel Type ......................... 82 Figure 38: Carbon Cap Percent Share of Electricity Generation by Fuel Type ........ 83 ix

PAGE 11

Figure 39: Carbon Cap Policy Scenarios New Capacity Additions .......................... 83 Figure 40: Carbon Cap Policy System Costs and Differentials ............................... 84 Figure 41: Carbon Cap Policy C02 Emissions Profile ............................................ 84 Figure 42: S02 and NOx Emissions Profile ............................................................ 85 Figure 43: Btu Tax C02 Emissions Profile .............................................................. 86 Figure 44: C02 Tax Scenario Generation by Fuel Type ......... ............................... 87 Figure 45: C02 Tax Scenario New Capacity Additions ........................................... 88 Figure 46: C02 Tax Scenario C02 Emissions and Tax Revenues Profile .............. 89 Figure 47: C02 Tax Scenario Criteria Pollutant Profile ........................................... 89 Figure 48: Biomass Co-Fire Cost and C02 Emission Profile .................................. 90 Figure 49: Natural Gas Sensitivity Generation by Fuel ........................................... 91 Figure 50: Fuel Cost Sensitivity New Capacity Additions ........................................ 92 Figure 51: Fuel Cost Sensitivity Total System Cost Comparison ............................ 92 Figure 52: Fuel Cost Sensitivity C02 Emissions Profile .......................................... 93 Figure 53: Load Forecast Sensitivity Generation by Fuel Type ............................... 94 Figure 54: Load Forecast Sensitivity System Cost Comparison .............................. 94 Figure 55: Energy Demand Forecast Sensitivity C02 Emissions Profile ................. 95 Figure 56: Mathematical Representation of System Objective Function ................. 96 Figure 57: Sensitivity Chart of Target ForecastTotal System Cost (2035) ........... 97 Figure 58: Price/Demand Trade-Off Curve ............................................................. 98 Figure 59: Total System Fuel Consumption under Standard and Elastic Demand 100 Figure 60: C02 Emissions Profile under Standard and Elastic Demand ............... 101 Figure 61: Scenario Comparison ChartCosts & C02 Emission Differential from BAU ................................................................................................................ 103 X

PAGE 12

LIST OF TABLES Table 1: Summary of Recent Air Emissions Legislative Initiatives ............................ 3 Table 2: Colorado Net Generation by Fuel Type (1990-2005) .................. ............. 10 Table 3: Colorado Historical C02 Emissions (1990-2002) ...................................... 12 Table 4: Colorado 1990 Carbon Dioxide Emissions by Sector ................................ 13 Table 5: Carbon Dioxide Emissions from Colorado Utilities Fossil Fuel Use (1990) 13 Table 6: Forecast of Carbon Dioxide Emissions from Colorado Utilities Fossil Fuel Use through 2015 ............................................................................................. 13 Table 7: Colorado Utilities Carbon Dioxide Emissions Compared to 1990 Level .... 14 Table 8: Xcel Energy (PSCo) C02 Intensity ........................................................... 15 Table 9: Percent of Total Emissions from U.S. Electric Generation Technologies .. 17 Table 10: Metro Denver Voluntary Air Pollution Reduction Program Results .......... 18 Table 11: Metro Denver Voluntary Emissions Reduction Program Cost Components ......................................................................................................................... 19 Table 12: Colorado RPS Requirements .................................................................. 27 Table 13: RAMATS Colorado Energy and Demand Forecast ................................. 30 Table 14: CEF Colorado Energy and Demand Forecast.. ....................................... 31 Table 15: Colorado End-Use Sectors Share of Total Energy Requirements ........... 32 Table 16: Colorado Energy Forecast (Low, Base, High) ......................................... 33 Table 17: Colorado Demand Forecast (Low, Base, High) ....................................... 33 Table 18: Colorado SeasonalfTOD Sectoral Load Distribution ............................... 35 Table 19: Colorado Existing Generation Mix & Emissions Rates (Input for 2005) ... 39 Table 20: Colorado RES Requirements and conditions .......................................... 41 Table 21: Top Five Retail Electricity Providers in Colorado (2005) ........................ .45 Table 22: Xcel Energy Cumulative DSM Programs Energy Saving in Colorado ..... 46 Table 23: Colorado Generating Capacity Retirement ............................................. 54 Table 24: Thermal and Renewable Resources Cost and Performance Data .......... 55 Table 25: Technology Specific Hurdle Rates .......................................................... 57 Table 26: Colorado Transmission Constrained Paths ............................................. 58 Table 27: Ten-Years Colorado Transmission Costs (Millions2004$) ................... 58 Table 28: Colorado Statistical Population & Generation Information ....................... 62 Table 29: Reference Scenario (BAU) Generation Mix by Fuel Type ....................... 64 Table 30: Percent Share of Generation by Fuel Type ............................................. 65 Table 31: Firm Load Obligations Compared to Model Output (GW) ........................ 68 Table 32: Base-Year (2005) Emissions .................................................... ..... ......... 68 Table 33: Reference Scenario DSM Costs ............................................................. 77 Table 34: Carbon Cap Policy Effective C02 Reduction ............................ ..... ......... 81 Table 35: Carbon Cap Policy Net Cost Increase from BAU (2005M$) .................... 82 Table 36: C02 Tax Costs and Tax Revenues ........................................................ 88 Table 37: Monte Carlo Simulation Inputs ................................................................ 96 xi

PAGE 13

Table 38: Elastic Demand Consumer/Producers Surpluses .......... ...................... 100 xii

PAGE 14

PREFACE There is widespread scientific consensus that global warming is occurring and poses serious risk of adverse climate change from growing greenhouse gas emissions -in particular carbon dioxide (C02). Climate change is a long-term global problem and its economic and environmental impacts on society may take many years to develop fully. Certain countries have begun to adopt C02 mitigation policies to comply with the Kyoto Protocol. In the United States, some states and regions have begun to adopt C02 mitigation policies in order to stay current with possible future global C02 market and mandatory C02 regulatory regime. Such early action mitigation policies allow states, regions, and power companies to adjust gradually to changing economic conditions that may arise from carbon mitigation policies. However, most states including Colorado lack the tools needed to assess the tradeoff between economic costs and society gains from emission reduction required for sustainable energy planning. In 2004, Colorado voters passed Amendment 37 creating a Renewable Energy Standard (RES) for the State, which required 10% of regulated utilities retail energy be produced from renewable energy resources. In 2006, the Colorado legislature passed new laws that encouraged the development of integrated gasification combined-cycle generation. In 2007, new bills were passed that doubled the amount of RES to 20% for regulated utilities and added a 10% requirement for non-regulated utilities (Cooperatives and Municipalities). This bill also encouraged the development of new transmission infrastructure to support the development of new renewable energy resources and established energy efficiency and Demand-Side Management (DSM) goals for the regulated utilities. Also in 2007, The Governor of Colorado announced a statewide plan to reduce C02 emissions by 20% from 2005 actual emission levels by 2020 and 80% by 2050. These new legislative actions in Colorado have created challenges for both the regulators and the utilities. Most significantly, the new legislation has replaced earlier "least-cost" and "fuel neutrality" utility planning goals with the idea of "cost effective resource planning," which takes into consideration the costs and benefits of adding more renewable resources and DSM programs to the utility's resource plan for resource acquisition. This study is the first direct statewide assessment of new legislative mandates for more renewable and energy efficiency measures to reduce greenhouse gas emissions from Colorado's power sector. xiii

PAGE 15

1 INTRODUCTION 1.1 Global Greenhouse Gas initiatives The Intergovernmental Panel on Climate Change's Second Assessment Report (IPCC, 1996) concluded that "the balance of evidence suggests that there is a discernible human influence on global climate" [1]. The IPCC, 1996 report formed the basis for the Kyoto Protocol, a global initiative on climate change developed by the United Nations Framework Convention on Climate Change (UNFCCC), which assigns its signatory nations mandatory goals for the reduction of greenhouse gas emissions. Signed in December 1997, the Kyoto Protocol went into effect on February 2005 [2]. The objective of Kyoto Protocol is the "stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner'' [3]. Those countries that ratify the protocol commit to reduce their emissions of Carbon Dioxide (C02) and five other greenhouse gases, or to engage in emissions trading if they maintain or increase emissions of greenhouse gases. The Kyoto Protocol now covers more than 160 countries globally and over 55% of global GHG emissions [4]. The Kyoto Protocol establishes key fundamental principles for the signatory countries to follow in the first commitment period (2008-2012): 1 Governments are separated into two general categories: developed countries (Annex I countries) who have accepted GHG emission reduction obligations, (with the exception of the United States and Australia) and who must submit an annual GHG inventory; and developing countries (Non-Annex I countries), who have no GHG emission reduction obligations but may participate in the Clean Development Mechanism. Any Annex I country that fails to meet its Kyoto obligation will have to submit 1.3 emission allowances in a second commitment period for every ton of GHG emissions they exceed their cap in the first commitment period. During the first period, Annex I countries are required to reduce their GHG emissions by an average of 5% below their 1990 levels. Reduction limitations expire in 2013. 1 The second commitment period will be established in subsequent revisions.

PAGE 16

Kyoto includes "flexible mechanisms," which allow Annex I economies to meet their GHG emission limitation by purchasing GHG emission reductions from elsewhere. These can be bought either from the main non-Kyoto compliant markets (such as UK Emissions Trading Scheme [ETS]; the European Union ETS; the Chicago Climate Exchange, or from projects that reduce emissions in non-Annex I economies under the Clean Development Mechanism or in other Annex-1 countries under the Joint Implementation [5]. In a recent study by the World Bank, the need for future action to reduce the risks of climate change has played a major role on the international agenda. A variety of approaches are being implemented to reduce carbon emissions, ranging from efforts by individuals and firms to initiatives at the city, state, regional, and global levels. Major global initiatives to reduce nations' climate footprint include the 1992 UN Framework Convention on Climate Change and its 1997 Kyoto Protocol and Europe's carbon constraint for electricity generators and industry under the European Union ETS [5a]. Carbon markets now form a new sector in the worldwide economy. A credible response to climate change, these markets also provide a new and powerful tool for future climate mitigation. The World Bank's study valued carbon market growth at US$30 billion in 2006, three times greater than in the previous year. The study also showed that the market was dominated by the sale and re-sale, under the EU ETS, of European Union Allowances (EUAs) at a value of nearly $25 billion. Project-based activities through the COM and Jl also grew sharply, rising to a 2006 value of about US$5 billion. In 2006, the Chicago Climate Exchange and the New South Wales Market also traded record volumes and values. 1.2 Greenhouse Gas Initiatives in the United States Currently the United States does not regulate C02 domestically and the EPA has not promulgated emission limits for C02. However, congressional attempts to mitigate carbon have begun to appear more frequently in legislative sessions (Table 1 ). The U.S. is a signatory but has neither ratified nor withdrawn from the Kyoto Protocol. 2 On July 25, 1997, before the Kyoto Protocol was finalized, the U.S. Senate voted unanimously (S. Res. 98) that the United States should not be a signatory to any protocol that did not include binding targets and timetables for developing as well as industrialized nations and which "would result in serious harm to the economy of the United States." A successor to the Kyoto Protocol with a global cap-and-trade system that would apply to both industrialized nations and developing countries is in the works and could be in place by 2009 [6]. 2 The signature is symbolic. The Kyoto Protocol is non-binding on the United States unless ratified. 2

PAGE 17

Table 1: Summary of Recent Air Emissions Legislative Initiatives Legislative Sulfur Dioxide Nitrogen Oxides Mercury (Hg) Carbon Dioxide Notes Initiatives (S02) (NOx) (C02) Acid Rain Control Nationwide 8 90 Nat i onwide 2 1 o EPA Promulgate Hg By 2009 EPA to $6000/ton penalty Act of 2005 ( 1 09lh m i llion tons reduced million tons for reduct i on identify credible for NOx nonCongress H R to 4 45 mill i ons for 201 0 2013 & 1 70 regulations for indicators to compliance 227 ) to reduce acid 2010 2013 & 3 0 million tons Eastern & western protect ecosystems Emission trading for depos i t i on under mill i on tons thereafter Regions for such as Rocky Hg not allowed CAA thereafter significant Mountains & others reduct i ons class I Clean Air Planning 4.5 million tons for 1.87 million tons for 24 tons for 2010Affected units Applies to >25 MW Act of2005 2010-2013 & 3.5 2009-2014 & 1 7 2014 & shall not 2006 emission mercury and NOx (Introduced in miHion tons 2014million tons exceed 50% of Hg rates determined allowance program House H R. 1873) 2016 and 2 25 thereafter in Coal or 41bsfTBtu by EIA used for NOx allowance of million tons and for 2015 & 2010-2014 and 1 51b/MWh of 3 thereafter thereafter 10 tons units emission years average & Hg and 30% of Hg In rates for 2001 used allowance 0000227 Coal or updated by for 2015 & lbs/MWh of 3 yrs EPA output-based thereafter avg penalty NOx rate $5000, Hg $10000 Clean Power Act of By 2010 reduce By 2010 reduce to By 2010 reduce to By 2009 reduce to Cap and trade for all 2005 (Introduced in nat i onal to 2 25 1 .51 mill i on tons 2050 mill i on tons 5 tons but Hg Applies to Senate S 150) after 2010 western >15 MW Allowances region to 2 75 & should be nonwestern region distingu i shable to 1.975 m illi on tons between western & nonwestern region Clean Smokestacks 75% reduction from 75% reduction from Reduction to 1990 90% reduction from Cap-and-trade for all Act of2005 the Phase II 19971evel level 19991evels pollutants but Hg (Introduced in requirements under House H R. 1451) title IV Clear Skies Act of 30 % by 2010 & 30% by 2010 & By 2010 lesser of Not cons i dered Safety valve ; 2005 (Introduced in 50 % by 2018 from 50 % by 2018 from unit s baseline allowances priced Senate S 131) baseline or most baseline or most allowable emi ssions for S02 at $2000 stringent Federal or stringent Federal or rate under NESHAP NOx at $4000 and State emission State emission or most stringent Hg at $2187.5 CPI limitation applicable limitation Federal/State adjusted. NOx zonal to baseline year applicable to emission limitation trades WRAP baseline year applicable to specific S02 trades baseline year Mercury Emission By 2010 & By 2010 & For 2009 subject to By 2010 & Act of 2005 thereafter 2 .75 thereafter 1 510 section 112 of CAA thereafter 2 050 (Introduced in million tons in million tons not to exceed 2 48 million tons Senate S.730) western region and grams per 1000 1 .975 million tons in MWh & no coal type nonwestern region differentiated And by 2010 thereafter 5 tons nationwide limit. n the United States there is a voluntary program in place to collect and report information on annual greenhouse gas emissions. The Energy Policy Act (EPA) of 1992 directed the Energy Information Administration (EIA) to establish a mechanism for "the voluntary collection and reporting of information on annual reductions of greenhouse gas emissions [7]. The EIA, which has gathered and reported GHG emissions for the period of 19942004 [8], reports that in 2004: The U.S. power industry emitted 2,298.6 million metric tons of carbon dioxide (million MTC02) from the combustion of fossil fuels (coal, oil and natural gas) during the generation of electricity. 3

PAGE 18

Electric power generation accounts for 38% of total U.S. C02 emissions. Electric power generation accounts for 32% of total U.S. greenhouse gas emissions. The report also indicates that, in 1990-2004: C02 emissions from the electric power industry have increased by 496.3 million metric tons, or 27 percent. This trend reflects rises in: o U.S. population, which increased by 18 percent, going from 248.7 to 293.7 million o Economic growth: GDP grew by about 51 percent o Corresponding increased electric power requirements. Yet the report also indicates, for the 1990-2004 period, C02 emissions intensity from electricity generation fell by 2.1 percent, going from 0.593 MTC02 per megawatt hour (MWh) generated in 1990 to 0.580 MTC02 per MWh generated in 2004. The drop in C02 intensity reflects the increased use of natural gas and nuclear power for electricity generation [9].3 Although there is no mandatory requirement in the U.S. to reduce GHG, many states and local governments have begun with their initiatives and are moving ahead with their own legislation on GHG reduction programs. For example, Regional Greenhouse Gases Initiatives (RGGI) is a Northeastern states program to reduce GHG emissions, and California recently passed the Global Warming Solutions Act of 2006, A.B. 32 aimed at reducing carbon emissions from sources within the state. At the local level, many cities such as the City of Denver, are establishing Climate Action Plans for their own GHG reduction programs [11]. 1.3 Regional Greenhouse Gas Initiatives (RGGI) RGGI is a regional initiative by Northeastern states to reduce GHG emissions. Currently, the RGGI has eight participating states: Connecticut, Delaware, Maine, Maryland, New Hampshire, New Jersey, New York, and Vermont, and Massachusetts, Pennsylvania, Rhode Island, and the District of Columbia have shown interest in joining. 4 Similar to the cap-and-trade program to control acid rain, the RGGI's cap-and-trade program seeks to reduce GHG emissions from electric power sector. It includes a carbon dioxide budget, state emission caps for fossil fuel-fired electric power plants of at least 25 MW of generating capacity, scheduled emission reductions, provisions for the use of offsets, and provisions for trading of carbon dioxide allowances. 3 In 2005, EIA stopped producing the report on the Voluntary Reporting of Greenhouse Gases Program [ 10). 4 Available online at: http://www.rggi.org/states.htm 4

PAGE 19

Starting in January 2009, each participating State has agreed to cap its GHG emissions from power production with the goal of stabilizing emissions by 2015 to the level of three years' average emissions (2002-2004), followed by a 10% reduction between 2015 and 2020 [12]. The proposal allows participants to purchase offsets to meet 50% of their emission reductions. RGGI energy modeling projects that, under RGGI policy scenarios, electricity prices will increase. Yet RGGI points out that: "policies to deliver meaningful end-use energy efficiency measures (both through RGGI and due to other state energy efficiency policies) are effective in sufficiently reducing total electricity usage by households so as to overcome the price increase impact of RGGI resulting in a net reduction in expenditures on average across households. The modeling methodology for RGGI GHG emissions reduction is discussed in Chapter 4 below. 1.4 California GHG initiatives California's Global Warming Solutions Act of 2006 (A.B. 32) aims to reduce carbon emissions from sources within the state. The 2006 Act limits the state's 2020 carbon emissions to 1990 levels (roughly a 25% cut) and implements a reporting system to monitor compliance.5 The Act specifies that all emissions from the generation of power consumed within the state are expected to be subject to the new laws. Because California imports power from neighboring states, emissions in those states may also be affected. In addition, through the West Coast Governors' Global Warming Initiative, California collaborates with Washington and Oregon on its greenhouse gas policy. The Act also includes an allowance trading program that has yet to be designed [13]. In 2006, the California Senate also passed a companion bill (S.B. 1368), which prevents load-serving entities from entering into long-term contracts for base-load generation unless the plants comply with a GHG performance standard. That standard will require that power imports from other states (currently about 25% of California power is imported) not exceed the rate of emissions for combined-cycle natural gas base-load generation. A "tax" would apply to power imports not meeting the standard [14]. 1.4.1 West Coast Governors' Global Warming Initiative The Governors of California, Oregon, and Washington have approved a series of recommendations for action to combat global warming and are working together on state and regional goals and strategies to combat global warming over the coming years. In November 2004, a staff report stated: [15] ... Global warming will have serious adverse consequences on the economy, health and environment of the West Coast states. These impacts will grow 5 California has negligible coal-fired generation. Most of California generation is from Natural Gas, Nuclear, or Renewable. 5

PAGE 20

significantly in coming years if we do nothing to reduce greenhouse gas pollution. Fortunately, addressing global warming carries substantial economic benefits. The West Coast region is rich in renewable energy resources and advanced energy-efficient technologies. We can capitalize on these strengths and invest in the clean energy resources of our region." The Governors of three states have committed to act individually and regionally to develop strategies that promote long-term economic growth, protect public health and the environment, consider social equity, and expand public awareness in order to reduce greenhouse gas emissions below current levels. 1.5 Research Motivation The motivation for this study derives from the fact that there is as yet no study of Colorado to identify the outlook for the state power sector's future C02 emissions, to estimate costs and benefits, or to consider primary air pollutants in addition to C02. Recently published reports have identified the state's growing need for electricity and plans on building more conventional, coal-fired power plants. Yet these plans of action fail to take into account the quantity and impact of C02 emissions, despite the looming probability of future mandatory limits on C02 emissions. See reports [19], [20], and [21]. This study is an attempt to provide a statewide energy planning and policy evaluation model that not only considers ways to respond to increased energy needs but also ways to decrease the sector's carbon and pollution footprint. 1.6 Research Objective Through this research, we have developed a modeling framework and a Colorado specific database to investigate scenarios of sustainable power generation. We used a MARKAL optimization model to: Model various scenarios for future sustainable energy production in Colorado Model ways to improve competitiveness of renewable energy Model ways to keep fossil fuel generation relatively competitive. Evaluate costs and benefits of alternative scenarios for statewide C02 mitigation targets Evaluate reduction of criteria pollutants (S02 and NOx) emissions as ancillary benefits. In order to investigate scenarios of future electric generation technologies and their impact on state's future GHG and air pollutant emissions, the following scenarios were developed and evaluated: Reference Scenario (Business-As-Usual) Advanced Emerging Technology Scenario Energy Efficiency Scenarios Regulatory Policy Scenarios Sensitivity Scenarios 6

PAGE 21

The research aimed to understand how technology and policy could affect Colorado's power sector C02 emissions intensity in the future. It also sought to develop an optimized assessment of power generation existing and future statewide to determine the best resource mix for sustainable energy in the future. 1. 7 Research Impact This study provides results of a Colorado-specific case study of the potential for efficient and clean energy technologies to address a number of energy-related challenges within the power sector facing the state. These challenges include climate change, recently promulgated Renewable Portfolio Standards, fuel price volatility, power transmission constraints, and inefficiencies in energy production. Some of the challenges are visible today and are being integrated into public policies at various local governmental levels (see, for instance, the Greenprint Denver Council Climate Action Plan). Others are emerging and have uncertain future outcomes. How the state will respond to them will affect Colorado's economy and the well being of its citizens. In considering the environmental impact of fossil-fueled electricity generation on the state's emissions levels, this study makes a major contribution to power sector planning. The study also explores clean energy technology and makes a strong case for the value of renewable energy in the state's energy program. Finally, the study identifies specific policy scenarios policymakers can use to design an appropriate environmental and economic response to Colorado's burgeoning energy needs. 7

PAGE 22

2 BACKGROUND ON ENERGY OUTLOOK 2.1 The U.S. Energy Outlook Recent volatility in energy prices has changed the future energy outlook. According to the Energy Information Administration (EIA)'s 2006 Annual Energy Outlook (AE02006), the cost of fuel delivered to electricity generators in 2025 will be higher than projected in the AE02005. The projection of increased generation of electricity from coal-fired plants and decline in natural-gas fired generation is, in part, due to higher natural gas prices and slower growth in electricity demand compared to the 2005 projection. The AEO 2006 projected electricity generation of 1,070 billion kilowatt-hours (kWh) from natural gas in 2025 is 24% lower than the year AE02005 projection of 1,406 billion kWh [16]. 2.1.1 The U.S. Carbon Dioxide Emissions AE02006 also projects carbon dioxide (C02) emissions from energy use to increase from 5,900 million metric tons in 2004 to 7,587 million metric tons in 2025 and 8,114 million metric tons in 2030, for an average annual increase of 1.2% per year. Projected increased C02 emissions from coal in 2030 is due to higher natural gas prices and the use of more coal generation to displace higher-cost gas generation. AE02006 also projects 6 gigawatt (GW) of new nuclear capacity additions, with no additional new nuclear plants after 2020 due to expiration of the Energy Policy Act of 2005 (EPAct2005) production tax credit [17]. 2.1.2 The U.S. Renewable Energy U.S. expected capacity expansion from renewable generating units is projected to be about 8 percent. The U.S.'s renewable electricity generation is projected to grow by 1. 7 percent per year to displace fossil-fueled electricity generation mainly due to higher fossil fuel prices. Growth is related to improved renewable technology and State Renewable Portfolio Standards (RPS). The expected affect of State RPS programs, which specify a minimum share of generation or sales from renewable sources, are included in this projection. The projection also includes the extension and expansion of the Federal tax credit for renewable generation through December 31,2007, as enacted in EPAct2005 [17]. 2.2 Colorado Energy Outlook Recent uncertainty and volatility in natural gas prices has also affected Colorado's future energy outlook. In 2005, after nearly three-decades, a new coal-fired power plant with a capacity of 750 MW was approved to be built in Colorado with a service date of 2009 [18]. A recent report by Western Resource Advocates also reported that a new 600 MW coal-fired power plant is proposed to be built in the southeast of 8

PAGE 23

Colorado [19]. A study by the Colorado Long Range Transmission Planning Group (CLRTPG) points out the possibility of building an additional 500 MW coal-fired unit in Brush, Colorado. The CLRTPG study is discussed in Chapter 5. Climate Alert Report, published by Environmental Defense, raises concerns about the new trend for utilities to build ever more conventional coal-fired plants. The report emphasizes its concern by stating: ... In the southwestern United States, the race is on between efforts to curb global warming pollution and proposals to build more than a dozen outdated, high polluting power plants" [20]. The proposed increased generation of electricity from conventional coal-fired plants and decline in proposed natural-gas fired generation in Colorado is, in part, due to increased demand for base-load generation and to higher projected natural gas prices in future. This trend is consistent with the national trend projected by DOE/EIA. In 2004, steam coal (43 percent) and natural gas (42 percent) fired power plants accounted for 85 percent of Colorado's installed power generation capacity. Of 42 percent gas generation capacity, 23% was combined-cycle generation capacity, and the remaining 19% combustion turbine generation capacity. Hydroelectric accounted for 6% and pumped storage facilities for 5%, whereas renewables (wind) and steam oil generating capability accounted for 2% each. (Figure 1) 211104 CalarmoO.rw.-. EIVT.-.alclgy"Jp Sl:ua5: Nr6:Jnll Elldric Ei"&gf r:---. Wrd 2%1 5% I Iii Slaom-Cool 43% Figure 1: Colorado Electric Power Generation Capability in 2004 In 1990, 92% of Colorado's net power generation was from coal-fired generating power plants, and 4% from gas-fired generation. Due to the economic expansion of last decade, Colorado experienced a high growth in use of electricity that resulted in a surge of installed gas-fired generation capacity to meet the increased demand. In 2005, power generation from gas-fired units increased to 24% of Colorado net power generation, while generation from coal-fired met 72% of Colorado's net power 9

PAGE 24

generation needs. Table 2 shows Colorado's net electricity generation by fuel type (percent share) in five year increments from 1990 to 2005. Table 2: Colorado Net Generation by Fuel Type (1990-2005) Average Annual 1990 1990 1995 2000 2005 Growth Share 1995 2000 2005 Fuel Type (MWh) (MWh) (MWh) (MWh) Rate(%) (%) Share(%) Share(%) Share(%) Coal 29,814,983 30,492,682 35,381,219 35,570,135 1.3% 91.6% 85.6% 80.1% 71.7% Oil 27,390 11,712 109,385 17,046 -2.5% 0.1% 0.0% 0.2% 0.0% Natural Gas 1,290,092 2,856,788 7,157,438 11,923,290 54.9% 4.0% 8.0% 16.2% 24.0% Other Gas 0 0 0 2,430 0.0% 0.0% 0.0% 0.0% Hydro 1,419,870 2,131,189 1,454,415 1,415,296 0.0% 4.4% 6.0% 3.3% 2.9% Renewable 28,990 32,910 17,914 810,561 179.7% 0.1% 0.1% 0.0% 1.6% Pump Storage -33,198 91,953 45,175 -122,063 17.8% -0.1% 0.3% 0.1% -0.2% Total Generation 32,548,127 35,617,234 44,165,546 49,616,695 3.5% 100.0% 100.0% 100.0% 100.0% Source: DOE/EIA Because coal prices are considerably lower than natural gas prices, coal-fired capacity is generally more economical to operate than natural-gas fired capacity. The Colorado share of natural gas generation was much less before the last decade, when a surge of new natural-gas fired plants were installed to meet needs and ensure reliability. These plants operate comparatively fewer hours than coal plants, coming into use only when electricity demand is high. (See Figure 2 for Colorado's net generation share by fuel type in 2005.) A recent report by the Colorado Energy Forum (CEF) raised the question of what needs to be done today to ensure that all Colorado families, businesses, and communities have affordable, reliable, and environmentally sound sources of electricity in the decades ahead. The report concluded that, with Colorado energy demand expected to grow at approximately 2% per year until 2025, the state's utilities will need significantly more and newer transmission infrastructure to meet demand. [21] There are over 60 electric distribution utilities serving end-users in Colorado. Aquila and Xcel energy (aka Public Service Company of Colorado) serve 59% of the state as regulated utilities under the jurisdiction of Colorado Public Utilities Commission. The other41% of the state's end-users are served by non-regulated Municipal (18%) and Cooperative (23%) utilities [21]. Colorado's electric end-users consist of three main demographics: residential (34%), commercial (41%) and industrial (25%) (See Chapter 5 for details). 10

PAGE 25

Colorado Net Gel lenition Share 1-+jdo, 2.9% OherGas, 0.0% 1\Etlr.ll Gas, 24.0% 01,0.0% Renelll.slje, 1.6% PuTp Staage, 0.2% Figure 2: Colorado Net Generation in 2005 The CEF report projected Colorado's electricity need as 4,900-7,000 megawatts (MW) of new generation by 2025. To meet projected growth, the report suggests a generation portfolio mix of approximately 3,000 MW of base-load power, 1,500 MW of intermediate power and 1,350 MW of peaking power. The CEF report, however, does not offer any recommendations on how the state should address its growing energy needs, nor does it offer any preferred future generation resource mix from the currently available choices -traditional electric generating resources, emerging advanced resources (integrated gasification combined cycle coal technologies, advanced combined cycle with carbon capture, and sequestration technology), renewable resources (wind, solar, geothermal), or energy efficiency. Rather, it outlines the magnitude of the problem and leaves it to the citizens of Colorado to determine the specific set of generation resources and energy efficiency measures required to address their growing need for power [21]. 2.2.1 Colorado Sectoral Carbon Dioxide Emissions In 1990, Colorado's total C02 emissions from all sectors were 66.11 MMT, of which electric power's share was 30.96 MMT (47%) and transportation's was 18.90 MMT (29%). (Table 3) 11

PAGE 26

Table 3: Colorado Historical C02 Emissions (1990-2002) Colorado C02 Emissions from Fossil Fuel Combustion Million Metric Tons C02 (MMTC02 ) Sector/Year Year 1990 1991 1992 1993 1994 1996 1996 1997 1998 1999 2000 2001 2002 Commercial 3.96 4.26 4.09 4.24 4.04 4.01 4.21 4.18 3.78 3.81 3.76 4.38 4.29 Industrial 7.00 7.91 8.13 8.84 8.34 8.20 8.45 8.80 9.25 9.03 9.60 12.96 12.33 Residential 5.29 5.78 5.51 8.09 5.88 8.10 8.45 8.27 5.94 8.39 8.82 7.28 7.49 Transportation 18.90 19.08 19.31 21.23 21.51 22.15 22.57 22.41 23.50 25.05 25.31 26.61 26.41 Electric Power 30.96 30.30 31.24 31.81 33.14 32.18 33.87 34.06 34.50 34.75 38.27 40.42 39.22 Total All Sectors 66.11 67.31 68.27 72.22 72.71 72.64 76.64 76.72 76.96 79.03 83.77 91.62 89.74 Source:http://epa.gov/dimatechange/emissions/downloads/C02FFC_2002.xls See Figure 3 for C02 emissions level distribution by five main sectors: residential, commercial, industrial, transportation, and electric utilities. 45.00 4000 35.00 30.00 25.00 MMTC02 2 0 00 15.00 10 .00 5 00 0.00 "' _,, .J--1-1-r ] 1 Colorado C02 Emlulona from Fo .. n Fuel Combuatlon Source: EPA/Climate Change Emissions 1 -1 -11---1 -1-1 1 ----, -J 1 1 1 ] ] ff-l l --1111---= f--f-l J 1 1990 1991 1992 1993 11194 1995 1996 1997 1998 1999 2000 2001 2002 IJCommerdel lndusbilll DResidential OTransportation Etectric Power Figure 3: Colorado Fossil Fuel C02 Emissions Colorado Department of Public Health & Environment (CDPHE) Greenhouse Gas Emissions Inventory & Forecast (1990 through 2015) show Colorado C02 emission for 1990 for all sectors at 78.72 million tons (71.56 MMT). The C02 emissions from Colorado power sector for 1990 were 38.97 million tons (35.43 MMT). See Table 4. 12

PAGE 27

Table 4: Colorado 1990 Carbon Dioxide Emissions by Sector SECTOR CARBON DIOXIDE PERCENT OF (TONS) TOTAL Utilities 38,966,038.07 49.6% Transportation 19,430,412.68 24.7% Industrial 9,826,549.78 12.5% Residential 6,052,128.37 7.6% Commercial 4,445,451.91 5.6% Total 78,720,580.81 100% CDPHE estimates show that, to generate electricity in 1990, Colorado utilities used 380.4 trillion Btu of coal and 5.1 trillion Btu of natural gas. (Table 5) Table 5: Carbon Dioxide Emissions from Colorado Utilities Fossil Fuel Use (1990) TYPE OF FUEL FUEL CONSUMPTION C02 EMISSIONS (MILLION BTU) (TONS) Biomass* 29,223.10 22.90 Bituminous Coal 380,424,360.00 38,666,331.95 Natural Gas 5,150,000.00 299,683.22 Total 385,603,583.10 38,966,038.07 Source: Colorado Greenhouse Gas Emissions Inventory & Forecast (1990 through 2015). Colorado Department of Public Health & Environment, Revised Oct. 2002. *Biomass is no longer included in the greenhouse gas inventory per EPA's State Worlcbook Volume VIII. CDPHE 2002 forecasts that by 2015, Colorado utilities will require 42% more coal and natural gas, with a corresponding rise in C02 emissions According to this scenario, power sector C02 emissions measured at 38.9 million tons in 1990 will reach 55.4 million tons by 2015 (Table 6). Table 6: Forecast of Carbon Dioxide Emissions from Colorado Utilities Fossil Fuel Use through 2015 FUEL ENERGY CONSUMPTION C02 EMISSIONS (MILLION BTU) (TONS) Bituminous Coal 540,659,105.21 54,952,591.45 Natural Gas 7,319,180.00 425,909.79 Other 41,531.16 32.55 Total Projected for 2015 548,019,816.40 55,378,533.80 Source: Colorado Greenhouse Gas Emissions Inventory & Forecast (1990 through 2015), Colorado Department of Public Health & Environment, Revised Oct. 2002. From 1990 to 1997, Colorado C02 emissions from utilities increased 7.58 %, rising from 38.97 million to 41.92 million tons. By 2005, utilities C02 emissions were 45.17 million tons (41.06 MT), or 15.92% over 1990 levels.6 Yet these figures do not take 6 Source of C02 emissions for 2005 is from EPAEmissions Tracking System. Total C02 tonnage does not include purchased power C02 Emissions. 13

PAGE 28

into account the fact that Colorado is a net importer of electricity. When C02 emissions are added to for imported electricity, the utilities' C02 footprint is even larger (Table 7). In 2007, Coloradans alarmed by escalating C02 emissions, passed a Renewable Energy Standard (RES) designed to bring C02 emissions far below levels projected by CDPHE (See 2.2.3 below). Table 7: Colorado Utilities Carbon Dioxide Emissions Compared to 1990 Level Year Coal Coal C02 Gas Gas C02 Total C02 % Consumption Emissions Consumption Emissions Emissions Increase (mm BTU) (tons) (mm BTU) (tons) (tons) form 1990 1990 380,424 360 38 666 332 5,150 000 299 683 38, 966 0 1 5 -1997 408 ,901, 240 41, 560 722 6 180 000 359 620 41, 920 342 7 58 2005 392 402 546 40 227 ,881 83, 147 0 7 9 4 942 296 45 170 177 15 .92 . .. U S EPAEm1ss1ons Trac kmg System T otal C02 tonnage does not mclude purchased power C02 Em1ss1ons In this study, the Reference Case is benchmarked to 43.6 MMT C02 emissions for 2005 including about 1. 78 MMT C02 for power imported into Colorado. In the 1990 Cap Carbon Policy Scenario, C02 emissions are capped at the 1990 level of 30.96 MMT. 2.2.2 Xcel Energy (PSCo) Carbon Dioxide (C02) Intensity Xcel Energy's Colorado operation is the Public Service Company (PSCo), which currently serves close to 60% of Colorado's electricity needs. In 2006, As part of the Greenprint Denver Council working group, Xcel calculated PSCo's C02 intensity metric. According to thse calculations, in 2005 PSCo's owned generation C02 intensity was 2,067 lbs/MWh. When combined with its purchased power 1,242 lbs/MWh, the company's C02 intensity dropped to 1 ,7481bs/MWh (Table 8). 14

PAGE 29

Table 8: Xcel Energy (PSCo) C02 Intensity Xcel Energy (PSCo) Energy Resources Mix C02 Intensity By Source Class and Energy Type (2005) Generation Sources C02 Tons MWh C02% MWh% (Lbs/MWh) Owned ETS Coal Units 20,933,481 17,557,807 65.6% 481% 2,385 1 Owned ETS Gas Units 1,971,936 4,324,392 6.2% 11.8% 912 Owned non-ETS Coal Unils 242,341 153,423 0.8% 0.4% 3,159 Owned non-ETS Gas Units 3,215 3,009 0.0% 0.0% 2,137 Owned non-ETS Hydro Units 0 197,468 0.0% 0.5% 0 2 Owned non-ETS Non-Emittor Unrts 0 160,472 0.0% 0.4% 0 OWned G_,.uon subtollll 23,150,973 22,396,571 72.5% 81.3% 2,08 Purchased Coal 3,175,342 3,028,211 9.9% 8.3% 2,097 Purchased Gas 5,599,478 10,109,452 17.5% 27.7% 1,108 Purchased Hydro 0 552,931 0.0% 15% 0 3 Purchased Zero Emittor 0 436,744 0.0% 12% 0 Pure haMel Power SUbtollll e,n4,120 14,127,331 27.5% 31.7% 1,242 PSCo Energy Resources Total 31,925,793 36,523,909 100.0% 100.0% 1,748 Source: Chapman, D. Xcel Energy's C02e Intensity Metric; Xcel Energy, 2006. Collected as part of the University of Colorado Denver working group with the Denver Greenprint Council As can be seen in Figure 4, coal fired power (owned) accounts for some 65.6% of PSCo's C02 intensity, gas-fired power plants (owned) for 17.5%, coal generation (purchased) for 9.95%, and gas generation (purchased) for 6.18%. Xcel Energy (PSCO) Energy Resource Mix C02 Intensity By Source ct .. end Energy Type (2005) O.C. Sou"'": EPAETli(Enaalon Troclltne Syslllm) & Non-ETS Ownod nonTS Coal Unilo,0.71% Figure 4: Xcel Energy (PSCo) Energy Resource Mix C02 Intensity 15

PAGE 30

2.2.3 The Colorado Renewable Energy Standard In 2004 Coloradans passed "Amendment 37," a voter initiative that established the first Renewable Energy Standard (RES) in the nation. The 2004 RES applied to the state's two rate-regulated utilities: Xcel Energy (PSCo) and Aquila [22]. It allowed other Colorado covered utilities (serving 40,000+ customers) to opt out of the RES. It also allowed for exempt utilities, with a majority vote involving a minimum of 25 percent of the utility's customers, to opt in [23]. In 2007, Colorado revised and extended the 2004 RES requirements in House Bill 1281. This Bill increased requirements for the rate-regulated utilities in 2015 from 10 to 15%, and in 2020 and thereafter to 20%. The new legislation also requires cooperatives and municipally owned utilities to include renewable energy in their resource portfolio at the less stringent rate of 10% of retail sales by 2020. House Bill 1281 applies to every provider of retail electric service in the state of Colorado except municipally owned utilities that serves 40 thousand customers or less. House Bill 1281 defines 'Renewable Energy Resources' as: solar, wind, geothermal, biomass, new hydroelectricity (with a nameplate rating of 10 megawatts or less), and hydroelectricity in existence on January 1, 2005 (with a nameplate rating of 30 megawatts or less). Fossil and nuclear fuels and their derivatives are not eligible energy resources. The Bill caps the retail rate impact of RPS in Colorado at 1%. A number of other states have now instituted renewable fuel goals similar to those enacted in Colorado. In 2004, New York set its renewable energy production goal at 25% of by 2013. In 2005, Vermont set a voluntary goal of 10% of total electricity sales from renewable sources, a goal that will become mandatory if it is not met by 2012 [24]. 2.3 Air Quality Impact of Electricity Generation Electricity generation from current fossil fuel technologies produce a variety of atmospheric pollutants, including criteria pollutants (S02 and NOx), particulates, C02, and mercury. Table 9 (from EPA sources) shows the percentage of different pollutants produced by generating electricity, and how these variously affect ambient air quality, air toxicity, and climate change. Together S02 and NOX contribute to acid rain, while S02 also contributes to PM2.5 and NOX to ozone formation, further degrading ambient air quality. The fact that reducing GHG emissions can also reduce S02 and NOx emissions, and thereby improve air quality, is an important policy consideration. 16

PAGE 31

Table 9: Percent of Total Emissions from U.S. Electric Generation Technologies Impact Ambient Air Quality Toxic Climate Acid Rain Change Pollutant PM10 PM2 5 S02 NOx Hg C02 N20 Emission 16 3 65 20 43 38 4 % Source: EPA [28) 2.3.1 State's Electric Power Sector Air Emission Regulations The U.S. EPA and several states have recently enacted air emission regulations governing the emission of NOx, S02, and mercury from power plants. 7 North Carolina's Clean Smokestacks Initiative has announced compliance plans for the installation of S02 scrubbers, selective catalytic reduction (SCR), and selective non catalytic reduction (SNCR) NOx-removal technologies [16]. 2.3.2 Colorado Electric Power Sector Air Emission Regulations In 1998, Colorado's Senate Bill 98-142 authorized voluntary air pollution reduction from stationary sources owned by Xcel (PSCo) in metropolitan Denver [26], to be fully paid for by PSCo ratepayers. 8 The overall impact of this Metro Emissions Reduction act has been positive. S02 emissions in metropolitan Denver have been reduced from historical highs by 62.7%, and S02 emission are capped at 10,000 tons per year (Figure 5). 7 The Clean Air Interstate Rule (CAIR) covers S02 and NOx emissions in the Eastern U.S.; the Clean Air Mercury Rule (CAMR) covers mercury emissions nationwide; and the Clean Air Visibility Rule (CAVR) requires certain units-depending on their visibility impactsto install pollution controls in certain areas of the country. EPA issued both CAIR and CAMR air emissions regulations in 2005. 8 Xcel Energy is recovering the costs of implementing the Voluntary Emissions Reduction Agreement through a rate rider, Metro Emissions Reduction Air Quality Improvement Rider (AQIR), mechanism spread over a fifteen year period. Such mechanisms allow utilities to proactively seek improvements in operation of power plants in order to reduce emissions. 17

PAGE 32

30 000 25,000 )( 20, 000 ri 0 1/) 15 000 10 000 5 000 Metro Denver Coal-Fired Power Plants Emissions (Arapahoe Cherokee Valmont) Wrth Anlpoh oe 1&2 R o plocomen1 with Gu Geno1'1111on 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Data Compiled from EPA ETS t.:i:02(Tons) -+-NOx(Tons) & 000 000 8 000 000 7 000 000 6 000 000 N 5 000 000 8 4 000 000 3 000 000 2 000 000 1 000 000 Figure 5: Metro Denver Voluntary Air Pollution Reduction Program Source: EPA-ETS NOx emissions, which were not capped, were voluntarily reduced by 11.8%. For the same period, however, C02 emission increased slightly (0.3%) proportional to heat input increases (Table 10). Table 10: Metro Denver Voluntary Air Pollution Reduction Program Results Metro Denver Emissions HEAT INPUT 502 NOx C02 (Tons) And Heat Input (mmBtu) (Tons) (Tons) Previous Three Years Avg (2000-2002) 83 ,871, 318 26 787 17, 522 8 540 886 Three Years Avg ( 2003 2005) 84 097 600 10 000 1 5,450 8 517 529 Percent Reduction (Increase) ( 0 3%) 62 .7% 11. 8% (0. 3%) Reduct i ons (Increase ) ( 226 282 ) 16 788 2 ,071 ( 23 356 ) Figure 6 shows 802, NOx, and C02 emissions from all coaland gas-fired power plants in Colorado. As of 2005 in general, 802 emissions had decreased, NOx emissions stayed almost the same, and C02 emissions increased to above 45 million tons (41.06 MMT). 18

PAGE 33

.. 0 z 100000 ...; eoooo 0 II) 40000 20000 Coal & Gas F i red Power Plants Em i ssions (Colorado East & Wes t) 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Data Ccmpiled from EPA E7S 1---so2(Tono) --+-N0x(Tons) C02(Tons)j Figure 6: Colorado Power Plants Air Emissions Source: EPA-ETS N 25000000 0 (.) 20000000 15000000 10000000 Table 11 shows the cost of the Air Pollution Control (APC) component of the Metro Denver Voluntary Emissions Reductions Program. About 46% of this cost is attributable to retrofitting, 19% to fixed operation and maintenance (O&M) costs, and 16% to variable O&M costs. Recognizing the complexity of the problem, Coloradans have shown growing interest in strategies that address the problem of criteria pollutants, air toxins, and greenhouse gases simultaneously [36]. Table 11: Metro Denver Voluntary Emissions Reduction Program Cost Components Metro Denver Voluntary Emissions Reduction Program Costs ( 98$) 1998 2017 %of NPV ($1000) Total Emission Control Cap ital Revenue Requi rements 94 898 46 4 Emission Control Fixed O&M Costs 38 967 19 0 Emission Cont rol Variable O&M Costs 32,381 15. 8 Emission Control Heat Rate Impacts (Fuel) 1 882 0 9 Arapahoe Repl acement Capita l Revenue 28 269 1 3 8 Requi remen t Arapahoe Replacement Fue l & Var O&M Costs 6 129 3 0 Arapahoe Replacement Fixed O&M Costs 2 043 1 0 Total $204,569 100 0 19

PAGE 34

To ensure minimum funding for "cost-effective energy efficiency" programs, which would ultimately reduce greenhouse gas production statewide, the state could establish a similar ratepayer supported program. In 1996, California established a public goods charge (AB1890) that ensured minimum funding levels for "cost effective conservation and energy efficiency." By 2000, California's program had proved so effective that the state's extended it through the year 2011 and passed an additional natural gas surcharge (AB1002) for similar purposes. Also in 2000, California passed the Energy Security and Reliability Act (AB970), which directed the state's Public Utilities Commission to establish a distribution charge to provide revenues for a self-generation program. California also issued a directive to develop more energy-efficient and cost-effective electricity generation methods and to address the state's reliability concerns. 9 In the spring of 2001, California set up a new state agency the Consumer Power and Conservation Financing Authority ("CPA"). Created to encourage energy efficiency, conservation, and the use of renewable resources, the CPA was authorized to issue up to $5 billion in revenue bonds to finance energy efficiency programs and self generation activities.10 Colorado's regulators and policymakers already have some regulatory mechanisms in place to develop a framework for a continued energy efficiency funding. They can levy a Public Benefit Charge ("PBC"). A small carbon tax, based on consumption, could be collected from electricity users. These funds could then be used to generate additional funding to support more programs for energy efficiency and renewable energy, which would cut C02 emissions in the long run .. 9 California Standard Practice Manual, Economic Analysis of Demand-Side Programs and Projects, October, 2001. Available online at: http://www.energy.ca.gov/. 10 See 2004-2005 budget analysis, available online at: http://www.lao.ca.gov/analysis 2004/Resources/res 15 8665 an104.htm 20

PAGE 35

3 MODELING METHODOLOGY 3.1 MARKAL Model This research utilizes the MARKAL modeling framework to investigate scenarios of future electric generation technologies and their impact on the environment. 11 MARKAL is a model that represents an energy system from the extraction or import of fuels, through their conversion to useful forms, to their use to meet end-users (i.e., residential, commercial, industrial, and transportation) demands. MARKAL (acronym for MARKet Allocation) is a bottom-up, dynamic linear programming model.12 The MARKAL model determines the least-cost pattern of technology investment while meeting the required energy demands and model constraints, and then calculates the resulting environmental impact such as greenhouse gas emissions [27a]. MARKAL model assumes perfectly competitive markets for energy carriersproducers maximize profits and consumers maximize their collective utility. The result is a supply-demand equilibrium that maximizes the net total surplus (i.e. the sum of producers' and consumers' surpluses). The model computes an inter temporal partial equilibrium on energy markets, which means that the prices and quantities of various commodities are in equilibrium at all times, i.e., in each time period the suppliers produce exactly the quantities demanded by the consumers.13 The objective of the MARKAL model is to minimize the discounted total system cost for a region (or set of regions if multiple regions are modeled) obtained by adding the discounted periods' total annual cost. The total includes annual operating costs, annualized investment costs, and a cost representing the welfare loss incurred when demands for energy services are reduced due to their price elasticity. This objective is equivalent to maximizing the total surplus (consumers' plus producers' surpluses) [27b]. Figure 7 shows the price/demand trade-off curve where consumers' and producers' surpluses are maximized at an equilibrium quantity [37]. 11 MARKAL model capabilities include; linear programming (LP) application focused strictly on the integrated assessment of energy systems, non-linear programming (NLP) formulation which combines the 'bottom-up' technology model with a 'top-down' simplified macro-economic model, stochastic programming to address future uncertainties, mixed integer programming techniques to model endogenous technology learning, and multiple regions modeling (NLP/LP). The MARKAL source code is written in the Generalized Algebraic Modeling System (GAMS) language. The model's documentation is available online at: http://www.ecn.nl/unit bs/etsap/. 12 Dynamic here means that all investment decisions are made in each period with full knowledge of future events. 13 Partial equilibrium here means that the model computes both the flows of energy forms and materials as well as their prices. so that the suppliers of energy produce exactly the amounts the consumers are willing to buy. 21

PAGE 36

B 'C IL. Producer surplus s Equlllbnum p01nt Q Quantity 8 Figure 7: Price/Demand Trade-Off Curve The building blocks depicted in Figure 8 represent MARKAL model stylized power sector network, referred to as a Reference Energy System (RES) consisting of energy carriers, conversion or resource technologies, and energy services. The two basic components of an energy system model are commodities and processes. Commodities represent energy carriers (e.g., fuels, emissions, energy, etc.) flowing through an energy system. 14 A process transforms commodities from one form into another (e.g., power plants transform fuels into electricity). This user defined network includes all energy carriers involved with primary supplies (e.g., mining or import of fuel), conversion and processing (e.g., power plants), and end use demand for energy services. The demand for energy services may be disaggregated by sector (i.e., residential, commercial, industrial) and by specific functions within a sector [28]. Constraints Resource Technologies (e.g., Wind or Solar) Figure 8: Generic Power Sector RES 14 An energy carrier (or energy source) is anything in the energy system containing usable energy to produce another energy carrier (e.g. coal or gas used to produce electricity) or to produce usable heat or mechanical movement via certain technologies (e.g. gasoline, electricity, wood). An energy service is a commodity representing a demand for some useful service, such as heating of dwellings, etc. 22

PAGE 37

3.2 MARKAL Model Applications MARKAL energy and environment modeling framework has widely been accepted within the U.S. energy and environment modeling community. The U.S. EPA has developed a national MARKAL database and is working to regionalize the database into nine census regions [29]. In addition, the U.S. Department of Energy's Energy Information Administration (EIA) recently has adopted the MARKAL framework as the basis for its System for the Analysis of Global Energy Markets (SAGE) model. SAGE is used to produce EIA's annual International Energy Outlook. Altogether, MARKAL and its variants are used in approximately 40 countries around the world. International research community has used MARKAL to develop strategies for addressing climate change and other environmental challenges. Gomez used MARKAL in surveying Technological Learning in Energy Optimization Models [30]. Makela used MARKAL to model the Nordic Electricity Production System [31]. And, Gielen used MARKAL to examine the integration of energy and materials systems engineering for GHG emissions mitigation [32]. MARKAL allows user to model energy, environmental, and policy issues, such as greenhouse gas emissions mitigation policies, to examine system-wide emissions limits on an annual basis or cumulatively over time. It also allows users to model the imposition of a carbon tax, or other fee structures. As a result, various costs for carbon may be generated for different levels of emission reductions. In this way, future technology configurations are generated and may be compared. MARKAL is a demand-driven energy-economic model, which means that all the specified energy demands have to be satisfied. The user specifies the energy system structure, including resource supplies, energy conversion technologies, end use energy demands, and the technologies needed to satisfy these demands. The user also defines technology fixed and variable costs, technical characteristics (e.g., conversion efficiencies), availability, performance attributes, and pollutant emissions. The specification of new technologies, which are less energyor carbon intensive, allow the user to explore and evaluate the effects of these choices on total system costs, changes in fuel and technology mix, and the levels of greenhouse gases and other emissions [28]. 3.3 EPA National MARKAL model EPA's National MARKAL model is a comprehensive energy and economic model that simulates a national energy system by representing the interactions between resource supply (e.g., fuel), conversion processes (e.g., power plants), end-use technologies (e.g., heat pumps), and demand for energy services (e.g., space heating). The EPA's National MARKAL model represents the U.S. as one region. EPA has adopted the MARKAL model to assess current and future energy technology options. EPA's national MARKAL model determines the least-cost pattern 23

PAGE 38

of technology investment and utilization required to meet specified demands while satisfying model constraints (e.g., emissions caps), and calculates the resulting criteria pollutant and greenhouse gas emissions through 2030 [35]. 3.4 EPA Regional MARKAL model EPA has sponsored development of regional version of EPA's national MARKAL model. The regional modeling effort has provided an example of how the MARKAL model can provide useful analyses and tools to states and regions that need to make energy/technology decisions. The value of regional or state level MARKAL model as a valuable planning tool has been recognized by the EPA [33]: ... The U.S. EPA also recognizes the needs for integrated energy and environmental planning tools at the state and regional level. The advantage of a regional model is that it will accurately reflect the policy initiatives that are being designed and implemented by the states, using appropriate cost and performance characterization of the technologies that are available at the state and regional level." Northeast States for Coordinated Air Use Management (NESCAUM) has developed and run the MARKAL model and technology database, NE-MARKAL. Each of RGGI New England states is modeled as its own region, with a focus on air quality and climate change.15 Building on the lessons from NE-MARKAL, the U.S. EPA has been partnering with the bi-partisan Northeast-Midwest Institute and Ohio State University to build an Ohio-MARKAL model, with NESCAUM playing a supporting role to ensure compatibility. EPA is also developing a nine-region MARKAL database (EPA9R), to address regional differences explicitly, that will account for variations in supply, demand, and technologies between the nine U.S. Census regions. EPA is expecting to use the EPA9R database to examine renewables in the future [29]. 3.5 MARKAL Model Application to Colorado Currently, there is no energy and environment model and technology database developed for Colorado. This study is the first power sector and environment modeling study, the Colorado Energy and Environment model, analyzing Colorado's power sector for a sustainable energy future. Colorado is part of the14 states Western Electricity Coordinating Council16 a region of the North American Electric Reliability Council (NERC) 17 Most of the modeling 15 Available online at: http://www.nescaum.org/topics/ne-markal-model 16 Western Electricity Coordinating Council, http://www.nerc.com/regional/wecc.html. 17 NERC Regions, http://www.nerc.com/regional. 24

PAGE 39

performed by various agencies includes Colorado with other western states. For example, DOE's Annual Energy Outlook modeling using NEMS18 represents Colorado as part of the Rocky Mountain Power Area. EPA's Integrated System Analysis Workgroup (ISA.:W) which contributes to the Air Quality Assessment is also developing a nine (9) region technology assessment and corresponding emission growth rates from the scenario analyses using MARKAL model which Colorado will be part of a broader region combined with other states [29]. Currently, there is no Colorado-Specific energy and environment technology assessment model that can assess Colorado's growing electricity needs and its corresponding emission growth rates. This study is the first to develop an energy and environment technology assessment modeling framework for Colorado. 18 The U.S. Department of Energy, National Energy Modeling System (NEMS). 25

PAGE 40

4 MODELING METHODOLOGY FOR COLORADO 4.1 Colorado Energy Model and Database Development This study seeks to develop a working portfolio of resources and public policies to foster clean energy technology solutions to the energy and environmental challenges facing the state. These challenges include air pollution, greenhouse gas emissions, and inefficiencies in the production and use of energy. The purpose of the study is to estimate the costs and benefits of alternative sets of policies to accelerate clean and sustainable energy technology solutions in Colorado. 4.2 Energy Demand Forecast Energy planning models require energy demand forecast for all the years in the planning horizon. There is no statewide long-term energy demand forecast available for Colorado that can be used for this study. Utilities in Colorado perform their own long-term forecast for their own use. The long-term forecast of the two regulated utilities (Xcel Energy and Aquila) that serve almost 60% of the state's load is publicly available. The other 40% of the state's load is served by cooperative and municipally owned utilities with limited publicly available long-term forecasts. Energy and Demand forecast is discussed in the next chapter. 4.3 Existing Generation Resources Data Sources Based on EIA's 2005 Annual Electric Generator Report (EIA-860), there are 281 small and large electric power generators in Colorado.19 The report shows that 18 of those generators are not connected to the electrical grid, 10 are retired, and 41 are in "Cold Standby" status. The National Electric Energy Data System ("NEEDS") database also contains the generation unit records used to construct the "model" plants that represent existing and planned/committed units in EPA modeling applications of Integrated Planning Model. 20 NEEDS includes basic geographic, operating, air emissions, and other data on generating units. The following data are available from NEEDS: Coal Supply and Transportation Assumptions Natural Gas Assumptions Federal and State Emission Regulations and Enforcement Actions Cost and Performance of Generating Technologies and Emission Controls Sulfur Dioxide (S02), Nitrogen Oxide (NOx), and Heat Rates 19 EIA-860, Annual Electric Generator Report, is available online at: http://www.eia.doe.gov/cneaf/electricitv/page/eia860.html 20 National Electric Energy Data System (NEEDS) is available online at: http://www.epa.gov/airmarkets/progsregs/epa-ipm/index.html#needs 26

PAGE 41

Power System Operating Characteristics and Structure Electric Generating Unit Inventory NEEDS database generation unit records are utilized to construct the model's database for existing generating units in Colorado. Next chapter discusses in detail the existing and near future approved and planned generating resources. 4.4 Renewable Portfolio Standards (RPS) requirements By 2020, about 16 percent of the expected retail sales of Colorado shall include electric power generation from renewable generating units. 21 This is based on retail sales of all qualified utilities within the state acquiring the maximum renewable energy under the Colorado RPS requirements. The amount of energy attributable to RPS requirements is estimated for both Investor Owned Utilities (IOU) and Non-lOU utilities for input to the MARKAL model. Table 12 shows total RPS renewable energy requirements for both IOU and Non-lOU utilities in Colorado. Table 12: Colorado RPS Requirements IOUII RPS Effective RPS Year (GWh) (%of Load) 2005 -2008 1,864 3.40% 2011 4,268 7.20% 2014 4,589 7.20% 2017 7,774 11.40% 2020 11,624 16.00% 2023 12,338 16.00% 2026 13,051 16.00% 2029 13,765 16.00% 2032 14,478 16.00% 2035 15,192 16.00% The cost and benefit of RPS coupled with its competitiveness with other least cost generation technologies to meet Colorado's energy requirements is quantified within the model. A Rule-Based scenario is designed to capture the RPS requirements within the model. The percent requirement is modeled as a floor (i.e., a lower bound since its mandated) for the renewable generation in Colorado. The Rule Based scenario also recognizes the fact that the RPS requirements for solar generation shall include 4% of RPS requirement from solar of which 2% shall be from distributed solar (i.e., rooftop solar). See section 5.8.1 for more discussion on RPS requirements. 21 The 16% is the weighted average ofRPS requirement from Regulated and Non-Regulated utilities. It is assumed that regulated utilities will serve 60% of Colorado load with 20% RPS requirement and Non-regulated utilities serve 40% of Colorado load with 10% RPS requirement in 2020. 27

PAGE 42

5 SCENARIO ANALYSIS This section provides the results of scenario analysis by showing how technology evolution, energy efficiency and carbon policy could impact state's future power sector C02 intensity, and provides a statewide optimized assessment of power sector's existing and future generation resource mix for a sustainable energy future. This study establishes a database and utilizes MARKAL energy model to: Model alternative scenarios for sustainable energy future for Colorado's power sector while improving the competitiveness of renewable energy, and relatively keeping fossil fuel generation competitive. Evaluate costs and benefits of alternative scenarios for statewide C02 mitigation targets with criteria pollutants (S02 and NOx) emissions reductions as ancillary benefits. 5.1 Approach In order to investigate scenarios of future electric generation technologies and their impact on state's future GHG and air pollutant emissions, the following scenarios were developed: Reference Scenario (Business-as-Usual) Advanced Emerging Technology Scenario Energy Efficiency Scenarios Policy Scenarios, and Sensitivity Scenarios The study concentrates on the electric power system of Colorado and presents the development of a supply-side energy system incorporating Renewable Portfolio Standards, Demand-Side Management and Energy Efficiency measures. The focus of the work is to demonstrate the current status of power sector in Colorado and quantify the pathways for sustainable future energy system. The modeling horizon is 30 years. A year is divided into three seasons with spring and fall seasons combined. Seasons are divided into three time fractions; day, night, and peak hours. Since Colorado is a net importer, power imports from neighboring states are modeled to account for utility (take or pay) fixed contracts and other imports under pure economic conditions. Within the main scenarios, total of 12 other scenarios, variant from the main scenarios, are modeled and analyzed, covering the time span from 2005 to 2035. The Reference Scenario (Business-As-Usual, "BAU") represents the most probable development of the power system under present known conditions while the other scenarios serve mainly for variation from BAU to show possible pathways for a 28

PAGE 43

sustainable development of power system with main focus on clean energy technologies and mitigation of GHG emissions. 5.2 Reference Scenario The development of energy scenarios allow a way to analysis and examine a range of resource portfolios and policies for consideration of alternative possibilities. An important step for any system planning modelling exercise is to establish a baseline scenario that represents a reasonable progression of the system's advancement into the future years taking into account certain aspects of the current and future conditions. Reference scenario serves as the basis for the subsequent analysis of alternate technology and policy scenarios In preparation for this study, a Reference scenario has been established by: developing state-wide energy and demand forecast for each of three sectors (Residential, Commercial and Industrial); adopting forecasts of energy supply prices from the DOE/EIA and the regulated utilities' filings before the state's Public Utilities Commission (PUC); establishing state's existing power plants installed capacity, coupled with Independent Power Producers (IPP) installed capacity ; establishing state's mandated Renewable Portfolio Standards (RPS) requirements for all regulated and non-regulated utilities; establishing state's mandated Demand-Side Management (DSM) and Energy Efficiency (EE) requirements for all regulated utilities; establishing known near-term power plants retirements through state's PUC and utilities Resource Plans; establishing approved and proposed future power plants through state's PUC and utilities Resource Plans, and establishing assumptions for "guiding" model choices in situations where there are limitations on system evolution that inhibit the selection of ideal economic choices MARKAL is a least-cost optimization model for long-term energy system planning. Therefore, it is necessary to establish within the model the resources bounds and restrict some aspects of model choices to better reflect the conditions as the most likely evolution of state's electric power system, assuming a Business-as-Usual (BAU) perspective. BAU assumes a continuation of current energy policies using existing resources and adding planned and future conventional resources to meet electricity requirements of the state Embedded within this assumption are limitations on how much the energy system will remain similar to what it is now without intervention Representing each of these important aspects of the state's energy system in the model determines the nature of the Reference scenario. These are discussed in detail in the following sections. 29

PAGE 44

5.3 Demand Forecast Energy planning requires energy demand forecast for those years in the planning horizon. There is no statewide long-term energy demand forecast available for Colorado that can be used for this study. Utilities in Colorado perform their own long term forecast for their own use. The long-term forecast of the two regulated utilities (Xcel Energy and Aquila) that serve almost 60% of the state's load is publicly available. The other 40% of the state's load is served by cooperative and municipally owned utilities with limited publicly available long-term forecasts. In order to develop a statewide energy and demand forecast for the planning horizon of 30 years, two recent studies were utilized; 1) the RMATS Study and 2) the CEF Report. 5.3.1 RMATS Study In 2003, Governors of Wyoming and Utah announced the formation of the Rocky Mountain Area Transmission Study (RMA TS) [25]. The sole purpose of RMA TS was to conduct an analysis of generation and transmission alternatives within the region based on data, assumptions, and scenarios developed by the participating stakeholders. The RMATS region covers the States of Colorado, Idaho, Montana, Utah and Wyoming. RMATS performed a Forecast of Energy and Demand for Colorado in September 2005. RMA TS forecast combined total energy and demand requirements for Colorado East and West for the month of July (Summer Peak). Table 13: RAMATS Colorado Energy and Demand Forecast RMA TS Forecast (September 2005) Energy (GWh) Peak(MW) 2008 2013 58,146 9,750 65,422 11,146 Table 13 shows RAMATS forecast for Colorado total energy requirements of 58,146 and 65,422 GWh for years 2008 and 2013, respectively. For the same years, RMATS forecast Colorado peak demand of 9,750 and 11,146 MW, respectively. 5.3.2 CEF Report The Colorado Energy Forum also performed a study of Energy and Demand forecast for Colorado a year later in September 2006 [21]. CEF also projects the combined total energy and demand requirements for Colorado East and West. Table 14 shows the CEF forecast for total Colorado energy requirements of 52,656, 64,662, and 78,351 GWh for years 2006, 2015, and 2025, respectively. For the same years, CEF forecast a Colorado peak demand of 10,080, 12,400, 15,114 MW, respectively. 30

PAGE 45

Table 14: CEF Colorado Energy and Demand Forecast Colorado Energy Forum Forecast (September 2006) Energy ( GWh ) Peak(MW) 2006 2015 2025 52, 656 10, 080 64 662 12, 400 78 ,351 15, 114 A trend analysis was performed to curve fit historical data from DOEIEIA combined with the projected energy and demand data from RMA TS and CEF. Colorado Elecbic Demand Forecast: UNEAR Trendline 25.000.---------20.ooo+----------i 10.000 j------,--__.,."C-5.000 Figure 9: Colorado Peak Demand Trend (2005 -2035) Figures 9 and 10 depict Colorado Peak Demand and Energy Forecast, respectively, using actual historical data for 2000-2005 combined with sparse projection data for 2008 and 2013 from RMATS, and sparse projection data for 2006, 2015, and 2025 from CEF. 31

PAGE 46

Colorado Elecbic Energy Forecast: UNEAR Trendline 120,000 -------------------------------, i 100,000 60,000 40.000 20.000 Figure 10: Colorado Energy Trend (2005-2035) Results of Trend Analysis were used to build Colorado Demand and Energy requirements for each Sector. The End-Users in Colorado are represented by three sectors; Residential, Commercial, and Industrial sector. The relationship of each sector to total energy requirements were developed from available historical data from DOE/EIA for 2000-2005. Table 15 shows the distribution of total energy requirement among three sectors. Table 15: Colorado End-Use Sectors Share of Total Energy Requirements End-Use Percent Share of Total Energy Requirements Residential Commercial Industrial 34 41 25 The results of Trend Analysis show an average annual growth rate of 2.0% and 1.9% for energy and demand, respectively. See Tables 16 and 17 for Colorado energy and demand forecast, respectively, utilized in the model as Reference Scenario (BAU). 32

PAGE 47

Table 16: Colorado Energy Forecast (Low, Base, High) Colorado Energy Forecast (GWh) Year Low (1.5%) Base (2.0%) High (2.9%) 2005 48,353 48,353 48,353 2008 53,138 54,811 55,270 2011 55,565 59,271 60,395 2014 58,103 63,731 65,995 2017 60,757 68,191 72,115 2020 63,533 72,651 78,802 2023 66,435 77,110 86,109 2026 69,469 81,570 94,094 2029 72,643 86,030 102,819 2032 75,961 90,490 112,353 2035 79,431 94,950 122,771 It is projected that Colorado will need to add to the existing 2005 installed capacity of 11,232 MW-new generation resources of 3,361 MW and 7,196 MW for years 2014 and 2026, respectively.22 Table 17: Colorado Demand Forecast (Low, Base, High) Colorado Demand Forecast (MW) Year Low (1.5%) Base (1.9%) High (2.5%) 2005 9,664 9,664 9,664 2008 10,114 10,390 10,417 2011 10,575 11,176 11,218 2014 11,059 11,962 12,081 2017 11,564 12,747 13,010 2020 12,092 13,533 14,010 2023 12,644 14,319 15,088 2026 13,222 15,105 16,248 2029 13,826 15,890 17,497 2032 14,457 16,676 18,842 2035 15,118 17,462 20,291 A Low and a High projection are also estimated for the sensitivity analysis to alleviate the uncertainties with long term energy and demand projections. The Low energy projection results in an average annual growth rate of 1.5%, whereas a High projection provides an average annual growth rate of 2.9%. 22 This need is based on the forecasted demand for each year plus a 22% reserve margin. For example, for 2014, the need of 3,361 MW is calculated by taking the forecasted demand of II ,962 MW plus 2,631 MW for reserve margin less 11.232 MW of existing installed capacity in 2005.This capacity estimate does not include any transmissions losses which usually run between 6-8%. 33

PAGE 48

Colorado Bectric Energy Forecast 140,000 120,000 r -100,000 80,000 60,000 40,000 20,000 0 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 -+-l0111(1.5%) -----Base (2.0%) High(2.9%) Figure 11: Colorado Energy Demand Projection (Low, Base & High) Figures 11 and 12 depict Colorado Energy and Demand Forecast (Low, Base, and High) corresponding with Tables 16 and 17 above. 25,
PAGE 49

Mountain region data from EIA was utilized to represent the sectoral demand load pattern for Colorado. EIA utilizes aggregated data corresponding to the categories end-users. There are four seasons of 3 months each: (1) December, January, February; (2) March, April, May; (3) June, July, August; and (4) September, October, November. There are 3 "time-of-day" categories: midday, morning/evening, and night. Thus, there are 12 categories to match to each sector. There are three seasons: summer, winter, and the "shoulder'' season which include spring and fall seasons. The "peak" hours is represented by using the top 1 % of loads in each of the 12 categories thus 3 peak categories. Table 18 shows 12 time slices load distribution for Colorado. Table 18: Colorado Seasonal/TOO Sectoral Load Distribution Suson Time-of-Day_ Residential Commercial Industrial Summer Day 0.137 0.198 0.123 Summer Morning/Evening 0.097 0.060 0.080 Summer Night 0.026 0.019 0.046 Summer Peak 0.004 0 004 0.003 Winter Day 0.091 0.131 0.089 Winter MorninWI:vening 0.140 0.081 0.109 Winter Night 0.031 0.016 0.045 Winter Peak 0.003 0.003 0.003 Spring/Fall Day 0.187 0.301 0.203 Spring/Fall Morning/Evening 0.223 0.147 0.201 Spring/Fall Night 0.050 0.031 0.092 Spring/Fall Peak 0.006 0.008 0.006 Total 1.000 1.000 1.000 For the nine time-slice modeling, morning/evening category could be part of "day" and part of "night", thus it was split between day and night providing nine time-slices. Figure 13 shows Colorado nine time-slice load distribution patterns utilized in the model for Reference scenario. 35

PAGE 50

Fes. O:rrm o Ind. Figure 13: Colorado Nine Time-Slice Proxy Load Distribution 5.5 Supply Options Load Management In MARKAL a conversion plants (e.g., coal-fired power plant) can produce electricity in each time division up to a level governed by the annual availability factor (AF) of that plant. For all existing and new power plants certain AF is determined based on available data and are utilized in the model. Instead of a fixed and constant availability throughout the year, seasonal and time-of day dependent values may be assumed by means of division-dependent to reflect resource availability for renewable power plants (e.g., hydro, solar, wind). In this case, the production in each time division cannot exceed either of the levels given in the two conditions described above. The actual level of production for certain plants is then established as part of the solution of the MARKAL model, subject to these constraints and the demand load pattern. In order to limit the load following characteristics of specific power plants, e.g. to avoid unrealistic operation patterns, so-called externally Load Managed plants can be modeled as well. Their production in each time division is fixed by means of annual time-sliced capacity factor (CF) parameters, but not both that is with AF. 5.6 Fuel Supply Prices Fuel supply price input to the model are those developed by Xcel Energy for Colorado. Data for fuel supply prices were gathered from Energy Information Agency (EIA) and Xcel Energy's 2007 Colorado Resource Plan. Xcel Energy used various sources of data to compile and develop its fuel prices. For example, for gas prices, it 36

PAGE 51

used a blend of New York Mercantile Exchange, EIA, and other sources. Figure 14 show the comparison of gas price projections by EIA and Xcel Energy. Natural gas price forecast developed by Xcel Energy is higher than EIA and more representative of fuel market in the west and therefore were adopted and utilized as input for Colorado model. 14.00 12.00 10.00 I 8.00 6.00 4.00 2.00 0.00 11111t1.a1 Gas PHce Forecast Cor"llaison For Colorado ...--------- -" ____. __. Figure 14: Natural Gas Prices Forecast The same reasoning goes for other fuel types. For example, Xcel Energy developed coal prices for its Colorado operation using various sources as well. Both Powder River Basin (PRB) and Colorado-Wyoming (CO-WY) coal types are utilized in the model with prices adopted from Xcel Energy coal prices. Coal prices for both PRB and CO-WY coal resources are assumed to be the same. Figure 15 shows the comparison of coal price projections by EIA and Xcel Energy. 1.80 1.60 11.40 1.20 1.00 I 0.80 0.60 0.40 0.20 0.00 .. CCXII Plice Forea.t COIlJBII iaon For Colorado .... ..... --------j I 2005 2008 2011 2014 2017 2020 2023 20:26 2029 2032 2035 -+-BA(05$) --lUll Figure 15: Coal Prices Forecast 37

PAGE 52

Nuclear fuel prices were also adopted from Xcel Energy fuel prices. Figure 16 shows the comparison of nuclear fuel price projections by EIA and Xcel Energy. 1.60 1.40 1.20 I 1.00 o.eo 0.60 0.40 0.20 0.00 J'llJclear Fuel Price Forecast Con1aison For Colorado -------/_ / r/ ----2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 Figure 16: Nuclear Prices Forecast Following are the sources of data for fuel prices: EIA price forecast from AE02007 I o Report No. DOE/EIA-0383(2007), Gas Table 108, Coal Table 15, Full Report Release date: February 2007 AE02007, at 88, Nuclear fuel costs rise steadily to $0.62 per million Btu in 2030. Xcel Energy's 2007 Colorado Resource Plan, Price Forecast, Figure 1. 7 of Volume 1. 5. 7 Existing Installed Capacity Table 19 shows the aggregated generating capacity in Colorado during the base year 2005. The total installed capacity in Colorado was 11,232 MW which include 5,143 MW of Coal-Fired plants (1,733 MW of Bituminous coal, 3,410 MW of Sub Bituminous coal). Colorado also has 4,226 MW of Gas-Fired power plants (1,760 MW of Combined Cycle and 2,460 MW of Combustion Turbine) and 107 MW Gas Fired Steam plants, and 276 MW of Internal Combustion plants. The remaining includes 643 MW of Hydro, 563 MW of Pumped Storage, 265 MW of Wind, and 10 MWof Solid Waste. See Table 19. Appendix A lists all the existing units utilized in the model as part of the Reference Scenario. Appendix B lists all generating units operating characteristics for year 2005 which include: number of hours each unit operated, energy input, power output, heat rate, and emissions factors. The emission factors shown in Table 19 are the 38

PAGE 53

aggregation of emission factors from all units categorized by technology and fuel type. Table 19: Colorado Existing Generation Mix & Emissions Rates (Input for 2005) POWER PLANTS CAPACITY HEAT-RATE S02 NOX C02 /PARAMETERS (GW) (BTU/KWH) (KT/PJ) (KT/PJ) (KT/PJ) (PJ/PJ) OUTPUT OUTPUT OUTPUT Steam-Coal 1.733 10,617 0.3008 0.4908 272 Bituminous 3.111 Steam-Coal 3.410 10,473 0.4668 0.4088 270 Sub-Bituminous 3.069 Steam 0.107 13,387 0.0 0.3110 252 Natural Gas 3.923 Combined Cycle 1.760 7,398 0.0 0.0193 Ill Natural GAS 2.168 Combustion Turbine 2.466 11,215 0.0 0.0716 175 Natural GAS 3.287 SmaiiiC 0.276 15,280 0.006 0.3110 252 NGAS/Oil 4.478 Hydro 0.6429 ---Pumped-Storage 0.5625 ---Muni.-Solid Waste 0.0098 ----Wind 0.2647 ----Total Capacity 11.2319 5.8 New Regulatory Landscape for Renewable and DSM/EE Since 2004, Colorado voters and General Assembly have passed a number of Bills aimed at clean energy technologies and energy efficiencies with the goal of sustainable energy future and the reduction of greenhouse gas emissions from Colorado power sector. In 2004, Colorado voters passed Amendment 37 creating a Renewable Energy Standard ("RES") for the State which required 1 0% of regulated utilities retail energy to be produced from renewable energy resources. In 2006, Colorado Legislature passed new laws that encouraged the development of integrated gasification combined cycle ("IGCC") generation in Colorado. In 2007, new Bills doubled the amount of retail renewable energy to 20% for regulated utilities and added 1 0% requirement for non-regulated utilities (Cooperatives and Municipalities), encouraged the development of new transmission infrastructure to support the development of new renewable energy resources, and established energy efficiency and Demand-Side Management (DSM) goals for the regulated utilities (Investor Owned Utilities, "IOUs"). Also in 2007, The Governor of Colorado announced a statewide plan to reduce by 2020 carbon dioxide emissions by 20% from 2005 actual emission levels and to reduce C02 emissions 80% by 2050. 39

PAGE 54

These new legislative actions have created new challenges for both the regulators and the utilities. In its recent rule making, Colorado Public Utilities Commission (PUC) has recognized the significant impact of the new legislations on the utilities resource planning process and has eliminated the guiding principles of "least cost" planning and fuel neutrality with the concept of "cost effective" resource planning giving consideration of the costs and benefits of adding more renewable resources and DSM programs to the utility's resource plan for resource acquisition. This study is the first direct statewide assessment of new legislative mandates for more renewable and energy efficiency measures to reduce the state's greenhouse gas emissions. 5.8.1 Renewable Portfolio Standards In 2004, Colorado's new Renewable Energy Standard (RES), the first in the nation, enacted through a voter initiative "Amendment 37". The 2004 RES applied to two rate regulated utilities in the state, Xcel Energy (PSCo) and Aquila [22]. It allowed other Colorado covered utility (40,000 or more customers) to opt out of the RES, or an exempt utility to opt in, with a majority vote involving a minimum of 25 percent of the utility's customers [23]. In 2007, HB07-1281, increased the amount of electricity a utility must generate or cause to be generated from renewable energy resources (Renewable Energy Standard "RES" or Renewable Portfolio Standard "RPS"). The previous RES established by the voter-approved in 2004 Ballot Amendment 37, required utilities to meet a 10% RES by 201 5 and required 4% of that standard to be obtained from solar energy sources. HB07-1281 doubled these requirements by mandating that by 2020 qualifying regulated retail utilities (Investor-Owned Utilities "IOU") meet a 20% RES. This legislation continued the requirement of Amendment 37 for the IOUs to satisfy 4% of the RES from solar resources. Also in 2007, another Bill was passed, SB07 -1 00, which provided a mechanism for the designation of energy resource zones and the development of additional transmission infrastructure for delivery of from those zones energy (e.g., from small remote wind farms) to the load centers of the utilities. The Bill applies to each provider of retail electric service in the state of Colorado other than Municipally Owned utilities that serve forty thousand customers or less. See Table 20 for Colorado RES requirements. 40

PAGE 55

Table 20: Colorado RES Requirements and conditions YEAR REQUIREMENTS ACCEPTS OUT OF CREDIT ENACTED EXISTING STATE TRADING CAPACITY SUPPLY 2004 I 0% of rate regulated utilities retail electricity sales in Yes Yes Yes Colorado shall include renewable energy by 2015; 4% of requirement must be solar of which 2% shall be distributed solar 2007 Rate regulated utilities retail electricity sales in Colorado Yes Yes Yes shall include Renewable Energy or energy efficiency, or a combination of both as follows: 3%-2007 5% 2008 through 20 I 0 10%-2011 through 2014 15%-2015 through 2019 20% 2020 and thereafter Cooperatives and Municipally Owned utilities retail electricity sales shall include renewable energy as follows: I% 2008 through 20 I 0 3%-2011 through 2014 6%-2015 through 2019 I 0% 2020 and thereafter; 4% of renewable requirement must be solar of which 2% shall be rooftop solar The eligible "Renewable Energy Resources" are defined in the Bill as solar, wind, geothermal, biomass, new hydroelectricity with a nameplate rating of ten megawatts or less, and hydroelectricity in existence on January 1, 2005, with a nameplate rating of thirty megawatts or less. Fossil and nuclear fuels and their derivatives are not eligible energy resources. The retail rate impact of RPS in Colorado is defined in the Bill to be caped at 1%. This study does not attempt to gauge the rate impact of RPS exogenously since it is assumed that the model is a least cost optimization model and internalizes the least cost resource options (conventional or renewable) within the model. More States have instituted renewable goals similar to those of Colorado. In 2004, New York enacted a goal to have 25% of its generation from renewable by 2013. In 2005, Vermont enacted a goal that all growth, up to 10% of total electricity sales, shall be from renewable and the goal becomes mandatory if not met by 2012 [24]. 5.8.2 DSM and Energy Efficiency (DSM/EE) Energy efficiency is characterized as an alternative to energy supply options, such as conventional power plants that produce electricity by conversion technologies from fossil or nuclear fuels. Demand-side management resources act to reduce the demand for electric power and include a variety of measures such as energy efficiency, demand response, and energy conservation. There are two types of demand side resources: peak shavers and energy savers. Peak Shavers are used to reduce a customer's demand and energy requirements during peak hours. Energy savers are used to reduce energy over all periods of the year. An example of an energy saver would be replacement of incandescent light bulbs with more energy 41

PAGE 56

efficient compact fluorescent (CFL) bulbs to reduce energy consumption throughout the year. 5.8.2.1 DSM Cost Effectiveness In 1990s, the cost effectiveness of utility DSM programs was debated in various rate making proceedings. Historically, evaluation of energy efficiency and conservation programs are supported by engineering economics analyses where the present value of saved energy is estimated to exceed the initial DSM capital investment. However, more traditional economic analysis based on market failures, economic efficiency and the appropriate accounting of cost and benefits are also suggested for evaluation of energy conservation programs [38]. Several comprehensive studies using large databases of DSM technologies have published cost of DSM based on impact evaluation of DSM programs. In general, the studies have indicated the levelized cost of typical DSM programs cost utilities (Utility Cost) around $25-35 per megawatt hour saved on average, and have a TRC (Total Resource Cost) of $40-60 per megawatt hour saved on average [39]. 23 Generally speaking, DSM displaces gas-fired generation and sometimes coal generation as well. The major contribution of DSM/EE to any electric power system is the avoided cost of capacity and reduced fossil fuel consumption for generation, or the avoided purchased power cost. 24 DSM also has a major contribution to avoid the associated emissions of power generation such as C02 and criteria pollutants. 5.8.2.2 Screening of DSM Options: Cost Effectiveness Tests In 1987, California developed standards for screening of DSM options which later was used by most jurisdictions including Colorado. 25 The practice manual defines five different benefit-cost tests: the Participant Test, the Ratepayers Impact Measure Test (RIM), the Total Resource Cost (TRC) Test, the Societal Test, and the Utility Cost (UC) Test. The Participant Test measures the difference between the quantifiable costs incurred by a DSM participant and the subsequent benefits received by that participant. The RIM Test primarily measures the impact of DSM programs on the utility rates. The 23 Most of the large-scale studies were perfonned in early 1990s and the unit cost ofDSM savings in tenns ofUC or TRC was in 1990s dollars. Most recent papers are still referring to these numbers as the cost ofDSM programs. See M. Curtis 2004 paper on "Energy conservation in electric utilities: an opportunity for restorative economics at SaskPower," Technovation 24, at 399. 24 DSM measures also contribute to savings in incremental transmission and distribution investment costs that utilities avoid in investing. 25 The standards were developed by the California Energy Commission (CEC) and the California Public Utilities Commission in a process called the "California Collaborative Process" and published as the "Standard Practice Manual: Economic Analysis of Demand-Side Management Program." Available online at: www.energy.ca.gov/greenbuilding/documentslbackground/07-J CPUC STANDARD PRACTICE MANUAL. PDF 42

PAGE 57

TRC Test (also called the All Ratepayers Test) compares the total costs of DSM programs (including costs incurred by the utility and the participant) and the avoided costs of energy supply. In a fact, the TRC Test is a summation of the Participant Test and the RIM Test. That is, benefits are still the total avoided supply costs, but costs are now the sum of the costs incurred by the customer and by the utility. The Societal Test includes the quantified effects of environmental costs (i.e., externalities costs) in the costs and benefits analysis of DSM programs. Finally, the Utility Cost Test measures the utility's avoided costs (i.e., costs related to fuel, operation and maintenance, and capacity costs) against the DSM program costs including rebates and administrative costs but not the customer costs. Designing a DSM program that passes all standards tests is a difficult task. There are times that one test may pass the standards but others may fail. Figure 17 shows a case where the RIM Test is passed but the TRC Test is failed mainly due to the fact that the TRC is more than the benefits that is the avoided cost. D fA, I :, --r -c .. .,../ C Avol-'' C...ta .. -'' DSM -Dill / ,. wl DSM Coo; --__l_ I I I -..: kWh Sales DSM example showing TRC Test fails (D above A), but RIM Test passes TRC=To181 Resource Coot: RIM= Ratepayer lmpad Measure Figure 17: DSM RIM Test Example26 Figure 18 shows the opposite of the above case that is the TRC Test passes but the RIM Test fails mainly due to the fact that high DSM costs causes higher average rates for non-participants. 26 Swisher, Joek N. and et al, 1997. Improving Energy Efficiency and Protecting the Environment. UNEP Collaborating Center on Energy and Environment, November. 43

PAGE 58

It is most common to adopt TRC Test results because the benefits are still the total avoided supply costs, but the costs are the sum of the costs incurred by the utility and by the customers. This concept is adopted in this study to evaluate the cost effectiveness of state-wide aggregated DSM and Energy Efficiency measures taken by cities and local entities to redesign building codes and mandate energy efficiency programs within their jurisdictions to stay current with voluntary climate change initiatives. 27 1/1 Q) ::J c:: Q) a::: A,T :::. DSM Avoided :; eo.ta : Com Y ::_B -,-) __ 1_ llllorw t DSII -...: :..--llWh UWild kWh Sales DSM example showing TRC Test passes (D below A), but RIM Test fails TRC=Tolal ReSOIICe Cost. RIM= Ratepayer Impact Measure Figure 18: DSM TRC Test Example 5.8.3 Regulated utilities DSM in Colorado Colorado has two investor-owned utilities (IOU) subject to rate making regulations under the Colorado Public Utilities Commission (PUC). Both IOUs serve approximately 60% of the state's customers and provide about 59% of the electricity sales. The top 5 providers of Colorado's retail electricity in 2005 are shown in Table 21. 27 See Denver Greenprint report available at: http://www.greenprintdenver.org/ and Colorado Climate Action Plan available at: http://www.colorado.gov/energy/inluploaded pdf/ColoradoCiimateActionPian OOI.pdf 44

PAGE 59

Table 21: Top Five Retail Electricity Providers in Colorado (2005) Utility Ownership Sales (GWh) Xcel Energy IOU 26,481 City of Colorado Springs Muni 4,479 Intermountain REA Coop 1,929 Aquila IOU 1,824 City of Fort Collins Muni 1,393 Top Five Total Sales 36,106 Colorado Total Sales 48,353 .. Source: EIA State Electnc1ty Profiles Regulated utilities in Colorado, namely Xcel Energy and Aquila, have a major role to play in energy efficiency and conservation programs. Xcel Energy has been involved in DSM programs since 1980s through rate making regulations. As part of 2003 Least-Cost Planning (LCP), a Comprehensive Settlement Agreement was approved by the Colorado PUC requiring Xcel Energy to provide more DSM programs to its ratepayers with the approval of its proposed first coal-fired power plant in more than two decades. 28 Xcel Energy committed to undertake a total of 320 MW of demand reduction and BOO GWh of energy savings over the 8-year period (i.e., 100 GWh per year or 0.38% of annual sales) beginning in 2006 and ending in 2013. The total cost of this undertaking was proposed for approval at $196 million (1996 dollars). In 2007, Colorado General Assembly passed the Energy Efficiency Bill, known as HB07 -1 037.29 The Bill requires Colorado PUC to establish energy savings and demand reduction goals (e.g., sets minimum goals) for regulated utilities (IOUs) to acquire through energy efficiency conservation, load management and demand response programs. The impact of these goals is to reduce the energy and capacity that a utility would have traditionally planned to serve through supply-side resources. The Bill also allows for utility investments in cost-effective electric DSM programs to be more profitable to the utility than any other utility investment that is not already subject to special incentives. The legislation also specifies that the goal of DSM shall be consistent with allowing all classes of customers an opportunity to participate in DSM programs and be consistent by giving due consideration to the impact of DSM programs on non-participants and on low income customers which basically means no rate increases due to increased DSM measures. 5.8.4 Impact of New Legislation on DSM in Colorado In 2007, Xcel Energy offered an Enhanced DSM Plan to its customers. For the period 2009-2020, in addition to 2003 LCP Agreement, Xcel Energy will spend $738 28 See Xcel Energy's Certificate of Public Convenience and Necessity for Comanche 3 Pulverized Coal Power Plant before the CPUC, Dockets 04A-214E, 04A-215E, and 04A-216E. 29 House Bill 07-1037, ''CONCERNING MEASURES TO PROMOTE ENERGY EFFICIENCY, AND MAKING AN APPROPRIATION THEREFOR", enacted 2007. 45

PAGE 60

million (2006 dollars) on more DSM programs to achieve 2,350 GWh (i.e., about 200 GWh per year) of energy savings. Table 22 shows Xcel Energy's Annual Incremental DSM Programs Energy Savings in addition to programs in place in 2006. The annual incremental energy savings will peak at 1,284 GWh or 3.6% of total retail sales in year 2021 and will start decreasing thereafter unless more DSM programs are added.30 Table 22: Xcel Energy Cumulative DSM Programs Energy Saving in Colorado Programs HB 07-1037 Total In Current Programs Energy Years Plan (as of May 07 Savings (GWh) 2007 99 99 2008 202 202 2009 313 313 2010 430 430 2011 552 552 2012 670 670 2013 779 779 2014 767 101 868 2015 752 197 949 2016 735 292 1027 2017 718 384 1102 2018 700 472 1172 2019 683 543 1226 2020 670 600 1270 2021 670 614 1284 2022 670 569 1239 2023 657 527 1184 2024 621 483 1104 2025 613 407 1020 2026 605 322 927 Source: Xcel Energy, June 2007 The cumulative annual energy savings in Table 22 are modeled in Reference Scenario representing BAU. For example, for year 2020, total energy savings of 1,270 GWh is modeled as DSM contribution to energy savings at the penetration rate of 25% Residential and 75% Commercial. Xcel energy performed a comprehensive DSM study suggesting the DSM penetration distribution rate in Colorado is 25n5 between residential and commercial customers, respectively [40]. Figure 19 shows the cumulative energy savings within the Xcel Energy's system from current and proposed DSM action plans. The total energy savings peaks at 1,284 GWh in year 2021 and begins to decrease thereafter unless more DSM investments are made. 30 The amount of cumulative peak energy savings of I ,284 G Wh in year 2021 translates into 3.6% of Xcel Energy's forecasted retail energy requirement of 35,834 GWh in 2021. See Xcel Energy's 2007 Colorado Resource Plan. Vol. 2 Technical Appendix, at 2-129. 46

PAGE 61

( 14(1) 12!D 1!XD liD liD 4(1) 2!D 0 -CISM-2IXB Solllorrwt CISM_H!-A "' .... "' ij ij ij a -----
PAGE 62

benefits by using less fossil fuel generation mostly gas generation and the associated C02 and NOx emissions related to gas fired generation technologies. 115 10 5 Plltlllc S.rvta. Clarnpmy ol Color..to Al1rtu.l AwldMI Bnlulana C..ta Sues: 2fX11 CEMOx:lel Figure 21: Xcel Energy DSM Avoided Emissions Costs 5.9 Integrated Gasification Combined Cycle (IGCC) IGCC is a power generation process that integrates a gasification system with a conventional combustion turbine combined cycle power block. The IGCC advanced technology is commercially available for generating electricity with coal with the promise of substantially reducing air emissions, water consumption, and solid waste production from coal power plants. The gasification system converts coal into a gaseous "syngas" which made of hydrogen and carbon monoxide. The combustible syngas is used to fuel a combustion turbine to generate electricity, and the exhaust heat from the combustion turbine is used to produce steam for steam turbine cycle and the gasification process [41]. The IGCC technology is in its infancy with a relatively high cost option. The first large-scale IGCC plant of 629 MW was proposed by American Electric Power to be operational in New Haven, West Virginia at a cost of $2.23 billion (I.e., $3,545/kW) and recently was approved by West Virginia PUC. 31 In addition to being affected by the current increases in construction labor and material costs, an IGCC project also must absorb the costs of C02 capture and sequestration as well as costs involved in first generation design. According to Electric Power Research Institute (EPRI) Technical Assessment Guide, the cost estimates made early in the development period of a new technology are low by a factor of two or more (expressed in constant dollars) compared with the 31 Source: e-newsletter from POWER magazine. June 2007 and March 2008. 48

PAGE 63

actual costs of the first commercial version of the technology. EPRI's TAG refers to this effect as the learning curve phenomenon which affects the costs to decline after a new technology is commercialized and improved versions of the new technology are built [42]. In 2006, Colorado Legislature passed a new law (HB06-1281) that encouraged the development of integrated gasification combined cycle ("IGCC") generation in Colorado. The Bill directed Colorado PUC to consider proposals by electric utilities to fund, and construct an "IGCC" project with C02 capture and sequestration (CCS). Colorado PUC has defined "Section 123 resources" within its Resource Planning Rules as meaning "new energy technology or demonstration projects including new clean energy or energy efficient technologies and IGCC projects." 32 This legislation encouraged electric utilities to investigate the development of IGCC projects. Xcel Energy in its 2007 Colorado Resource Plan has proposed one 600 MW IGCC plant with 50% CCS to be constructed in Colorado by 2016. Reference Scenario (BAU) captures this IGGC plant with the start date of 2017. Plant's characteristics including investment and O&M costs are adopted from Xcel Energy's 2007 Resource Plan. A number of coal-related provisions such as a loan guarantee for an IGCC plants were authorized by EPAct2005. These provisions have spurred some activity and interest by the utilities. Xcel Energy has proposed building a 600 MW IGCC facility in Colorado (owning 150 MW) as the Western IGCC Demonstration Project.33 The Xcel Energy's proposed IGGC project is captured in this study by modeling a 600MW IGCC unit with an in-service date of 2017 both in the BAU scenario and BAU plus Advanced Technology scenario. 5.10 Renewable Technologies Renewable electricity generation encompasses a collection of technologies including: Biomass-fired generators Landfill methane Solar generators; o Photovoltaics technology (rooftop PV and central PV) o Solar thermal storage Geothermal power Wind turbines 32 Section 123 refers to Colorado Revised Statutes. See C.R.S. for energy emcient technologies under section 402-123( I) and IGCC projects under section 40-2-123(2). 33 A, suggests recent discussions among the industry experts indicate that 300-350 MW IGCC may not be economically feasible. The size ofXcel's IGCC was first suggested at 300 MW range but in its 2007 Resource Plan a 600MW IGGC plant is considered. 49

PAGE 64

Just a few years back, generating electricity from renewable energy sources were considered more costly than from conventional sources, in particular natural gas generation. Recent advances in renewable technologies coupled with incentives and tax credits from federal government and high cost of natural gas have made renewable sources increasingly competitive with conventional sources. Each of above technologies has found market opportunities especially when supported by government policies and incentives. Absence government incentives, a few are able to compete directly with available best conventional generation technologies. The push towards more renewable generation over the next decade will significantly change the cost and availability of renewable technologies, as increased demand induces more research and development that will bring many improvements in renewable technologies resulting in lower costs and better performance. As discussed in IGCC section above, the learning curve phenomenon will cause the costs to decline as the industry matures and improved versions of the new technologies are built. 5.1 0.1 Biomass Biomass is plant matter such as trees, grasses, agricultural crops or other biological material. It can be used as a solid fuel, or converted into liquid or gaseous forms, for the production of electric power, or heat. 34 Biomass facilities for electricity generation are often considered base-load plants with capacity factors of 85% or better. Electricity generation from biomass resources already accounts for a significant amount of total renewable electricity generation. There are three types of biomass generation: Co-firing of biomass in combination with coal in existing coal-fired power plants Direct firing of biomass in dedicated systems, and Gasification of biomass for combustion in gas turbines Currently, biomass fuel sources are typically converted to electricity through combustion in an internal combustion engine or a steam boiler. Although most combustion units are direct-fired, research is under way to more fully develop biomass gasification processes: the ultimate product of these two processes is a biogas that can be combusted. Co-firing biomass with coal is currently used in handful of power plants around the country. Colorado Springs Utilities (CSU) recently conducted a study to burn biomass with its coal-fired units. The study confirm the benefits of biomass co-firing by stating "Co-firing with biomass fuels will generally result in lower 802, NOx, C02, mercury and other emissions than firing with 1 00 percent coal. Percent reductions in 34 See NREL Biomass research link at: http://www.nrel.gov/biomass/ 50

PAGE 65

emissions are generally proportional to the co-firing percentage (heat input basis). Reductions in NOx have been more difficult to predict accurately." The study conducts interviews with plant operators and confirms that biomass co-firing has minimal impact on the system operation by stating ... few impacts on system operation (in either a negative or positive manner) during co-firing tests conducted over three months at the Ottumwa facility. The only significant impact was a reduction in the percentage of S02 emissions during co-firing equivalent to the reduction in percentage of coal fed to the boiler." [43] Studies show that co-firing 5-15 percent biomass mixture with coal requires only minor burner and feed intake modifications to existing units with minimum capital costs of less than $50/kW to achieve a co-firing at a level of 1 0 percent. Co-firing entails no loss of efficiency and can contribute to the reduction of C02 and criteria pollutants emissions. C02 emissions from the combustion of biomass is generally considered C02 neutral; however, increasing attention is being paid to the neutrality of C02 emissions from forest products given the long time periods for conversion of atmospheric C02 to plant material. Any marginal increases in operating costs associated with the biomass resources are offset by the value of the emissions reductions benefits from co-firing biomass. National Renewable Energy Laboratory's 2003 biopower technical assessment concludes that the marginal O&M costs for biomass co-firing, exclusive of feedstock costs, are 0.23 cents per kWh less than those burning coal [44]. In this study, biomass is modeled as stand alone biomass gasification combined cycle and also with co-firing option with all bituminous coal fired units in Colorado at the ratio of 10 percent biomass and 90 percent coal. The competitiveness of biomass generation technologies heavily depends on the price and availability of biomass fuel resources. For biomass generation to become a major player in the RPS, the prices for biomass feedstock will need to be competitive with coal prices at near or under $2.00 per million Btu (mmBtu). Biomass supply curve developed by the U.S. Department of Energy (DOE) estimates that biomass fuel resources are available at prices around $2.00-2.50/mmBtu.35 In this study, the DOE average price of $2.25/mmBTu escalated at an inflation rate of 1.5% over the planning horizon is adopted. It is also assumed that the reductions in marginal O&M costs as reported by DOE offsets any capital improvement or retrofit costs for co-firing which is reported to be less than $50/kW. 5.1 0.2 Geothermal Geothermal resources convert extremely hot underground geothermal water into electricity and are generally run as base-load facilities. Three basic technologies exist to extract heat from underground reservoirs for the production of electricity: 35 U.S. DOE Biomass Program publications: Available online at: 51

PAGE 66

direct steam, flash steam, and binary cycle; each technology is best suited to the type of reservoir available. Based on surveys conducted to date, Colorado's relatively low-temperature geothermal resources are expected to be best utilized with binary cycle technology. Given the nature of the fuel source and the reliability of the power block, geothermal-generated electricity is considered a base-load resource with capacity factors between 85-95%. Relatively low levels of C02 and 502 are emitted from direct steam and flash steam facilities [45]. Geothermal technologies are modeled as base-load power plants. 5.1 0.3 Solar Recent report by the Western Governors' Association fYYGA) projects as much as 8,000 MW of solar capacity could be installed in the Western states with a combination of distributed solar electricity systems and central concentrating solar power (CSP) plants by 2015, and an additional 2,000 MW of solar thermal systems could be installed in the same timeframe. WGA further projects by 2015, the cost of electricity from future CSP plants should be competitive with plants burning costly natural gas, and distributed systems should have declined in price to the point that they should be able to produce electricity below retail utility rates in most parts of the West [45]. Colorado has over 300 days of sunshine per year, making it an ideal location for solar photovoltaics and solar thermal technologies [46]. Colorado requires 4% of RPS requirements must be from solar and 2% of that must be from "on-site" solar systems (PV system) located at customers' facilities. 5.1 0.3.1 PV and Solar Thermal Central station solar power technologies include both solar thermal electric and photovoltaic (PV) generators. The vast majority of the central station solar projects underway or actually deployed today are concentrating solar power (CSP) technologies, which as a class include all the thermal generators as well as concentrating PV. Flat-plate PV can also be used for utility-scale systems, but the much higher energy market values of distributed generation make it the more attractive deployment mode for flat plate PV today. As PV costs decline and its market volume grows, central station flat plate PV deployment will become more commonplace [45]. The WGA report cites a Solar Task Force survey of the CSP industry indicating capability to produce over 13 GW by 2015 if the market could absorb that much. The Solar Task Force also projects that, with a deployment of 4 GW, total nominal cost of CSP electricity would fall below 1 0/kWh. Analysis shows that CSP at 1 0/kWh is equivalent to a blended base load-peak value of natural gas generation at a fuel cost of $7/MMbtu. Achieving 4 GW of CSP deployments by 2015 from the current 354-MW base requires growth similar to that of the PV and wind industries in the past decade. 52

PAGE 67

5.10.4 Wind Wind power has come a long way in the past decade. Today, advanced wind turbines make a major contribution in meeting renewable generation requirements of the utilities. In 2007, the total installed U.S. wind energy capacity reached 13.9 GW and the Colorado's share was 1.07 GW.36 In 2001, the Colorado PUC ordered Xcel Energy to include a 162 MW wind plant as part of its integrated resource plan [47]. The PUC concluded that the wind plant would cost less than new gas-fired generation under reasonable gas cost projections by stating: "We find that adding Enron's Lamar wind energy bid to PSCo's preferred resource plan is in the public interest and comports with the IRP rules. This determination is based solely on our finding that the acquisition of the Lamar facility will likely lower the cost of electricity for Colorado's ratepayers. After a careful analysis of the economics of the wind bid, we find that it is justified on purely economic grounds, without weighing other benefits of wind generation that could be considered under the IRP rules." Since last large wind farm, Xcel has added 835 MW more wind resources to its resource portfolio to meet the minimal non-solar levels of the RPS requirements through 2020 and displace fossil-fired generation which reduces both gas and coal burns for electric production and the associated C02 emissions. Xcel has also performed wind integration study to look at the cost of a 20% capacity penetration level of wind. This 20% capacity penetration equates to about 1.400 MW of wind, or about 350 additional MW of wind on Xcel system in Colorado. In its 2007 Resource Plan Xcel states: ... We do not believe that this 20% capacity penetration is a ceiling on the amount of wind that the system can accommodate. However, we do need to perform further studies and to look beyond the concepts in our studies to date to determine how we might modify system operation or plant generation in ways to allow cost effective integration of wind resources. Therefore wind additions beyond 2015 should be considered an indication of a desire to add additional wind resources in that tie frame rather than a firm commitment for those additions." 5.11 Near-Term Power Plants Retirements Colorado's recent legislative mandates on power sector industry to add more energy efficiency programs and renewable technologies to their resource portfolio coupled with the statewide targets to reduce carbon dioxide emissions, have heightened 36 Infonnation from American Wind Energy Association accessed March 2008. Available online at: www .awea.org. 53

PAGE 68

utilities awareness on greenhouse gas emissions and utilities have started voluntarily to plan responsibly and retire old coal-fired plants. Xcel Energy in its 2007 Resource Plan proposes to retire six existing generation units and re-power them with natural gas combined cycle technology. The six units to be retired are listed in Table 23. In the Reference Scenario (BAU), a total of 353 MW of capacity is retired from the base year total installed capacity in 2005. Table 23: Colorado Generating Capacity Retirement Fuel Type Plant Capacity (MW) Date Bit. Coal Cameo 1 24 12/2010 Bit. Coal Cameo2 54 12/2010 Steam Gas Zuni 1 39 12/2009 Steam Gas Zuni1 2 68 12/2012 Sub-Bit Coal Arapahoe 3 47 12/2012 Sub-Bit Coal Arapahoe 4 121 12/2012 Xcel Energy in its Resource Plan states the retiring these four coal units and re powering them with a 480 MW Combine Cycle (CC) is expected to reduce C02 emissions by 1.4 million tons per year. 37 These units are modeled as retired in the Reference Scenario and the decision to replace the units is made by the model which in the Reference Scenario (BAU) would be the same as Xcel Energy's decision that is a replacement of equal amount of capacity with conventional CC technology, however in Advanced and policy scenarios the decision would be based on the economics and carbon policy constraints. 5.12 Approved and Proposed Future Power Plants Due to uncertainty and volatility in natural gas prices the coal-fired generation has re entered as a viable option in the utilities' near-term resource portfolios. The Xcel Energy's new coal-fired power plant with a capacity of 750MW approved to be built in Colorado with a service date of 2009 is captured as an investment of $1.3 billion plus the transmission interconnection and delivery cost (i.e., $2020/kW in 2005$) in the model [18]. In addition, the possibility of building an additional 500MW coal-fired unit in Brush, Colorado (aka, Pawnee II) by 2014, reported in the study by Colorado Long Range Transmission Planning Group (CLRTPG), is not modeled but the model is allowed to choose economically the least-cost plant. Another 600MW coal-fired power plant, reported by Western Resources Advocates to be built in the southeast Colorado by Tri-State Generation and Transmission is not modeled as near-term definite resource addition. 5.13 New Power Plants All new capacity decisions will depend on the costs and operating efficiencies of different options, fuel prices, and the availability of incentives such as Federal 37 See Xcel Energy 2007 Resource Plan at 54

PAGE 69

production tax credits for investments in wind technologies, and the carbon policy constraints. Natural gas plants are generally the least expensive capacity to build with lower C02 emissions but are characterized by comparatively high fuel costs. Advanced clean technologies like IGCC, nuclear, and renewable plants are typically more expensive to build but have relatively low operating costs and, in addition, receive tax credits under EPAct2005, and meet the low carbon policy objectives. The CEF report suggests a generation portfolio mix of approximately 3,000 MW of base-load power, 1,500 MW of intermediate power and 1,350 MW of peaking power to meet the growth. For this study, technology choices for new generating capacity will be made to minimize cost while meeting state's renewable and energy efficiency mandates coupled with climate action plan and other policy objectives. The model will make the choice of technology for capacity additions based on the least expensive options available subject to resource and policy constraints [21]. Table 24: Thennal and Renewable Resources Cost and Perfonnance Data Emlulon Rmes' Copital COl NO. SOl S..ru or Billa DOEll lA ad Modeled Power Galeralio c ... Heal Role' AF VAROM FXDOM EPA-NM 11pdatnl frv u auled Tdro 4l 10 BJ 21 441 1420 lhdro PS 4l J1l4 IJ 2.65 16.71 New Cool IGCC wilh lO% CCS 4 001 40 10 oz 17 J.Ol 17.14 I 041 04216 0 7064 Xool EncrxY" NowAdv CT 520 )0 I llJ 92 21) 119 921 0.017) NowAdv CC 127 )0 7,211 9) )09 9 42 16l 0.0714 Xcel EnOIXY' cc JO 7 99 94 049 ll7l Ill 0.1532 NowCC Ill JO 7,463 95 211 IJ 19 119 0.3413 Xcel E.ncqy CT JO 10 lZl 94 0 Ill 651 1,271 0.561) Newer 659 JO 10,459 91 195 4.31 I 246 0.5175 Xool Enorav' Now Goo IGCC wilh 90% CCS I 124 )0 79l2 91 2.93 19.95 16 0.0794 Goo Slcom )0 1),390 92 O.lZ 016 1517 24151 New Ad" Nuc:ICII' 2197 40 10 liZ 92 0.60 li.OO PV c .. urol )1)0 JO 10,213 196 PV Roolloo 7,519 )0 10,213 1.96 Sol ThcnoaJ Zl39 )0 IO 213 4Jll Wmd I 690 20 10 283 23.24 Xool Enc:rn" Cool Bu:d Imports" 2 159 Gosa-tlmoons Ill Notes CC .. Combined Cycle CT-Combusaion Turbine PCPulvcn,...Cool IGCC = ln....-cd GM1f1<11lion Combined ComJ Pulverized coal unit by X,;:el f.ncrsy \\1lh no S02 mel N(h. imp.:l (net of other 2 umts) PS = Pumped S10r11c Hydro F AF,. Availebility Factor Hca Rae "' Rmewllbles' hca I'll:! .-e .. cquiulcnt heal rau: Cl!pit.al COSI Updaacd CIIPilal ODIIS include trmmnssion mel dcli\uy costs. For Sol. fln.t )"'CIII" coSI is shown. subscqucnl years cotls are lower. Wind cos1s include PTC Emission = Source of power piEts emissions is EPA-ETS (Emaion Tracking SysLCm) lm.pons unports m: tnn.smisston constnincd 5.100 GWh pcryc. Xed Energy-"' Operates as Public Scrncc Company ofColor.to fila:l its 2007 Raoun::c PIE wilh CoLondo PUC on No,. 2007 55

PAGE 70

The MARKAL database for this research contains 14 generating technology options for future capacity needs consideration. The model covers 30 years study period from 2005 to 2035 in three years increments. Table 24 shows the conventional thermal and advanced technologies including Renewable technologies utilized in the Reference Scenario and Advanced Technology Scenario. The cost and performance characteristics of some of these resources were updated from Xcel Energy 2007 Resource Plan.38 The complete lists of input data gathered from sources such as DOE and EPA models are given in Appendix C. 5.14 Discount Rate and Inflation Rate Utilities use financial market risk measures to determine cost of capital or the discount rate for calculation of net present value of proposed capital investment decisions. By definition the appropriate discount rate for an investment is the opportunity cost of capital the rate of return that investors expect in capital markets for the same degree of risk as the risk associated with the project being considered [48]. The discount rate is considered as global parameter within the MARKAL model to represent the time value of capital for energy systems investment from the societal point of view. The discount rate used in the Reference Scenario (BAU) is assumed to be at 7.5%. This is consistent with largest utility in Colorado, Xcel Energy's current discount rate of 7.88% based on after-tax weighted average cost of capital. As discussed in previous sections, lOU's serve close to 60% of Colorado's load and are the main drivers in generation resources capital investment in Colorado. Inflation rate of 1.5% is assumed for all commodity prices beyond 2008. MARKAL also uses technology specific discount rate (or "hurdle rate"). In this study, the hurdle rate is based on the technology's life and the debt and equity capital structure for investment. EIA's fixed charge factors used for AE02007 within the National Energy Modeling System is utilized to represent the technology specific hurdle rate.39 Table 25 shows EIA's FCF for various technologies based on capital structure 45% debt and 55% equity and 38% tax rate and the respective depreciable life of each technology. The factors are utilized as technology hurdle rate for each technology within the model. 38 It is assumed that Xcel Energy's data is more up-to-date than other sources and therefore is adopted for some technologies in this study. 39 Data received through personal communication with ElA Staff Laura Martin. 56

PAGE 71

Table 25: Technology Specific Hurdle Rates r1xea Charge Plant Type Factors Scrubbed Coal New 16.23% Integrated Gas Comb Cycle 16.23% IGCC w/Sequestration 16.23% Conv Combustion Turbine 13.68% Adv Combustion Turbine 13.79% Conv Gas/Oil Comb Cycle 15.42% Adv Gas/Oil Comb Cycle 15.42% Adv CC w/Sequestration 15.45% Fuel Cells 15.42% Advanced Nuclear 17.63% Biomass (Wood) 13.24% Geothermal 12.68% Hydroelectric 15.81% Wind 11.26% Solar Thermal 12.02% Photovoltaic 11.07% Source: DOEIEIA 5.15 Transmission Constraints and Infrastructure Improvement Costs The electric system in Colorado is covered by two control areas or regions: Colorado East (the front range) and Colorado West (west of the continental divide). Power flows into and out of Colorado are constrained by a set of transmissions lines. Power flow into Colorado from north is constrained by transmission lines limits known as TOT-31imits, from the southwest (Four Corners) by TOT-2A limits, and from the West by TOT-1A limits.40 The transmission between two Colorado regions is constrained by TOT -5 limits. Figure 22 shows Colorado transmission Constrained diagram. Figure 22: Colorado Transmission Constrained Diagram 40 The term 'TOT" is short for Total transfer capability of a set of transmission lines over a geographically defined boundary. 57

PAGE 72

Table 26 shows the limits of the Colorado Transmission Constrained Paths. MARKAL is not an hourly model but allows imports and exports from and to a region to be modeled. Colorado does import and export power but in general is a net importer of power. The imports into Colorado are limited to TOT3 limits based on an annual average capacity factor of 40% or 5, 1 00 GWh per year. Table 26: Colorado Transmission Constrained Paths PATH PATH DESCRIPTION RATING TOT1A Utah to Western Colorado E toW: 650 MW TOT2A Four Corners to Southwest Colorado N to S: 690 MW TOT3 Wyoming to Northeast Colorado N to S : 1,450 MW TOT5 Western Colorado to Eastern Colorado Wto E : 1 675 MW Source: CCPG In 2004, CLRTPG42 forecast that over the next ten years, the demand for power will grow by 25% in Colorado's Front Range.43 To meet such a demand, the CLRTPG study forecast that over 2,750 MW of new generation resources will be added in the Front Range and robust high-voltage transmission will be needed to convey the power to major delivery points. Table 27 shows the overall transmission investment estimated by the CLRTPG representing a combination of budgeted and unbudgeted projects. Table 27: Ten-Years Colorado Transmission Costs (Millions-2004$) Entltv Scenario 1 2760MW Scenario 2 2760MW Scenario 3 2760MW AQUila $37 9 $25.6 $37.9 csu $41.1 $23.4 $41.1 PRPA $60.0 $60.0 $60.0 PSCo $443.8 $227.6 $4772 TSGT $138.2 $75.3 $138.2 Western $66.0 $1033 $102.1 Tote I $786.9 $616.1 $866.6 Source: CLRTPG 41 Colorado Coordinated Planning Group for Transmission, http://ccpg.basinelectric.com/ 42 The Colorado Long Range Transmissions Planning Group (CLRTPG) consists of six entities. Western Area Power Authority and five Load-Serving Entities in Colorado; Aquila Networks, Colorado Springs Utilities, Platte River Power Authority, Tri-State Generation and Transmission, and Xcel Energy/Public Service Company of Colorado. CLRTPG was initiated in January 2004 to jointly explore the potential for the development of a "back bone" transmission network in the State of Colorado that could benefit all electric LSE's in the state. 43 Colorado Long Range Transmissions Planning Group, Colorado Long Range Transmission Planning Study, April 2004. CLRTPG available online at: http://www.rmao.com/wtpp/CO Transmission Planning Group.html 58

PAGE 73

Scenario 1 modeled the majority of new generation in the southern portion of the Front Range of Colorado which include Xcel Energy's new coal-fired power plant's transmission investment requirement. Scenario 2 modeled the majority of new generation in the Northern Front Range of Colorado. Finally, Scenario 3 modeled a balanced generation pattern (combination of Scenarios One and Two) in the Front Range of Colorado. CLRTPG report concludes ... the costs for Scenario Three may be more reflective of the actual long-term costs." For energy modeling, Scenario Three results are modeled as the transmission investment requirement of power generation capacity addition in Colorado. The input to the model for transmission investment for periods "between" 2011-2017 is $322 million (2005$) per GW. For all other periods, a uniform transmission investment requirement of $70/KW is modeled.44 Transmission constraints from north (i.e., TOT3) have also been incorporated into the model as the upper bound of power generation imports into Colorado. 5.16 Resource Bounds MARKAL model is an optimization model that is each decision variable has certain specified bounds and may fall between three categories: equal to its lower bound, or equal to its upper bound, or strictly between the two bounds. The following resource bounds are incorporated in the model. 5.16.1 Power Imports As discussed in section 5.15 above, Colorado does import and export power but in general is a net importer of power. The imports into Colorado are limited to TOT3 limits based on an annual average capacity factor of 40% or 5, 1 00 GWh per year. 5.16.1.1 Market Prices of Imports Prices for imports are adopted from electric market prices developed by Xcel Energy for on-peak and off-peak periods using the implied market heat rates at three locations (south of Colorado at 4 Corners, west of continental divide at Craig, and southwest power pool) and the estimated gas market prices.45 The prices for Hydro and Renewable are assumed to be at 80% of off-peak and 120% of on-peak prices, respectively. However, for this study Hydro and Renewable import limits are set at zero only coal and gas based generations are allowed to compete with other resources. All market electric prices beyond 2030 were escalated implicitly at 2.33% 44 Xcel Energy's 2007 Colorado Resource Plan models $70/kW as transmission investment requirements. 45 Figure 1.7-2 of Volume I ofColorado Resource Plan filed November 15.2007 before Colorado PUC. 59

PAGE 74

based on the natural gas escalator as adopted by Xcel Energy. Figure 23 shows Market Prices forecast for import of power into Colorado. 5.16.2 Biomass Limits The Western Governor's Association's Biomass Task Force estimates that Colorado has a potential generating capacity of 436 MW. In this study, about 173 MW of total 436 MW Colorado potential biomass generating capacity is utilized for co-firing with existing 1,733 MW bituminous coal fired capacity at ratio of 1 0 percent biomass and 90 percent coal. The remaining 267 MW of biomass potential generating capacity is made available for biomass gasification technology. 100.00 140.00 120.00 I 100.00 00.00 EO.OO 40.00 20.00 0.00 Bectric Mar1c:et Price Forecast For Imports to Colora:to _.... .......------"""" ........--..-....... .....---.. ....... #' #' ,tp #' ./' ,f --+-Feak Of-Feak Figure 23 Market Prices Forecast for Power Imports to Colorado 5.16.3 Geothermal Limits The Western Governors' Association's Geothermal Task Force estimates Colorado Geothermal capacity at 70 MW based on existing Colorado reservoir data. The study indicates the first 20 MW of Geothermal can be developed at $80/MWh and an additional SOMW of capacity could be developed at $200/MWh or less. Geothermal limit is set at 70 MW and the cost and performance data is shown in Appendix C. 5.16.4 Solar Limits A Rule-Based constraint is designed to capture the RPS requirements in Colorado within the model. The percent requirement is modeled as a floor (i.e., a required bound since it is mandated) for the renewable generation in Colorado. The Rule Based constraint also recognizes the fact that the RPS requirements for solar generation shall include 4% from solar of which 2% shall be from distributed solar (i.e., rooftop solar). 60

PAGE 75

5.16.5 Wind Limits The Rule-Based constraint for renewables allows non-solar renewables to fulfill the RPS mandated requirements after taking under consideration 4% share for solar technologies. Since other renewable technologies such as geothermal and biomass have limited availability in Colorado, wind technology captures the majority of RPS requirements which reaches 16% of total electric sales by 2020. In the BAU scenario the RPS requirements is modeled as shown in Table 12. In Advanced Technology and Carbon Policy scenarios the RPS requirements is considered as the floor (i.e., the lower bound) but wind penetration is caped at 33% of total electric retail sales in 2035. This is mainly due to recent electric utilities independent studies that intermittent nature of wind generation could only be integrated into the utility system up to certain percentage of the utility generation. Beyond certain limits, the integration of wind generation becomes more costly thus less economical. For example, Xcel Energy recently performed a wind integration study for wind integration of 10% (722 MW), 15% (1038 MW), and 20% (1444 MW) into Xcel energy's system in Colorado and reported different integration costs and limits for its Colorado operation [49]. In this study, for the carbon tax scenario, the wind constraint had to be relaxed to see the impact of the carbon tax on the entire system. In the following section, the results of scenario analysis are documented. 61

PAGE 76

6 SCENARIO DESIGN AND ANALYSIS RESULTS This study looks at Colorado's energy needs, existing power system, and the many factors involved in achieving maximum economic and societal benefits from electricity production at minimum cost. In addition to traditional considerations of fuel type and generation capacity, policymakers today must take into account the affects of renewable portfolio standards, demand-side management strategies, and various measures to improve energy efficiency. Based on our in-depth analysis and projections, we offer several quantifiable pathways Colorado could use to achieve sustainable energy production and minimize harmful emissions in the future. Colorado's burgeoning population and fast-growing economy trend toward ever increasing demand for energy services in 2005-2035 (Figures 9 and 1 0). During the 1990s Colorado's population grew by over 30%. In July 2006, the state had 4. 75 million residents and a population growth rate that is third in the nation. Over the next decade, moreover, it is projected to add a million new residents. 46 In 2005, per capita electricity usage in Colorado was 10.4 MWh a year and sales were 0.217 KWh per state GOP. In 2004, the state registered 828 MT of C02 per each GWh of electricity generated. Yet by early 2007, Colorado's renewable generation capacity (excluding hydro) amounted to only 298 Mw. (Table 28) Table 28: Colorado Statistical Population & Generation lnfonnation47 Population July 1, 2006 4,753,377 KWh sales per person (2005) 10,365 KWh sales per dollar of state gross 0.217 Domestic _product 2005 (2006$) C02 emissions per GWh generated 828 2004 (metric tons) MW of renewable energy generating 298 Capacity early 2007 (excluding Hydropower) Demand for energy services drives estimates of future requirements for electricity generation over time, and concomitantly, of required added power plant capacity (see Section 5.3 above). Figure 24 shows the composition of aggregated demand for 46Colorado Alliance for Immigration Reform. Available online at: http://www.cairco.org/data/data co.html 47 Western Resources Advocates, available online at: http://www.westemresourceadvocates.org/medialpdf/State%20Ciean%20Energy%20Policies%20May%202007.p M 62

PAGE 77

energy by sector. In Colorado, the commercial sector dominates usage, followed by the residential and industrial sectors. Colorado SadDral IJ&mlnd F'onlc:aa 1CIXXIHD 9XID.OO ,-8XID.OO r7!XID.OO r-! -IDXD.OO r-!DXD.OO f-=.. 1 -4CIXD.OO DXD.OO 1 -1 -1 -1 -1 --a:xm.oo r--11XID.OO -rr--r-0.00 z::a; :2IX8 :2011 :2014 2017 2lliD 2IJ23 2IJ26 :2ll29 :2032 2035 II R!sidErtiBI Onm!rtial D h1sbiBI Figure 24: Colorado Sectoral Energy Demand 6.1 Energy Supply With primary energy use in the sate consistently on the increase, by 2035 Colorado will need 87% more electricity. For the domestic and imported supply of energy for the BAU scenario, see Figures 25. liD 7tD liD !Ul ilCD 3D = a.v =-ttt-f-f--t-1-.111_0>11 c Mll"'LO>ol I:Rl olD. c Mri Solid-.-Qlo il 1CII% Ill% BJ% -70% ED% 5011. 40% 3J% 211% 1CI!Io aEbnel cab_Bl_O:III c Elecln:ity f--t-1 Figure 25: Reference Scenario (BAU) Primary Fuel Consumption 63

PAGE 78

Coal currently provides the bulk of the state's energy generation, followed by natural gas, with a minimal role for oil and liquid fuel. Other sources (such as biomass) will play a role in the out years. At night electricity is used to pump water to storage facilities for generation during the day peak hours. Two new coal-fired power plants, with in-service dates modeled as 2011 and 2017, will increase coal generation from 2005 levels. While coal currently dominates electricity generation, therefore, it is projected to play an even larger role into the future, peaking in 2035 at 46,000 GWh of the state's total generation requirements (Table 29, Figure 26). 12!DD 1!IIID IIIID IIIID a "'" ,. Table 29: Reference Scenario (BAU) Generation Mix by Fuel Type DOEIEIA Model Outout CGWhl FueiTvoe 2005 2005 2020 2035 COAL 35,570 37,328 45,947 45,947 HYDRO+ PS* 1,415 3,499 3,615 3,632 GAS 11,923 11,366 16,489 22,585 OIL 17 400 0 0 RENEWABLE+MSW 810 875 12,699 33,006 IMPORTS 2,403 1,417 0 PS* = Pumped Storage with 2,172 GWh Generation included in Hvdro + PS. EIA data is onlv for Hvdro Generation BAU B...ECTRICITY GBERAT"Iall (GIIh) BAU BedrlcltyGeiWnllk:w1 %Stae d TaiBI I --1 i c--I I I li --w Ill fffff--1--f-----.f-, 1 1-i -1- ,. --1--1-1--1 -1-----1 ---f-1 -1--1-1-1-1 --1---1 -1--1--1 -f-f--f-1 -11-1--10% 1-1--1-1-I-I-1-I-f-I-1-1 -1-:2IXl5 2IXI! 2011 2014 2017 2!J20 2IJZ3 Zllll 202!1 2!ll2 2!135 :2IXl5 2IXI! 2011 2014 2017 2!J20 2IJZ3 Zllll 21129 2!ll2 2!135 cGAS DOL NPCRTS! Figure 26: Reference Scenario (BAU) Electricity Generation 64

PAGE 79

In this scenario, gas generation increases over the years and doubles by 2035. This is mainly due to the fact that all new capacity additions will be gas-fired. The hydro generation of 3,500 GWh in the reference scenario includes 2,172 GWh of generation from pumped-storage facilities in the peak hours. Table 29 also shows that the Reference Scenario (BAU) generation level is within +/-4% of reported DOE/EIA generation levels, which means the Reference Scenario (BAU) is calibrated very closely to the actual status in 2005 and reflects a realistic scenario for Colorado's power sector in the future. In 2005, coal (including about 4% of coal-based generation imports into the state) accounted 71% of the Colorado's total electricity generation. Yet in the BAU scenario, the projected share of renewables consistently increases over the years until in 2020, they meet the RPS requirement of around 16% of production (see Section 5.8.1 and Table 30). It should also be noted that in the out years, when renewable (especially wind) technologies will be more competitive, the model shows the use of renewables well above RPS requirements. It should be noted that in the model, RPS requirements were set as a floor, not a ceiling. Table 30: Percent Share of Generation by Fuel Type Fuel Type 2005 2020 2035 COAL 66.7% 57.2% 43.6% HYDRO+ PS 6.3% 4.5% 3.4% GAS 20.3% 20.5% 21.4% OIL 0.7% 0.0% 0.0% RENEWABLE+MSW 1.6% 15.8% 31.3% IMPORTS 4.3% 1.8% 0.0% Other major sources of energy for electricity are oil and imports. In the BAU, oil consumption for electricity generation in 2005 was less than 1% and came mostly from small power generators owned by utilities or municipalities. Imports (4% in 2005) generally come from existing long-term utility contracts. By 2020 when most of these will have expired, imported energy is expected to account for only 1.8% of consumption, and it is projected be completely phased out by 2035. In the Reference Scenario (BAU), renewables and gas-fired generation will gradually add to Colorado's generation fleet. It is therefore projected that coal consumption will decrease to 57% by 2020 and to 44% by 2035. The BAU projects: retiring 350 MW of old coal-fired generation in the next decade; adding new pulverized coal-fired generation of 750 MW (in service 201 0); adding 600 MW of IGCC technology with 50% carbon capture technology (in service 2017). The source of hydro generation includes both private and federally owned hydro and pumped-storage facilities, which accounted for 6.3% of generation in 2005 and is 65

PAGE 80

projected to drop to 4.5% in 2020 and 3.4% by 2035. This is due to the fact that there are currently no plans to build hydro power plants in the next 30 years, mostly because of licensing requirements. Other sources of renewables energy such as wind and solar, are projected to cover most of the increased need for electricity in the next 30 years. In 2005, renewables (mostly wind generation) only covered 1.6% of total energy needs. Wind generation is projected to increase to 8% by 2020 and to 24% by 2035. There was no significant solar generation in 2005, but by 2020 solar is projected to account for 3.1% of total Colorado's electricity generation, and by 2035 it should reach 4.3%. This is in part the effect of RPS requirements, which require utilities to acquire or generate a certain percentage of their sales from renewable sources (see Tables 12 and 20). 6.2 Installed and New Capacity Addition In 2005, the total installed generating capacity in Colorado was 11.22 GW, of which 45.8% was coal-fired, 38.6% gas-fired, 2.4% oil-fired, 5.7% hydro, 5.0% pumped storage, and 2.4% wind generation capacity (Figure 27) .. "' IS ,. 1-1 -li -r--rr-1-1-.. 1--, -f-.,_..,.. o....cR> .,..._cr 1 l I ; I '" ""' .... E ,. ,. ------.. F.. f--f-t--f--:-- 81_0. o 9.DJ!I._O. DO [JL D HI'(A) .a: .,..,_a: cr .,..,.._cr oFY_Qrlnl a........, Figure 27: Reference Scenario (BAU) Installed and New Capacity Additions 66 I .. 1--

PAGE 81

In 2020, coal's share of capacity is projected to drop to 30. 7%, of which 6.5% belongs to new coal-fired units. Gas-fired generation capacity increases to 43.4%, of which 20.6% belongs to old combined cycle (CC) and combustion turbine (CT) generating capacity, with 17.4% new CC, and 5.4% new CT. Renewables share generating capacity should increases to 19%, of which: 14.9% is wind technology, 3.5% solar technology, and 0. 7% biomass and geothermal technology. By 2035, the renewables should account for 33% (of which, 27% is from wind technology}, 4.3% solar technology, and 1.7% biomass and geothermal technologies. At the same time, coal-fired generating capacity should fall to 20.8%, while gas-fired generating capacity will keep its share at 41.4% (23.8% new CC, 3.6% new CT) of the state's total generating capacity.48 Figure 27 shows the portfolio of generation mix-which includes a sizeable amount of renewable technology by 2035 to meet the state's projected needs, while meeting the RPS requirements, and least-cost criteria of the model. By 2035, added new wind capacity (about 8.2 GW), which will accounts for 27% of total generating capacity, can be given credit for only <1 GW of effective load-carrying capability. The correct assessment of capacity credir for wind-generated power has been the subject of disagreement in many jurisdictions. Capacity credit is based on a reliability metric, known in the industry as a plant's 'effective load-carrying capability' (ELCC). Several studies applied the reliability metric to wind power plants to assess their effective capacity credit [58]-[59]. Theses studies estimate wind power's capacity credit at 20-40 percent of the rated capacity of the wind plant. Colorado adopted ELCC in the first competitively bid, large-scale wind project to come before the Colorado PUC [47]. Xcel Energy in Colorado uses 12% as its ELCC for wind plants and in resource plans. Figure 27 for a side-by-side comparison, that shows a very different profile for wind generation depending n the method used to calculate capacity credit. When the current wind technology capacity credit is applied, Colorado's total amount of new capacity added in each period meets the state's firm load obligation, the projected demand forecast, plus a planning reserve margin of 22% to insure reliability of power supply (Table 31 ). Because our model uses a higher margin to cover losses and any difference between levelized summer demand and actual peak demand, it builds in 8-10% more capacity than the forecasted-demand-plus-22% planning-reserve method.49 48 In Reference Scenario (BAU), model prefers CC technology to CT from total system cost point of view and inefficiency ofCT technology. In Advanced technology scenario, model prefers advanced CT technology again because of total system cost point of view and more efficient CT technologies. 49 Transmission losses of6.5% are used as input to the model. 67

PAGE 82

Table 31: Firm Load Obligations Compared to Model Output (GW) Forecast + 22"A. Planning Year Forecast Reserve Model Output 2008 10.39 12.68 13.99 2011 11.18 13.63 14.82 2014 11.96 14.59 15.89 2017 12.75 15.55 16.83 2020 13.53 16.51 17.84 2023 14.32 17.47 18.85 2026 15.10 18.43 19.92 2029 15.89 19.39 21.05 2032 16.68 20.34 22.07 2035 17.46 21.30 23.06 In the model, a reserve margin of 35% is chosen for Colorado grid, as a percentage of the summer peak demand. Note that the reserve given by the user in the model is typically much larger than prevailing rule-of-thumb values used by the electric utilities (e.g., 22% used above as planning reserve which was added to demand forecast and compared with new capacity additions built by the model). Reason for this is that the reserve margin in MARKAL also encompasses the difference between the levelized summer (or winter) demand and the actual peak occurring on one moment in that same period when the demand is actually the highest. 50 6.3 Reference Scenario (BAU): Projected Emissions Profile The model's C02 emission output for 2005, the model's base year, closely matched DOE/EIA and EPA reported emissions (Table 32). The model also includes all C02 emissions from coal or natural gas based electricity imported into Colorado. Model emissions of criteria pollutants are also within close range of reported emissions. Table 32: Base-Year (2005) Emissions EMISSIONS Source S02 (kt) NOx(kt) C02*(kt) OOEIEPA Reports 54-68 41,000 42,000 Model Output 55.2 58.7 43,605 Note: C02" Emissions include Imports Emissions of 1, 781 kt. Figure 28 shows the Reference Scenario (BAU) C02 emissions profile for all installed and new capacity additions. By 2020, the level of C02 emissions increases by 17% from 2005 levels to 51 122 MT. This is mainly due to increased demand for energy and the addition of two new power plants, one with no carbon-capture 50 See MARKAL user manual. Available online at: http://www.etsap.org/tools.htm 68

PAGE 83

technology and the other an IGCC plant with 50% carbon-capture. The level could have been still higher if RPS mandates were not in place. In the BAU, the RPS required 16% effective rate is reached by 2020, so that by 2035, C02 emissions increase to only 52,149 MT, which is up 2% from 2020 levels BAU BAU Poww ........ C02 Enillaloow PrCIIIIe (MT) C02 Enillaloow Prall .. dT ..... b F ... Typ. OIIIJ) 1<1l% r= = r """' t---I"'! I--I-5001) j .,. r-110% -1 --1-711% GJI) 110% --:DJI) 1 50% --1 ---21111) -1 -30% -I--I--I--I-20% ,_ 1---1--,_ 1---1---1-1CIDl ,___ --t--1 -t--I I I r: I I 10% -r 0 0% ;ms 2IXIB 2011 2014 3:117 3:13:1 2023 2028 21129 21132 21136 ;ms 2111! 2011 2014 2017 2020 2023 2026 2029 2DJ2 2035 IIO!n6_)1t8 .!C3:'1::_XI:III 0 EI:_R=' D ...... sa_EM PP .CrnG_Jilall O::X:_XI:III D ...... SJ:)_EII. R=' a a: D--0: OCT .,__cr DCI a: 0 Nlw'_OC DCT .,..,_CT DO.,._, c a:.l Qrl c.. np:.t DO.,._, 0 Gwlln1JOI1 0. Gwl h'pcwt Figure 28: Reference Scenario (BAU) Projected C02 Emissions Profile In Colorado, coal generation contributes more than 80% of C02 emissions throughout the modeling horizon. Because coal units are the most economical to operate, the model uses all installed and new coal power plants to their maximum level of availability at all hours. Under the Reference Scenario (BAU), the C02 levels reach 52,000 MT by 2035. But as RPS requirements are increasingly met over time, fewer fossil-fueled generation capacity is added and more demand is met by renewable technologies. This change in the power generation mix puts the brakes on the increase in C02 emissions in Colorado, but RPS requirements alone cannot decrease the C02 emissions level unless other constraints such as carbon policy scenarios are introduced. In the following sections carbon policies and their impact on the pattern of coal units utilization is discussed. 6.4 Advanced Technology Scenario This study evaluated a total of 6 advanced technologies:* Pulverized coal with carbon capture and sequestration (CCS). Equipped with (CCS), conventional pulverized coal-fired technology is considered to be at 50% of C02 capture capability. 69

PAGE 84

Integrated Gasification Combined Cycle (IGCC) technology is a power generation process that integrates a gasification system with a conventional combustion turbine. It uses coal or natural gas to generate electricity from a combined-cycle power block. Coal-fired IGCC technology is considered to have 50% CCS capability. Natural gas-fired IGCC is considered to have 90% CCS capability. High efficiency advanced combined cycle. Powered by natural gas High efficiency combustion turbines. Advanced nuclear technology. *Although state utilities are seriously considering nuclear generation as a viable option to reduce C02 emissions in the near future coupled with advanced CCs and CTs, their first availability is projected no earlier than 2014 which is the first year availability incorporated in the model. See Table 24 and Appendix C for detailed cost and performance data on all technologies considered in the model. 6.4.1 Energy Supply Comparing the reference (BAU) and advanced technology scenarios (Figure 29), we see that advanced technologies use less fuel (in this case, natural gas) to generate the same amount of electricity. Adv..-d TechllllogyS..IW"Io 120000 ,....-------------, 11Xl000 1-----r-----n-1 eoooo------eoooo -1 -BI'U BI'I.J+.IIdlllEOi ----10% -- lD'II. D Hft:R:l 0 G'IS 0 Q. 0 FUaMB.BOIH 0 M'tRr Figure 29: Advanced Technology Scenario Electricity Generation It should be noted that in the out years, as more advanced technologies become available thus more cost-effective, their share of the energy production increases. 70

PAGE 85

6.4.2 New Capacity Additions Figure 30 compares the BAU and advanced technology scenarios with regard to new capacity additions. Beginning in 2014, when new technologies are projected to become more widely available, the model incorporates the use of more Combustion Turbines (CT). With higher efficiencies and lower costs51, advanced CTs can operate fewer hours than are projected for conventional CCs in the reference scenario (BAU), and renewables technologies supply the energy formerly produced by the now displaced CCs (Figure 29). Advanced Tectnology Scerato NBwCI!fa:lty A:llllkrB 15t------J or--II a.::rr-_a: O::rrB_)f,al!ll 0 O:X:_)foel 0 Gl!!dtemlll oA&_cr .Adtl_a:; .O::Iw_oc oel::r'PI_cr FV_
PAGE 86

6.4.3 System Costs and C02 Emissions Profile According to the advanced technology scenario, as more energy-efficient technologies are introduced into the generation mix, both system cost and the C02 emissions profile will be slightly reduced in the out-years. The discounted total system cost reduces by 1.1% compared with the BAU scenario, and the growth in C02 emissions will slow slightly as renewable energy sources begin to displace fossil-fueled CCs' generation and more advanced CTs are used to meet peak demand (Figure 31 ). ----_r-----I r----:-g --.I -T--caa.-Figure 31: Advanced Technology Scenario Costs and C02 Emissions Profile In the remaining portions of this study, the advanced technology scenario is considered as BAU on the assumption that, as advanced technologies become available, they will be incorporated during the normal course of business to improve efficiency and lower costs. 6.5 Energy Efficiency and C02 Emissions Reduction In the 1980s, energy sector economists introduced the paradigm of energy-efficient resource use. More efficient means of production would generate 'negawatts,' argued Emory Lovins, and thereby reduce the need for new plants to meet end users' power demands [56]. While utility companies have been well aware of the energy efficiency argument for the past three decades, 52 deregulation and increasing competition in the electric utility industry during mid-1990s led many to cut costs and spend less on demand side management (DSM) [39]. In the national (and international) debate on strategies 52 Utilities have been involved in energy efficiency programs since 1973 Arab oil embargo which continued throughout 1980s and 1990s but started dwindling down during the electric industry restructuring. Recent uncertainties regarding greenhouse gas regulations appears to jump started energy efficiency programs by states and utilities responding more favorably to energy efficiency programs again. 72

PAGE 87

to reduce greenhouse gas emissions, DSM programs are considered viable options for reducing C02 emissions in utilities resource plans [50]. The terms 'energy conservation' and 'energy efficiency' are often used interchangeably in policy discussions, although they are not in fact the same. Efficiency entails doing more with less, while conservation means doing without. Energy efficiency is the ratio of energy services output (say, electricity) to energy input (say, coal) and is a measure of how much energy is produced for every unit of energy consumed to make it. Better technology is required to improve energy efficiency, while to raise the level of energy conservation requires regulation and a change in consumer lifestyles and energy consumption behavior [51]. Some argue that consumption has increased because production has become more efficient, resulting in lower prices. Boardman comments that substantial improvements in energy efficiency have been passed on to the general economy in the form of greater productivity: "The substantial improvements in energy efficiency have been absorbed into more and larger product. At some stage, society needs to recognize that ever-higher standards of living are threatening our ability to limit climate change and, therefore, reducing our future quality of life." [52] Herring similarly argues that higher energy efficiency has a 'rebound effect' of driving up higher consumption [51]. Three categories of rebound effect have been described: [53]: 1. Direct effects, which stem from consumers' natural tendency to use more of any low cost product or service; 2. Indirect effects, such as lower energy costs' spur to the economy, which creates more income and leaves more income available to spend on other products and services-some of which (such as travel) consumes more energy; 3. Economy-wide effects, long-term changes in the economy caused by technological innovation and changes in consumer preferences and behaviors, which is brought about by the substitution of relatively cheap energy for other factors of production. According to Herring, even cost-effective and energy efficient lighting has a rebound effect. Studies of the Compact Florescent (CFL) bulbs, for instance, suggest that about a third of users choose to leave CFL lights on longer and to install additional CFL lighting, because it uses less energy, in the garden or for security. See [51], p. 201. Herring concludes with the question "Does innovation to produce more energy efficient products and systems lead to lower energy consumptions? ... This depends 73

PAGE 88

upon the extent of the 'rebound,' which effect is difficult to measure on a national or macro-scale." In Colorado, the regulated utility Xcel Energy was required to conduct a market potential assessment of various DMS and energy efficiency measures. 53 The goal was: "To conduct a market study to determine levels of efficiency available for various customer classes, the costs associated with such measures and whether such levels of DSM are cost-effective and available in Colorado." The study estimated potential savings in electricity and peak demand from DSM measures in Xcel Energy's Colorado service territory.54 The study covered savings in new and existing residential and nonresidential buildings, as well as from making industrial processes more energy efficient. The original study was restricted to DSM measures presently available commercially and covered the 8-year period 2006-2013. This period was later extended to 2015 to allow Xcel Energy's resource planning activities to harvest the results. Primary data collection for the study involved 300 residential on-site surveys, 152 commercial on-site surveys, and 193 vendor telephone surveys. Secondary sources included several internal Xcel Energy studies and data, as well as a variety of information from third parties. The study identified baseline end-use and developed estimates of effects from future energy efficiency gains using varying DSM programs. As part of the baseline, the study identified the types and approximate size of various market segments with the greatest DSM potential in Xcel Energy's Colorado service territory. These characteristics then served as inputs for a modeling process. The bulk of the analytical work for this study was carried out using a model developed for studies of energy-efficiency potential. The model was a spreadsheet model that integrated data on technology-specific engineering with that on customer behavior and utility market saturation, load shapes, rate projections, and marginal costs. The study used the Total Resource Cost (TRC) test screen DMS measures and considered electric utility avoided-cost benefits only [40]. To evaluate and justify potential benefits from energy conservation, researchers generally use a supply-side planning model where utility marginal costs are the yardstick against which conservation program costs are judged [54]. Under this framework, technologies or practices that reduce energy use through efficiency are characterized as "liberating" supply for other useful energy demands. These energy efficient technologies are therefore thought of as a supply resource and plotted on an 53 Xcel Energy is doing business as Public Service Company of Colorado. 54 Colorado DSM Market Potential Assessment, KEMA, 2006. 74

PAGE 89

energy supply curve. In this study, we used this technique to assess the cost and benefits of DSM and energy efficiency measures. 6.5.1 Statewide DSM/EE Plans In past years, Colorado's regulated and non-regulated utilities have actively pursued DSM programs. For example in 1993, Colorado utilities spent 0.40 percent of their revenues for DSM programs, with an estimated savings of 0.53 percent of sales. DSM activities dropped off in late1990s, mostly due to increased competition in the electric utility industry following restructuring era [39]. In 1998, for example, Colorado spent 0.11 percent of state electricity sales revenues for DSM programs, which resulted in estimated savings of 1.29 percent of sales. Comparing 1998 to 1993 DSM activities; a total of 0.29 percent reduction in DSM related expenditures from 1993 spending levels occurred in 1998 but, the savings as percent of sales were higher, 0. 73 percent change from 1993 sales level [55]. 6.5.2 Cities and DSM/EE Projects and Costs Aside from utility DSM programs, cities and municipalities that provide electric services to end-users have also been active in DSM/EE programs for many years. For example, Seattle City Light (SCL) has been active in energy conservation programs since 1979 and continues to provide energy efficiency programs to residential customers. In 2002-2004, SCL sought to secure Bonneville Power Administration funds, under the conservation agreement for residential energy sector, by focusing on conservation projects eligible for power purchase offsets. In 2004, SCL secured 1.1 MW from energy-saving residential projects. SCL reports that it cost $102/MWh on average to accomplish Energy Conservation for Multifamily Residential housing during 1986-2004 and $79/MWh on average for the Built Smart project of 1992-2004 [57]. Recently, many cities and municipalities that do not provide electric services to their citizens have become actively involved in DSM/EE programs to provide incentives to reduce electricity consumption. One direct benefit of reduced consumption is a smaller greenhouse gas footprint for these cities [e.g., Boulder, Aspen, and Seattle]. 6.6 Aggressive DSM/EE Scenario This study modelled two aggressive DSM scenarios independently. Both posit that that recent climate action plans at the city and state level require Colorado utilities and municipalities to work together to reduce electricity consumption, particularly if they are to achieve the target C02 reductions by specified dates. First, it is assumed that Colorado utilities' efficiency programs will result in the reduction of 300 GWh a year, beginning in 2008 and accumulating over the planning horizon. It is further assumed that the penetration rate in the three major energy using sectors is 75

PAGE 90

commercial (65%), residential (25%) and industrial (10%).55 In this aggressive scenario, the accumulated total energy savings over the planning horizon would be 9,000GWh. It was further assumed that cities and municipalities will institute more stringent building codes and energy conservation programs, which will result in a 1% per year reduction in energy demand. The goal is to reduce electricity consumption by 30% by 2035. The same sectoral distribution factors commercial (65%), residential (25%) and industrial (10%) used in the first aggressive DSM scenario obtains for the second scenario as well. Over the planning horizon for this scenario, The accumulated total energy saving would be 14,200 GWh 2020 and 28,400 GWh by 2035.56 This conservation goal is consistent with SWEEP 7 four-prong energy efficiency goals for 2005-2020. The reductions goal for each of proposed program is as follows: DSM (7,323 GWh) Building Code (1834 GWh) Lamps Standards (3784 GWh) Industrial Option (1733 GWh) TOTAL= 14,674 GWh by 2020 [NREL study for RMCO]. 58 As discussed above, the average DSM/EE costs (including rebates and administrative costs paid by municipalities to accomplish energy efficiency programs) is on the order of $60-90, depending on the location and the type of conservation program. For example, Seattle City Light's average conservation program cost the city some $90 per MWh for residential customers (cost to utility not TRC) [57]. 55 Recent study by Xcel Energy shows a penetration rate of75/25% for commercial and residential customers, respectively. 56 The total conservation of28,400 GWh is 30% of Colorado total base case energy demand forecast of95,000 GWh for 2035. 57 Southwest Energy Efficiency Project, http://www.swenergy.org/ 58 NREL performed a spreadsheet analysis for Rocky Mountain Climate Organization (RMCO) incorporating Southwest Energy Efficiency Program proposed DSM/EE measures. 76

PAGE 91

Table 33: Reference Scenario DSM Costs Annual Maximum Average Hour Year ($/kWh) ($/kWh) 2008 0.053 0.105 2011 0.037 0.118 2014 0.036 0.102 2017 0.052 0.126 2020 0.070 0.169 2023 0.093 0.226 2026 0.125 0.303 2029 0.168 0.405 2032 0.224 0.543 2035 0.301 0.727 Source: Xcel Energy DSM Docket In this study, Xcel Energy's maximum-hour avoided marginal energy prices are used as the Total Resource Cost of the DSM programs, since Xcel Energy is the largest electric utility actively pursuing DSM measures in the state. (Table 33) 6.6.1 DSM/EE Scenario Electricity Generation Figure 32 compares generation levels with DSM/EE measures and BAU with no such measures. 2ID5 2IXJ3 201, 2014 2017 2CI20 2ll23 2ll26 2029 2D32 2lll5 N:lllr: BA.ll n:uit.x.::.l CEM+ 81"1_CEM EWJ (ro [9.4 SCIU+Yaii_CSJI EI'U"+-T ........._ BA.r+.Ad.IT+EE (3J%by21ll5) Figure 32: DSM/EE Scenarios Total Electricity Generation As expected, aggressive DSM/EE measures substantially reduced total energy generation, with the most aggressive Energy Efficiency program cutting energy generation by 30% in 2035. Potential costs and benefits, in the form of C02 reductions, are discussed in the following sections. 77

PAGE 92

In considering electricity generation by fuel type, the patterns of generation from coal do not change much when DSM and other energy efficiency measures are introduced, because coal is already the least-cost fuel and is not at margin to be affected by such measures. The major expected contribution of aggressive DSM/EE measure is a reducing investment in new capacity and lowering fuel costs associated with generating units that are at the margin, such as natural gas CT generating units (Figure 33). Figure 33 also shows the level of generation from natural gas and renewables. As more aggressive DSM/EE measures are introduced, less generation from gas-fired units are used. The same is true with renewables. Since the amount of renewables is based on RPS requirements (a percent share of total sales of electricity), as energy demand diminishes over time, less is correspondingly required from renewable sources. 120000 100000 80000 40000 20000 0 -eER;V s=Fia811C'f SCENARIOS EleclrtdlyGiit...-la I 1: Full T)'pt ---1-t---1-eo.u BAdfT
PAGE 93

Including wind technology, the need for new capacity decreases by an astonishing 59 percent by 2035. 15.a:D 10.a:D 5.a:D O.a:D DSM_Eragy Blk:lelq Sea artos Adllllal-. -----d oiGCC lt>ol ocx: <%:5 Goiitwna oMICT Figure 34: Aggressive DSM/EE Scenarios New Capacity Additions 6.6.3 DSM/EE System Cost-Benefits Figure 35 compares the reference scenario (BAU) with the total discounted system cost reductions realized in all DSM/EE scenarios. The level of cost-benefits increases as the level of DSM/EE measures increases. For example, if we implement 300 GWh per year DSM/EE measures within the state, end-users will save a total of more than 7 percent. Additional emissions reduction benefits are discussed in the next section. DSMEnergy Eflic:iency Coot-0 -c-'1(, 2 ,CXXl ...,..!,_ ._ 2 4 Roduction Reduction ......... .....,._T DSMIEE Scenllrioo !ram BAll lramBAU I -4,CXXl .,. (!aO<&H>EM) ... -...,._ T BAU noDSM) 0 .O ,CXXl 1BAU+Xce4 OSM -527 % f ... 2 BAU+Xcel (DSM+EnhDSM -7211 U1'1!o -8,CXXl 3--BAU.+AdvT -1 2110 -251% -10,CXXl-4-BAUAdvT+OSM .44'1(, .....,_ T 5-BAU"+AdvT+EE (30'1!o by 2035) ,8110 % -12.CXXl EE ,._,.. ..,. I NH. EW..1 hctda bdh t::DAR-c:vwn: F1gure 35: DSM/EE Cost Benefits (2005M$) 79

PAGE 94

Ratcheting up DSM/EE measures to 1% per year reduction in energy consumption would bring at least a 19 percent cost reduction from BAU's discounted total system cost, due in large part to avoided investment in new power plants and associated operating costs. 6.6.4 DSM/EE C02 Emissions Profile Over the course of the planning horizon, currently proposed DSM measures would have minimal impact on C02 emissions. Combining both Xcel Energy's DSM measures and the Advanced Technology scenario, for instance, would reduce overall C02 by 0.6%. By contrast, any of the aggressive DSM/EE scenarios would reduce C02 emissions substantially. Reductions in C02 emissions are a societal benefit that would be realized in addition to the cost savings described above. (Figure 36) CXII---+-a-tJ(rmCB4 .,..,..,....T BA.U no OSM +Xcel OSY BAU+Xcel OSM+EnhOSM BAUAdvT BAUAdvTOSU 300GW BAUActvT+EE 30% b 2035 C02 RedUCtion A.:luc::tion fnmBAU (01) (no DSM) 0 -1,580 .()J -2,155 .() ... -J 315 .() .... -13,1SO -2o4CMI -39,238 -71% Figure 36: DSM/EE C02 Emissions Profile 6. 7 Carbon Policy Scenarios With rising awareness of greenhouse gases' potential impact on the environment, a number of legislative efforts are now under way to reduce C02 emissions. But while C02 regulation has been roundly discussed, there is as yet no clear consensus on what will likely be accepted. What is likely is that within a few years the picture on C02 regulation will become clearer than it is today. As discussed in chapter 1, there is no federal requirement in the U.S. to reduce GHG, but states and local governments have begun to institute GHG reduction initiatives on their own. The Regional Greenhouse Gases Initiatives (RGGI) is one such program formed by the Northeastern states, and California recently passed the Global Warming Solutions Act of 2006, A B. 32 to reduce carbon emissions from sources within the state. At the local level, many cities (such as the City of Denver) are establishing Climate Action Plans for their own GHG reduction programs [11]. 80

PAGE 95

To control GHG emissions from the Northeast's electric power sector, the RGGI has proposed a 'cap and trade' program similar to the one used to control acid rain. Each participating state has agreed to cap its GHG emissions from power production, setting the goal of stabilizing emissions by 2015 at the average level 2002-2004. The RGGI will then seek to reduce emissions by 10% by 2015, and by another 10% by 2020 [12]. 6.7.1 Carbon Cap Recently the United States' Senate introduced the Low Carbon Economy Act of 2007, S. 1766, which establishes a mandatory greenhouse gas (GHG) allowance program to maintain covered emissions at approximately 2006 levels in 2020, 1990 levels in 2030, and to cut emissions to at least 60 percent below 1990 levels by 2050. The Colorado Climate Action Plan sets a goal for the state to achieve an economy wide reduction in C02 emissions by 20% below 2005 levels in 2020 and by 80% below 20051evels in 2050. The plan calls for utilities statewide to reduce emissions and consumers to change the way they use energy. For example, the Action Plan calls for significant customer and government initiated reductions in energy usage including improvements in lighting performance, a call for industrial users' efficiency, and changes in building codes. To accomplish these goals, this study uses an aggressive energy efficiency scenario, which incorporates changes in usage driven by customers. 59 To assess the impact of proposed climate action plans, we analyzed two scenarios: one to reduce the C02 level by 1 0% and another by 20% by 2020. We also looked at ways to achieve 1990 levels by 2035 (Table 34). With the for 10% reduction goal, total net reduction of C02 emissions over the planning horizon was 18.7 percent, while with a 20% reduction goal C02 emissions were reduced 25.5, similar to the plan to achieve 1990 C02 levels, which achieved a 25.1 percent emissions cut. Table 34: Carbon Cap Policy Effective C02 Reduction C02 Carbon Policy Reduction %Reduction Scenarios (kt) from BAU BAU + 1 0% by 2020 102,437 18.7% BAU + 20% by 2020 139,738 25.5% BAU + 1990 Level 137,493 25.1% The net system-wide additional cost to implement these carbon policies were mainly related to adding more renewables and efficient, less carbon intensive, new 59 Colorado Climate Action Plan available online at : hnp://www.colorado.gov/energy/in/uploaded pdf/ColoradoCiimateActionPian OOI.pdf 81

PAGE 96

generating capacity. Achieving a 10% reduction in C02 emissions by 2020 cost roughly 1 0 percent more than implementing the BAU scenario, while the cost of achieving a 20% reduction was 15 percent higher, and or achieving 1990 emission levels 13 percent higher. (Table 35) Table 35: Carbon Cap Policy Net Cost Increase from BAU (2005M$) Cost Diff. % Carbon Scenarios From BAU Increase BAU+10% by 2020 5,198 10.4% BAU+20% by 2020 7,439 14.9% BAU+1990 Level 6,712 13.4% Figure 37 shows the electricity generation by fuel type for carbon policy scenarios compared to BAU. Carbon Polley Scerwtos ElaclrldiV Gal &i6Ji I by Fuill'ype 12!XIXl ,---------------., I 1CDXD f-----! !IXXD ---DCa!ll Dt-t,c:hJ.Gas Figure 37: Carbon Cap Policy Electricity Generation by Fuel Type As more stringent carbon policies are instituted, the level of coal-fired generation falls. With a 10% cap at 2005 levels, coal's share of total generation drops from over 40% to under 40% by 2020. With a 20% cap at 2005 levels, it falls to 30% by 2020, and with a cap based on achieving 1990 emissions levels, it falls to less than 30% by 2035. In the scenario where C02 emissions reach 1990 levels by 2035, nuclear capacity becomes competitive with other generating technologies, in large part because it is second to renewables as a source of C02-free generation. However, if restrictions on renewables (especially wind) are relaxed from 11 GW to higher levels in 2035, growth in nuclear capacity is delayed to future years. 82

PAGE 97

100'/a !IJ% 60% 40% 20% 0% Carbon Polley Sc8l artos EIKirtclly Gllownlllon bv F.-Type BAU BOU + 10% 17;' 2020 BtU+ 20%17;' 2ll2D BAU + 199l L....a ac.c. ot-t,cto Gee N;.ae&" al Figure 38: Carbon Cap Percent Share of Electricity Generation by Fuel Type The portfolio of new capacity additions is also affected by different carbon cap scenarios. The more stringent the carbon cap policy, the less fossil-fueled generation technologies are used, and the more renewable technologies are added to the generation mix (Figure 39). 25 15 10 ---=-1----; 0 J1 II il c.ocr OGI'S_O:S ON.J:iea'" 0 P\I_CI!n"aa 0 PV_Fb::Jrq:J 0 ScEr_Ttwmlll 0 Wn::l Figure 39: Carbon Cap Policy Scenarios New Capacity Additions As stringent carbon cap scenarios require that even advanced CC and CT technologies be replaced, natural gas IGCC technology also enters the generation mix. In fact, the entrance of IGCC technology is a sign that renewable technologies 83

PAGE 98

have reached their limits in meeting future carbon regulations alone. In this study wind technology's limit is set at 33% of total generation capacity mix in 2035, and that the limit for solar thermal generation is 1 GW in 2035. Wind integration and solar thermal site considerations therefore restrict the addition of renewables, even where they are the most cost effective choices The next choice for less carbon intensive and cost effective technology appears to be natural gas IGCC technology with 90% carbon-capture sequestration. When considering total discounted system costs and cost differentials for the three carbon cap policies, it appears that meeting the 2005-level cap (BAU+20%) by 2020 costs more than implementing the other two cap scenarios. c.-..-y-...... l """ """' -r--H ('""' Cost Dltf. I...., 1 -f-. Carbon Scenarios From BAU "Aolncrwase BAU+10% by 2020 5198 10. 4% = iJ-1---: BAU+20% by 2020 7439 14. 9% BAU+1990 LtM!t 6,712 13. 4% 1---: -"'"' .. :IJll) ..., Figure 40: Carbon Cap Policy System Costs and Differentials It is also estimated that meeting the 1990-level cap by 2020 will cost the system about 1.5% less than meeting the 20% cap by 2020(Figure 40). JO(II) >--------------------------------:ztm 2CD5 201 1 2014 2017 2IJlO 2DZ'l 20iB 2D2Il 2DJ2 Figure 41: Carbon Cap Polley C02 Emissions Profile 84

PAGE 99

When considering the C02 emissions profile of the three carbon cap policies, it should be noted that implementing a 1990 cap by 2035 would allow for more gradual investment in less carbon-intensive technologies, resulting in a less discounted total system cost over the planning horizon (Figure 41 ). 6.7.2 Co-Benefit of Carbon Cap Another important benefit of instituting a carbon policy is reduction of such criteria pollutants as S02 and NOx emissions. As shown in Figure 42, the emission profile for both S02 and NOx reduces in 2011 as carbon policies take effect. This is a co benefit of emission policies designed to control other pollutants and entails no additional cost. A beneficial byproduct of constraining fossil-fueled generation is a drop in NOx emissions of 30-52% and of S02 emissions of 20-42% percent. Scenarios NOX 502 J! 45f-----10% Cap 30.2% 19.6% JOf----20% Cap 43.6% 31.5% 1990 Cap 52.3% 41.9% Figure 42: S02 and NOx Emissions Profile 6.7.3 Carbon Tax We devised two scenarios to assess the impact of two possible carbon tax policies. One is a Btu tax applied upstream based on the heat content of fossil fuel production. The other is a tax applied downstream based on C02 output from the use of fossil fuels. Btu Tax: This is an energy tax based on heat content -the British Thermal Units (Btu) generated by particular fuels. As introduced in the United States in the1990s, the Btu tax would have been imposed on electricity generation by coal, natural gas, petroleum products, and imported electricity at a base rate of 25.7 cents per million Btus (p/MBtu).60 The tax would have had a neutral impact on a regional basis and would have affected the market shares of energy sources equally. In the event, the Btu tax met with strong opposition in the United States and was never adopted. 60 The proposed Btu tax by the Clinton Administration would have applied to nuclear-generated electricity or hydroelectricity if adopted. We did not apply Btu tax to nuclear or hydro electricity in this study. 85

PAGE 100

In our Btu tax scenario, we applied a tax of 50 cents (p/MBtu) upstream (double what was proposed in the 1990s) to all fossil fuels used to generate electricity. Figure 43 shows the C02 emissions profile following imposition of such a Btu tax. It should be noted that, over the planning horizon this Btu tax would increase the discounted total system cost by more than 8% while reducing total C02 emissions by less than 1%. This is mainly due to the fact that the Btu tax has a neutral impact on the system and affects all fossil fuel costs and C02 emissions proportionally. 31 55, ().X) 50,0Xl 45,0Xl 40,0Xl 35,0Xl :Jl,c ---#' ,/' .I #' ,p9' J' Figure 43: Btu Tax C02 Emissions Profile C02 Tax: Another tax proposal is a C02 tax that would impose a specific dollar amount per metric tonne of C02. This tax is based on the amount of C02 emitted and would apply to electricity generated by fossil fuels (coal, natural gas, petroleum products, and imported electricity). Evaluating how a C02 tax of $20-$70 per tonne downstream would affect total system C02 reduction, we noted that although large tax revenues were generated, the tax did little to shift fuel usage.61 Once constraints on renewables were relaxed, however, the system reacted to a $58/t tax by building more wind power plants and retiring old coal-fired generating units. 61 We did not model carbon capture and sequestration (CCS) retrofits option for cost comparison on the existing coal-fired power plants. This option is a viable option once CCS retrofit technology is developed 86

PAGE 101

Figure 44 shows the level of electricity generation by fuel type associated with four different sizes of C02 tax. As C02 tax rises above $50/t, the system begins to reduce its use of coal. Coal, the most carbon-intensive fuel, begins to be replaced with renewables, particularly wind generation, and imported electricity levels also diminish as C02 taxes increase. 12aXJO 1CXXXXl 8CJCXX) 0 Carbon Tax Scenarios Gel l8l1ltion by Fuel Type WO Wind Unit ---r--, -I BIIU BIIU + S20 Tax BIIU + $50 Tax BIIU + $160 Tax BIIU + $70 Tax (J Ccel 0 Gls 0 0 N.J:::Iea" 01 0 h-p:rts Figure 44: C02 Tax Scenario Generation by Fuel Type Once a C02 tax is imposed and constraints on renewables are relaxed, fuel types and new capacity additions begin to shift within the system (Figure 45). At $20/t C02 tax, the system reacts neutrally and no major shift in new capacity additions take place. At $50/t C02 tax, the system reacts minimally to replace new fossil-fueled capacity (e.g., CTs) with renewables. At $60/t C02 tax and higher, the system shifts dramatically toward more renewables and reduced fossil-fueled new capacity. It should be noted, however, that the relationship between renewable and fossil-fueled capacity is not one-to-one. Because it takes more new wind capacity to replace every unit of fossil-fueled capacity lost (due to lower capacity credit assigned to wind). The impact of integrating large new wind capacity into the system, moreover, has yet to be evaluated. Other options-such as energy efficiency measures and retrofitting existing coal-fired generating units with carbon capture and sequestration technology-also need to be considered. 87

PAGE 102

35 25 20 15 10 -5 0 Carbon Tax Scenarios New Capadly adclllons woWndUrrit 1 I--1-----::-f--I I .'' """"'"'= r;, + $20 Tax I BAU +$50 Tax BAU + seo Tax BAU+$70 Ill Elarass_O:::: CaT6_)fcel 110:::: cr 0 S:lla"_lhermll 0 Vllnd OGX_Xcel 0 o::::_a:::s Geclherrral o l'tN_CT 11/>tN_O:::: PV_Oertral PV_Fb:Jitap Figure 45: C02 Tax Scenario New Capacity Additions Tax Different levels of tax on C02 produce very different effects on the discounted total system cost, amount of tax revenues, and C02 reductions. When the C02 tax is set at or below $50/t, the system reacts minimally to reduce C02 emissions. Although the tax generates over $21 billion over the planning horizon, the system will invests only 1.5% more to achieve 4% reduction in C02. As the C02 tax is increased to >$50/t, however, the system invests proportionally more on renewables, resulting in 17% higher costs and a 40% reduction in C02 emissions. Table 36: C02 Tax Costs and Tax Revenues C.rbon Dlacou-%Coat C02 Tal Dllr. from C02 Av-T81 TotalS .... Dltr. from Ravenua BAU -uctlon C02% C02 Coat Sc:enlrloa Coat BAU C2001MSI Ccoat+T81l Ckll Reduction CS/11 BAU 50007 BAU+$20 50057 01% 8 874 17.8% .2% -1380 BAU+SSO 50,759 15% 21410 44.3% -24 286 -4.4% -913 BAU+$80 58683 17.3% 17 253 51.8% -208825 -39.6% -125 BAU+$70 64,047 28.1% 14,334 56.7% -294,173 -53.9% -96 Under this scenario, any reductions in tax revenues come about because consumption of fossil fuel has been reduced by the introduction of renewables. With a C02 tax of $70/t, undiscounted tax revenues from C02-emissions decline sharply 88

PAGE 103

beginning in 2017, when more renewables are online to displace fossil-fueled generation (Table 36 and Figure 46). tiDD 40D) I I DID ------1CEDJ -----lt----" 2l:lfi 2(111 2011 2014 2017 2IJ2I) 2ll2'3 2lliB 2llZ9 2llll 2ID5 -... ........._ BAD1 _. BA12D BA150 BAlaJ --+--BAT10 I 2IXIl -------- -j i 1!m !!I I = l '::'' H : 2IJl5 2101 2011 2014 2011 2020 2023 za 20l9 2m2 2m5 BA12D BA.150 BA"RRl 8AT70 Figure 46: C02 Tax Scenario C02 Emissions and Tax Revenues Profile The emissions profile for the criteria pollutants 802 and NOx mirrors that for C02 emissions from coal-fired power plants (Figure 47). 802 Errialon L.-vel BAU (BADH)va.CXIZTa-2D -... ZJJS 2D11 2D14 2D17 2lli!D 2lZl 3JiB zw xm 2DS -"' NlX Er1'w.lon UMII BAU(BADH)va. CXIZTa-2XJ5 2CDI!I 2)11 4D14 2D17 2lli!D :2D23 2D2B Zl29 :2D32 3l3l5 -... EW>i BATXI BA150 8A.18) ......... BAT70 Figure 47: C02 Tax Scenario Criteria Pollutant Profile 6.7.4 Biomass Co-Fire In this study, biomass was modeled first in a stand-alone biomass gasification combined cycle and then considered as a co-firing option, with all bituminous coal fired units in Colorado posited at a ratio of 10% biomass to 90% coal. Clearly, the 89

PAGE 104

competitiveness of biomass-generation technologies depends heavily on the price and availability of biomass fuel resources. For biomass generation to become a major player in the RPS, the prices for biomass feedstock need to be competitive with coal price, which is at near or under $2.00 per million Btu (mmBtu). In this study, the DOE average price of $2.25/mmBTu was escalated over the planning horizon to reflect an inflation rate of 1.5%. In Colorado, when biomass is co-fired with all bituminous coal-fired units at the ratio of 10 percent biomass to 90 percent coal, the total reduction in C02 emissions is estimated to be around 1.1 %. Over the planning horizon, this amounts to 6,000 kt but increases the discounted total system cost by $2.2 billion, and a cost for C02 reduction of roughly $370/t. To make biomass co-firing a viable option for C02 reduction, prices for biomass feedstock must be competitive with coal (Figure 48) Figure 48: Biomass Co-Fire Cost and C02 Emission Profile 6.8 Sensitivity Analysis 'Sensitivity analysis' generally means variations in output following changes in a model's inputs. In evaluating sensitivities within the Colorado model, we considered both single (parametric) and multiple (global) variables. Single variable sensitivity analysis is used to evaluate the response to changes in a single input (such as the cost of natural gas) while holding all other inputs constant. Multiple variable sensitivity analysis is used to evaluate the relationship of multiple inputs and outputs [29]. Analyzing multiple variable sensitivities involves the perturbation of multiple model inputs simultaneously and the evaluation the effects of each input, singly and together, on model outputs. Within the Monte Carlo simulation, input perturbation is determined by a random number generator [60]. The Monte Carlo simulation is a statistical sampling technique used to obtain a probabilistic approximation of the solution for a particular model. The goal of a Monte Carlo simulation is to identify key sources of variability and uncertainty and to 90

PAGE 105

quantify the relative contribution of these sources to the overall variance and range of model results. Researchers and analysts most often use the Monte Carlo simulation to evaluate the expected impact of risk (the probability of an undesirable outcome) and policy changes involved in decision making [61]. 6.8.1 Gas Prices Sensitivity Analysis There are large variations in projecting the natural gas prices. Utilities use various sources of data to compile data and develop projections for fuel prices. In making projections for the price of natural gas, for example, Colorado's Xcel Energy used a blend of data from the New York Mercantile Exchange, EIA, and other sources. Figure 14 compared gas price projections by the EIA and by Xcel. Since Xcel's higher forecast is generally considered more representatives of actual fuel market prices in the West, it was adopted as the basis for the Colorado model. For sensitivity analysis with regard to the effect of gas prices on variations in fuel consumption and new capacity additions, we inflated the Xcel natural gas price by 50% (Semi-High gas price scenario) and then by 100%.(high gas price scenario) as input into the model. 12IDD 11I111l IIIIIl 0 Fuel Coat Senalllvtty Qoa....,, by Fu.l TYPe --1-.. 0:111 0 f*1>. a. 0 01 0-0 in1U1a Figure 49: Natural Gas Sensitivity Generation by Fuel As natural gas prices increase and coal-fired, hydro units and imports reach maximum levels, renewables are brought in to fill the gap created by displaced natural gas generation (Figure 49). 91

PAGE 106

"' "' 10 -i fiiU2 --""'""' oo::E:..}'cll ./ldl_cr ai'OI_CC a: ocr a::_czs 0 RI_Fb::lqt 0 Sdlr_......._ D W1d Figure 50: Fuel Cost Sensitivity New Capacity Additions As was noted before, the renewable electricity generation in the model is constrained at 33% of total generation in 2035; forces the system to begin investing in more renewable generation resources in earlier years for higher renewable generation after all coal-based import limits are exhausted. In the high gas scenario, for example, all imports limits are reached in all periods and the system begins investing in more renewables as early as 2011 to compensate for reduced investment in advanced CTs (Figure 50). In the semi-high gas price scenario, the total discounted system costs increase by 10%, and in the high gas price scenario, they increase by 20% (Figure 51). F .... OaatSendlvil 111111 1111111 {;OSt tv ---., Gas Sensitivity Increase Increase """" -r-----t: t--t---High Gas 10,093 20.2% t-1---Semi High Gas 5,538 10.0% """ t-1--t---1CID) t--t-----r-----0 1IW .. """ ....,_,.""" Figure 51: Fuel Cost Sensitivity Total System Cost Comparison 92

PAGE 107

In the high gas cost scenario, C02 emissions drop over the planning horizon by as much as 3.5%, while in the semi-high gas cost scenario, they drop by less than 1%. Higher gas prices have minimal impact on C02 emissions because most natural gas-fired units (CTs) operate with low capacity factor throughout the year, coming online to meet system peak demand only. In addition, carbon intensity is much lower in high efficiency gas units than in fossil-fueled generating units and carry a much smaller carbon footprint. (Figure 52) C02 % lltedu-lnlm A..-Oaa llenollivlty --(ld) IIAU Cool (Ill) High G11 -18 302 -3.5% -523 Semt-Htgh G11 -4,788 -0.9% -1,181 Figure 52: Fuel Cost Sensitivity C02 Emissions Profile 6.8.2 Load Forecast Sensitivity Analysis Energy planning models require forecasts of energy demand for all the years in the planning horizon. To alleviate the uncertainties necessarily involved in projecting energy demand over the long term, we devised a low and a high energy needs projection for the sensitivity analysis. The low energy projection resulted in an average annual growth rate of 1.5%, while the high projection forecast an average annual growth rate of 2.9%. (Figure 53) In the event, Colorado's economy and population have grown faster than expected and energy consumption and end-user demand have grown along with it. With renewables capped at 33% of total generation in 2035, the state needs to build more fossil-fueled power plants to meet demand increases. In a slow-growing economy, or where less energy is used due to conservation or high electricity prices, demand for energy demand decreases and the state will need to build fewer power plants, including renewables, to meet end-user needs. (Figure 49) 93

PAGE 108

,....., 131111 111DD IIIII) 31111 --2 r ft Figure 53: Load Forecast Sensitivity Generation by Fuel Type The power generation system's sensitivity to load will be a major factor determining future costs and benefits. In the high load forecast, the system incurs more than 14% in discounted total system costs. The opposite is true where the load forecast is lower than BAU, and the system's discounted total costs are reduced by more than 15% (Figure 54). llor-..ay.tlnolyolo ---%1ncr .. se fromBAU 14.2% -15.4% Figure 54: Load Forecast Sensitivity System Cost Comparison With regard to its C02 emissions profile, Colorado follows the expected pattern. A higher load forecast increases the C02 emissions profile by more than 4%, while a low load forecast decreases the system's C02 emissions profile by more than 4%. (Figure 55) 94

PAGE 109

Eiwfgy!Jitnwa-
PAGE 110

Table 37: Monte Carlo Simulation Inputs nput Rllntl' M-llnputa Unit. Default Low High Dncrlplion NabJ,.I Gal Cost IMIPJ 0 -1.2 24 Natur11l Gae Price New Genernon T IChnology Oiacount Rote 0.075 0075 02 T ochnology-Spoc.fic hullle note R.,_ble Growth Rote Gruwth note of .-..-bios Wind % arowth 20 0 30 Solllr-PV % arowth 30 0 30 SolllrThltrTnol % arowth 10 0 20 Biomul % gruwth 15 0 20 C.Doc:itvl:rrit Pook .. pocity lor technology Wind G0835 0 11 Solllr PV G-0.03 0 0.5 PV-RooftDD ot 0.250 SolllrThltrTnol G-0.025 0 1 Goothormol Gig8wotta 0 0 0.07 lliomllu Gig8wotta 0 0 0.44 The fuels tracked in this simulation included coal, natural gas, and oiL The types of renewable technologies included biomass, geothermal, solar, and wind power. Other inputs included: forecasts of natural gas cost, restraints on growth bounds of renewable technologies; and technology-specific hurdle rates_ Figure 56 shows the mathematical representation of an electrical system cost optimization, as objective function, formulated in the Monte Carlo simulation [631Simulation Model Xjv Capcity Cjv Capacity Cost Yjv Electric output Fjv Fuel cost Q Energy demand min. Total Cost= (Fjv Yjv + VAROM) + Xjv + FXDOM) s.t., O=O Y>=Q Figure 56: Mathematical Representation of System Objective Function 6.8.4 Uncertainty of Model's Input/Output Uncertainty in a forecast arises from the combined uncertainties of all its assumptions and the way these are weighted in the formulas used in the modeL An assumption might have a high degree of uncertainty, for instance, yet have little effect on the forecast because it is not weighted heavily in the model formulas. On the other hand, an assumption with a relatively low degree of uncertainty might influence a forecast greatly_ Sensitivity refers to the amount of uncertainty in a forecast that is caused by the uncertainty of an assumption as well as by the model itself [601-Figure 57 shows the model's sensitivities measured against total costs by rank correlation coefficients. Positive coefficients indicate that an increase in the 96

PAGE 111

assumption is associated with an increase in the forecast. Negative coefficients imply the reverse. In this case, natural gas prices have a positive coefficient, or very strong correlation, with total costs. By contrast, wind growth rate has a negative coefficient, meaning that increased amounts of wind generation capacity in the system will result in less total cost. 07 011 -00 03 QICoet -02 .()5 .. I I I I I I 05 Figure 57: Sensitivity Chart of Target ForecastTotal System Cost (2035) 6.9 Elastic Demand Elastic demand modeling allows us to obtain partial equilibrium between supply and demand in a region's energy system. Partial equilibrium represents the situation in which consumer and producer surplus are maximized. The total surplus of an economy is the sum of the suppliers' and the consumers' surpluses. In MARKAL, the term 'supplier' means any economic agent that produces (and sells) one or more commodities (e.g., an emission permit, an energy service). A 'consumer' is a buyer of one or more commodities. Some agents may be both suppliers and consumers, but not for the same commodity. The Reference Energy System for a given commodity defines a set of suppliers and a set of consumers. Generally, the set of suppliers of a commodity are represented by their inverse production function, where the marginal production cost of the commodity (vertical axis) is plotted as a function of the quantity supplied (horizontal axis). In MARKAL, the supply curve of a commodity is not explicitly expressed as a function of factor inputs (such as aggregate capital, labor, and energy in typical production functions used in the economic literature), It is rather represented as the inverse step-wise constant and increasing supply function of each factor. This is because in Linear Programming, the shadow price of a constraint remains constant over a certain interval and then changes abruptly, giving rise to a stepwise constant functional shape. Each horizontal step of the inverse supply function indicates that the commodity is produced by a certain technology or set of technologies in a strictly linear fashion. As 97

PAGE 112

the quantity produced increases, one or more resources in the mix (either a technological potential or some resource's availability) is exhausted. At this point, the system must start using a different (generally more expensive) technology or set of technologies and will produce additional units of the commodity only at higher unit cost. Each change in production mix therefore generates one step of the staircase production function with a value-inverse demand function, which is a step-wise constant, decreasing function of the quantity demanded. As shown in Figure 58 the supply-demand equilibrium is at the intersection of the two functions, and corresponds to an equilibrium quantity Q(e) and an equilibrium price P(e), which means that at this price, suppliers are willing to supply the quantity Q and consumers are willing to buy exactly the same quantity Q. 3 'C IlL Producer surplus s Equllrbnum pornt Quantity Figure 58: Price/Demand Trade-Off Curve The concept of total surplus maximization extends the cost minimization approach upon which bottom-up energy system models are based. These types of models have fixed energy service demands, and are therefore content to minimize the cost of supplying these demands. In contrast,-the MARKAL demands for energy services are themselves elastic to their own prices, allowing the model to compute the supply demand equilibrium. Each energy service within the MARKAL model has several attributes that describe (a) the amounts of service to be satisfied at each time period, (b) the seasonal/time of-day nature of these electricity requirements, and (c) the price-elasticity of the demand and the allowed interval of demand variation. In policy runs, the mix of inputs required to produce one unit of a sector's output is allowed to vary according to defined elasticities of substitution. 98

PAGE 113

MARKAL is a technology-explicit, partial equilibrium model that assumes price elastic demands and competitive markets with perfect foresight (resulting in marginal-value pricing). In MARKAL, the surplus function is derived from the demand and supply functions that link prices and quantities for different economic agents. The equivalence programming of the MARKAL-ED (elastic demand) is based on the following considerations:62 Aggregate demand curve. On the aggregate demand curve, the price corresponding to a given quantity represents the willingness to pay to get one more unit of the product. It reflects the value of the product to the consumer given the available quantity. The supply curve. The supply curve represents the long-run marginal production cost of a firm with the total cost of producing the output being minimized. It gives the minimum price at which suppliers are willing to supply the quantity. Maximum net benefit. The maximum net benefit is obtained when the marginal unit is just beneficial, that is, when the marginal production cost and the price (the value for the consumer) are equal. In MARKAL-ED the surplus function is linearized to obtain a linear formulation of the non-linear objective function for the LP (Linear Programming) solver. 6.9.1 Elastic Demand Fuel Consumptions Figure 59 compares total system fuel consumptions under various scenarios for both standard and elastic demand conditions. Under the standard condition, most of the evaluated scenarios show the same level of fuel consumption except for the high load forecast and Aggressive DSM/EE scenarios. The high load forecast establishes the upper boundary and the Aggressive EE scenario the lower boundary of total system fuel consumption, while the DSM and low-load forecast scenarios fall in between but have lower consumption lower than in the BAU scenario. Under the elastic demand condition, all but the high load and Aggressive EE scenarios fall under the BAU level of fuel consumption. This exercise shows that under the elastic demand condition, the system responds to long-run marginal production costs and adjusts system demand to maximize producer/consumer surpluses, which results in an optimized level of fuel consumptions and lower emissions and costs. 62 MARKAL users' manual. 99

PAGE 114

250 200 Standard vs Bastlc Darrand Tctal System Fuel ConsU11Jtion (PJ) Elastic IJernn:l 10".::02 --20"ACC2 1990C02 -BAU DSM EE -Hit! Gas Serri_Hg, Gas S20Tax $50Tax $60Tax $70Tax Hg, Forec:ast LoN Foret:aSI Figure 59: Total System Fuel Consumption under Standard and Elastic Demand Table 38 shows the elastic demand adjustments to areas under the consumer/producer curves of modeled scenarios to arrive at total surplus maximization and cost minimization. Table 38: Elastic Demand Consumer/Producers Surpluses Elastic Demand Surpluses ll.iOnaumariProducer Surplua 2005M$ Scanarloa BAU 0 Carbon Polley 1 0% below 2005 by 2020 -4.539 20% below 2005 by 2020 -6.370 1990 level by 2035 -5.882 DSMIEE Prgrama DSM (300 GWhiYr) 2.524 EE (1% per year) 8.840 Fuel Coat Sanaltlvlty High Gas -8.856 Semi-High Gas -5.119 Forac:ast Sanaltlvlty High Energy Forecast -6.888 low Energy Forecast 7.681 100

PAGE 115

Figure 60 shows the C02 emissions profile for 13 scenarios compared to BAU for both standard and elastic demand. In high load scenarios show the highest C02 emissions (Figure 60). The lowest C02 emission profile was the $70/t C02 tax scenario, followed by the $60/t, 20% cap, the1990-level cap, and 10% cap on C02 emissions. (It should be noted that emission profiles under elastic demand conditions are slightly lower and more uniformly distributed, since fuel consumptions were optimized to account for demand elasticity in response to long-run marginal costs.) It should also be noted that the sharp decline in C02 emissions under the C02 tax of $60/t and $70/t was due to the accompanying sharp decline in use of existing coal fired power plants, which, with no use limits, were replaced with higher penetration of renewables -in particular wind technology. As noted before, wind use for all scenarios (except the C02 tax scenario) was constrained at 33% of total generation Standard vs Elastic Demmd C02 Enissions Profile foob Wlrd l...lmls 55, aD 50,aD 45,aD 40,aD 35,aD li! 30,aD 25,aD 20,aD 15,aD 10,aD 5,aD -:----__ ---....._ 0 Elastic Clerra"d -1 OO/cC02 :ZOO/cC02 1990CXl2 -BAU --DSM -EE -HghGas -Serri_Hgh Gas $20Tax $50Tax $60Tax $70Tax -Hgh Forecast l...oNForecast Figure 60: C02 Emissions Profile under Standard and Elastic Demand 101

PAGE 116

6.10 Scenario Comparison and Conclusion For this study, five main types of scenario were developed, and analyzed: Reference Scenario (Business-as-Usual) Advanced Emerging Technology Scenario DSM/Energy Efficiency Scenario, and Regulatory Policy Scenarios, and Sensitivity Scenarios The study focused on Colorado's electric power system and supply-side energy system incorporating Renewable Portfolio Standards, Demand-Side Management strategies, and Energy Efficiency measures. The work aims to demonstrate the current status of the state's power sector and to quantify pathways for achieving sustainable energy production in the future. Within the five main types of scenario, a total of 17 scenarios were modeled and analyzed (including sensitivity scenarios), covering the period from 2005 to 2035. The Reference Scenario (Business-as-Usual) represented the most probable development of the power system under present known conditions. The other scenarios served mainly to show possible ways toward the sustainable development of a power system that incorporates clean energy technologies and mitigate GHG emissions. In all scenarios, attempts to mitigate C02 emissions impose considerable costs except energy efficiency and DSM measures (Figure 61 ). While energy efficiency and DSM measures represent the most promising pathways for reducing discounted total system costs and at the same time reducing C02 emissions. However, using these methods, projections of C02 levels in the future will never meet any of C02 cap requirements. With DSM and energy efficiency measures, total net reductions in C02 by 2035 will range between 0.3-7percent. The system cost reduction is the most promising between 1-19 percent over the 30 years planning horizon. Although they cost more than BAU, both the carbon cap policy scenario and carbon tax policy scenario modeled were able to reduce C02 emissions to expected levels by 2020 or 2035. In the two scenarios analyzed, one reduces C02 emissions by 10% and the other by 20% by 2020. The net impact of the 10% reduction s a 19 percent reduction in C02 at 1 0 percent higher cost, while in the 20% policy scenario, the net impact is a 25.5 percent reduction in C02 emissions at 15 percent higher cost. Analyses of carbon tax policy showed that, when renewable resources are constrained, where wind power, for example, s bounded at 11 GW and Solar Thermal at 1 GW by 2035, carbon taxes at various level did not respond to taxes. We examined four carbon tax levels ($20, $30, $40, and $50 per ton) and found the system minimally reactive to taxes. According to the model, huge tax revenues were 102

PAGE 117

generated with virtually no impact on carbon-intensive electricity generation. By contrast, when bounds on renewable were relaxed and carbon taxes slightly raised, the system quickly reacted to slow generation from carbon-intensive generating facilities. The system showed sensitivity to a carbon tax of $58 per ton, when coal based generation began to be phased out in favor of adding renewable (wind) capacity. Scenario Colaparison Chart Costs & C02 Errissions fromBAU C02 Cap -60 . 0'/o -40 0'/o -310'/o -20.0'/o DSM/EE CXl2 Tax* CEM'EE CXl2 Cap CXl2 Tax x Gas Sensiti\Aty Faecast Sensiti\Aty Figure 61: Scenario Comparison Chart-Costs & C02 Emission Differential from BAU 103

PAGE 118

Sensitivity analyses had both positive and negative effects on the system. Gas price sensitivity reduced the C02 emission profile but increased net system costs. Load forecast sensitivity analysis showed a direct relation between C02 emissions, cost, and energy demand, with low forecasted demand leading to reducedC02 emissions and system costs and high forecasted demand leading to increases in both C02 emissions and costs. This study provides a foundation for policymakers to assess the implications of strategies designed to mitigate carbon dioxide emitted by Colorado's power sector and to identify policies that will provide sustainable energy at a reasonable cost in the future. It uses an optimization model to evaluate different energy production mixes and the costs and benefits of different policy options. The study makes a substantial contribution to the development of state power system geared not only to meet future energy needs and keep down costs but also to make the best possible use of policies and clean energy technologies to mitigate the emission of pollutants and greenhouse gases. 6.11 Future Research Work One of the most important criteria of using MARKAL model for this study is that it's expandable to other sectors of economy. One logical addition to this model is to develop and disaggregate the sectoral demand into more refined demand devices such as space cooling, space heating, or office buildings or appliances. Another most import remaining work is to add the transportation sector followed by mining and oil and gas industry to quantify their level of upstream greenhouse gas emissions. Together, the model can provide a statewide energy system model and can be used as policy research toll to assess climate change and greenhouse gas initiatives. 104

PAGE 119

APPENDIX A: MODEL INPUT FOR BASE-YEAR (2005) POWER PLANTS CAPACITY AND HEAT RATE List of Colorado Installed Capacity-Base-Year (2005) Cepilclty Heat Rata Technology Type PlentNeme Owner (MW) Type Fuel Type (mmBTu/MWh) FouiiFuel Cameo_ I PSCo 24 ST-Coal Brt 12.764 Cameo_2 PSCo 54 ST-Coal Brt 10.994 Nuda_1-4 TSGT 100 ST-Coal Brt 11.827 Hayden_ I PSCo 205 ST-Coal Brt 10.357 Hayden_2 PSCo 300 ST-Coal Brt 12.246 Cherokee_! PSCo 115 ST-Coal Bit 10.807 Cherokee_2 PSCo 120 ST-Coal Bit 10.470 Cherokee_3 PSCo 165 ST-Coal Bit 10.145 Cherokee_4 PSCo 388 ST-Coal Bit 9.532 Valmont_5 PSCo 199 ST-Coal Bit 9.391 WN Clark_ I WPE 19 ST-Coal Bit 13.056 WN Clark_2 WPE 24 ST-Coal Bit 13.056 Trigen Colorado IPP 20 ST-Coal Bit 10.864 1733 10.82 Martin_Orake_5 csu 47 ST-Coal Sub_Brt 11.678 Martin_Drake_6 csu 79 ST-Coal Sub_Brt 11.182 Martin_Drake_7 csu 133 ST-Coal Sub_Brt 10.427 Nixon_ I csu 208 ST-Coal Sub_Brt 10.492 Pawnee PSCo 505 ST-Coal Sub_Brt 10.430 Craig_ I TSGT 428 ST-Coal Sub_Brt 10.307 Craig_2 TSGT 428 ST-Coal Sub_Brt 10.423 Craig_3 TSGT 408 ST-Coal Sub_Brt 10.155 Rawhide PRPA 270 ST-Coal Sub_Brt 10.415 Arapahoe_3 PSCo 47 ST-Coal Sub_Brt 11.808 Arapahoe_4 PSCo 121 ST-Coal Sub_Brt 11.251 Comanche_ I PSCo 366 ST-Coal Sub_Brt 10.453 Comanche_2 PSCo 370 ST-Coal Sub_Brt 10.470 3410 10.47 Zuni_ I PSCo 39 ST-NGA NGA 13.387 Zuni_2 PSCo 68 ST-NGA NGA 13.387 107 13.39 Ft_St_ Vrain_CC_1 PSCo 297 cc NGA 7.591 Ft_St_ Vrain_CC_2 PSCo 128 cc NGA 7.662 Ft_St_ Vrain_CC_3 PSCo 131 cc NGA 7.558 Ft_St_ Vrain_cc_ 4 PSCo 135 cc NGA 7.556 Front_Range_1a IPP 132 cc NGA 7.274 Front_Range_1 b IPP 133 cc NGA 7.274 Front_Range_tc IPP 196 cc NGA 7.274 Rocky_Mtn_EC_1a IPP 143 cc NGA 7.274 Rocky_Mtn_EC_1b IPP 143 cc NGA 7.274 Rocky_Mtn_EC_1c IPP 322 cc NGA 7.274 1780 7.398 105

PAGE 120

APPENDIX A: (CONT.) List of Colorado Installed Capacity-Base-Year (2005) ConL, I.Oapilcny t1HI HIIW Technology Type Plant Name Owner (MW) Type Fuel Type (mmBTu/MWh) Manchief_1 IPP 132 CT NGA 10.953 Manchief_2 IPP 132 CT NGA 10.953 Arapahoe_5 IPP 39 CT NGA 9.960 Arapahoe_& IPP 39 CT NGA 9.960 Arapahoe_7 IPP 45 CT NGA 9.960 Blue_Spruc:e_1 IPP 138 CT NGA 10.587 Blue_Spruc:e_2 IPP 138 CT NGA 10.587 CPP_Brush IPP 68 CT NGA 8.650 Brush_4D IPP 130 CT NGA 9.960 Limon_1 TSGT 67 CT NGA 12.269 Limon_2 TSGT 67 CT NGA 12.269 Frank Knutson_1 TSGT 67 CT NGA 12.797 Frank Knutson_2 TSGT 67 CT NGA 12.797 Plains_End_1 IPP 113 CT NGA 9.580 Founlain_Valley_GT _1-0 IPP 236 cc NGA 10.685 FI_Lupton_ GT _1 PSCo 50 CT NGA 10.316 FI_Luplon_GT_2 PSCo 50 CT NGA 15.500 Martin_Drake_1 csu 5 CT NGA 12.560 Martin_Drake_3 csu 5 CT NGA 12.150 Martin_Drake_ 4 csu 11 CT NGA 12.100 Rawhida_A1 PRPA 60 CT NGA 15.049 Rawhida_B2 PRPA 60 CT NGA 15.049 Rawhida_C3 PRPA 60 CT NGA 15.049 Rawhida_E4 PRPA 66 CT NGA 15.049 Rifte_1 IPP 68 cc NGA 11.860 Nixon_GT_1 csu 30 CT NGA 17.512 Nixon_GT_2 csu 30 CT NGA 17.512 Thenno_lnd IPP 129 CT NGA 9.400 Thenno_Carb IPP 150 CT NGA 9.400 Thenno _Greeley IPP 32 CT NGA 9.700 Thenno_UNC IPP 69 CT NGA 8.650 CU_Cogen IPP 31 CT NGA 8.428 Valmon\_7 IPP 41 CT NGA 10.668 Valmont_8 IPP 41 CT NGA 10.668 2481 11.215 Alamosa_GT_1 PSCo 12 CT NGA 14.450 Alamosa_GT_2 PSCo 14 CT NGA 15.467 Birdsall_1 csu 16 CT NGA 14.500 Birdsall_2 csu 17 CT NGA 14.500 Birdsall_3 csu 23 CT NGA 14.500 Brighlon_1 PSCo 50 CT NGA 16.318 Brighton_2 PSCo 50 CT NGA 16.318 Fruita_GT PSCo 15 CT NGA 14.992 Pueb1_6 WPE 28 CT NGA 14.500 Charokee_Dies PSCo 6 CT DSL 14.500 Bullock PSCo 12 IC DSL 14.500 Dalla_IC TSGT 5 IC DSL 14.500 WPE_Diesei_IC WPE 18 IC DSL 14.500 Rocky _Ford TSGT 10 IC DSL 15.806 278 106

PAGE 121

APPENDIX A: (CONT.) List of Colorado Installed Capacity-Base-Year (2005) Cont, 1:;8PIICilY Technology Type P .. ntNeme (MW) Agregeta Foull Plents Coai_ST_Brt 1733 Coai_ST _Sub_Brt 3410 Totlll Colli 5143 NGA_ST 107 NGA_CC 1760 NGA_CT 2466 TotiiiNGA 4333 NGAIDSL_CT/IC Totlll 278 Totiii(Foull Fuel) 9,752 Hydro Foothills Hydro Plant 3.1 Stron1ia Springs Hydro Plant 1 Dillon Hydro Plant 1.8 Williams For!( Hydro Plant 3 North For!( Hydro Plant 5.5 Boulder Canyon Hydro 5 Georgetown 0.8 Georgetown 0.8 Palisade 1.6 Palisade 1.6 Salida 0.8 Salida 0.6 Shoshone 7.5 Shoshone 7.5 Manitou Springs 2.5 Manitou Springs 2.5 Ruxton Par!( 1 Vallecito Hydroelectric 0.4 Vallecito Hydroelectric 2.5 Vallecito Hydroelectric 2.5 Redlands Water & Power 0.6 Hydro Plant 2.5 Blue Mesa 43.2 Blue Mesa 43.2 Estes 17.2 Estes 17.2 Estes 17.2 Morrow Point 86.6 Morrow Point 86.6 Big Thompson 5.2 107

PAGE 122

APPENDIX A: .. (CONT.) List of Colorado Installed Capacity-Base-Year (2005) Cont., "8PIICRY Technology Type Plant Name (MW) Green Mountain 13 Green Mountain 13 Marys Lake 9.3 Flatiron 43 Flatiron 43 Pole Hill 38.2 Lower Molina 4.8 Upper Molina 8.6 Hillcrest Pump Station 2 Boulder City Lakewood Hydro 3.5 Boulder City Belasso Hydro 3 Taylor Draw Hydro Facility 2 Boulder City Silver Lake Hydro 3.3 Crystal 30 2.2 2.2 4 Ames Hydro 3.7 Tesla 25 McP'1.2 Towaoc 11.4 Ruedi 5 Total Colorado Hydro 842.9 Cabin Creek 162 Cabin Creek 162 Flatiron 8.5 Mount Elbert 115 Mount Elbert 115 Total Pump-Storage 5412.5 Wind Lamar Plant 4 Lamar Plant 1.5 Ponnequin Phase 1 5.2 Ridge Cnsst Wind Partners 7.55 Colorado Gnsen Holdings LLC 162 Ponnequin 9 Ponnequin 15.4 Spring Canyon 60 Total Colorado Wind 284.115 MSW Metro Wastewater Distrid 9.8 Totlll (Non-Foull Fuel) 1,480 Totlll Colorado 11,232 108

PAGE 123

APPENDIX 8: EXISTING POWER PLANTS EMISSION FACTORS Colorado Fossil-Fueled Electric Generation Emissions (2006) CumuloiMI CumuiiiiMI Cu-IMI C-IMI A .. rago 802 A .. rago C02 A .. rago N01 AA..,.go A .. rago TOCII An.,..IHtlt Annu11802 AnnuoiC02 AnnUli NOll Emlaolon Emlulon Emluion ..... 802 C02 NOI OpeniUng Input Emlulono Emllllonl Ernllllonl Rl .. Rl .. -Copoclty WI. HtltRI .. Emllllon Emllllon Emil lion PllntNirno Co.,..ny Houn FuoiT..-em-) (toni) (toni) (toni) Cllllmmlltul (lblmm8tu) CMWI Awrago ,_, Rl'" CktiPJI Rl'" CktiPJI Rl'" CktiPJI CoiiUnlta Clmoo2 PSCc 8 512 .41 llituminoua 4,ot8, 182 2,101 1 420 473 8 731 5 I .Ozt 20S 2 0 357 54 0 .03 10 ,8114 1 4248 284 0 .41144 en.,.,. PSCc 7 705 34 lliNmnoua 8,154,781 2.185 2 834 ,1144. 2 1,439 8 0 531 204.8 0 3S3 liS 0 .07 10,807 0 .722t 2711 0 4801 CII-H2 PSCc 8 428 .23 llituminoua 9,881,1182 2,442 1 1182,304. 9 3,383.5 O S04 20S O 0 .8119 120 0 .07 10 470 0 88S3 270 0 8218 CII-H3 PSCc 7 300 18 llitumnoua 10,758,231 704 0 1,100,488 7 1,820.3 0 131 204 8 0 338 185 0 10 10,145 0 1873 281 0 4325 Ch-H4 PSCo 8 013 12 !litumnoua 27,248,288 1 ,7411. 9 2 788,504 5 4,157 8 0 128 204 7 0 30S 388 0 24 9,532 0 1542 248 0 3884 Hoyden 1 PSCc 8 S22 .11 llituminoua 19,317,348 1,297 8 1,981,1182. 4 4,094.5 0 134 20S 2 0 424 205 0 12 10 3S7 0.17S3 288 0 5531 Hl)ldon 2 PSCc 8 332 07 llituminoua 24,238,730 1,583. 5 2,488 889 9 3,881.3 0 131 20S 2 0.32t 300 0 18 12 248 0.202t 317 o .soee Nucla TSGT 8,125 78 llituminoua 9,325,981 1,401 7 957 53S 7 1,921.7 0.301 20S 3 0.412 100 0 08 11,827 0.4478 308 0 .8140 Valmont PSCo 8,858 .85 llituminoua 15,810,832 878. 7 1 822, 190 9 2,514.1 0 .111 20S 2 0.318 188 0 12 9 391 0.1315 243 0 3782 Bll Cool Clpaclty 14,341 13,115,213 24,044 1141 0 33 10 ,511 0.3001 272 0.4t0t Atapohol3 PSCo 7,528 34 Subbiluminouo 3,788,821 840. 1 388,878 8 1,448.7 0 .4119 20S 2 0 788 47 0 .01 11,801 0 742 30S 1 142 "'-hol4 PSCc 7 125 94 Subbiluminoua 7,581,824 1,471 5 778,898 0 889. 7 0 388 20S 2 0 234 121 0 04 11,251 0 .5411 291 0 .332 Comoncho 1 PSCc 8 017 30 Subbilumnoua 25 853,048 0,813 0 2 ,828, 704 3 4 0S7 8 0 518 204 9 0 .318 388 0 .11 10,453 0 .879 270 0 417 CCmoncho 2 PSCo 7 070 12 Subbilumnoua 25 533,282 U2t. 8 2 ,814,0111. 8 3 ,1113. 5 0 53S 204 8 0 307 370 0 .11 10 ,470 0 701 270 0 404 Craig 1 TSG 8 ,8112. 33 Subbiluminoua 41,82t,888 1,057 4 4 271,224 2 5 ,823. 5 O .OS1 20S 2 0.280 428 0 13 10,307 0.0111 288 0 .383 Craig2 TSGT 8 372 .43 Subbiluminoua 39,421 782 1,001 3 4 044 ,873. 2 5 ,415. 5 O .OS1 20S 2 0 275 428 0 13 10 ,U3 0 017 289 0 381 Craig3 TSGT 7 ,118. 15 Subbiluminoua 31,8119,0311 2,010 1 3 272 837 4 8 487 7 0 128 20S 2 0 408 408 0 .12 10 155 0181 282 0 .511 p-PSCo 7 188 38 Subl>ituminoua 34 ,507,1811 11,248 1 3 532 021 5 3,888 1 0 .852 204 7 0 213 505 0 15 10 ,430 0.857 289 0 .278 Merlin Dnlko 5 csu 7.583 .25 Subbiluminoua 3 ,858, 884 1,340 3 374 834 4 788 2 0 733 20S O 0 420 47 0 .01 11,878 1 .078 302 0 .818 Merlin Dnlko8 csu 8.380 .25 Subl>ituminoua 7,411 277 2 ,1130. 0 759,839 0 1,408 3 0 .7111 20S 1 0 .380 79 0 02 11, 182 1 114 289 0 53S Merlin Dnlko 7 csu 8, 7511. 50 Subl>ituminoua 12,5117, 145 4 894 1 1 291 527 2 2.787 9 0 777 20S 1 0 .439 133 0 04 10 427 1 021 2811 0 .577 R....,ido PRPA 7,515 73 Subbituminoua 22,783,530 875 5 2 337,590 1 3,729 1 0 077 20S 2 0327 270 0 .01 10,415 0 .101 2811 0 .42t RayO Nixon csu 7,837 75 Subbiluminoua 18,844,908 3,750 8 1 707,111 4 2,188.1 0 451 205 1 0 281 208 0 .01 10 482 0 598 271 0 344 Bub-Bit. Cool Clpaclty 44,MI 21,000,124 42,524 3410 0.17 10,473 0.4811 270 0.4011 Cool Clpaclty 8 Pollution 51,310 41,115,417 11,511 5051 Goo Unlta Combined Cycln Fort St. PSCo 8,223 15 NGA 10,745,839 3 2 838,809. 2 161.2 0 .001 118 .11 0 030 204 0 12 7591 0 001 114 0 029 FottSI. Vrain 3 PSCc 8 194 75 NGA 10,888 788 3 4 853, 102 7 148.9 0 001 118 9 0 027 259 0 15 78112 0 001 115 0 028 Fort 51. Vrain 4 PSCc 8.188 17 NGA 10,1138, 184 3 2 849, 922 3 78 3 0 001 1189 0 014 228 0 13 7558 0 001 113 0 014 F"'"IRongo1 FrinWPSCC 5 ,771.81 NGA 7,441 ,880 2 3 442 ,2588 114 7 0 001 118 9 0 031 231 0 13 7274 0 001 109 0 028 F"'"t Rongo2 FrinwPSCo 8 205 25 NGA 7,971 ,870 2 4 473 755 7 143 8 0 001 118 8 0 038 231 0 13 7274 0 001 109 0 033 Rocl
PAGE 124

APPENDIX B: (Cont.) cu-c ..... -Cumu..atw C-A-802 A-COZ A-NOll AAA-TCIIIII Annu.l-Ann,...eoz AnnuoiCOZ Ann,... NOll -Emluloft Emlulon ..... eoz coz NOll 0 ......... lnpul Emlulofto Emlulofto Emlulofto -CopKIIJ Wt. ---Emlulon Emlulon PloniNMio Componr Houro F .... Trpo ........ ltonol """"' """"' llblnlmBiul llblnlmBiul llbhnmBiul IIIWI A-,_, -llltiPJI -llltiPJI -llltiPJI Com-Tu-.v.p.hoe CT5 Fl1ng/PSCo 3,153.10 NGA 1.042,518 0.4 81,155.4 12.0 0.001 118.8 0.023 38 0.02 8880 0.001 148 0.028 .v.p.hoe CT8 Fl1ng/PSCo 3,132.08 NGA 1,072,008 0.4 83,707.2 11.8 0.001 118.8 0.022 31 0.02 8880 0.001 148 0.028 Ar.po'-CTI Fl1ng/PSCo 3,132.08 NGA 1,072,008 0.4 83.107.2 11.8 0.001 118.8 0.022 45 0.03 8880 0.001 148 0.028 BlueSprucoCT1 Fl1ng/PSCo 1,280.41 NGA 1,703,715 1.2 101,488.3 21.8 0.001 118.1 0.034 138 0.08 105117 0.002 151 0.045 BlueSprucoCT2 Fl1ng/PSCo 873.88 NGA 1.2011.088 2.8 72.401.8 222 0.005 120.1 0.037 138 0.08 105117 0.008 1eo 0.048 -3GT2 Fl1ng/PSCo 18.31 NGA 7.218 0.0 428.8 0.4 0.000 118.8 0.111 88 0.04 8850 0.000 130 0.121 -4GT4 Fl1ng/PSCo 302.88 NGA 138,805 0.0 8,308.7 0.0 0.000 118.8 0.000 85 0.04 8880 0.000 148 0.000 -4GT5 Fl1ng/PSCo 270.88 NGA 131,188 0.0 7,788.8 0.0 0.000 118.8 0.000 85 0.04 8880 0.000 148 0.000 Founlllln Volo)'CT1 Fl1ng/PSCo 2,211.75 NGA 818,127 0.1 38,783.0 31.8 0.000 118.8 0.103 38 0.02 10885 0.000 1eo 0.138 Founloln Voller CT2 Fl1ng/PSCo 1,473.50 NGA 451.230 0.1 28,815.1 21.3 0.000 118.8 0.084 38 0.02 10885 0.001 1eo 0.127 Founloln Voller CT3 FrlngiPSCO 2,128.50 NGA 818,521 0.1 38,838.0 30.5 0.000 118.8 0.088 38 0.02 10885 0.000 1eo 0.133 Founloln Voller CT4 Fl1ngiPSCo 1,380.00 NGA 387,884 0.1 23,851.8 18.3 0.001 118.8 0.087 38 0.02 10885 0.001 1eo 0.131 Founlllln Vole)' CT5 FrlngiPSCO 1,808.50 NGA 511,885 0.1 30,408.7 28.7 0.000 118.8 0.104 38 0.02 10885 0.001 1eo 0.140 Founloln Vole)' CTB Fl1ng/PSCo 1,30125 NGA 382.015 0.1 22.103.8 18.5 0.001 118.8 0.087 38 0.02 10885 0.001 1eo 0.130 F,..k KnuiiOn BR1 TSG 228.25 NGA 151.013 0.0 8,875.7 2.3 0.000 118.8 0.030 87 0.04 12117 0.000 182 0.048 Fronk KnuiiOn BR2 TSG 11125 NGA 130,815 0.0 7,n4.2 1.8 0.000 118.8 0.024 87 0.04 12117 0.000 182 0.038 Urnon L 1 TSG 20225 NGA 141,101 0.0 8,385.2 1.8 0.000 118.8 0.021 87 0.04 12288 0.000 184 0.042 Urnon L2 TSG 181.00 NGA 138,083 0.0 8,082.1 1.8 0.000 118.8 0.028 87 0.04 12288 0.000 184 0.043 MonchloiCT1 Fl1ng/PSCo 382.00 NGA 438,137 0.1 28,037.8 13.1 0.000 118.8 0.080 132 0.08 101153 0.001 184 0.083 MonchloiCT2 FrlngiPSCO 4411.00 NGA 471,852 0.1 28,081.4 14.1 0.000 118.0 0.080 132 0.08 10853 0.001 184 0.082 RowllldeCTA PRPA 58.88 NGA 40,018 0.0 2.381.7 0.8 0.000 118.8 0.040 eo 0.03 15048 0.000 225 0.018 RowllldeCTB PRPA 155.12 NGA 108,730 0.0 8,342.8 1.7 0.000 118.8 0.032 eo 0.03 15048 0.000 225 0.080 RowllldeCTC PRPA 108.81 NGA 18,782 0.0 4,583.2 12 0.000 1188 0.031 eo 0.03 15048 0.000 225 0.058 R-CTD PRPA 52.32 NGA 35,588 0.0 2,115.2 0.8 0.000 118.8 0.034 88 0.04 15048 0.000 225 0.084 RorDNtmnCT2 csu 8825 NGA 25,444 0.0 1,512.4 0.8 0.000 118.8 0.047 30 0.02 11512 0.000 282 0.104 RorDN-CT3 csu 54.50 NGA 14,741 0.0 818.3 0.3 0.000 118.8 0.041 30 0.02 11512 0.000 282 0.080 VolmoniCTI Fl1ng/PSCo 8822 NGA 24.807 00 1,410.5 1.4 0.000 118.8 0.112 41 0.02 10888 0.000 1eo 0.151 V-CTB Fl1ng/PSCo 103.88 NGA 28,411 0.0 1,1411.8 1.4 0.000 118.8 0.085 41 0.02 10888 0.000 1eo 0.128 I 111,1H ZTI 1711 a.118 0.0711 TOM 11,331 11,111,m 17,131 -..ronno U,M3 41,111,11.'1 11,11.'1 110

PAGE 125

APPENDIX C: MODEL INPUT ASSUMPTIONS AND RESOURCE BOUNDS Technology Investment Cost Technology ($/kw) 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 Biomass_cc 3313.0 3313.0 3302.0 3302.0 3280.0 3302.0 3258.0 3231.0 3255.0 3255.0 PC_Coai_CCS 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 3769.0 Com3_Xcel 2020.0 2020.0 2020.0 2020.0 2020.0 2020.0 2020.0 2020.0 2020.0 2020.0 IGCC_Xcel 4008.0 4008.0 4008.0 4008.0 4006.0 4008.0 4008.0 4008.0 4008.0 Geothermal 3641.0 3641.0 3641.0 3641.0 3641.0 3641.0 3641.0 3641.0 3641.0 3641.0 IGCC_CCS 4008.0 4008.0 4008.0 4006.0 4006.0 4008.0 4008.0 4006.0 4006.0 Adv_CT 519.6 519.6 519.6 519.6 519.6 519.6 519.6 519.6 519.6 Adv_CC 627.0 827.0 827.0 827.0 827.0 827.0 827.0 827.0 827.0 cc 885.0 885.0 885.0 885.0 885.0 885.0 885.0 885.0 885.0 885.0 CT 659.0 659.0 659.0 659.0 659.0 659.0 659.0 659.0 659.0 659.0 Gas_IGCC_CCS 1124.0 1124.0 1124.0 1124.0 1124.0 1124.0 1124.0 1124.0 1124.0 Adv_Nuclear 2897.0 2897.0 2897.0 2897.0 2897.0 2897.0 2897.0 2897.0 2897.0 PV_Cantral 3830.0 3793.0 3593.0 2941.0 2941.0 2941.0 2941.0 2941.0 2941.0 2941.0 PV_Roollop 7519.1 8379.5 5239.9 4565.6 3891.2 3891.2 3891.2 3891.2 3891.2 3891.2 Sol.-_ Thermal 2539.0 2466.0 2348.0 2348.0 2106.0 2106.0 2106.0 2106.0 2106.0 2106.0 Wind 1890.0 1690.0 1890.0 1890.0 1690.0 1690.0 1690.0 1690.0 1690.0 1690.0 1890.0 Notes: CC = Combined Cycle CT = Combustion Turbine PC = Pulverized Coal IGCC = ln1egrated Gasification Combined Cycle Com3 = Pulverized coal unrt by Xcel Energy 111

PAGE 126

APPENDIXC: (Cont.) Variable O&M Costs Technology ($/GJ) 2005 2008 2011 2017 2020 2023 2026 2029 2032 2035 New Biomass CC 0 14.94 14.94 15.67 16.82 17.77 18.13 18.71 18.73 18.73 Muni Solid Waste 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 New PC with CCS 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 2.9389 Com3Xcel IGCCXcel Co-Fire 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 Bit Coal Steam 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 Sub Bn Coal Steam 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 0.7714 DSF S1eam 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 DiaseiiC 2.46&4 2.46&4 2.46&4 2.46&4 2.46&4 2.46&4 2.46&4 2.46&4 2.46&4 New Geolhennal 6.3556 6.3558 6.3556 6.3556 6.3558 6.3558 6.3556 6.3556 6.3556 6.3556 Hydro 1.2455 1.2455 1.2455 1.2455 1.2455 1.2455 1.2455 1.2455 1.2455 HydroPS NewCoaiiGCC NewAIJvCT 0.7859 0.7859 0.7859 0.7859 0.7859 0.7859 0.7859 0.7859 0.7859 NewAIJvCC 0.8583 0.8583 0.8583 0.8583 0.8583 0.8583 0.8583 0.8583 0.8583 cc 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 0.1359 NewCC 0.7806 0.7806 0.7806 0.7806 0.7806 0.7806 0.7806 0.7806 0.7806 0.7806 CT 0.029 0.029 0.029 0.029 0.029 0.029 0.029 0.029 0.029 0.029 0.029 NewCT 2.2086 2.2086 2.2086 2.2086 2.2086 2.2086 2.2086 2.2086 2.2086 2.2086 NewGasiGCC 0.8147 0.8147 0.8147 0.8147 Gas Steam 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 0.1458 New AIJv Nuclear 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 0.1667 PV_Central 0 0 0 0 0 0 0 0 0 0 PV_Rooftop 0 0 0 0 0 0 0 0 0 0 Solar_ Thennal 0 0 0 0 0 0 0 0 0 0 Wind 0 0 0 0 0 0 0 0 0 0 0 Notes: $/GJ = 0.0036"$/KWh CC = Combined Cycle CT = Combustion Turbine PC = Pulllerized Coal IGCC = ln1egreted Gasification Combined Cycle Com3 = Pulllerized coal unn by Xcel Energy PS = Pumped Storege 112

PAGE 127

APPENDIXC: (Cont.) Fixed O&M Costs Technology (S/kW/yr) 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 NBiomass CC 50.17 50.17 50.17 50.17 50.17 50.17 50.17 50.17 64.71 86.52 Muni Solid Waste 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 96.4781 NPC with CCS 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 46.2143 Com3Xcel 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 IGCCXcel 17.14 17.14 17.14 17.14 17.14 17.14 17.14 Co-Fire 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 Bit Coal Steam 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6438 15.6436 15.6436 15.6436 15.6438 Sub Bit Coal Steam 15.6436 15.6436 15.6436 15.6436 15.6436 15.6436 15.6438 15.6436 15.6436 15.6436 15.6436 DSFSteam 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 DieseiiC 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0 0 0 0 0 0 0 0 0 0 Hyllltl 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 14.2025 Hyllltl PS 16.7086 16.7088 16.7086 16.7086 16.7086 16.7086 16.7088 16.7086 16.7086 16.7086 16.7088 N-CoaiiGCC 17.14 17.14 17.14 17.14 17.14 17.14 17.14 17.14 17.14 N-AdvCT 8.8671 8.8671 8.8871 8.8671 8.8671 8.8671 8.8671 8.8871 8.8671 N-AdvCC 9.4173 9.4173 9.4173 9.4173 9.4173 9.4173 9.4173 9.4173 9.4173 cc 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 15.7508 N-ee 13.193 13.193 13.193 13.193 13.193 13.193 13.193 13.193 13.193 13.193 CT 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 6.5109 N-CT 4.3137 4.3137 4.3137 4.3137 4.3137 4.3137 4.3137 4.3137 4.3137 4.3137 N-GasiGCC 19.9501 19.9501 19.9501 19.9501 19.9501 19.9501 19.9501 19.9501 19.9501 Gas Steam 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 0.861 N-Adv Nuclear 58 58 58 58 58 58 58 58 58 58 58 PV_Central 8.9648 8.9648 8.9648 8.9648 8.9648 8.9648 8.9648 8.9648 8.9648 8.9648 PV_Rooftop 0 0 0 0 0 0 0 0 0 0 Solar_ Thennal 43.5494 43.5494 43.5494 43.5494 43.5494 43.5494 43.5494 43.5494 43.5494 43.5494 Wind 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 23.2443 Notes: CC = Combined Cyde CT = Combustion Turbine PC = Pulvarized Coal IGCC = Integrated Gasification Combined Cycle Com3 = Pulverized coal unit by Xcel Energy PS = Pumped Storage 113

PAGE 128

APPENDIX C: (Cont.) Technology Heat Rate Heat Rate Heat Rate Technology Fuel (PJ/PJ) (Btu/kWh) Biomass CC Biomass 3.013 10,283 Muni Solid Waste Solid Waste 3.791 12,939 New PC with CCS Sub_Bit Coal 3.324 11,343 Com3Xcel Sub_Bit Coal 2.541 8,672 IGCCXcel Sub_Bit Coal 2.989 10,202 Co-Fire Bit Coal & Biomass 3.111 10,618 Bit Coal Steam Bit Coal 3.111 10,618 Sub Bit Coal Steam Sub_Bit Coal 3.069 10,474 Disstilate Fuel Oil Distillate Fuel Oil 3.784 12,916 DieseiiC Diesel 3.784 12,916 Geothermal Geothermal 3.013 10,283 Hydro HYDRO 3.013 10,283 HydroPS Electricity 1.100 3,754 New Coal IGCC Sub Bit Coal 2.989 10,202 NewAdvCT Natural Gas 2.506 8,553 NewAdvCC Natural Gas 2.133 7,281 Existing Gas CC Natural Gas 2.168 7,399 NewCC Natural Gas 2.187 7,463 Existing Gas CT Natural Gas 3.084 10,525 NewCT Natural Gas 3.065 10,459 New Gas IGCC Natural Gas 2.330 7,952 Existing Gas Steam Natural Gas 3.923 13,390 New Adv Nuclear Nuclear 3.080 10,512 PV_Central Solar 3.013 10,283 PV_Rooftop Solar 3.013 10,283 Solar_ Thermal Solar 3.013 10,283 Wind Wind 3.013 10,283 Notes: CC = Combined Cycle CT = Combustion Turbine IC =Internal Combustion PC = Pulverized Coal IGCC =Integrated Gasification Combined Cycle Com3 = Pulverized coal unit by Xcel Energy PS = Pumped Storage 114

PAGE 129

APPENDIX C: (Cont) Fuel Cost Fuel Type (SIGJ) 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 Sub_Bil Coal 1.00 1.05 1.10 1.13 1.15 1.18 1.21 1.25 1.30 1.35 140 Disstilate Fuel Oil 5.24 7.42 7.26 8.55 8.47 6.58 8.74 6.88 7.10 7.11 7.22 Diesel 8.68 9.70 9.83 9.07 9.07 9.32 9.85 9.98 10.03 9.87 9.88 Coal Base Imports 11.11 12.50 13.89 14.30 14.72 16.11 18.08 21.11 23.34 25.58 27.22 Gas Based Imports 18.67 18.89 20.58 19.45 20.84 22.22 25.00 28.33 32.78 36.11 38.41 High Gas Based Imports 18.67 28.90 41.12 38.90 41.88 44.44 50.00 58.66 65.58 72.22 78.82 Mod Gas Based Imports 16.67 23.78 30.84 29.18 31.26 33.33 37.50 42.50 49.17 54.17 57.82 Natural Gas Step 1 7.95 7.50 6.80 6.50 7.50 8.00 8.50 9.50 10.20 11.00 12.20 High_ Gas Step 1 7.95 10.78 13.80 13.00 15.00 16.00 17.00 19.00 20.40 22.00 24.40 Mod_ Gas Step 1 7.95 9.08 1020 9.75 11.25 12.00 12.75 14.25 15.30 18.50 18.30 Nuclear 0.46 0.80 1.00 1.20 1.15 1.17 1.21 1.25 1.30 1.37 1.40 Biomass Stap1 2.13 2.23 2.33 2.44 2.55 2.87 2.79 2.92 3.05 3.19 3.33 Biomass Stap2 2.13 2.23 2.33 2.44 2.55 2.67 2.79 2.92 3.05 3.19 3.33 Bit_ Coal 1.00 1.05 1.10 1.13 1.15 1.18 121 1.25 1.30 1.35 1.40 Muni_ Waste 1.00 102 1.03 1.05 1.08 1.06 1.09 1.11 1.13 1.14 1.18 Natural Gas Step 2 7.95 7.50 8.80 6.50 7.50 8.00 8.50 9.50 10.20 11.00 1220 High_ Gas Step2 7.95 10.78 13.80 13.00 15.00 16.00 17.00 19.00 20.40 22.00 24.40 Mod_ Gas Stap2 7.95 9.08 10.20 9.75 11.25 12.00 12.75 14.25 15.30 18.50 18.30 Not81 SIGJ = 0.948"$/mmBtu 115

PAGE 130

APPENDIX C: (Cont.) Emission Rates Output Rate (at the production level) Technology PC_Coal50% CCS Com3_Xcel IGCC_Xcel Bit Coal Steam Sub Bit Coal Steam Disstilate Fuel Oil DieseiiC Coai_IGCC 50%CCS AdvCT AdvCC cc NewCC CT NewCT Gas IGCC 90% CCS Gas Steam Coal Based Imports Gas Based Imports Notes: CC = Combined Cycle CT =Combustion Turbine IC =Internal Combustion PC = Pulverized Coal C02 (lb/MWh) Output 1,167 2,159 1,048 2,159 2,143 2,000 2,000 1,048 921 865 881 889 1,278 1,246 86 1,587 2,159 881 NOx (lb/MWh) Output 0.373 0.000 0.427 3.895 3.181 2.468 2.468 0.429 0.087 0.071 0.153 0.341 0.568 0.517 0.079 2.415 IGCC =Integrated Gasification Combined Cycle Com3 = Pulverized coal unit by Xcel Energy 116 S02 (lb/MWh) Output 0.619 0.000 0.705 2.387 3.705 0.195 0.195 0.706

PAGE 131

APPENDIX C: (Cont.) Power Import Bounds Import (Gwhlyr) Bound 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 Coal Utility Contracts FX 444.4 4444 444.4 444.4 Coal Based UP 1416.7 1416.7 1418.7 1418.7 1416.7 14187 1418.7 1416.7 1416.7 1416.7 1416.7 Gaa Based UP 1418.7 1418.7 1416.7 1418.7 1416.7 1418.7 1418.7 1416.7 1416.7 1416.7 1416.7 Hydro UP 0 0 0 0 0 0 0 0 0 0 0 Rer.wable UP 0 0 0 0 0 0 0 0 0 0 0 Note: Fixed bourKI is far utitity fixed (take or pay) contacts 117

PAGE 132

APPENDIX C: (Cont.) Investment Bounds Investment Technology Unit Bound 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 New Biomass CC GW UP 0.1 0.44 Existing Muni Solid Waste GW UP 0 0 0 0 0 0 0 0 0 0 0 Com3Xcel GW FX 0 0 0.75 0 0 0 0 0 0 0 0 IGCCXcel GW FX 0 0 0 0 0.6 0 0 0 0 0 0 Existing Bit Coal Steam GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Sub Bit Coal Steam GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Disstilate Fuel Oil GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Diesel IC GW UP 0 0 0 0 0 0 0 0 0 0 0 New Geothermal GW UP 0.03 0.04 Existing Hydro GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Hydro PS GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Gas_CC GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Gas_CT GW UP 0 0 0 0 0 0 0 0 0 0 0 Existing Gas_Steam GW UP 0 0 0 0 0 0 0 0 0 0 0 PV_Central GW LO 0.02 PV_Rooflop GW LO 0.01 Solar_ Thermal GW UP 0.1 1 Wind GW UP 0 0.83 11 Notes: CC = Combined Cycle CT = Combustion Turbine PC = Pulverized Coal IGCC = Integrated Gasification Combined Cycle Com3 = Pulverized coal unit by Xcel Energy PS = Pumped Storage 118

PAGE 133

APPENDIXC: (Cont.) Energy Efficiency and DSM (Maximum Hour Marginal Avoided Costs) Scenario Technology ($/GJ/yr) 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 300 GWhlyr Commercial 29.2 32.8 28.3 35.0 48.9 62.8 84.2 112.5 150.8 201.9 Xcel DSM Commercial 29.2 32.8 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 Xcel Enh_DSM Commercial 29.2 32.8 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 '%peryrEE Commercial 29.2 32.8 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 300 GWhlyr Industrial 29.2 32.8 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 '% peryrEE Industrial 29.2 328 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 300 GWhlyr Residential 29.2 32.8 28.3 35.0 48.9 62.8 84.2 112.5 150.8 201.9 Xcel DSM Residential 29.2 32.8 28.3 35.0 46.9 62.8 84.2 112.5 150.8 201.9 Xcel Enh_DSM ReSidential 29.2 32.8 28.3 35.0 48.9 62.8 84.2 112.5 150.8 201.9 1%peryrEE Residential 29.2 328 28.3 350 469 62.8 842 112.5 150.8 201.9 119

PAGE 134

APPENDIX C: (Cont.) Renewables Growth Rates Parameter GROWTH GROWTH GROWTH GROWTH GROWTH Parameter GROWTH_ TID GROWTH_ TID GROWTH_ TID GROWTH_ TID GROWTH_ TID Notes: Technology Biomass PV_Central PV_Rooftop Solar_ Thermal Wind Technology Biomass PV_Central PV_Rooftop Solar_ Thermal Wind Value 1.15 1.30 1.30 1.10 1.20 Value 0.007 0.100 0.025 0.100 0.200 Growth = Maximum annual growth rate in capacity Growth_ TID = Incremental capacity over and above growth constraint 120

PAGE 135

APPENDIX C: (Cont.) Technology Hurdle Rates and Global Discount Rate Technology Hurdle Rate Biomass_CC 0.132 PC_Coai_CCS 0.162 Com3_Xcel 0.162 IGCC_Xcel 0.162 Geothermal 0.126 IGCC_CCS 0.162 Adv_CT 0.138 Adv_CC 0.154 cc 0.138 CT 0.138 Gas_IGCC_CCS 0.154 Adv_Nuclear 0.177 PV_Central 0.111 PV_Rooftop 0.111 Solar_ Thermal 0.120 Wind 0.112 Global Discount Rate 0.075 121

PAGE 136

REFERENCES [1] The Intergovernmental Panel on Climate Change publications is available online at: http://www.ipcc.ch/ [2] Kyoto Protocol is available online at: http://unfccc. int/resource/docs/convkp/kpeng. htm I [3] Article 2: Objective of Kyoto Protocol, available online at: http://unfccc. int/essential background/convention/background/items/1353. php [4] Kyoto Protocol: Status of Ratification, 10 July 2006. Available online at: http://unfccc.int/files/essential background/kyoto protocol/application/pdf/kpsta ts.pdf [5] Zobaa, A. E., and S. McConnach "International Response to Climate Change: An Overview", IEEE Trans. On Power systems, February 2006. [5a] The World Bank, State and Trends of the Carbon Markets 2007, World Bank Institute, Washington, D.C., May 2007. [6] BBC News, "Politicians sign new climate pact", February 16, 2007. Available online at: http://news.bbc.co.uk/1/hi/sci/tech/6364663.stm [7] Title XVI, Section 1605(b) of the Energy Policy Act of 1992. [8] EIA Online Archives of previous Voluntary Reporting of Greenhouse Gases for 1994-2004 are available at: http://www.eia.doe.gov/oiaf/1605/1605b old.html [9] EIA's Voluntary Reporting of Greenhouse Gases for 2004, available online at: http://www.eia.doe.gov/oiaf/1605/vr04data/pdf/chap2.pdf [1 0] Report# DOE/EIA-0608(2005). See EIA online notice of not producing the Voluntary Reporting of GHG at: http://www.eia.doe.gov/oiaf/1605/vrrpt/ [11] The Greenprint Denver Council draft Climate Action Plan is available online at: http://www.greenprintdenver.org/docs/Greenprint Council Report.pdf 122

PAGE 137

[12] RGGI staff working group recommendations at a glance, available online at: http://www.rggi.org/docs/rggi glance 8 24 05.pdf [13] The California Global Warming Solutions Act of 2006. http://www.leginfo.ca.gov/pub/05-06/bill/asm/ab 00010050/ab 32 bill 20060927 chaptered.pdf [14] The California Senate Bill1368 was passed in September 2006. http://www.energy.ca.gov/ghgstandards/documents/sb 1368 bill 20060929 c haptered.pdf [15] West Coast Governors' Global Warming Initiative Staff Recommendations to the Governors. Available online at: http://www.climatechange.ca.gov/westcoast/documents/index.html [16] Annual Energy Outlook 2006, AEO 2006 (with Projections to 2030), DOE/EIA0383(2006), Energy Information Administration, U.S. Department of Energy, Washington, DC. Available online at: http://www.eia.doe.gov/oiaf/aeo/index.html [17] The complete text of the Energy Policy Act of 2005 is available online at: http://frwebgate.access.gpo.gov/cgibin/getdoc.cai?dbname=1 09 cong public I aws&docid=f:publ058.109.pdf [18] Report of Xcel Energy Comanche Station, unit 3, near Pueblo, Colorado, available online at: http://www.xcelenergy.com/XLWEB/CDA/0.3080, 1-11 1875 24319-13543-2 30 79-0,00.html [19] Western Resource Advocates and SWEEP, Tri-State Generation and Transmission Association's Resource Plan-Analysis & Alternatives, April 2006. Available online at: http://www.westernresources.org/energy/pdf/WRA SWEEP%20Tri-State%20Resource%20Pian%20Analysis.pdf [20] Milford, J. and et al., Climate Alert-Cleaner Energy for the Southwest, Environmental Defense and Western Resource Advocates, 2007 available online at: http://www.westernresourceadvocates.org/media/pdf/CiimateAiertReport.pdf [21] The Colorado Energy Forum, Colorado's Electricity Future-A Detailed Look at the State's Electricity Needs and Electricity's Economic Impacts, September 2006. Available online at: http://www.coloradoenergyforum.org/studies.asp [22] For the complete text of the Colorado RES rules, see Colorado PUC web site, http://www.dora.state.co.us/ puc/electric/EiectricRules.htm 123

PAGE 138

[23] For the complete text of Colorado's enacted Renewable Energy Standard, see Colorado StatutesfTITLE 40 UTILITIES/PUBLIC UTILITIES/General and Administrative/ ARTICLE 2 PUBLIC UTILITIES COMMISSION/40-2-124. http://www.dora.state.co.us/puc/rulemaking/Amendment37/Amend37RelatedSt atutes.htm [24] Energy Information Administration, "State Renewable Energy Requirements and Goals: Status Through 2003", Annual Energy Outlook 2005, DOE/EIA0383(2005) (Washington, DC, February 2005), pp. 20-23, online at: www.eia.doe.gov/oiaf/archive/aeo05/index.html [25] RMA TS study is available at: http://psc.state.wv.us/htdocs/subregionai/FinaiReport/rmatsfinalreport.htm [26] For the complete text of enacted law, see Part 12 to Article 7 of Title 25 and Article 3.2 of Title 40 of Colorado StatutesfTITLE 40 PUBLIC UTILITIES COMMISSION/. [27a] Loulou, Richard, Gary Goldstein, and Ken Nobel, Documentation for the MARKAL Family of Models, Energy Technology Systems Analysis Programme, October 2004. Available online at: http://www.etsap.org/tools.htm [27b] AdJ. Seebregts, Gary A. Goldstein, and Koen Smekens, Energy/Environmental Modeling with the MARKAL Family of Models, International Resources Group Ltd., Washington, DC 20036. [28] Goldstein, Gary A., and L.A. Greening, "Energy Planning and the Development of Carbon Mitigation Strategies Using the MARKAL Family of Models." International Resources Group, Washington, DC. [29] EPA, 2006. MARKAL Scenario Analyses of Technology Options for the Electric Sector: The Impact on Air Quality, EPA/600/R-06/114, September. [30] Gomez, T.L., 2001. Technological Learning in Energy Optimisation Models and Development of Emerging Technologies, Sc.D. dissertation, Swiss Federal Institute of Technology Zurich. [31] Makela, Jussi, 2000. Development of an Energy System Model of the Nordic Electricity Production System, Master thesis, Helsinki University of Technology, Department of Engineering Physics and Mathematics, December. 124

PAGE 139

[32] Gielen, D., 2000. Materia/ising dematerialisation: integrated energy and materials systems engineering for greenhouse gas emissions mitigation, Ph.D. Thesis, ISBN: 90-5155-008-1. [33] Willson, E. and et al., Supporting Regional Multi-Pollutant Analysis and Decision Making: Developing a Regional MARKAL Energy-Environment Model in New England, EPA Office of Research and Development /National Risk Management Research. [34] Seebregts, A.J. and et al., 1999. Endogenous Technological Change in Energy System Models-Synthesis of Experience with ERIS, MARKAL, and MESSAGE, ECN-C-99-025, April. [35] EPA, U.S. National MARKAL Database Documentation, February 2006. [36] Rubin, E.S., M.B. Berkenpas, A. Farrel, G. A. Gibbon and D. N. Smith, "Multi Pollutant Emission Control of Electric Power Plants", Proceedings of EPA DOE-EPRI Mega Symposium, Chicago, IL, EPRI, Palo Alto, CA (August 2001). [37] Field, Barry C., 1997. Environmental Economics: An Introduction. 2"d edition, McGraw Hill. [38] Southerland, R., 1996. The economics of energy conservation policy, Energy Policy, Vol. 24, No.4, pp. 361-370. [39] Nadel, Steven and H. Geller, 1996. Utility DSM: What have we learned? Where are we going? Energy Policy, Vol. 24, No.4, pp. 289-302. [40] KEMA, 2006. Colorado DSM Market Potential Assessment, Volume I & II, KEMA Inc., Oakland, California, March. [41] Rosenberg, William G. and et al, 2004. Deploying /GCC in this Decade with 3Party Covenant Financing, John F. Kennedy School of Government, Harvard University, July. [42] Technical Assessment Guide, 1993. Electric Power Research Institute, Volume 1, at 8-9. [43] Colorado Springs Utilities, 2006. Biomass Co-firing Conceptual Design Study, Final Report, November. [44] Bain, R.L. and W.A. Amos, 2003, Highlights of Biopower Technical Assessment: State of the Industry and the Technology, National Renewable Energy Laboratory, April. Available online at: 125

PAGE 140

webdev. its. iastate.edu/webnews/data/site biorenew reading/8/webnewsfilefiel d file/BioPowerTechAssess. pdf [45] Western Governors' Association, 2006, "Clean and Diversified Energy Initiative", Solar Task Force Report, January. [46] Land and Water Fund of the Rockies, 2002, Renewable Energy Atlas of the West, July. Available online at: www.energyatlas.org [47] Lehr, R. et al, 2001, Colorado Public Utility Commission's Xcel Wind Decision, National Renewable Energy Laboratory, September. [48] Electric Power Research Institute, 1984, Choice of Discount Rates in Utility Planning: A critique of Conventional Betas as Risk Indicators for Electric Utilities, February. [49] EnerNex Corporation, 2006, Wind integration study for Public Service Company of Colorado, Final Report, May. [50] Myers, Ed and Grace Hu, 1999. Demand-Side Carbon Reduction Strategies in an Era of Electric Industry Competition, The Electricity Journal, January/February 72-81. [51] Herring, H. and R. Roy, 2007. Technological innovation, energy efficient design and the rebound effect, Technovation 27, 194-2-3, on-line via ScienceDirect, [52] Boardman, B., 2004. Achieving energy efficiency through product policy:the UK experience, ENVIRONMENTAL Science and Policy, on-line via ScienceDirect, April. [53] Dimitropoulos J and S. Sorrell, 2006. The rebound effect: theoretical basis, extensions and limitations. Proceeding of the 291 h IAEE International Conference, Germany, June. [54] Hirst, Eric, et al, 1986. Energy Efficiency in Buildings: Progress and Promise, American council for an Energy-Efficient Economy, Washington, D.C., at 179. [55] Nadel, Steven, 2000. Utility Energy Efficiency Programs: A Brief Synopsis of Past and Present Efforts, American council for an Energy-Efficient Economy, Washington, D.C., at 6. [56] Lovins, A and P. Joskow, 1994. The great negawatts debate, The Electricity Journal, May 29-61. 126

PAGE 141

[57] Seattle City Light, Energy Conservation Accomplishments: 1979-2004, Active Residential Programs, 2004. [58] Milligan, M. and B. Parsons, 1999, A Comparison and Case Study of Capacity Credit Algorithms for Wind Power Plants, Journal of Wind Engineering, 23(30, at 159-166. [59] Billinton, R. and et al., 1996, A Sequential Simulation Technique for Adequacy Evaluation of Generating Systems Including Wind energy, IEEE Transaction Energy Conservation, 11(4), at 728-734. [60] Evans, James R. and David L. Olson. Introduction to Simulation and Risk Analysis. Prentice Hall, 1998. [61] U.S. Environmental Protection Agency, Guiding Principles for Monte Carlo Analysis, EPN630/R-97/001 [62] Crystal Ball, Decisioneering, Inc. Denver, CO, 2005. [63] EPRI, 1982. Mathematical Generation Planning Models Using Decomposition and Probabilistic Simulation, Palo Alto, CA, August. 127