63215 E N E R G Y A N D M I N I N G S E C TO R B OA R D D I S C U S S I O N PA P E R PA P E R N O. 2 1 N O V E M B E R 2 0 1 0 Impacts of Transmission and Distribution Projects on Greenhouse Gas Emissions Review of Methodologies and a Proposed Approach in the Context of World Bank Lending Operations Marcelino Madrigal Randall Spalding-Fecher The Energy and Mining Sector Board DISCLAIMERS The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. CONTACT INFORMATION To order additional copies of this report and associated tools, please contact the Energy Help Desk: 202-473-0652, energyhelpdesk@ worldbank.org. This paper is available online: www.worldbank.org/energy/. The material in this work is copyrighted. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with com- plete information to the Copyright Clearance Center, Inc, 222 Rosewood Drive, Danvers, MA 01923, USA, fax 978-750-4470. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street N.W., Washington, D.C., 20433, fax 202-522-2422, e-mail: pubrights@worldbank.org. Copyright © 2010. The World Bank Group. All rights reserved. Design and layout: Nita Congress Paper No. 21 NOVEMBER 2010 Impacts of Transmission and Distribution Projects on Greenhouse Gas Emissions Review of Methodologies and a Proposed Approach in the Context of World Bank Lending Operations The Energy and Mining Sector Board Contents Foreword ...................................................................................................................................... vii Acknowledgments........................................................................................................................ ix Abbreviations ............................................................................................................................... xi Executive Summary .................................................................................................................... xiii 1. Introduction ............................................................................................................................. 1 Importance of T&D in the World Bank Energy Portfolio ........................................................................................................2 Significance of the Electricity Sector in Global GHG Emissions .............................................................................................3 Objective of This Study ..................................................................................................................................................................9 2. GHG Accounting Principles Relevant for T&D Projects ....................................................... 11 Corporate and National Inventories versus Project-Level Net Accounting ........................................................................ 12 Additionality and Net Emissions Accounting ......................................................................................................................... 14 Project Boundaries and Double Counting .............................................................................................................................. 16 3. Categorization of Project Types and Emissions Impacts .................................................... 21 The Structure of T&D in World Bank Lending Operations .................................................................................................. 21 Project Categorization by Objective ......................................................................................................................................... 22 Categorization of Emissions Impacts........................................................................................................................................ 23 Relevant GHG Methodologies Reviewed ................................................................................................................................ 25 4. Direct Nongeneration Impacts of T&D Projects .................................................................. 27 Embodied Emissions in Construction Materials .................................................................................................................... 27 Energy Use in Construction....................................................................................................................................................... 29 Land Clearing............................................................................................................................................................................... 29 SF6 Fugitive Emissions ................................................................................................................................................................ 30 N2O Emissions from Corona Discharge................................................................................................................................... 34 Summary of Direct Nongeneration Emissions Impacts ......................................................................................................... 35 5. Generation Emissions Impacts of T&D Projects .................................................................. 39 Technical Loss Reduction ........................................................................................................................................................... 39 Increased Reliability ................................................................................................................................................................... 41 Distribution Capacity Expansion .............................................................................................................................................. 42 Electrification ............................................................................................................................................................................... 42 Transmission Capacity Expansion ............................................................................................................................................ 43 Cross-Border Trade ..................................................................................................................................................................... 44 Summary of Impacts on Power Generation: Direct and Indirect ......................................................................................... 48 v 6. Recommended Approach ..................................................................................................... 53 Recommended Project Boundary ............................................................................................................................................. 53 Step 1. Determine Which Direct Nongeneration Emissions Will Be Included .................................................................. 55 Step 2. Calculate Direct Nongeneration Emissions for the T&D Projects ........................................................................... 56 Step 3. Determine How Baseline and Project Emissions for Power Generation Effects Should Be Calculated ............ 61 Step 4. Calculate Baseline Power Generation Emissions for the T&D Projects .................................................................. 61 Step 5. Calculate Project Power Generation Emissions for the T&D Projects .................................................................... 67 Step 6. Summarize GHG Emissions Impacts ........................................................................................................................... 73 7. Case Studies ........................................................................................................................... 75 Case Study 1: Ethiopia-Kenya Power Systems Interconnection Project .............................................................................. 75 Case Study 2: Energy Access Scale-Up Program, Kenya ........................................................................................................ 77 Case Study 3: Eletrobras Distribution Rehabilitation Project, Brazil ................................................................................... 86 Summary of Results and Conclusions from the Three Case Studies .................................................................................... 89 8. Conclusions ............................................................................................................................ 93 Importance of Net Emissions Accounting and Including Power Generation Emissions Impacts ................................... 93 Implementation Issues: Level of Effort, Data Collection, and Uncertainty ......................................................................... 93 Lessons for the Bank’s Overall Effort on GHG Accounting under the SFDCC .................................................................. 95 Annexes A: Data Tables for Methodology Proposals........................................................................................................................... 97 B: World Bank T&D Projects................................................................................................................................................ 101 Glossary ......................................................................................................................................107 References ..................................................................................................................................109 Boxes 1.1: Example of the Importance of T&D Investments to Power Sector GHG Emissions Reductions in India .................7 2.1: GHG Protocol Overall Principles for GHG Accounting ................................................................................................ 11 4.1: Example of Embodied Emissions in Long-Distance Transmission Line ..................................................................... 28 4.2: The Corona Effect ................................................................................................................................................................ 34 4.3: Example of Direct Nongeneration Emissions from a Typical Transmission Project ................................................. 36 5.1: Cross-Border Trade and GHG Emissions Example: Cambodia-Vietnam ................................................................... 45 5.2: Illustration of Sample Grid Emission Factor Calculations in the Draft Approved Methodology from NM0269/ NM0272 ................................................................................................................................................................................ 49 5.3: Example of Impact of Generation Emissions from a Typical Transmission Project .................................................. 52 Figures 1.1: Sectoral Breakdown of WBG Energy Lending, FY2003–09 .............................................................................................3 1.2: Electricity Grid Components ................................................................................................................................................4 1.3: GHG Emissions for the World by Sector and Country Income Level ............................................................................5 1.4: Global Growth in Carbon Dioxide Emissions by Sector and Region..............................................................................6 1.5: Life-Cycle GHG Emissions for Electricity by Fuel Type: 2005 .........................................................................................6 1.6: T&D Losses by Region, Technical and Nontechnical ........................................................................................................8 1.7: Share of Technical and Nontechnical Losses in Selected African Utilities .....................................................................8 Contents vi 1.8: Evolution of the Transmission System and Power Generation Capacity in Brazil ........................................................9 2.1: Sources of Electricity System Emissions: Life-Cycle Phase versus Value Chain Step................................................. 17 2.2: Potential Project Boundary for Nongeneration Emissions from T&D Projects ......................................................... 18 2.3: Possible Impacts of T&D Projects on Generation and Other Value Chain Stages ..................................................... 19 2.4: Potential Baseline and Project Emissions Sources for Assessing Net Emissions Impacts on Generation ............... 20 4.1: Potential Emissions Sources for Direct Nongeneration Emissions from T&D Projects ............................................ 27 4.2: Life-Cycle GHG Emissions for Long-Distance Transmission of Solar Power for North Africa to Europe ............ 28 6.1: Recommended Project Boundary for T&D Projects ..................................................................................................... 54 6.2: Recommended Baseline and Project Emissions Sources for Assessing the Impacts of Emissions on Generation 54 6.3: Decision Tree for SF6 Calculation Approach ................................................................................................................... 57 6.4: Decision Tree for Technical Loss Reduction Projects..................................................................................................... 61 6.5: Decision Tree for Increased Reliability Projects .............................................................................................................. 62 6.6: Decision Tree for T&D Capacity Expansion Projects..................................................................................................... 62 6.7: Decision Tree for Electrification Projects ........................................................................................................................ 62 6.8: Decision Tree for Cross-Border Trade Projects .............................................................................................................. 63 Tables E.1: Categories of T&D Project Impacts on GHG Emissions Used in This Study...............................................................xv 1.1: WBG Energy Portfolio by Financing Source, FY2003–09 ($ millions) ..........................................................................2 1.2: Sectoral Breakdown of WBG Energy Lending, FY2003–09 ($ millions)........................................................................3 2.1: Project Boundary Definitions from Transpower New Zealand’s Carbon Footprint .................................................. 13 3.1: Categories of T&D Project Impacts on GHG Emissions Used in This Study.............................................................. 24 3.2: GHG Measurement Methodologies for the Direct Nongeneration Emissions Impacts of T&D Projects .............. 25 3.3: GHG Measurement Methodologies for the Generation Emissions Impacts of T&D Projects ................................. 25 4.1: IPCC Default Emission Factors for T&D Equipment .................................................................................................... 31 4.2: Characteristics of SF6-Containing T&D Equipment....................................................................................................... 33 4.3: SF6 Fugitive Emissions from the Power Sector in Selected Countries.......................................................................... 33 4.4: Inclusion of Different Emissions Sources in Direct Nongeneration Emissions Methodologies and Case Studies ..... 35 5.1: Possible Impacts of Different T&D Project Categories on Power Generation ............................................................ 50 5.2: Baseline and Project Scenarios for Impacts of T&D Investments on Power Generation .......................................... 51 6.1: Questions to Determine Which Direct Nongeneration Emissions Calculation Modules to Apply ......................... 55 6.2: Default Emission Factors for SF6 Losses in Operation ................................................................................................... 58 6.3: Relationship between Power Rating and SF6 Capacity for T&D Equipment .............................................................. 59 6.4: Decision Matrix for Whether to Use Build Margin as Part of Baseline (Importing Country) Electricity Emission Factor ................................................................................................................................................ 73 6.5: Example of Summary for T&D Project GHG Emissions (all tCO2 over project life) ................................................. 73 7.1: Summary of GHG Impacts for Ethiopia-Kenya Power Systems Interconnection Project (tCO2) ............................ 77 7.2: Summary of GHG Impacts for Kisii-Awendo Line (tCO2) ............................................................................................ 81 7.3: Summary of GHG impacts for Eldoret-Kitale Line (tCO2) ........................................................................................... 86 7.4: Summary of GHG Impacts for Eletrobras Distribution Rehabilitation Project (tCO2) ............................................. 89 7.5: Summary Results for Three Case Studies (tCO2) ............................................................................................................ 90 A.1: Carbon Density in Biomass Types .................................................................................................................................... 97 A.2: Default Emission Factors for Generator Systems in Small-Scale Diesel Power Plants for Three Load Factor Levels (kg CO2e/kWh) ....................................................................................................................... 99 A.3: Default Energy Efficiencies of Different Power Plant Types (%)................................................................................. 100 Contents vii Foreword The Strategic Framework for Development and report presents a methodological approach that can Climate Change serves to guide and support the be used in the context of WBG T&D lending opera- operational response of the World Bank Group tions to determine the most important impacts of (WBG) to new development challenges posed by T&D projects on GHG emissions in the power sec- global climate change. Under the framework, the tor. WBG committed to developing and testing green- In addition to helping our staff involved in T&D house gas (GHG) emissions accounting methods at operations determine these impacts, we hope that the project level to improve the Bank’s and its clients’ our report will contribute to the ongoing debate knowledge base, capacity, and access to additional on developing comprehensive and effective sector- climate finance. based methodologies for GHG emissions account- The energy sector is an important source of emis- ing. sions globally. The WBG has surpassed its com- We would like to take this opportunity to thank mitment to support renewable energy and energy Jamal Saghir. This work was conducted under his efficiency activities, which will directly contribute leadership during his tenure as director of the for- to lower-carbon energy sector development. A con- mer Energy, Transport, and Water Department. siderable portion of the Bank portfolio supports transmission and distribution (T&D) infrastructure, which is fundamental to increasing and expanding access to modern energy services. Lucio Monari Manager, Energy Anchor Unit (SEGEN) This report aims to contribute to an understand- Sustainable Energy Department ing of the GHG implications of T&D projects. The November 2010 ix Acknowledgments This report was prepared by a team of World Bank de Gouvello, Richard Hosier, and Masami Kojima. staff and consultants led by Marcelino Madrigal We are also thankful for the overall guidance for this (Senior Energy Specialist) of the Sustainable Energy work provided by Lucio Monari, SEGEN’s sector Department Energy Anchor unit (SEGEN). Bank manager. Valuable comments have been provided by staff who contributed to this work include Pedro a number of people in the Bank, including Sameer Antmann (Senior Energy Specialist), Gabriela Akbar, Lucas Bossard, Harikumar Gadde, and Elizondo (Senior Energy Specialist), and Nataliya Monali Ranade. Other contributors from SEGEN Kulichenko (Senior Energy Specialist). The princi- include Ashaya Basnyat, Jie Li, Varun Nangia, and pal consultant for this work was Randall Spalding- Xiaolu Yu. The team is grateful for the collaboration Fecher, Senior Advisor: Carbon & Energy Southern of the task team leaders of the pilot projects consid- Africa, Poyry Energy Management Consulting ered in this study: Paivi Koljonen, Luiz T. Maurer, (Sweden), who was assisted by Francois Sammut, and Leopoldo Montanez. We thank the Energy Senior Project Developer, Carbon Limits (Norway). Sector Management Assistance Program (ESMAP) The team appreciates the insightful comments and for providing financial support for the development guidance of World Bank peer reviewers Christophe of this work. xi Abbreviations AC alternating current km kilometer AM approved methodology kt kilotonne AMS approved methodology, small-scale ktCO2 kilotonnes carbon dioxide CDM Clean Development Mechanism kV kilovolt CEET Carbon Emissions Estimator Tool kW kilowatt CO2 carbon dioxide kWh kilowatt hour CO2e carbon dioxide equivalent m meter DC direct current MtCO2 million tonnes carbon dioxide FSR feasibility study report MVA mega volt amperes FY fiscal year MW megawatt g gram MWh megawatt hour GEF Global Environment Facility N2O nitrous oxide GHG greenhouse gas NM new methodology proposal gWh Gigawatt hour SF6 sulfur hexafluoride ha hectare SFDCC Strategic Framework for Development IFC International Finance Corporation and Climate Change IGES Institute for Global Environmental t ton Strategies T&D transmission and distribution IPCC Intergovernmental Panel on Climate tCO2 tonne carbon dioxide Change tCO2e tonne carbon dioxide equivalent kg kilogram WBG World Bank Group All dollar amounts cited are U.S. dollars. xiii Executive Summary The Strategic Framework for Development and sector. The focus is understandable, given the large Climate Change (SFDCC) approved in 2008 guides share of international investment going into the and supports the operational response of the World power generation subsector, and because the major- Bank Group (WBG) to new development challenges ity of emissions from the power sector are a result of posed by climate change. One activity pursued by the operation of power plants. the SFDCC is to further develop and test methods However, focusing on direct emissions from to analyze climate risks and greenhouse gas (GHG) the different subsectors within the power sector emissions at the project level. The SFDCC empha- underestimates the impact of T&D investments sizes the need to improve GHG accounting activities at the project level to understand the implications of on GHG emissions. One reason for this is that the World Bank’s interventions. anywhere from 7 to 20 percent or more of the electricity generated is lost through technical line The SFDCC established that GHG accounting losses in the T&D system. T&D losses vary con- activities should be carried out as an analytical exer- siderably by country, ranging from 7 to 8 percent cise and not as a business requirement of the project in North America and Europe to 15 percent or preparation or approval process. The framework more in Central and South America. In response prescribes a net emissions approach, which com- to demand from developing countries, World Bank putes emissions reductions or increases by compar- financing for energy infrastructure development ing emissions in a “without project” scenario and a has increased significantly in recent years, reaching “with project” scenario. Additionally, GHG account- $8.2 billion in fiscal 2009. T&D accounts for $6.1 ing activities should be seen by all stakeholders as billion, or 22 percent, of all energy sector lending credible, transparent, feasible, harmonized, and in the past seven years. demand driven. Study Objective Importance of Transmission and The objective of this study is to contribute to the Distribution in the World Bank SFDCC goal of improving GHG accounting in the Portfolio and Power Sector Emissions energy sector by reviewing, assessing, and recom- The power sector is one of the largest sources of mending GHG accounting methodologies for elec- GHG emissions, accounting for more than a quarter tricity T&D projects. Existing methodologies are of global GHG emissions. Emissions from the power examined to test whether they can provide simple sector have grown dramatically in recent decades, and accurate estimates of net project emissions. particularly in developing countries. Most GHG In addition, the study identifies and conceptually emissions accounting analysis for the power sector designs a methodological approach for T&D proj- has focused on emissions from the combustion of ects. The study focuses on the T&D sector due to fossil fuels in power plants rather than the emis- its importance in the World Bank’s energy lending sions from the transmission and distribution (T&D) portfolio and the lack of comprehensive meth- xv odologies to determine the impact of such inter- to recognize the different mechanisms by which ventions on GHG emissions. The study builds on T&D investments can affect emissions from power existing information and relies on methodologies generation plants. In some cases, these modules are developed under different climate finance mecha- simpler than similar CDM baseline methodolo- nisms such as the Clean Development Mechanism gies. There are several reasons for this. First, the (CDM). objective of this study is to provide methodologies that can be used ex ante to estimate GHG impacts. The study also considers some of the fundamental Consequently, they do not include a monitoring principles in other accounting procedures, such as methodology. Second, the methodologies must rely corporate GHG accounting. Methodologies that on data traditionally collected during project prepa- have the objective of emissions accounting for cli- ration and appraisal, which generally do not provide mate finance mechanisms need to have specific the level of detail a dedicated carbon finance feasi- characteristics, such as additionality and ex post bility study would require. Third, the GHG emis- monitoring. These methodologies must calculate sions impact assessment is not used for generating a project’s emissions reductions or increases by credits or securing carbon revenue. Therefore, the estimating the project’s net emissions impact. Most level of accuracy required is not as stringent as for corporate GHG accounting methodologies estimate carbon finance projects. and report a corporation’s emissions inventory, similar to how the Intergovernmental Panel on Climate Change (IPCC) methodologies are used for Review of GHG Accounting national GHG inventories. These methodologies do Methodologies not require additionality tests. The study investigates The survey of methodologies and case studies elements of both emissions accounting approaches indicates that direct nongeneration emissions for due to the increasing need to understand the carbon T&D projects are well covered by many existing intensity of the World Bank portfolio and to fulfill approaches. There is broad consensus on the type the SFDCC’s objective of performing accounting of emissions that are relevant and their estimation at the project level using a net emissions approach. methodology. Estimating these impacts requires The study aims to identify methodologies that can additional data beyond that typically available dur- provide simple, rapid, and accurate estimates of ing project preparation and appraisal. net emissions impacts in the context of the project preparation cycle. There is less experience in the analysis of the impacts of T&D projects on emissions from power generation in terms of net emissions impacts. Diversity of T&D Projects and Several potential impacts have no accepted estima- Associated GHG Impacts tion methodologies at all. The direct impact on World Bank T&D project interventions are very power generation of technical loss reduction and different from traditional private sector or CDM the indirect impact on generation of electrifica- transactions. T&D projects are quite diverse in tion are noted in several methodology guidelines terms of the technologies supported, the objectives and international studies. However, the impacts being pursued, and the scope of the intervention of T&D projects that seek to increase reliability in the context of larger utility investment plans or capacity have not been analyzed for their GHG and multidonor financing. These challenges make contributions. Cross-border trade, although raised analyzing the GHG impacts of World Bank proj- by several proposed CDM methodologies, does not ects difficult. The proposed approach is modular in have an accepted standard for GHG impact analy- order to accommodate the diversity of projects and sis. Executive Summary xvi Significance of Net Emissions facilitates the analysis of a variety of project objec- Approach tives by type and the identification of different mechanisms by which the intervention can affect One of the most important conclusions of this work emissions from power generation. is that the impacts of T&D projects on generation emissions are likely to be much higher than those on Three categories of emissions impacts from GHG direct nongeneration emissions. In some cases, the projects are delineated, as shown in table E.1. In net emissions impacts could be negative (that is, the these definitions, the physical boundary of the T&D project would contribute to reducing overall system project (as opposed to the boundary in terms of emissions), even though direct nongeneration emis- emissions sources) consists of the physical site(s) sions are positive. As a result, assessing the impacts where the project will be constructed. Examples on power generation may be even more important would be substations, transmission lines, and the than calculating direct nongeneration emissions of right-of-way corridor for a transmission expansion T&D projects. Leaving out the impact on generation project. emissions in the GHG analysis could significantly Actions outside the physical boundary of the project underestimate the impact of T&D projects on GHG could include investment in power generation and emissions. changes in dispatch or in the operation of nongrid generators or energy sources. Since indirect impacts Summary of Approach will occur only if these other actions take place, The proposed approach links T&D project objec- these emissions are not fully attributable to the proj- tives to their potential impacts on GHG emissions. ect, although the project contributes to these emis- This approach facilitates rolling out GHG account- sions reductions or increases. Direct emissions can ing in the context of current practices for technical be attributed to the project. All impacts are analyzed and economic evaluation of World Bank projects. over the same project life that is used in the techni- Different modules to assess GHG impacts are pro- cal and economic analysis performed during the posed for different objectives. The approach also Bank’s project appraisal. Table E.1: Categories of T&D Project Impacts on GHG Emissions Used in This Study Category of emissions impact Description Direct nongeneration effects Similar to standard corporate or national inventory. Emissions that occur within the physical boundary of the T&D project, and possibly through the life cycle of that equipment. Direct generation effects Effect on short-term and/or long-term generation emissions that does not require any other actions outside the physical boundary of the T&D proj- ect. This would be the case for technical loss reduction projects. Indirect generation effects Effect on short-term and/or long-term generation emissions that requires actions outside the physical boundary of the T&D project. This would be the case for increased reliability, capacity expansion, electrification, and cross-border trade. Source: Authors’ analysis. Executive Summary xvii Categories of Emissions Impacts on the SF6 capacity of the new equipment installed. Where these data are not available, default values for Emissions impacts caused by T&D projects can be high- or medium-voltage system components may categorized based on the location of the emission- be used. If projects do not install any new equip- altering activity in relation to the defined boundary ment, this emissions source should not be included, of the T&D project, and by the location of the activ- because the SF6 emissions from existing equipment ity with respect to the physical site of power genera- would have occurred even without the project. tion. Generation Emissions Impacts Direct Nongeneration Emissions The project boundary for generation emissions GHG emissions resulting from operations within impacts is the physical site of the power genera- the project boundary, but emitted as a result of tion plants connected to the grid, as well as captive activities occurring outside the physical site of power or off-grid power generation plants that may be generation, are classified as direct nongeneration displaced as a result of the T&D project. Upstream emissions. These emissions include the following: impacts on fuel extraction or transportation are not Embodied emissions from construction materials. taken into account, nor are downstream impacts in Energy use in construction. electricity consumption. Land clearing emissions. To assess the net impact on power generation, dif- Sulfur hexafluoride (SF6) fugitive emissions. ferent modules for baseline and project emissions Embodied emissions from construction materials: are applied to the projects according to the multiple This source set is included where there are sufficient objectives and characteristics of the projects. A project data on construction materials required series of decision trees identifies the modules to be and their origin. This is likely to be a small emis- applied. A given project might have several objec- sions source, and generally, projects at early stages tives or impacts. In this case, each module would be of development will not have a detailed inventory of applied separately to the project. For example, an the materials required. electrification project might have land clearing, SF6- containing equipment installation, and displacement Energy use in construction: This source set is of an identified minigrid or isolated generators. The included only where there are sufficient project data baseline for each power generation impact and proj- on fuel usage in the construction phase. This is likely ect type is described below. to be a small emissions source, and not all projects will have a detailed estimation of the equipment fuel Direct Generation Effects usage during construction. Actions that result in an increase or decrease in Land clearing emissions: Land clearing could be emissions within the T&D project boundary and a significant source of emissions, depending on occur at the physical site for power generation are the vegetation type. The area to be cleared and the classified as direct generation effects. This includes carbon density of the biomass to be cleared should technical loss reduction. be available in the feasibility studies or can be esti- Technical loss reduction: The baseline is the quan- mated during project preparation. tity of electricity lost through technical losses prior Sulfur hexafluoride (SF6) fugitive emissions: to the project. Baseline emissions are the product of These emissions are generally small, but could be historical electricity losses and the marginal emis- significant for projects that install high-voltage sion factor of the grid. Project emissions are electric- equipment. The calculations are preferably based ity losses after implementation of the project multi- Executive Summary xviii plied by the marginal emission factor of the grid. If Electrification: The baseline is the alternative a detailed load flow and power generation model are sources of power for the customers who will be con- available, these projections are used instead to assess nected to the grid, as identified in the technical and the generation emissions impact. economic analysis of the project. As with capacity expansion and increased reliability, if no alternative Indirect Generation Effects sources are identified in the technical and economic Actions that result an increase or decrease in emis- analysis, the baseline is zero emissions. Baseline sions partially due to the T&D project but outside emissions are the product of the increased supply the project boundary, and occur at the physical site of electricity multiplied by the emission factor of for power generation are classified as indirect gen- the alternative power source. Project emissions are eration effects. These include the following: the increased supply of electricity multiplied by the marginal grid emission factor, adjusted for incre- Increased reliability. mental technical losses where necessary. For cases T&D capacity expansion. where a single new plant supplies the incremental Electrification. power, the plant’s emission factor is used. The major Cross-border trade. limitation in this case is the lack of an accounting Increased reliability: Where the project technical approach for the displacement of nonelectric energy and economic analysis specifies what power source sources. This is an area that has not been addressed would have been used when grid power was not by existing methodologies and is beyond the scope available, baseline emissions are the product of the of this report. The World Bank has recently com- emission factor of this source and the increased missioned a major study on a CDM baseline meth- power delivered (once reliability improves). If no odology for rural electrification that will explore the alternative source of power is identified in the proj- quantification of these impacts. ect documents, baseline emissions are zero. Project emissions are the increased grid power generation Cross-border trade: Baseline emissions are the multiplied by the marginal emission factor of the product of the incremental traded electricity mea- grid. sured at the receiving substation and the marginal grid emission factor for the importing country. T&D capacity expansion: Where the project tech- Project emissions are the product of the incremental nical and economic analysis specifies what power traded electricity measured at the receiving substa- source would have been used if grid power were tion and the marginal grid emission factor for the not supplied, baseline emissions are the product of exporting country, adjusted for losses on the new the emission factor of this source and the quantity line. For cases where a single new plant supplies the of electricity supplied by the new T&D capacity. In incremental exported electricity, the plant’s emission other cases, where an alternative is not specified, factor is used for project emissions. If a detailed load baseline emissions are zero because there would flow and power generation model is available for have been no power supply without the capacity both grids, these projections are used to assess the expansion. Project emissions are the quantity of emissions impact of the project. electricity supplied by the new system multiplied by the marginal grid emission factor. For cases where a single new plant supplies the incremental power, Findings of the Pilot Exercise that plant’s emission factor is used. If a transmis- The proposed approach was piloted to estimate the sion line connects two previously separate grids, the GHG impacts of the T&D interventions included cross-border trade module is applied instead of the in three World Bank loans at different stages of capacity expansion module. preparation. These projects comprised a proposed Executive Summary xix transmission interconnection between Ethiopia the direct nongeneration emissions modules. For and Kenya; a distribution loss reduction project in instance, although embodied emissions from mate- Brazil; and a power generation, transmission, and rial construction are known to be small as supported distribution access scale-up program in Kenya. by the pilot projects, not all projects will be able to Despite the small sample, the projects reflect to a determine these emissions during project prepara- great extent the variety of project types supported by tion, since the amount of material required and their the Bank. respective manufacturing sites are often not known until the project implementation phase. Some of the The three cases explored indicate that direct non- issues concerning data availability are discussed fur- generation emissions are relatively small compared ther below. to the direct and indirect impacts on power genera- tion. This is supported by evidence from the study Data, Baselines, and Uncertainty literature review. In all cases, direct nongeneration emissions range from 0 to 6 percent of generation For direct nongeneration emissions accounting, impacts. For example, the direct nongeneration the quantity of construction materials required for emissions for the interconnection between Ethiopia different projects is not usually known with cer- and Kenya are estimated at +804 kt of carbon diox- tainty at the project preparation time because the ide (CO2), largely from land clearing, while the detailed feasibility studies have not been completed. indirect impact on power generation is estimated The relatively small size of this impact would not at −69,812 ktCO2 because of the displacement of merit additional effort by the project teams. While power from a higher emissions grid. For one of the land clearing is generally covered in the environ- transmission projects in Kenya, direct nongenera- mental and social impact assessments, the project tion emissions are estimated at +14 ktCO2, while documentation should clarify the IPCC-defined the direct generation impact is −38 ktCO2 and the vegetation types so the correct emission factor indirect generation impact is +392 ktCO2. The T&D can be used. A lack of detailed data on equipment rehabilitation project in Brazil results in a direct containing SF6 is a gap that must be addressed, generation impact of −571 ktCO2 and an indirect particularly for projects that include high-voltage generation impact of −145 ktCO2, and it has negli- equipment. Existing environmental and social safe- gible nongeneration emissions. guards require regulated handling of SF6, but there These examples show that T&D interventions con- is no requirement to quantify the fugitive emissions tribute to reduced emissions, especially when sys- or specify the characteristics of all equipment being tems are interconnected to make better use of power installed. generation sources when reliability is improved or There will be uncertainty in estimating the direct technical losses are reduced. Thus, achieving the nongeneration GHG impacts of the project, but development objectives of T&D projects can also these impacts are generally small compared to gen- lead to emissions reductions. Projects may also con- eration emissions impacts. The default factors in the tribute to increased emissions, especially when they proposed approach may lead to overestimation of increase T&D capacity to serve increased demand emissions (for example, SF6 national default factors growth that will otherwise not be served by other or high carbon density values for land cleared in energy sources. case of vegetation type uncertainty would overesti- While current project preparation procedures mate emissions). Erring on the high side for these already provide most of the data that are crucial relatively small sources is preferable to underesti- in net impacts estimation, collection of data will mating them, but the best solution is to collect the require improvements over time, especially for data during project preparation. Executive Summary xx There are some data requirements for estimating of power is a major constraint to development, and the impact on generation emissions that teams many large industrial projects would not be imple- preparing projects need to understand. Although mented without significant T&D capacity expan- grid emission factors from the Institute for Global sion. The practical issue is whether the project team Environmental Strategies CDM database or a regis- can provide projections that identify the demand tered CDM project can be used, project teams could that would not be satisfied if a capacity expansion consider collecting primary data during project project was not implemented—that is, suppressed preparation. Large high-voltage interconnection demand. This uncertainty is not particular to the projects that conduct a power generation simulation proposed approach, but is true for any type of tech- for their economic analysis will already have this nical and economic assessment of projects and for information. However, lower-voltage distribution other accounting methods. The present proposal is and electrification projects will generally not per- a practical compromise—that is, to use the emis- form such an analysis and, as a result, grid emission sions of the alternative power sources identified by factors will need to be estimated. the project team while performing the technical and economic appraisal of the projects as the base- One important challenge in assessing net impacts line emissions, and to use a zero emissions baseline of increased reliability, technical loss reduction, and where there is no alternative source identified. capacity expansion projects is the clear separation Although the latter case will result in higher emis- of the impacts of these objectives both theoretically sions estimates, it is preferable since impacts will be and practically. While load flow and long-term eco- estimated on the conservative side. nomic dispatch simulations could provide reliable information to supply all the modules, such simula- Estimating emissions impacts is straightforward tions are not carried out for all types of projects. If for cross-border trade projects when load flow and the impacts on losses and reliability are determined long-term dispatch modeling data exist. This is likely separately, it is essential that the teams use consis- to be true for some large high-voltage intercon- tent baselines and project scenarios. For instance, if nection projects, but certainly not all (for example, the impact on the project’s technical losses is esti- smaller-scale projects that interconnect small distri- mated for an entire network, the impact of the proj- bution zones). In these latter cases, the challenge lies ect on increased transmission capacity should also in deciding whether the marginal emission factors be analyzed for the entire network. for the grid accurately represent the impacts on dis- patch caused by the project. For capacity expansion projects—and, to a lesser extent, for electrification projects—a source of The main challenge in electrification lies in the iden- uncertainty is how the baseline captures alternatives tification of a method to address the displacement of to the grid. In other words, if a capacity expansion fuels other than electricity. The approach presented project were not implemented, would the custom- in this report considers only the displacement of ers find other sources of an equivalent amount of other electricity sources. A separate work by the power? This is both a question of principle and World Bank is looking at this topic. It aims to review of practice. The principle issue is that economic the literature on rural electrification to determine development will drive the need for more power whether there are consistent patterns of baseline that must be provided by the grid or other sources. energy use and shifts in post-electrification patterns Even if those alternatives are not currently in place, across different countries and regions. This will be to exclude them from the baseline would be, in the first such effort to address fuel displacement in essence, to assume that demand for power is not the context of carbon financing or carbon account- growing. At the same time, the reality is that the lack ing. Executive Summary xxi Lessons Learned and the Way of the clients whose projects were considered in the Forward pilot process. While the proposed methodology is ready to be deployed for T&D projects in the con- Understanding the GHG emissions implications of text of current project preparation practices, it must World Bank interventions supports the process of be emphasized that the activities concerning meth- identifying lower-carbon options, facilitates the use odology development for GHG accounting and test- of emerging clean technology and climate funds, ing are, as described by the SFDCC and endorsed by and increases the capacity of World Bank staff and the board, an analytical exercise. The formal adop- clients. With this in mind, the proposed approach tion of GHG accounting procedures for Bank opera- has been designed to suit the structure of World tions may require some uniformity and consistency Bank projects in order to facilitate its implementa- across all sectors. As the work on piloting GHG tion in the context of existing project preparation accounting in other sectors moves forward, a Bank- practices, while capturing most relevant impacts on wide proposal on GHG analysis would be proposed GHG emissions from T&D interventions. to the Board as envisaged by the SFDCC. The proposed approach would not impose a signifi- Besides contributing to the SFDCC objectives, this cant additional burden on project preparation, and work contributes to the ongoing debate on method- it could be applied to projects if a specific mandate ology development that seeks a more comprehen- or incentives—such as climate financing—are intro- sive, easy-to-use, and reliable sector- and subsector- duced.1 The latter has generated interest from some based GHG accounting process across multilateral development banks. Moving away from the com- 1 For a typical project with two components, performing plexities and associated costs of project-based GHG accounting with the proposed approach should require accounting methodologies is an approach also being about 10 days of work in coordination with the team members performing the project’s technical and economic evaluation. seriously considered for a reformulated CDM. Executive Summary xxii 1. Introduction The Strategic Framework for Development and methods for GHG emissions analysis. It emphasizes Climate Change (SFDCC) approved in 2008 guides the need for GHG accounting activities to follow a and supports the operational response of the WBG net emissions approach, discussed in more detail to new development challenges posed by climate below, which computes emissions reductions or change. Different initiatives are supported by the increases by comparing emissions in a “without strategic framework. One of these activities is the project” scenario and a “with project” scenario.1 improvement of the knowledge and capacity of Additionally, GHG accounting activities should be WBG staff and clients to analyze development- seen by all stakeholders as credible, transparent, climate links at the global, regional, country, sector, feasible, harmonized, and demand driven. The and project levels. Further development and testing purpose of the activities is to of methods to analyze climate risks and greenhouse build staff and client capacity for carbon analysis gas (GHG) emissions at the project level is critical to prepare for a carbon-constrained future; to achieving this objective. Specifically, the strategic framework (World Bank 2008a) mentions the fol- gather information to understand better the lowing: implications of possible new approaches; …the WBG is developing methods to analyze cli- identify low-cost mitigation opportunities across mate risks and GHG emissions at the project level in GEF [Global Environment Facility] and carbon operations, especially in sectors that may be cur- finance projects. Their application will extend, for rently overlooked, that is, beyond energy and learning and information purposes, to a larger pool transport; of projects. The Bank will select pilot projects on a demand basis, and will work in close coopera- facilitate an analysis of alternatives; and tion with clients and local institutions. The IFC [International Finance Corporation] will progres- help promote the efficient use of emerging cli- sively apply these tools to its projects to inform the mate funds, including the Clean Technology dialogue with its private sector clients on climate Fund.2 related business opportunities and risks. This is an analytical exercise. It is neither a business require- ment, nor it will [sic] be used for decision-making about projects using traditional WBG financing instruments. By the end of the piloting period, a proposal will be prepared for Board consideration 1 on the future applications of the tools for GHG “…An emerging approach in all…sectors is to undertake analysis appropriate for Bank and IFC business GHG assessment, focusing on net emissions from a project, as part of a broader analysis of all project benefits and external models, client needs, and available climate financ- costs, including a range of externalities…. This would allow the ing instruments. analysts to place the GHG analysis of a project in the context of its development impact and assess the trade-offs where appli- The strategic framework sets out important prin- cable” (World Bank 2008a, p. 72). ciples that will guide the development and testing of 2 World Bank (2008a), p. 73. 1 Importance of T&D in the World Bank In the lending portfolio for the last seven years, Energy Portfolio transmission and distribution (T&D) projects com- prise $6.1 billion, or more than 22 percent of all In response to demand from developing countries, energy sector lending (see table 1.2 and figure 1.1). WBG financing for energy infrastructure develop- According to the formal classification for World ment has increased significantly in recent years, Bank lending, T&D projects are associated with reaching $8.2 billion in fiscal 2009 (see table 1.1).3 new network capacity expansion or rehabilitation Energy infrastructure projects seek to increase of existing T&D systems. These are projects that energy access; develop renewable energy and energy efficiency; and leverage private sector participation have new T&D equipment associated with network in energy generation, transmission, and distribution, capacity expansion. T&D rehabilitation projects, including through effective public-private partner- even if they implicitly result in loss reduction, are ship arrangements. included in this category if the energy efficiency component cannot be clearly disaggregated from network expansion or load increase. If the financing 3 Sourced from http://go.worldbank.org/ERF9QNT660. for energy efficiency components of T&D rehabilita- Table 1.1: WBG Energy Portfolio by Financing Source, FY2003–09 ($ millions) Institution FY2003 FY2004 FY2005 FY2006 FY2007 FY2008 FY2009 World Bank 1,176 921 1,868 3,155 2,016 4,512 6,548 IBRDa 468 259 593 1,565 504 2,674 3,569 IDAa 560 535 712 1,441 1,070 1,420 2,155 GEFb 55 62 105 51 128 145 84 Otherc 93 64 458 98 314 272 740 IFCd 638 705 764 1,308 1,170 2,923 1,647 MIGAe 556 73 232 190 417 110 33 WBG energy total 2,370 1,699 2,864 4,653 3,604 7,545 8,228 Source: World Bank calculations. a The International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA) together make up the World Bank. b The Global Environment Facility (GEF) provides grants and concessional loans to help developing countries meet the costs of measures designed to achieve global environmental benefits. The World Bank is one of the three implementing agencies of the GEF. c Other includes guarantees, carbon finance, special financing, and recipient-executed activities. Concerning carbon finance, the World Bank Carbon Finance Unit uses funding contributed by governments and companies in Organisation for Economic Co-operation and Development (OECD) countries to purchase project-based GHG emissions reductions in devel- oping countries and countries with economies in transition. Clean Technology Fund financing is not included in FY2009 financing figures. d The International Finance Corporation (IFC) provides loans, equity, and technical assistance to stimulate private sector investment in developing countries. e The Multilateral Investment Guarantee Agency (MIGA) provides guarantees against losses caused by noncommercial risks to investors in developing countries. 1. Introduction 2 Table 1.2: Sectoral Breakdown of WBG Energy Lending, FY2003–09 ($ millions) Sector FY2003 FY2004 FY2005 FY2006 FY2007 FY2008 FY2009 Energy efficiency 177 92 217 761 262 1,192 1,701 Large hydropowera 23 83 538 250 751 1,007 177 New renewable energyb 206 138 246 344 421 473 1,427 Oil, gas, and coal (upstream) 333 496 578 1,074 627 981 1,032 Other energyc 816 370 278 248 375 903 1,752 Thermal generationd 599 272 100 511 360 957 936 T&D 216 248 906 1,465 809 2,031 1,204 WBG energy total 2,370 1,699 2,864 4,653 3,604 7,545 8,228 Total low carbone 406 350 1,237 1,660 1,440 3,003 3,305 Total accesse 794 537 1,136 1,018 1,239 2,284 2,201 Source: World Bank calculations. a Large hydropower refers to hydropower projects larger than 10 MW. b New renewable energy refers to all renewable energy, excluding hydropower projects larger than 10 MW. c Other energy includes energy policy support projects. d Thermal generation includes all new fossil fuel power plants, including high-efficiency fossil fuel power plants (super- and ultra-critical power plants). e Low-carbon projects include renewable energy projects, energy efficiency, power plant rehabilitation, district heating, and biomass waste energy. Access projects include projects aimed at increasing access to electricity services. These categories are not mutually exclusive, as some projects are classified as blended low carbon and access. For IDA countries, access includes all generation, transmission, and distribution projects, as they are all needed for increased electrification. For IBRD countries, only projects specifically aimed at increasing electricity access (for example, rural electrification projects) are included. tion projects can be disaggregated, they are classified Figure 1.1: Sectoral Breakdown of WBG Energy as supply-side energy efficiency. Lending, FY2003–09 $ billions 9 Significance of the Electricity Sector New renewable energy 8 Hydropower > 10 MW in Global GHG Emissions 7 Energy efficiency Thermal generation Electricity grid systems are typically divided into 6 Oil, gas, and coal T&D generation, transmission, and distribution (see fig- 5 Regulation and reform 4 ure 1.2). While power generation investments can 3 be clearly distinguished in any grid, the boundary 2 1 between T&D is not always consistent across coun- 0 tries. Different countries use different voltage levels FY03 FY04 FY05 FY06 FY07 FY08 FY09 for this distinction. For one country, a line operating Source: World Bank calculations. at 69 kV could be considered part of the transmis- sion system, while for another country any line at 1. Introduction 3 Figure 1.2: Electricity Grid Components Generation Transmission Distribution Source: Brown and Sedano 2004. 69 kV could be considered part of the distribution ing transmission lines, either within a connected system. Some classifications state that the transmis- grid or across a national boundary, could also have sion system ends at the substation where the voltage this impact if the increase in capacity is significant is stepped down from 138 kV to less than 100 kV enough. However, if they only reduce technical (usually less than 50 kV). There are cases, however, losses, they may not affect the mix of operational where relatively long-distance lines may operate generating plants. below 100 kV, or where distribution lines within an The transmission system includes lines and sub- urban area could be more than 100 kV. For the pur- stations. Substations may contain transformers, pose of analyzing GHG emissions impacts, the exact switches, circuit breakers, voltage regulators and boundary between these systems is not as important capacitors, power factor correction devices, and as how these investments affect generation of power. storage devices. Direct current (DC) transmission The boundary between T&D is essentially the trans- systems also include a rectifier to convert generator formers that operate one voltage level above those alternating current (AC) power into DC power, and for individual households. an inverter to convert the DC power to AC power It is also useful to distinguish those transmission when it enters the distribution system. The lines investments that seek to interconnect two previ- would normally be mounted on steel lattice towers ously isolated networks (either within a country or because underground lines require cooling and are between countries) versus those aimed at upgrading much more expensive. and strengthening existing transmission lines. New The distribution system includes all the equipment interconnectors may significantly affect not only from the transmission substation to the individual the flow of power between countries or subnational customer’s meter. The feeder lines of a distribution regions, but also the dispatch of grid-connected system would operate between 2.4 kV and 33 kV. plants, resulting in a change in the mix of power Network equipment would comprise distribution plants supplying the grid at any given time. This can substations, pole-mounted transformers, low-volt- have a major impact on GHG emissions if hydro- age distribution wiring, and sometimes electricity electricity displaces coal- or gas-fired power in a meters. The distribution substations would have thermal power–dominated grid. Upgrades to exist- transformers, switches, and circuit breakers or fuses. 1. Introduction 4 Emissions from the Power Sector Figure 1.3: GHG Emissions for the World by Sector and According to the 2010 World Development Report: Country Income Level Globally, power is the largest single source of a. World greenhouse gas emissions (26 percent), followed Waste & wastewater, by industry (19 percent), transport (13 percent), 3% and buildings (8 percent), with land-use change, agriculture, and waste accounting for the bal- Land use ance…. The picture varies, however, across income change & groups. High-income country emissions are forestry, Power, 25% 17% dominated by power and transport, while land-use change and agriculture are the leading emissions sources in low-income countries. In middle-income Agriculture, countries, power, industry, and land-use change are 14% Transportation, the largest contributors—but with land-use change 13% emissions concentrated in a handful of countries (Brazil and Indonesia account for half the global Industry, 19% land-use change emissions). Power will most likely continue to be the largest source, but emissions are expected to rise faster in transport and industry Res. & com. buildings, 9% (World Bank 2010). b. High-income countries This is illustrated in figure 1.3. In addition, emis- Agriculture, 8% sions from the global power sector have grown dra- Other, matically in recent decades, particularly in develop- 18% Indus- try, 15% ing countries (see figure 1.4). Transpor- Power, 36% tation, Most of the GHG analysis of the power sector has 23% focused on emissions from combustion of fossil fuels in power plants, rather than issues in T&D c. Middle-income countries (see, for example, Bosi and Laurence 2002; Kartha, Lazarus, and Bosi 2004; Sharma and Shrestha 2006; Land use GHG Protocol 2007). Many of the early CDM Other, 14% change & forestry, methodologies were also related to power genera- 23% tion. The methodologies being developed in the Power, 26% Agriculture, Activities Implemented Jointly trial period for 14% project-based emissions trading under the United Nations Framework Convention on Climate Change Industry, 16% also focus on power generation. For this reason, Transportation, 7% the second consolidated methodology approved by the CDM Executive Board was for grid-connected d. Low-income countries Other, 14% renewable power projects that displaced grid-con- Power, 5% Land use nected fossil fuel plants. The focus is understand- Transportation, 4% change & able, given the large share of international invest- Industry, 7% forestry, 50% ment going into the power generation subsector, and Agriculture,20% the fact that most of the emissions from the power sector come from the operation of power plants (see Source: World Bank 2010. figure 1.5). 1. Introduction 5 Figure 1.4: Global Growth in Carbon Dioxide Emissions by Sector and Region Million metric tonnes CO2 12,000 Developing countries Rest of world 10,000 China Developed countries Rest of OECD countries 8,000 United States 6,000 4,000 2,000 0 2000 2005 2000 2005 2000 2005 2000 2005 Electricity & heat Transportation Industry Building use +21% +12% +21% +6% Source: Herzog 2009. Note: OECD = Organisation for Economic Co-operation and Development; CO2 = carbon dioxide. Figure 1.5: Life-Cycle GHG Emissions for Electricity by Fuel Type: 2005 g CO2e/kWh 1,000 900 CO2e, from other processes CO2 operation, from fuel 800 700 600 500 400 300 200 100 0 d ga te e sil llin ic, sil hou c, cy eam ar cle , r st nal ce e or st nal cy al oi w ura titu er p ai st lta el id ico e ico s cle m co CH s -sh cle rv or olt n ll riv n bu er fu ox bu er ry vo io io m at bs Ph n n t P se Nu ea rd m nt on m nt ,s am tov a of tic o lid oo l fro n , su re Ha co P i ul ot co i as d, n so o HP o, m Ph H ru as G in dr C s, ,C og W o, st Ga s, Hy dr as Ga Bi Hy og Bi Source: Bauer et al. 2008. Note: Hard coal is for Germany, all others are for Switzerland. CHP = combined heat and power. 1. Introduction 6 Given the focus on power generation, much less in Central and South America. In developing coun- analysis of the impacts of T&D investments on GHG tries, however, a substantial portion of these losses emissions has been done, and particularly on how are “nontechnical” (that is, electricity is consumed, these investments affect the rest of the power sector. but the utility does not receive revenue because it The standard guidelines for the power sector (for is not being metered or it is being taken illegally, example, GHG Protocol 2005b; IPCC 2006b) gener- among other reasons). Figure 1.7 presents an exam- ally say very little about emissions related to T&D, ple of this taken from a survey of African utilities. which is part of the rationale for this study. Considering a major developing country such as Focusing on direct emissions from the different India, where technical T&D losses are 29 percent subsectors within the power sector underestimates (South Asia Sustainable Development Department the impact of T&D investments on GHG emissions. 2009) and power sector emissions in 2006 were One reason for this is that anywhere from 7 percent 744 Mt of carbon dioxide (CO2) (WRI 2006), these to more than 20 percent of the electricity generated losses amount to 217 MtCO2. In China, where is lost through technical line losses within the T&D power sector emissions were 3,000 MtCO2 in 2006 system. Box 1.1 illustrates the importance of T&D (WRI 2006) and technical losses were 18 percent investments on emissions with a World Bank analy- (IEA 2009), these losses would be responsible for sis of the mitigation options for the Indian energy 552 MtCO2. This is larger than the total national sector. GHG emissions (2005) from France, South Africa, or Ukraine (WRI 2006). T&D losses vary considerably by country. As shown in figure 1.6, losses range from 7 to 8 percent in An additional dimension of the impacts of T&D North America and Europe to more than 15 percent investments on GHG emissions that has been largely Box 1.1: Example of the Importance of T&D Investments to Power Sector GHG Emissions Reductions in India A World Bank analysis of low-carbon options for the Indian economy concludes that “reducing technical T&D losses is one of the most cost-effective means of improving power sector performance while simultaneously reducing CO2 emissions. Reducing technical losses is in fact equivalent to adding new capacity with no increase in CO2 emissions.” The table below shows the impact of advancing or delaying by five years the implementation of the T&D loss reduction program assumed in the baseline power sector development plan (scenario 1) on CO2 emissions and total investment over a 25-year period, assuming that the same amount of grid electricity will be supplied to end users in all cases. If the program is accelerated 5 or 10 years, emissions and investment require- ments decline significantly. T&D loss reduction Change in CO2 emissions Change in investment implementation 2007–31 (Mt) 2007–31 (billion 2007 rupees) Accelerated by 10 years −568 −94 Accelerated by 5 years −248 −6 Delayed by 5 years 1,392 227 Source: South Asia Sustainable Development Department 2009. Note: The years are fiscal years. The total investment covers all investments needed to supply the same amount of electric- ity to consumers as in scenario 1 and includes life extension, efficiency improvement, and new plant construction. 1. Introduction 7 Figure 1.6: T&D Losses by Region, Technical and Nontechnical 18.0% 16.0% North America Europe World Total 14.0% Asia & Oceania Middle East 12.0% Africa Eurasia 10.0% Central & South America 8.0% 6.0% 4.0% 2.0% 0.0% 1980 1981 1982 1983 1984 Eurasia 1985 1986 1987 1988 1989 Middle East 1990 1991 1992 1993 1994 1995 World Total 1996 1997 1998 1999 2000 2001 North America 2002 2003 2004 2005 2006 Source: Pinto 2010. overlooked is the importance of T&D investments in Figure 1.7: Share of Technical and Nontechnical Losses enabling renewable energy technologies. Renewable in Selected African Utilities sources of power are frequently located far from Utility A 17.8 17.5 consumption centers; bringing these sources to the market requires investment in T&D. This situ- Utility C 17.9 4.7 ation can be seen in different power sectors where Utility D 2.4 3.0 3.6 the existing or envisaged level of renewable power Utility F 12.0 6.6 11.9 sources is considerable. Consider, for instance, the Utility G 4.0 16.0 Distribution case of Brazil. As of 2006, about 90 percent of the Utility H 20.5 4.5 Transmission installed generation capacity was renewable, primar- Nontechnical Utility J 16.0 ily hydropower. These sources are located in river Utility K 8.9 3.4 basins across the country’s vast territory. Exploiting Utility L 12.6 4.0 these resources to maintain the large share of renew- Utility E 17.8 3.8 8.1 able energy in the system has required a constant 0 10 20 30 40 expansion of the transmission system, as shown in Losses (% of energy dispatched or purchased) figure 1.8. Source: Pinto 2010. Denmark benefits from a large interconnected system that facilitates the integration and manage- 1. Introduction 8 Denmark is an electricity corridor to and from the Figure 1.8: Evolution of the Transmission System and neighboring countries. The strength of such a trans- Power Generation Capacity in Brazil mission system has been crucial in maintaining sys- tem operation during conditions where wind power a. Transmission system infrastructure (networks above 230 kV) supply has declined sharply.5 The transmission sys- tem is also used to export excess wind power during Km (thousands) low-demand periods from Denmark to Norway and 80 20% 23% 3% 5% 14% 17% Sweden. 11% 60 Other developed countries have found that 40 achieving a high penetration of renewable energy 20 requires a well-developed transmission system. In the United States, a study directed by the U.S. 0 Department of Energy found that achieving a 2000 2001 2002 2003 2004 2005 2006 20 percent share of wind energy in the country 1999 exis ng grid Reinforcements and new installa ons would require investments of about $20 billion in the transmission system (U.S. DOE 2008). This b. Installed generating capacity is largely driven by the fact that wind resources MW (thousands) are mostly located in the Midwest, far from the consumption centers and existing transmission 80 Renewable systems. Similar findings have emerged for the 60 European integrated electricity market, where 40 achieving 20 percent renewable energy by 2020 will require considerable transmission investment 20 across borders.6 Conventional 0 2000 2001 2002 2003 2004 2005 2006 Objective of This Study Sources: Barroso and others 2007; MME and EPE 2006. This study seeks to contribute to the objectives out- lined in the SFDCC in the area of GHG accounting in the energy sector. The study concentrates on ment of a large amount of wind power generation, the T&D subsector for two reasons: (1) T&D proj- accounting for about 20 percent of electricity supply ects represent a considerable portion of the World in 2009, which is among the highest in the world. Bank’s energy portfolio, and (2) the implications The transmission interconnection capacity to its of T&D projects on GHG emissions have received neighboring countries (Germany, Norway, and less attention than power generation projects. Sweden) is about 5,780 MW, and the peak demand Renewable energy generation, energy efficiency, in the two Danish systems was about 6,500 MW and other off-grid projects have more available car- in 2009.4 The large capacity of the interconnec- bon finance–related methodologies than does the tions compared to internal peak demand is because T&D sector. The importance of the transmission system in achieving a lower-carbon power sector 4 Information from energinet.dk (the Danish transmission system operator) and the Danish Energy Agency; refers to 5 See, for example, Ackermann and others (2009) and CEPOS nameplate capacities. Actual interconnection capacity depends (2009). on network conditions and on the direction of the flow (imports 6 or exports). See May (2009). 1. Introduction 9 seems unquestionable. Understanding the implica- nents, such as additionality7 and ex post monitor- tions of T&D investment on GHG emissions in the ing. These methodologies must compute a project’s power sector and finding ways to measure these emissions reductions relative to a baseline, which impacts in the context of the SFDCC are the objec- means they compute the project’s net emissions. On tives of this report. the other hand, methodologies for corporate GHG reporting estimate a corporation’s direct emissions. The study reviews, assesses, and provides recom- Generally, such methodologies are similar to those mendations on methodologies for GHG accounting of the Intergovernmental Panel on Climate Change of electricity T&D projects. Existing methodologies (IPCC) for national GHG inventories and do not are assessed according to a set of selected principles require additionality tests. Given the strong cor- to test whether they can provide simple and accu- porate mandate for GHG accounting at the project rate estimates of net emissions at the project level. level specified in the SFDCC, and that project-level In addition, the study identifies and conceptually accounting for Bank projects is not intended for designs new methodologies that may be required to climate finance purposes, the study investigates ele- fulfill this objective. The study, along with the ana- ments of both accounting approaches. The outcome lytical efforts on GHG accounting in other sectors, should be methodologies that can provide simple assists in understanding the implications of future but reasonable estimates of net emissions impacts application of GHG analysis tools at the World for use in the project preparation cycle. Bank. The study builds on existing information and meth- 7 Additionality is defined by the UNFCCC as follows: A odologies developed under different climate finance CDM project activity is additional if anthropogenic emissions mechanisms, and considers some of the funda- of greenhouse gases by sources are reduced below those that would have occurred in the absence of the registered CDM mental principles in other accounting procedures, project activity. In other words, the project has lower emissions such as corporate GHG accounting. Methodologies than a counterfactual “baseline scenario”. Justifying additional- whose objective is emissions accounting for climate ity involved demonstrating that the project would not have happened without the benefits (financial and otherwise) of the finance mechanisms need to have specific compo- CDM. 1. Introduction 10 2. GHG Accounting Principles Relevant for T&D Projects The basic principles for GHG accounting are similar The Kyoto Protocol says that emissions reductions across many different sources, although they vary under the CDM must be “real, measurable, and somewhat according to the purpose of the meth- long-term.” The baseline methodologies used in the odology. The GHG Protocol, for example, identifies CDM must also follow principles included in the relevance, completeness, consistency, transparency, CDM Modalities and Procedures. These include and accuracy as key principles (see box 2.1). The estimating emissions reductions “in a transparent IPCC 2006 guidelines highlight that “good practice” and conservative manner” and “taking into account inventories are those that “contain neither over- nor uncertainty” (UNFCCC 2001). under-estimates so far as can be judged, and in which uncertainties are reduced as far as practica- The SFDCC provides some principles to guide the ble” (IPCC 2006a). This language reflects an empha- methodology development within this study. They sis on not just accuracy and completeness, but also closely follow the practice in the methodologies and on feasibility (“as far as practicable”). guidelines described above. The principles specified Box 2.1: GHG Protocol Overall Principles for GHG Accounting Relevance: Ensure the GHG inventory appropriately reflects the GHG emissions of the company and serves the decision-making needs of users—both internal and external to the company. Completeness: Account for and report on all GHG emissions sources and activities within the chosen inventory boundary. Disclose and justify any specific exclusions. Consistency: Use consistent methodologies to allow for meaningful comparisons of emissions over time. Transparently document any changes to the data, inventory boundary, methods, or any other relevant factors in the time series. Transparency: Address all relevant issues in a factual and coherent manner, based on a clear audit trail. Disclose any relevant assumptions and make appropriate references to the accounting and calculation meth- odologies and data sources used. Accuracy: Ensure that the quantification of GHG emissions is systematically neither over nor under actual emissions, as far as can be judged, and that uncertainties are reduced as far as practicable. Achieve sufficient accuracy to enable users to make decisions with reasonable assurance as to the integrity of the reported infor- mation. Source: GHG Protocol 2004. 11 in the SFDCC document are credibility, transpar- and regulations. Conversely, for a corporation doing ency, feasibility, and ease of harmonization. For GHG inventory reporting, it may be more important this particular study, the terms of reference suggest to prioritize transparency rather than accuracy since a greater emphasis on feasibility than ease of harmo- this allows all stakeholders the opportunity to easily nization, because these proposed methodologies will understand and replicate the reporting results. not be used for carbon finance project applications. The working definitions of these principles are as Corporate and National Inventories follows: versus Project-Level Net Accounting Credibility/accuracy: The assurance that the The methodologies for assessing the GHG emis- quantification of GHG emissions is systemati- sions impacts of projects and organizations typically cally neither over nor under actual emissions, as fall into two broad categories: corporate or national far as can be judged, and that uncertainties are inventories (sometimes called “gross emissions reduced as far as practicable. accounting”) and project-level net impacts account- Transparency: The addressing of all relevant ing. Corporate or national GHG inventories issues in a factual and coherent manner, based consider only the increases of emissions from the on a clear audit trail. Disclose any relevant activities within a specific project activity, company, assumptions and make appropriate references to or country. Net emissions accounting for a proj- the accounting and calculation methodologies ect, on the other hand, considers how the overall and data sources used. emissions of a larger system may change from the “without project” scenario to the “with project” sce- Feasibility: The ability for most of the calcula- nario, which may include decreases or increases in tions to be carried out using the existing data overall emissions as a result of the implementation that would normally be available through fea- of the project.1 The “without project” scenario is sibility studies and similar documentation pre- called a “business as usual,” “reference,” or “baseline” pared for World Bank projects, or for data to be scenario. The “with project” scenario is the scenario obtained relatively easily by the staff evaluating that includes implementation of the project, which these proposals. may lead to different emissions than the reference or business-as-usual scenario. An example would Ease of harmonization: The assurance of con- be installation of a more efficient fossil fuel–fired sistency with other widely used GHG accounting boiler. Operating the new, more efficient boiler will methodologies, taking into consideration how still create GHG emissions from the combustion of they may change over time. fossil fuels. Compared to the existing, less efficient Most GHG accounting systems follow similar prin- boiler, however, the project results in a net decrease ciples and acknowledge the tradeoffs among these in emissions, because emissions in the “without principles. For instance, a higher degree of accuracy project” scenario are higher than in the “with proj- may mean less transparency, because more sophis- ect” scenario. ticated methods and tools may be needed. The Inventory accounting is typically used for calculat- emphasis placed on the different principles should ing the GHG footprints (usually called “carbon be determined by the objectives of the GHG emis- sions accounting activity. For example, accuracy is a very important principle when accounting is 1 being used for climate financing purposes, because The SFDCC emphasizes the need for GHG accounting activities to follow a net emissions approach, which computes of the risk the higher crediting will compromise the emissions reductions or increases by comparing emissions from integrity of GHG emissions limitation agreements a “without project” scenario and a “with project” scenario. 2. GHG Accounting Principles Relevant for T&D Projects 12 footprints”) of companies and organizations, such tovoltaic panels and some other renewable electric- as the approaches described in the Greenhouse Gas ity technologies, where these technologies have Protocol Corporate Accounting Standard (GHG no GHG emissions in operation but may involve Protocol 2004) and other corporate carbon footprint substantial energy input to manufacture the com- models. The same approach is used for national ponents (Knapp and Jester 2001; Gagnon, Belanger, GHG inventories based on the IPCC Guidelines for and Uchiyama 2002). these inventories (IPCC 2006a). These methodolo- gies identify and provide tools to estimate all the This inventory approach is adopted by many of sources of emissions within a defined boundary, the companies in the power sector, including T&D whether this is a national boundary or company companies. The carbon footprint for Transpower, ownership boundary. Corporate inventory account- the national transmission utility of New Zealand, ing is not restricted to the physical boundary of the provides a useful example of project boundary set- project or company, but may also include increases ting for a corporate T&D gross emissions inventory. in emissions outside that boundary. For example, As shown in table 2.1, Transpower only considered the GHG Protocol Scope 2 emissions are from exter- fuel use within offices, owned vehicles, and sulfur nal power plants or other off-site energy production hexafluoride (SF6) in Scope 1. Technical losses for facilities that supply energy to the company, even the entire transmission system are not part of the though the power plants are not physically located at Transpower carbon footprint or gross emissions the company site. Furthermore, the GHG Protocol inventory. This is in line with guidance issued by the has a Scope 3 that can include other emissions United Kingdom’s National Grid (2008). Both com- increases upstream and downstream of the company panies argue that system technical losses should not (for example, emissions from producing the equip- be included in the utility’s carbon footprint because ment used by the company or emissions from com- this electricity is not purchased by the transmis- pany personnel traveling in vehicles not owed by the sion company, and the company cannot control the company). An important example of this in practice power generation sources, their geographical loca- has been the analysis of emissions from solar pho- tion, or the generation outputs. This may be the case Table 2.1: Project Boundary Definitions from Transpower New Zealand’s Carbon Footprint Scope 1: Direct emissions Petrol used in Transpower vehicles Diesel used in Transpower vehicles Bioethanol and biodiesel used in Transpower vehicles Diesel used in standby generators Reticulated gas in Transpower House SF6 losses from transmission equipment operation Scope 2: Electricity indirect Electricity purchased and used for Transpower’s own functions emissions Scope 3: Indirect emissions Staff business travel (taxis, rental vehicles, mileage claimed in private vehicles) and domestic and international air travel T&D losses from purchased electricity and reticulated gas used for Transpower’s own functions Office waste to landfill Electricity consumed to run lifts, common area lighting, and so on (for example, baseload electricity) in noncontrolled leased assets Source: Transpower 2009. 2. GHG Accounting Principles Relevant for T&D Projects 13 for a utility where all the technical loss improvement emissions beyond what would have happened measures that are financially viable with current anyway to receive credits for net emissions reduc- technologies and regulations have already been tions (Baumert 1999; Shrestha and Timilsina 2002). implemented. While the concept of baselines is always important for assessing a project’s net emissions impacts, how Net emissions accounting is typically used for cli- additionality is addressed is not as clear outside of mate change mitigation projects to demonstrate the carbon finance arena. This is discussed in more that they lead to a net decrease in overall national detail in the next section. emissions, even though there may be some GHG emissions associated with the project activity. Net While most projects in the energy sector will emit emissions accounting compares the total emissions GHG emissions, the net impact of the project may from the project scenario to the total emissions that be a reduction in GHG emissions if the project would have occurred in the same system without the scenario emissions are less than emissions from implementation of the project (that is, the baseline the baseline scenario. This does not mean that the scenario). All of the CDM methodologies, as well project has a negative emissions inventory, but that as projects in the voluntary carbon market, use net the total system emissions in the project scenario are emissions accounting. less than those in the baseline scenario. An example of the difference would be a gas-fired power station For projects using net emissions accounting to that emits significant GHG emissions, but that could qualify for carbon finance, both the baseline sce- have negative net emissions impact if it replaces a nario and the related concept of additionality are more carbon-intensive coal-fired power station. A critical. According to CDM rules, “the baseline for corporate emissions inventory for the utility owning a CDM project activity is the scenario that reason- this power station would still show positive emis- ably represents the anthropogenic emissions by sions, but the net emissions impact for the project sources of GHGs that would occur in the absence of investment could be negative. At the project level, the proposed project” (UNFCCC 2001). This means net emissions accounting gives a more compre- that the baseline is a hypothetical, or counterfactual, hensive picture of the impact of a project or inter- description of what would have happened without vention on overall national emissions. Therefore, project implementation (Spalding-Fecher 2002; Lee the SFDCC recommends studying net emissions and others 2005; Sharma and Shrestha 2006). This accounting approaches for World Bank–funded may or may not be similar to the current situation T&D projects. or historical emissions. For example, if the project is to replace industrial equipment with more efficient units, but the existing equipment has only one year Additionality and Net Emissions of useful life left, using the existing old equipment Accounting as the baseline for the future life of the project is Within the methodologies developed for carbon clearly not appropriate. This is also why the concept finance projects such as the CDM, there is a strong of additionality is important in the net emissions focus on tools and specific tests to prove additional- accounting methodologies used for carbon finance ity. Project proponents must justify that the project mechanisms. According to the CDM rules, “a would not have been implemented without the ben- CDM project activity is additional if anthropogenic efits of carbon financing to show that it is not part of emissions of GHGs by sources are reduced below the baseline scenario. Because the credits from these those that would have occurred in the absence of projects are used to offset emissions from other the registered CDM project activity” (UNFCCC countries or companies, without a strict additional- 2001). In other words, the project must reduce ity test, the purchasers of the credits would be emit- 2. GHG Accounting Principles Relevant for T&D Projects 14 ting more GHG emissions without compensating carbon savings, from this type of project would be for these emissions elsewhere. In the CDM frame- calculated from historical technical losses versus work, if a country purchased certified emissions technical losses after the project was implemented. reductions from a project that was not additional Lifetime energy and emissions savings would then and used these credits for compliance with their be annual savings multiplied by the economic life emissions reduction targets, they would not actu- of the new T&D equipment. But what if the exist- ally have met those targets because their emissions ing equipment was due to be replaced in any case were not offset by the business as usual CDM activ- in three years because it had reached the end of its ity (Greiner and Michaelowa 2001; Shrestha and useful life? In that case, should the baseline scenario Timilsina 2002; Tanwar 2007). itself include decreasing technical losses over time? This would reduce the calculated net impact of the Carbon finance projects evaluate additionality using project. a variety of tools and tests. The most commonly used tool is investment analysis, where project pro- The GHG Protocol for Project Accounting describes ponents provide a financial analysis of the project the typical “project-specific approach” to additional- showing that it is not viable without the revenue ity that is used in the CDM and many other carbon from the sale of carbon credits. The challenge is finance programs: how to objectively present the evidence for this financial analysis in a way that it can be audited by The project-specific approach to additionality aims a third party (Bode and Michaelowa 2003; Ellis, to identify a distinct baseline scenario specific to the project activity, in spite of subjective uncertain- Corfee-Morlot, and Winkler 2007). In practice, this ties involved in doing so. The reasoning behind this has been one of the most difficult issues to address approach is that a rigorously identified baseline in the CDM and similar programs (Ellis, Corfee- scenario is all that is necessary to establish addi- Morlot, and Winkler 2007; Schneider 2007). A tionality: if the project activity is different from its number of standard tools have been approved by the baseline scenario, it is additional. However, because CDM Executive Board, as well as guidelines on how identifying a baseline scenario always involves to apply these tools and what type of evidence may some uncertainty, many observers argue that this approach should be combined with explicit addi- be used in their application.2 tionality tests (GHG Protocol 2005a). The question of additionality does not arise for cor- The GHG Protocol also describes a second porate inventory accounting, because this approach approach, which is the “performance standard only reflects the actual emissions of the project (that approach” to additionality: is, there is no “without project” scenario). For net emissions accounting, a “without project” scenario, This is done by developing a performance standard, or baseline, is required to compare the project emis- which provides an estimate of baseline emissions sions to those that would have occurred without the that would otherwise be derived from baseline project. scenarios for each project activity. Under this approach, the presumption is that any project activ- As with all carbon accounting, additionality is inti- ity will produce additional GHG reductions if it has mately related to the selection of the baseline sce- a lower GHG emission rate than the performance nario. For example, consider a technical loss reduc- standard. A performance standard can provide a tion project that replaces substation equipment. consistent way to address additionality for a num- ber of similar project activities and avoids having Typically, the annual energy savings, and therefore to identify individual baseline scenarios. The chal- lenge is to set the performance standard at a suf- ficiently stringent level to ensure that, on balance, 2 See, for example, UNFCC 2008 and UNFCC 2010. only additional GHG reductions are quantified. 2. GHG Accounting Principles Relevant for T&D Projects 15 It is important to remember, however, that the GHG could in principle examine GHG emissions associ- Protocol for Project Accounting (GHG Protocol ated not just with inputs to the product, but also the 2005a) is designed—at least in part—to support inputs to those inputs, and so on up the product’s “value chain.” Generally, the cost and time require- projects that could generate carbon credits in the ments for this kind of analysis are prohibitive….The carbon markets outside of the CDM. As discussed secondary effects for many types of GHG projects earlier, the objective of this study is to propose can be relatively small, particularly for small proj- methodologies for T&D projects that will not be ects…. GHG project accounting requires decisions used for carbon financing or the creation of any about the trade off between accounting for second- tradeable carbon credits. World Bank–funded T&D ary effects and the time and effort required to do so (GHG Protocol 2005a). projects seek to address development objectives in the electricity sector, such as extending the coverage The World Bank’s Handbook on Economic Analysis of electricity, improving the reliability of services of Investment Operations (1996) states that choos- provided, or reducing electricity losses, among oth- ing the right project boundary for broader eco- ers. Even though the main objective is not to reduce nomic and environmental impacts of projects is emissions and receive any form of carbon credit, not always obvious, because these impacts may Bank interventions may have implications for GHG extend beyond the ownership boundaries of the emissions. In several cases, T&D projects could not project or the traditional financial analysis bound- be implemented without Bank support, at least at aries. the scale and scope defined with client counterparts in each project the Bank supports. In this sense, For T&D projects, two different dimensions must World Bank T&D investments can be said to be be considered for the physical project boundary “additional” in CDM terminology. (see figure 2.1). One is the stage of value chain for electricity supply, starting with the production Nevertheless, when developing baseline scenarios fuel for power stations, through power generation, for the T&D projects described later in this report, a T&D, and finally to consumption by the end user. key question is whether historical data are an accu- These activities, and the emissions associated with rate proxy for the baseline scenario. If technical loss them, would generally all be performed within the rates are changing dramatically (either increasing or same year.3 The most important distinction here is decreasing), it may not be appropriate to use these between impacts of generation emissions and non- for the baseline scenario against which a techni- generation emissions. In other words, T&D projects cal loss reduction project is compared. Similarly, will have impacts at the T&D value chain stage, for projects that replace T&D equipment, it may be but they will also have impacts in other value chain appropriate to limit the period over which emissions steps—particularly power generation. As discussed reductions are assessed to the remaining lifetime earlier, the explicit goal of many T&D projects is to of the equipment. This is standard practice in most affect power generation, so this category of emis- CDM baseline methodologies. sions impacts must be considered as part of the project boundary discussion. Project Boundaries and Double The second dimension is the life cycle over time of Counting all the equipment and facilities at each stage of the Setting the project boundaries is another critical ele- ment of any emissions accounting approach. As the GHG Protocol for Project Accounting notes: 3 Because power generation companies may stockpile some fuel, there will be a time delay between production of the fuel In a full “life-cycle analysis” of GHG emissions for and combustion in the power plant. This will generally not be a particular product (or project), for example, one more than a few weeks, however, for fossil fuels. 2. GHG Accounting Principles Relevant for T&D Projects 16 Figure 2.1: Sources of Electricity System Emissions: Life-Cycle Phase versus Value Chain Step Value chain Life-cycle phase Fuel supply Power T&D Consumption generation Materials Manufacture of Manufacture of Manufacture of Manufacturer of production metal, and so on metal, and so on metal, and so on materials Construct coal Construction of Construction of Construction of Construction mine and mining power lines/ factory, home, power stations equipment substations and so on Mining fossil fuel Use of power in Combustion in Transmitting Operation Transport fuel to cement factory, power plant power power plant home, school Disposal of Decommissioning substations/lines Source: Authors’ analysis. value chain. Whether it is a power station, transmis- T&D systems are only upgraded and replaced not sion line, or coal mine, all of these facilities have dismantled or removed. Whether the embodied input materials, a construction phase, an operational emissions in the materials used can be included will phase, and finally a decommissioning phase. While depend on the availability of data, and also on how these phases occur at different times—and the entire large these emissions are likely to be relative to emis- cycle may cover decades—they are all related to the sions in other phases. The potential project bound- ultimate production and delivery of electricity. Each ary for nongeneration emissions from a T&D proj- of the boxes in figure 2.1 will have GHG emissions ect is illustrated in figure 2.2. The practice of current from a variety of sources, as explained in more detail methodologies and companies in the industry of in “The Structure of T&D in World Bank Lending estimating nongeneration emissions is reviewed Operations,” page 21. in “The Structure of T&D in World Bank Lending Operations,” page 21. For calculating the nongeneration emissions impact of a T&D project, it would be ideal to include all Project Boundary for Generation Emissions of the life-cycle emissions for the T&D stage of the Impact value chain. This would include emissions related to In assessing the generation emissions impact of the manufacture of materials, as well as construc- a T&D project, the focus is on which parts of the tion and operation of the lines and substations. electricity system emissions are likely to change Decommissioning may be much more difficult to from the baseline scenario to the project scenario. estimate, and generally in developing countries, Within the nongeneration emissions project bound- 2. GHG Accounting Principles Relevant for T&D Projects 17 Figure 2.2: Potential Project Boundary for Nongeneration Emissions from T&D Projects Value chain Life-cycle phase Fuel supply Power T&D Consumption generation Materials Manufacture of Manufacture of Manufacture of Manufacture of production metal, and so on metal, and so on metal, and so on materials Construct coal Construction of Construction of Construction of Construction mine and mining power lines/ factory, home, power stations equipment substations and so on Mining fossil fuel Use of power in Combustion in Transmitting Operation Transport fuel to cement factory, power plant power power plant home, school Disposal of Decommissioning substations/lines Source: Authors’ analysis. ary defined in figure 2.2, there are no emissions stages of the value chain. The reason other life-cycle from the baseline scenario, since there would be no phases are not included for other value chain stages construction or operation at that site if the project is that these emissions are generally very small com- had not been implemented. Within those boxes, pared to emissions from operation. This is discussed therefore, the net emissions impact is based only on in more detail in chapter 5. the project scenario emissions—in other words, it is While power generation and fuel supply are clearly always an increase in GHG emissions equivalent to affected by many T&D projects, the impact on project emissions. downstream consumption is more complex. For Within the power generation subsector, however, example, if an investment in a new distribution line emissions could change significantly. As an example, and substation supplies power to a new cement discussed in more detail in the next chapters, a factory, the project scenario could include process technical loss reduction project does not have any emissions and fuel combustion emissions from that impact on emissions at the transmission or distribu- cement factory. For most T&D projects being ana- tion site, but it does reduce the amount of power lyzed by the World Bank, however, the consumer generation required to meet consumer demand. of the additional power is not specified and may Baseline emissions within power generation and be a mix of many households, business types, and fuel supply, therefore, could be significantly higher industries, so analyzing this would be very difficult. than the project scenario emissions in those boxes. In addition, there are no examples of CDM meth- Figure 2.3 illustrates the most important areas where odologies that take into consideration downstream T&D projects could affect emissions from other emissions from the consumption of a product pro- 2. GHG Accounting Principles Relevant for T&D Projects 18 Figure 2.3: Possible Impacts of T&D Projects on Generation and Other Value Chain Stages Value chain Life-cycle phase Fuel supply Power T&D Consumption generation Materials Manufacture of Manufacture of Manufacture of Manufacture of production metal, and so on metal, and so on metal, and so on materials Construct coal Construction of Construction of Construction of Construction mine and mining power lines/ factory, home, power stations equipment substations and so on Mining fossil fuel Use of power in Combustion in Transmitting Operation Transport fuel to cement factory, l t power plant power power plant home, school Disposal of Decommissioning substations/lines Source: Authors’ analysis. duced by the project activity. Since emissions from Double Counting combustion during power generation are the major For nongeneration impacts analyzed using a typical contributor to emissions from the energy sector corporate inventory approach, extending the project globally, it makes sense to focus on fuel combus- boundary from the physical T&D equipment site tion in power generation as the key value chain step to include construction and materials manufacture before the T&D operation. stages would essentially mean an overlap of emis- Based on current practice (see the following chap- sions estimates across sectors. In other words, if a ters) and the reasons explained earlier, the boundary construction company or steel tower manufacturing and review of potential impacts and methodologies company in that country also created an emissions for net impacts of T&D projects focus on combus- inventory, some of these emissions would overlap tion emissions at power plants. Downstream emis- with those that had been included in the T&D proj- sions as a result of energy consumption are not ect inventory. This is also the case for emissions taken into account. from power and heat consumption, because the emissions from power generation would also be Because the electricity supplied by a new T&D proj- attributed to the utility providing the power. ect could displace nongrid sources of energy (for example, captive/backup power or other fuels in the For assessing the generation emissions impact case of electrification), baseline and project emis- of T&D projects, there is an important overlap sions could be assessed for all of these sources (see between different projects and organizations within figure 2.4). the power sector, so it is not possible to simply sum 2. GHG Accounting Principles Relevant for T&D Projects 19 grids and one a new hydropower station in a sub- Figure 2.4: Potential Baseline and Project Emissions national grid that will now send more power over Sources for Assessing Net Emissions Impacts on the new line to another fossil fuel–dominated sub- Generation national grid. The net impact on emissions from the new transmission line could include the displace- Energy Baseline Project ment of fossil fuel power in one subnational grid by source scenario scenario hydropower in the other subnational grid, since it is emissions emissions the new line that allows this flow of power. The net emissions impact of the renewable power station, b Combustionn b Combustionn Grid power in grid in grid d however, might also be based on a baseline scenario power plant power plant of fossil fuel power if this was the predominant energy source on the existing grids. Therefore, add- Combustion Co b Combustion Co b ing the net emissions impacts of these two projects Captive in captive in captive power would overstate the total impact on national emis- power plant power plant sions, because some of the fossil fuel–fired electric- Non- Co Combustion of ombustion f Combustion of f ity savings claimed by the transmission upgrade are electrical other energy other energy also being claimed by the renewable power plant. energy sources sources Thus, while net emissions accounting is very valu- able on a project level, it could be misleading to Source: Authors’ analysis. use this approach to assess the impact of the entire World Bank lending portfolio. That said, the net emissions approach gives a much more comprehen- all of the net emissions impact for a total World sive means to assessing the overall impact of World Bank portfolio emissions impact. For example, con- Bank–funded projects, since it more accurately sider two projects funded in the same country, one reflects the impact of a given project across the a transmission line that connects two subnational entire energy sector. 2. GHG Accounting Principles Relevant for T&D Projects 20 3. Categorization of Project Types and Emissions Impacts The World Bank T&D project interventions are very priority projects from T&D utilities’ investment pro- different from traditional private sector or CDM grams. Project components and subcomponents are transactions. This section provides an overview of usually structured around the main objectives of the the diversity of the T&D projects at the Bank and lending operation (for example, increasing access their emissions impact based on the technologies and increasing transmission capacity). supported, the objectives being pursued, and the scope of the projects. Other factors play an important role in the structur- ing of lending packages. One of these is the need to perform discrete environmental or economic The Structure of T&D in World Bank analyses on site-specific investments as per World Lending Operations Bank operational procedures. For instance, a new The WBG’s lending portfolios include support for transmission line requires higher environmental and investments in a full range of electricity system com- social safeguards, whereas an existing substation has ponents: generation, transmission, and distribution. lower requirements. This usually leads to the separa- In addition to providing support for investments tion of two components for analysis during project in these areas, most operations would also include preparation and appraisal. A similar separation may components to support policy reforms, capacity be triggered by the need to analyze the economic building, and institutional strengthening. Between viability of different types of projects. Another factor fiscal 2003 and 2009, the World Bank approved that affects the definition of lending packages is co- 98 loans that had T&D components. This lend- financing. Cofinanciers, as well as the loan recipient, ing totaled $6.143 billion in 53 countries and some may have preferences for or restrictions on financing African regional projects (see annex B). certain types of projects. For instance, some finan- ciers may not be able to finance technical assistance Lending operations that support T&D investments with loan resources; others may have funds available usually support not a single project but a collection only for renewable energy projects. of projects. For instance, a lending operation could contain two components for T&D—one for a new World Bank T&D projects are therefore quite dif- transmission line in the interconnected system and ferent from typical carbon finance projects, such a second component to finance the expansion of as CDM projects, or typical private sector power several distribution substations in different areas of investments. In addition to the factors discussed the grid. Other operations support projects in all above, traditional carbon finance projects or private segments of the electricity sector. This type of opera- sector operations have clearer boundaries. A typical tion is more frequent in International Development carbon finance project could be a single wind farm, Agency countries, where pooling of resources a few minihydro projects, or a well-defined trans- among different donors is used to finance large-scale mission concession. CDM and private sector trans- investment plans. In IBRD countries, it is also com- actions are usually implemented by a single entity, mon to find loan packages that support some of the while World Bank loans may be supporting, at the 21 same time, investments implemented by different feasible and cover the effects of the variety of inter- agencies (for example, the ministry, a vertically inte- ventions that could be included in a loan. Even with- grated utility, a distribution company, and/or a rural out the inclusion of power generation, many T&D electrification agency). investments will have multiple impacts on the grid operation and therefore on GHG emissions. It is the Other aspects of World Bank operations that may objective of this work to identify methods that can factor into the applicability of existing methodolo- be used easily in the context of the loan preparation gies and the design of appropriate solutions for cycle. For this reason, and given the characteristics GHG accounting are the following: of World Bank interventions described above, the Technical diversity of projects: A project can discussion begins by categorizing projects accord- contain components at different voltages in ing to their objectives.1 The objectives largely define the system. Although all components may be the way in which projects are analyzed from the addressing a given strategy, components are ana- technical, economic, and environmental and social lyzed differently from a technical, economic, or safeguards perspectives, which provides a famil- environmental and social safeguards perspective. iar framework for project teams to analyze GHG implications. Linking project objectives to GHG Information availability: Current data availabil- implications or impacts could therefore be a suitable ity is driven by formal operational requirements, approach to feasibly start rolling out GHG account- which may affect the feasibility of some GHG ing of T&D interventions. accounting approaches. The amount and quality of data also depend on the risks each compo- nent may be facing (for example, environmen- Project Categorization by Objective tal, financial, or technical). For example, for a A review of all the World Bank loans approved from large transmission interconnector project, fairly fiscal 2003 to 2009 that included T&D components, detailed short-term and long-term load flow and as well as a review of the most frequently used indi- power system economic studies may be required cators in the results matrix of such loans, identified to appraise a project, while a substation upgrad- a set of project-level objectives tied to larger devel- ing could be assessed based on simpler data such opment goals common to the T&D portfolio. Such as substation capacity and local demand growth. project objectives can be related to specific impacts on GHG emissions, which could then be quanti- Timing and implementation readiness of invest- fied with specific GHG emissions accounting tools. ments: The subcomponents of an overall T&D Where a project has multiple objectives, multiple program may be rolled out over time, which tools or modules should be applied so all potential makes ex ante data availability a challenge. For GHG impacts can be captured. This will be particu- example, in a large rural electrification project, larly true when net emissions are assessed. it is likely that, at time of approval, only a subset (perhaps 10 percent) of the grid extension projects For the purposes of this report, projects are catego- have been identified at the level of engineering rized by the following objectives: detail required for implementation. The remain- der of the projects are designed and implemented Technical loss reduction: Reduce technical as the loan implementation progresses. losses in the transmission or distribution system so that less energy is lost between power genera- The combination and variety of World Bank projects in the electricity sector mean that the tools for esti- 1 From here on, project is used to describe the smallest compo- mating GHG impacts of T&D projects need to be nent or subcomponent in a loan whose technical and economic more comprehensive to make their implementation assessment is performed separately from the other components. 3. Categorization of Project Types and Emissions Impacts 22 tion and end users. The main impacts on GHG be greater than their nongeneration emissions. In would be the changes (reduction) in power gen- other words, although all of these projects will have eration. positive nongeneration emissions, the net impact of the project on emissions could be negative, so Increased reliability: Increase the reliability of that overall system emissions are lower after project electricity supply, so that consumers have fewer implementation. and/or shorter supply interruptions. The impact on GHG emissions could be increased grid gen- Categorization of Emissions Impacts eration and reduction of on-site (backup) power generation. The first distinction among the GHG impacts of T&D projects is nongeneration versus generation Distribution capacity expansion: Increase the impacts. The emissions at the physical T&D project overall capacity to distribute electricity, so that site do not have a corresponding baseline, since additional power generation can be supplied those activities would not have occurred without the to existing growing demand. An impact of this project. Assessing power generation impacts, on the objective would be an increase of grid genera- other hand, requires the development of a baseline tion, with displacement of other power sources. scenario to estimate the change in emissions from Electrification: Connect new consumers to the power generation plants before and after the project grid, thereby displacing other sources of electric- is implemented. ity (or even nonelectric energy sources). Within generation impacts, there is an important Transmission capacity expansion: Increase distinction between the different project categories the overall capacity to transmit electricity over and how these affect generation emissions outside significant distances, so that additional power the physical boundary of the transmission system. generation can reach different areas of the trans- Affecting generation output is one of the main mission system, such as distribution centers. This objectives of technical loss reduction, but this does would increase power generation and potentially not require direct actions to increase or decrease displace other power sources. generation output by generators. In other words, if a technical loss reduction project brings electricity Cross-border trade: Increase electricity trade production down by 2–3 percent while delivering between countries by constructing interconnec- the same amount of power to end users, no addi- tors between their national grids. This could also tional action is needed by the power generation occur within a single country, if two major grids subsector to achieve this reduction in energy use that were previously not connected can now and emissions. For increased reliability, capacity trade power through a new transmission line. expansion, electrification, and cross-border trade, This classification is not intended to imply that a however, these T&D projects will deliver more elec- particular project pursues only a single specific tricity to consumers and require actions by other objective. It is possible that a capacity expansion parts of the power sector. These actions could be project, for example, could have an impact on reli- additional investment or a change in operations, as ability, or that a technical loss reduction project in the reduced operation of backup power genera- could also be improving electricity access. tors after an increased reliability project is imple- mented. There will be cases where, in the short term, As discussed in the previous chapter, when assessing excess capacity in existing plants allows an increase the net emissions impacts of these different project in power generation without investment, but other types, one of the important features of T&D projects changes may be required such as dispatch rules. In is that their impact on generation emissions may the long term, a transmission interconnection can 3. Categorization of Project Types and Emissions Impacts 23 have important impacts on generation investment, power generation, changes in dispatch, or changes but actions in generation investment cannot be in the operation of nongrid generators or energy directly triggered by the transmission line. Without sources. new power generation, which will be built if many These definitions introduce an important distinc- other conditions exist, there would be no reason for tion in how T&D investments affect power genera- investment in large T&D systems. More importantly, tion. For instance, an international interconnection projects other than technical loss reduction have project could have impacts in power generation over impacts through displacing other power genera- the short and long terms. In the short term, existing tion sources outside the grid. This makes for some cleaner and cheaper power generation in one system uncertainty about the baseline, since the alternative could displace more polluting power generation in energy source must be identified to assess the net the other. This will not happen immediately, because emissions impacts. the generators will have to agree to new integrated Thus, two categories of GHG impacts on power gen- dispatch rules or other forms of dispatch coordina- eration by T&D projects are distinguished: direct tion. In the long term, the integrated market will generation effects and indirect generation effects. lead to an increase in generation capacity and effi- The description of the three categories of emissions ciency. However, for these new investments to mate- impacts is presented in table 3.1. The remainder of rialize, other financial, legal, and regulatory condi- this report is structured around these three catego- tions are required, which are outside the control of ries of emissions impacts. the T&D project investors and operators. In these definitions, the physical boundary of the Since indirect impacts will occur only if these other T&D project (as opposed to the boundary in terms actions take place, these emissions are not fully of emissions sources) consists of the physical site(s) attributable to the project, although the project con- where the T&D project will be constructed. An tributes to these emissions reductions or increases. example would be substations, transmission lines, Direct emissions can be attributed to the project. All and the right-of-way corridor for a transmission impacts are analyzed over the same project life used expansion project. Actions outside the physical in the technical and economic analysis performed boundary of the project could include investment in during the Bank’s project appraisal. Table 3.1: Categories of T&D Project Impacts on GHG Emissions Used in This Study Category of emissions impact Description Direct nongeneration effects Similar to standard corporate or national inventory. Emissions that occur within the physical boundary of T&D project, and possibly through the life cycle of that equipment. Direct generation effects Effect on short-term and/or long-term generation emissions that does not require any other actions outside the physical boundary of the T&D project. This would be the case for technical loss reduction projects. Indirect generation effects Effect on short-term and/or long-term generation emissions that requires actions outside the physical boundary of the T&D project. This would be the case for increased reliability, capacity expansion, electrifi- cation and cross-border trade. Source: Authors’ analysis. 3. Categorization of Project Types and Emissions Impacts 24 Relevant GHG Methodologies their objectives may differ (that is, corporate report- Reviewed ing or crediting), they provide important information on possible alternatives for the World Bank’s GHG Numerous methodologies, reports, and studies emissions accounting. These approaches are listed in address the GHG impacts of T&D projects. Although tables 3.2 and 3.3 and discussed in chapters 4 and 5. Table 3.2: GHG Measurement Methodologies for the Direct Nongeneration Emissions Impacts of T&D Projects T&D guidelines 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Vol. 3, Ch. 8.2 Emissions of SF6 and PFCs [perfluorocarbons] from electrical equipment (IPCC 2006c) Tools applied IFC Carbon Emissions Estimator Tool (IFC 2009) within the WBG Tools applied to Transpower (New Zealand) carbon footprint (Transpower 2009) power genera- Life Cycle Assessment of Aluminium Smelter in Greenland (Schmidt and Thrane 2009) tion, transmis- (uses “EcoInvent” as the source for T&D) sion, and dis- “Eco-Balance of a Solar Electricity Transmission from North Africa to Europe” (May tribution case 2005) studies Life Cycle Inventories of Energy Systems: Results for Current Systems in Switzerland and Other UTCE Countries (“EcoInvent”) (Dones and others 2007) Emissions of GHGs from the Use of Transportation Fuels and Electricity, Argonne National Laboratory (DeLuchi 1991) Source: Authors’ analysis. Table 3.3: GHG Measurement Methodologies for the Generation Emissions Impacts of T&D Projects Power sector GHG Protocol for Project Accounting (GHG Protocol 2005a) guidelines Guidelines for Quantifying GHG Reductions from Grid-Connected Electricity Projects (GHG Protocol 2007) Greenhouse Gas Assessment Handbook (World Bank 1998), Ch. 3.6.2 Guidelines for Energy Conversion and Distribution Projects Manual for Calculating GHG Benefits of GEF Projects: Energy Efficiency and Renewable Energy Projects (GEF 2008)a CDM baseline AMS II.A “Supply-side Energy Efficiency Improvements—Transmission and Distribution” and monitoring (ver10) methodologiesb AM0035 “SF6 Emission Reductions in Electrical Grids” (ver01) AM0045 “Grid Connection of Isolated Electricity Systems” (ver02) AM0067 “Methodology for Installation of Energy Efficient Transformers in a Power Distribution Grid” (ver02) AM0079 “Recovery of SF6 from Gas Insulated Electrical Equipment in Testing Facilities” (ver01) NM0272 “International Interconnection for Electric Energy Exchange” NM0269 “Reduction of Emissions through One Way Export of Power from Lower to Higher Emission Factor Electricity System” Source: Authors’ analysis. a The GEF manual does not cover T&D projects, but only investments in new renewable power and energy efficiency. b Approved methodologies can all be accessed at cdm.unfccc.int/methodologies/PAmethodologies/approved.html so no fur- ther reference is provided in this document. 3. Categorization of Project Types and Emissions Impacts 25 4. Direct Nongeneration Impacts of T&D Projects This chapter discusses each of the possible emissions Embodied Emissions in Construction impacts that should be included in the direct non- Materials generation impacts, for which type of investments they may be relevant, and how existing methodolo- The construction of T&D projects consumes large gies address these impacts. The direct nongenera- quantities of aluminum, concrete, other metals, and tion effects of T&D projects are based on emissions other building materials. All of these materials have sources in the construction and operation of the embodied emissions as a result of the energy used T&D system, as shown in figure 4.1. to produce them, meaning that the implementation of new T&D projects creates some upstream emis- sions in the manufacture of the materials used. The issue is whether the magnitude of emissions is likely to be great enough to merit the time and effort to Figure 4.1: Potential Emissions Sources for Direct calculate them. Nongeneration Emissions from T&D Projects Review of Existing Methodologies Life-cycle T&D International Finance Corporation Carbon phase emissions Emissions Estimator Tool (IFC CEET): This tool sources includes a section on embodied emissions from Manufacturer of construction materials, such as metals, composite Materials materials for lines, materials, plastics, and miscellaneous equipment. production substations, and so on The project proponent must supply the total quanti- ties of materials used, and the tool provides a table Land clearing of default embodied emission factors taken from the Construction Energy use in con- Agence Française de Développement’s “Première struction analyse des émissions des projets AFD.” SF6 fugitive EcoInvent: This is not a methodology, but rather a Operation emissions database that includes a variety of environmental Corona discharge impacts from the energy sector. The EcoInvent data- base includes T&D infrastructure requirements such as metal and wood, but does not appear to include SF6 fugitive Decommissioning emissions the embodied emissions in these materials. The database covers only European energy systems, so it reflects the power generation mix, T&D system Source: Authors’ analysis. characteristics, and material availability for Europe only. 27 May (2005): This analysis of life-cycle environmen- tal impacts of transmitting solar power from North Figure 4.2: Life-Cycle GHG Emissions for Long-Distance Africa to Europe includes embodied emissions in Transmission of Solar Power for North Africa to Europe materials based on the Umberto material flow soft- g CO2/kWh ware, using input data from the manufacturer of 20 the lines (ABB), supplemented with the Umberto Construction engineering and EcoInvent databases and other secondary 15 sources. Because this case study analysis was based on European-sourced materials, the EcoInvent 10 European energy database was appropriate for elec- tricity and other energy sources. Materials for the Disposal– high-voltage DC lines account for 0.4–0.6 kg CO2e/ 5 plant MWh, while operation of the line (ohmic resistance losses only) is 0.8–1.5 kg CO2e/MWh (see fig- 0 Line 1 Line 2 Line 3 ure 4.2). Ship transport The case study of a long-distance transmission line Truck transport Operation–HVDC between Ethiopia and Kenya presented in box 4.1, Overhead line based on the feasibility study report (FSR) for this Operation–plant Storage investment, shows that embodied emissions in a Submarine cable long-distance transmission line such as this are Solar field much less than 1 percent of typical fossil fuel power Source: May 2005. station combustion emissions. Box 4.1: Example of Embodied Emissions in Long-Distance Transmission Line The embodied emissions of materials are most likely to be significant in T&D projects that involve extensive infrastructure relative to the amount of power delivered, such as long-distance transmission lines. In addition, for projects with long line lengths, the materials in the lines will far outweigh the materials in substations and other equipment. An example of this is the Ethiopia Kenya power systems interconnection project (see chap- ter 7 for more detail). This project involves 1,200 km of double 772 mm2 line. The weight of this line, accord- ing to the manufacturer, is 1.91 t aluminum and 0.68 t steel per kilometer (Sural 2010). This amounts to 4,575 t aluminum and 1,628 t steel for the entire line. For embodied emission factors, the Global Emission Model of Integrated Systems database (Öko Institute for Applied Ecology 2009) provides 14.5 tCO2e/t aluminium (Germany) and 1.6 tCO2e/t steel (mix of electric arc furnace [EAF] and basic oxygen furnace [BOF] processes, Germany). This yields total emissions of 68,294 tCO2e. Over the lifetime of the line (2012–27), the projected electricity transmitted is 106,672 GWh (Fichtner 2008). Thus, embodied emissions are 0.64 kg CO2e/MWh. Given that fossil fuel power sources typically have emissions of 600–1,100 kg CO2e/MWh, this is one-tenth of 1 percent of those emissions. 4. Direct Nongeneration Impacts of T&D Projects 28 Assessment of Available Methodologies sions, because it is at the project site, even though it occurs before the actual operation of the T&D Calculating embodied emissions is straightforward project. This source of emissions is likely to be very if the underlying data for materials consumption small compared to the lifetime energy and emissions and emission factors are available. The challenge impacts of the T&D project. is that most T&D project appraisals would not contain a detailed materials inventory, since this is Review of Existing Methodologies only developed by a quantity surveyor after detailed design studies are complete—which would be well IFC CEET: This tool includes an equation for emis- after loan approval. sions from fuel consumption in mobile vehicles during construction. The project proponent must More importantly, the embodied emission factors supply the quantities of fuel used, and the tool pro- for materials are highly dependent on their source. vides a table of default calorific values and carbon For example, steel manufactured in Brazil will have emission factors taken from IPCC and the GHG much lower embodied emissions than steel manu- Protocol. factured in South Africa, since the grid emission factor in Brazil is almost 90 percent lower than in DeLuchi (1991): This comprehensive assess- South Africa (0.1 versus 1.0 tCO2/MWh). ment of electricity sector life-cycle environmental impacts from the Argonne National Laboratory Creating a database for embodied emissions of in the United States finds that emissions from the materials would clearly be beyond the scope of this construction of power plants, which are also highly report, or of most carbon accounting methodolo- material intensive, are equivalent to 3–5 kg CO2e/ gies, because of the complexity of life-cycle issues. MWh (table 13, p. 50), but does not include these in If this source is to be included, the emission factors the emissions from power stations. The study does must come from existing, reputable databases. The not provide a similar figure for T&D investments, databases for embodied emissions—for example, because it does not include construction emissions EcoInvent, the Global Emission Model of Integrated from T&D systems. Systems (Öko Institute for Applied Ecology 2009), the Inventory of Carbon & Energy (Hammond and May (2005): See page 28. Jones 2006)—generally focus on Europe, and so Assessment of Available Methodologies would need to be modified for materials sourced in developing countries. The methodological approach to construction emis- sions is straightforward, but calculating this source Where T&D projects are of sufficient scale to merit is only possible if the underlying data are readily the necessary data collection, this area of direct non- available, particularly data on the quantities of fuel generation emissions could be considered, but it will consumed by construction vehicles. Data for fuel not be possible for the majority of projects without calorific values and emission factors are available significant additional time and cost. from IPCC and GHG Protocol, but the quantity of fuel must come from the project documents. This Energy Use in Construction information is not something that is evaluated even during the detailed design phase of T&D projects. There is on-site energy use in the actual construc- tion of a T&D project, primarily in the form of transport fuel for construction vehicles and the ship- Land Clearing ping of components. This energy use could be con- New construction of long-distance lines, or even sidered a component of direct nongeneration emis- of distribution lines and substations, may affect 4. Direct Nongeneration Impacts of T&D Projects 29 carbon stored in biomass and soil. An obvious delivered in the first seven years, as the electrifica- example would be clearing forest for a long-distance tion program is rolled out, is 212,576 MWh, or transmission line, which would result in a one-time 50 kg CO2/MWh. If the last year of the program is release of the carbon stored in the vegetation. This used to approximate the ongoing delivery of power, impact would be common for new transmission the 20-year total would be 2,976,062 MWh, and the investments in areas with high forest cover, and pos- land clearing emissions would be 13 kg CO2e/MWh. sibly for electrification and distribution projects that Note that, unlike ongoing emissions such as SF6 or involve new feeder lines, but is unlikely to be impor- corona discharge, land clearing emissions per MWh tant for upgrading of T&D equipment to reduce are sensitive to the economic life used for the assess- losses and increase reliability. Some of the biomass ment of the T&D project. would grow back after line construction, although the amount and density would depend on the cli- IFC CEET: This tool was developed for any invest- mate and maintenance procedures for the line, as ment project undertaken by the IFC, and so covers well as on how high the line is. many sectors. It includes a section on land clearing that can be applied for any project type. Land clear- Review of Existing Methodologies ing emissions are the product of area cleared and AM45: Leakage emissions for electrification projects biomass density (above and below ground). The include emissions from transmission line construc- tool also includes a table of emission factors (above- tion.1 Leakages related to deforestation in the con- ground and below-ground biomass density) for a struction of interconnection lines are calculated as large variety of vegetation types, sourced from the follows: 2006 IPCC Guidelines for National Greenhouse Gas Inventories. LE1 = Adef × LC Assessment of Available Methodologies Where The data required for this component are land LE1 = Leakage emissions to be accounted for in the area cleared and the carbon content of the biomass first year of the project crediting period cleared. The land area cleared is directly proportional Adef = area of land deforested, in ha to line length, which would be reported in all project LC = carbon stock per unit area (above ground, documents. The default right of way required is not below ground, soil carbon, litter, and dead specified in the methodologies reviewed, because this biomass), in t of CO2 per ha can be dependent on infrastructure type. Right of ways for transmission lines can range from 150 to 200 This approach is also used by other proposed base- feet for 340–700 kV lines and from 60 to 150 feet for line methodologies, as well as the proposal from the 69–330 kV lines. Applying the biomass density from Methodologies Panel of the CDM Executive Board the IFC CEET requires an unambiguous definition of on transmission lines for cross-border trade. The the land type to be cleared. Because this is not always only registered project design document for AM45 given in the project documentation, it will be a source is “Celtins and Cemat Grid Connection of Isolated of uncertainty unless project proponents can provide Systems.” This Brazilian electrification program additional information. estimated land clearing emissions of 39,150 tCO2 in the state of Mato Grosso. The total electricity SF6 Fugitive Emissions Sulfur hexafluoride is used in insulation and cur- 1 The term leakage in the CDM rules refers to emissions rent interruption applications in both T&D systems impacts outside the defined project boundary. (IPCC 2006c). SF6 is used in gas-insulated switch- 4. Direct Nongeneration Impacts of T&D Projects 30 gear and substations, gas circuit breakers, and—less from electrical equipment. The Tier 1 approach, frequently—in high-voltage gas-insulated lines. which is the simplest, estimates emissions by mul- SF6 may escape as fugitive emissions during the tiplying default regional emission factors (provided manufacturing, installation, use, maintenance, and in the guidelines) by SF6 consumption by equipment disposal of this equipment. Distribution equipment manufacturers and/or the nameplate SF6 capacity of that is sealed may not emit any SF6 during use, but equipment at each life-cycle stage beyond manufac- transmission equipment often requires periodic turing in the country (see table 4.1). refilling and so has higher fugitive emissions during use. The amount of SF6 emissions during opera- While this is done at a national level in the IPCC tion and decommissioning is related to the num- Guidelines, the same principles could be applied at ber and type of equipment used, as well as to the a project or utility level. In other words, emissions maintenance and recycling procedures. This source could be estimated by multiplying nameplate capac- of emissions could occur in all project categories, ity of all equipment in use by the appropriate manu- depending on the type of equipment installed, refur- facturing, installation, use, and disposal emission bished, or maintained. factors. The magnitude of SF6 eemissions depends on The Tier 2 approach under the IPCC Guidelines is what equipment is used, how it is maintained, and the same as Tier 1, but the emission factors used operational factors. At a national level, countries must be country specific. In addition, there is a term report SF6 emissions from the power sector in their to include the SF6 recovery in retirement and dis- national emissions inventories, so this provides one posal. The Tier 3 method is a hybrid of emission fac- approach for estimating their magnitude. tor and mass balance approaches that can be imple- mented at a facility/utility/project level, and includes Review of Existing Methodologies separate equations for each stage of the equipment IPCC Guidelines: The 2006 IPCC Guidelines for life cycle. Depending on data availability, mass bal- National Greenhouse Gas Inventories provides three ance approaches may be used for some stages and approaches to estimating SF6 fugitive emissions emission factor approaches may be used for others. Table 4.1: IPCC Default Emission Factors for T&D Equipment Manufacturing Use/operation Disposal % consumed by %/year of nameplate % charge remain- Type of equipment Country manufacturers capacity losses ing at retirement Sealed-pressure SF6- Europe 7 0.2 93 containing equipment Japan 29 0.7 95 Closed-pressure SF6- Europe 8.5 2.6 95 containing equipment Japan 29 0.7 95 United n.a. 14 (including installation) Included in use States Gas-insulated transformers Japan 29 0.7 95 Source: IPCC 2006c. Note: n.a. = not applicable. 4. Direct Nongeneration Impacts of T&D Projects 31 AM35: The approved CDM baseline methodol- capacity of 1–2 percent, except for the United ogy AM35 “SF6 Emissions Reductions in Electrical Kingdom and Ireland, which are both reported Grids” provides a detailed utility-level accounting to have loss rates of 4 percent (Transpower 2009, for SF6 fugitive emissions based on a mass balance table 15.4). Note again that this database covers only approach. For both project and baseline emissions, European energy systems. the mass balance considers decreases in inventory, additions to inventory, subtractions or remov- Wartmann and Harnisch (2005): This study on als from inventory, retirement of SF6-containing reducing SF6 emissions reports typical quantities of equipment, and new SF6-containing equipment SF6 in different equipment types and the electrical purchased. Baseline emissions are from the mass capacity of that equipment in Europe (see table 4.2). balance of the last three years, while project emis- The study notes that the most important sources of sions are from monitored changes in the mass bal- emissions in the future will be sealed-pressure and ance in the relevant areas. To apply this approach to closed-pressure equipment. a project, the project must have its own dedicated May (2005): See page 28. This is based on inventory of SF6 cylinders, and purchases and dis- European energy systems. bursements of those cylinders must only be for the project. U.S. EPA (2006): This report is mentioned here because it estimated the total SF6 emissions from the AM79: While AM79 addresses SF6 emissions from power sector by country and region throughout the T&D equipment, it applies only to gas recovery proj- world. The estimate includes all T&D components, ects implemented at a site for testing gas-insulated as well as SF6 from manufacturing and disposal of electrical equipment. Because it deals only with the T&D equipment. Comparing these data to electric- recovery and reclamation of gas, it is not relevant for ity supply in selected countries, the emission fac- establishing direct nongeneration emissions from a tor for developing countries appears to be 2–3 kg new T&D project. CO2e/MWh; for industrial countries, it is less than AM45: Fugitive emissions are the product of the 1 kg CO2/MWh (see table 4.3). Note that the projec- quantity of SF6 leaks in equipment and the global tions for developing countries are based on electric- warming potential of SF6. The quantity of leaks is ity supply growth. determined using information from the equipment manufacturer and/or the quantity of SF6 injected Assessment of Available Methodologies into the equipment each year during routine main- Most of the methodologies reviewed rely on detailed tenance. data collection from the project proponent. If a proposed T&D project has a detailed projected SF6 Transpower: SF6 fugitive emissions are calculated inventory or list of SF6-containing equipment along using a mass balance approach. In other words, SF6 with capacity ratings, estimating fugitive emissions purchases less stock changes and disposal/recovery is relatively simple using default fugitive emissions is equal to the amount that must have been emit- rates from the IPCC, the Wartmann and Harnisch ted into the atmosphere. Transpower’s reported SF6 study, or a similar source. The leakage rates for the emissions in 2008/9 were 7,409 tCO2e. Based on closed-pressure equipment used in high-voltage the energy transmitted that year, 38,816 GWh, this lines are reported fairly consistently at 1–3 percent would be an emission factor of 0.19 kg CO2/MWh. per year across several sources. For the sealed-pres- EcoInvent: SF6 emissions from T&D aare included sure equipment used at lower voltages, leakage rates in the database as part of the life-cycle assessment, would be much lower; thus, these make a much based on yearly percentage losses from installed smaller contribution to total sector emissions. 4. Direct Nongeneration Impacts of T&D Projects 32 Table 4.2: Characteristics of SF6-Containing T&D Equipment Power SF6 capacity Average annual Share of EU SF6 Type of equipment rating (kg) emissions rate (%) emissions (%) Sealed pressure SF6-containing 1–52 kV 0.25–10 0.14–0.24 14 equipment Closed pressure SF6-containing > 52 kV 3–200 1.8 73 equipment Gas-insulated transfomers n.a. n.a. n.a. 0.2 T&D component manufacturing >1 kV <1% mass of n.a. 8 product High-performance power capacitors 1–5 kV n.a. n.a. 5 Source: Wartmann and Harnisch 2005. Note: High-performance power capacitors are mainly used in trains. Emissions from T&D components will be reduced by over 70 percent in 2010. EU = European Union; n.a. = not applicable. Table 4.3: SF6 Fugitive Emissions from the Power Sector in Selected Countries SF6 emissions, 2005 Domestic supply of Emission factor Country (MtCO2e) electricity (GWh) (kg CO2e/MWh) Brazil 1.37 483,974 2.83 China 6.79 3,268,918 2.08 India 2.00 808,153 2.47 South Africa 0.76 260,580 2.92 Africa total 1.52 621,206 2.45 United Kingdom 0.36 401,358 0.90 Switzerland 0.06 65,888 0.91 Germany 0.24 620,545 0.39 Sources: SF6 emissions, U.S. EPA 2006; electricity supply, IEA 2007. If the SF6 content is not known, or if there is no containing equipment. Nevertheless, some default detailed inventory of what type of equipment will be factor may be the only alternative where the project installed, the alternative is to use sectorwide default proponents do not have access to detailed data on the factors. These could present some challenges, because SF6-containing equipment to be installed. they combine higher-emitting, high-voltage equip- ment with lower-emitting, low-voltage equipment. It Note that the methodologies that consider SF6 emis- would be important to establish whether a particular sions from new investments implicitly assume that project was, in fact, installing equipment that contains all the SF6 fugitive emissions along the lines should SF6, because not all distribution projects will use SF6- be allocated to the T&D project. While this makes 4. Direct Nongeneration Impacts of T&D Projects 33 sense for T&D capacity expansion (new lines), interconnectors for cross-border trade, and electri- Box 4.2: The Corona Effect fication, it may not be appropriate for technical loss Corona is a phenomenon associated with all ener- reduction and increased reliability projects. A tech- gized transmission lines. Under certain condi- nical loss reduction project, for example, is unlikely tions, the localized electric field near an energized to replace all of the SF6-containing equipment in a T&D system. Strictly speaking, only the equipment conductor can be sufficiently concentrated to altered by the project would be part of the physi- produce a tiny electric discharge that can ionize cal project boundary, since SF6 emissions for other air close to the conductors. This partial discharge equipment would have existed both before and after of electrical energy is called corona discharge, project implementation. or corona. Several factors, including conductor voltage, shape, and diameter, and surface irregu- N2O Emissions from Corona larities such as scratches, nicks, dust, or water Discharge drops, can affect a conductor’s electrical surface gradient and its corona performance. Corona is High-voltage transmission lines can create nitrous the physical manifestation of energy loss, and oxide (N2O) from an effect called “corona discharge” can transform discharged energy into very small (see box 4.2).2 They are only present on the highest amounts of sound, radio noise, heat, and chemi- voltage lines, and thus would not be applicable to distribution investments or many transmission lines. cal reactions of the air components. Corona is well understood by engineers, and Review of Existing Methodologies steps to minimize it are a major element in the May (2005): This study notes that production rates design of extra high-voltage transmission lines are heavily dependent on weather conditions and (345–765 kV). Corona is usually not a design issue are basically higher in case of a high-voltage DC line for power lines rated at 230 kV and lower. Corona because of the formation of a space charge cloud. activity on electrical conductors surrounded by air Reported emissions for long-distance transmis- can produce very tiny amounts of gaseous efflu- sion lines are less than 1 kg CO2e/MWh from the ents: ozone and nitrogen oxides (including N2O). actual discharges of N2O and are not included in the life-cycle analysis results. May also states that, in Gaseous effluents can be produced by corona terms of load losses, “in the annual mean the corona activity on high-voltage transmission line electrical losses amount to approximately 2–3 kW/km for a conductors during rain or fog conditions, and can 400 kV system… [Earlier research] states 1–10 kW/ occur for any configuration or location. km for a 380 kV system and 2–60 kW/km for a 750 Source: CPUC 2005. kV system that strongly depends on the respective atmospheric conditions and can be neglected in this order of magnitude.” emissions from the U.S. electric power system in 1980 divided by the total power generation for that DeLuchi (1991): DeLuchi includes corona discharge year. DeLuchi notes that “fortunately, the most-likely of 3 kg CO2e/MWh for high-voltage transmission estimate is so small that it does not matter if it is lines, but states that this could be significantly larger. included in the total of greenhouse-gas emissions This result is based on a 1984 estimate of total N2O from electricity generation and use…. However, the maximum estimate of 61 g/kWh is of the same order 2 http://en.wikipedia.org/wiki/Corona_discharge; of magnitude as emissions from the nuclear-fuel www.archive.org/details/dielectricphenom028893mbp. cycle and, hence, cannot be ignored.” 4. Direct Nongeneration Impacts of T&D Projects 34 EcoInvent: The EcoInvent report (Dones and oth- influenced by local conditions. Unlike the other ers 2007) that explains the contents of the database direct nongeneration emissions impacts discussed, notes that “N2O emissions of the electricity high- there is no linear relationship between N2O corona voltage transmission due to corona effect are 5 kg emissions and other activity levels or T&D proj- CO2e/GWh. No country specific data are available.” ect specifications. Emissions will depend on the The global warming potential of N2O is 210, so this exact shapes and configuration of equipment, local is equivalent to 1.05 kg CO2e/MWh. weather conditions, and installation and mainte- While all three of these sources report average nance procedures. The effect is permanent only corona discharge emissions per megawatt-hour, under extreme design flaws and the right atmo- they also note that these emissions are not directly spheric conditions, and tends to be momentary proportional to electricity transmitted. Corona dis- (some days during the year) and transitory (a few charge depends on a variety of site-specific factors, seconds during over voltage conditions). from voltage levels to the specific technical charac- teristics and shape of components. Summary of Direct Nongeneration Assessment of Available Methodologies Emissions Impacts Very limited data are available on corona discharge Table 4.4 summarizes which direct nongeneration N2O emissions levels, or on how these levels are emissions sources are covered by the different meth- Table 4.4: Inclusion of Different Emissions Sources in Direct Nongeneration Emissions Methodologies and Case Studies Embodied Energy in N2O corona Source emissions construction Land clearing SF6 discharge Typical values <1 Not known Highly variable but > 0.2–3.0 1–3 (kg CO2e/MWh) 10 possible Studies addressing only direct nongeneration emissions IPCC N N N Y N IFC CEET Y Y Y Y N Transpower N N N Y N Aluminium smelter N N N Y N May (2005) Y Y N N Y EcoInvent ? ? N Y Y DeLuchi (1991) N N N Y Y Studies addressing impacts of generation emissions GHG Protocol–Electricity N N N N N AM45/NM0269 N N Y Y N AM35 N N N Y N Source: Authors’ analysis. Notes: N = not included in direct nongeneration emissions from T&D; Y = included in direct nongeneration emissions from T&D; ? = insufficient detail in report to determine if this source is included. 4. Direct Nongeneration Impacts of T&D Projects 35 odologies reviewed. It also includes some of the tion emissions, with most sources accounting for data on the magnitude of these sources, although 1–5 percent of the emissions of a typical fossil there were very few sources for these data. Given fuel–fired power station. The exception could be that typical oil and coal power stations would have land clearing, in areas where vegetation is dense, life-cycle emissions of 870–1,335 kg CO2/MWh required right of way is large, and lines are long (DeLuchi 1991), all the sources discussed here are relative to the power transmitted. likely to be less than 1 percent of power generation emissions, although land clearing is highly variable Accurate estimates of most of the direct non- and depends on local land conditions. generation components depend on having detailed activity data about the project which Box 4.3 illustrates the direct nongeneration emis- may not be collected in the normal project prep- sions sources discussed in this chapter, using a aration process. Examples include a construction hypothetical transmission investment. materials inventory, energy use by construc- tion vehicles, and SF6 nameplate capacity for In terms of project boundary and coverage of all equipment. The feasibility of including these sources, the review here suggests several conclusions components in the emissions inventory must be on direct nongeneration emissions impacts: assessed on a case-by-case basis. Supplementary Overall direct nongeneration emissions impacts data requests to project proponents may be are likely to be a small share of power genera- required for estimating these emissions sources. Box 4.3: Example of Direct Nongeneration Emissions from a Typical Transmission Project To illustrate the possible magnitude of direct nongeneration emissions from different T&D projects, consider a high-voltage transmission line that is 1,000 km long and has two side-by-side 500 kV lines. Over 20 years, the average flow of electricity is 6,944 GWh/year, for 138,898 GWh in total over the life of the project. The right of way is 60 m, and the land area cleared is 7,200 ha. If the line went through natural tropical forest, which has the highest biomass density (374 tCO2/ha above- and below-ground biomass), the total emissions from land clearing would be 2,693,800 tCO2. This is 19 kg CO2/MWh. Assuming that high-voltage equipment accounts for 75 percent of the SF6 fugitive emissions from T&D, the average emissions of SF6 for this type of line in Africa would be 1.52 MtCO2e SF6 from the African power sector divided by 621,206 GWh electricity supply multiplied by 75 percent, or 1.84 kg CO2/MWh. For embodied emissions, taking the line itself as the largest material component, and assuming 1.91 t alu- minium and 0.68 t steel/km of line (based on manufacturer specifications), total materials would be 4,575 t alu- minium and 1,628 t steel. Using embodied carbon factors of 8.2 tCO2e/t aluminium and 2.8 tCO2e/t steel, this is a total of 42,308 tCO2e, or 0.30 kg CO2/MWh. As discussed above, corona discharge is highly uncertain, but could be on the order of 1–3 kg CO2e/MWh. There are no data available on energy use in construction to estimate that component. Thus, the maximum total direct nongeneration emissions would be on the order of 25 kg CO2/MWh. However, if there was no land clearing, or the land was previously cropland or grassland, this figure would fall to 6 kg CO2e/MWh. 4. Direct Nongeneration Impacts of T&D Projects 36 Corona discharge is more complex, but it is not is particularly challenging because loan applica- directly related to traditional T&D project speci- tions are evaluated before the detailed engineer- fications, instead depending on many other local ing design studies that might include some of conditions and detailed manufacturing design this information have been completed. specifications. Including corona discharge with For SF6, in cases where detailed nameplate capac- any level of accuracy would therefore be very dif- ity data (or at least electrical capacity data) for ficult. new equipment are not available, estimating Upstream emissions from embodied emissions emissions accurately will be much more difficult. and energy use in construction are rarely covered National average emission factors would need to by any of the methodologies. They would form be allocated to high-, medium-, and low-voltage part of a more complete life-cycle analysis, which systems. In addition, not all projects will include is illustrated in two of the case studies. The fea- the installation of new SF6-containing equip- sibility of including this type of analysis in a ment, and projects that do not build new lines simple T&D project analysis tool is questionable, (for example, technical loss reduction) will not given the additional time and cost that would be affect SF6 emissions for all the existing equip- required to gather the data. This data collection ment. 4. Direct Nongeneration Impacts of T&D Projects 37 5. Generation Emissions Impacts of T&D Projects To assess the net impact on emissions from T&D and decreased off-grid generation. The focus of this projects, the impacts outside the direct nongenera- chapter is to understand how T&D projects that fit tion emissions project boundary must be assessed. the principal categories discussed earlier are likely to As discussed earlier, the most important net effects affect GHG emissions in power generation, and how are the impacts of T&D projects on the operation this has been quantified. of power generation plants, both grid-connected and captive/off-grid. Assessing these effects on Technical Loss Reduction power generation requires a net emissions approach because the change in emissions from the power The most common positive impact of T&D projects, generation system is the difference between the particularly upgrades or renovations of existing emissions from all power stations after the T&D T&D systems, is the reduction in technical losses investment (project scenario) versus what the total within the entire electricity grid system (see, for emissions from power generation would have been example, World Bank 2004 and 2008b). By upgrad- without the T&D investment (baseline scenario). ing transformers and other substation components, Each of the project categories discussed in chapter 3 performing additional maintenance, adding reactive will have different impacts on power generation. power, or other interventions, these project types result in lower technical losses, so that more of the As discussed earlier, direct generation effects are generated power is delivered to the consumer. This where a T&D investment reduces or increases power is true for both transmission projects and distribu- generation without requiring any other actions out- tion projects that are implemented within existing side the physical boundary of the project. In other systems, including a new transmission line added words, the investor in the T&D project determines along an existing line. the emissions impacts, without the need for action by any other actor. Indirect generation effects, by In a typical economic and financial analysis of a T&D contrast, require some action outside of the T&D upgrade project, one of the main sources of revenue project, either in terms of investment in power gen- would be reduced cost of power generation (or pur- eration or changing the operation of power genera- chased power) as a result of lower losses. If electricity tion plants (grid or off-grid). sales remain constant after the T&D upgrade and less power generation is required to deliver this electricity, All the project categories discussed in this chapter emissions from the operation of power plants on the also have direct nongeneration emissions, as illus- grid clearly are reduced. If the reduction in techni- trated by the net emissions methodologies presented cal losses is accompanied by increased sales with the in table 4.4. For example, an electrification project same amount of power generation, however, what that involves the installation of new SF6-containing does the additional electricity displace? equipment will have direct nongeneration emis- sions from fugitive emissions, even though the net Given the difficulty of assessing whether increased emissions impacts include increased grid generation electricity sales would displace other energy sources, 39 the current practice of T&D project analysis of AMS II.A: For retrofit projects, baseline emissions assuming that loss reductions result in lower power are the product of historical technical losses and generation is the most appropriate approach. If there the emission factor for the grid. The emission fac- is detailed power system modeling in the feasibility tor for the grid is determined by AMS I.D., which study for a T&D project, this analysis would exam- provides two options: (1) weighted average emis- ine how power generation, power flows, losses, and sions from all grid-connected plants or (2) a “com- overall system performance would be affected by the bined margin” approach from the CDM Executive investment project. This would also show in more Board (UNFCCC 2009d). Another option for the detail how the reduction in losses would affect gen- energy baseline is to determine technical losses eration in nonmarginal power plants (for example, of existing equipment based on standards and/or large base-load plants). In the absence of a detailed manufacturer ratings rather than actual measure- power generation and transmission model, the sim- ments. For radial networks where no standard is plest approach would be to assume that only the set available, other peer-reviewed approaches may of marginal plants is affected. be used to estimate baseline technical losses. For greenfield projects where there no T&D equipment A further issue is whether the T&D upgrade could is currently in place, the baseline is determined by affect construction plans for new power plants by the standards, manufacturer ratings, or other peer- increasing effective supply, and therefore delaying reviewed methods. the need for new power generation. While new plants could, in principle, be delayed, this impact would be The emissions reductions are limited to the date at relatively small because the technical loss improve- which equipment would normally be replaced or ment generally accounts for only a small percentage retrofitted. Note that this methodology does not of total power generation, so it will not entirely apply to introduction of capacitor banks and tap- replace new construction in a growing economy. changing transformers, because their impact on losses is more complex. Review of Existing Methodologies AM67: The methodology essentially covers a subset None of the methodologies reviewed consider any of technical loss reduction projects, where the losses changes in grid plant dispatch or merit order as a are achieved by installing higher-efficiency trans- result of reduced technical losses. The more recent formers in the existing distribution system. Baseline CDM methodologies include a marginal approach emissions are the product of “no-load” technical to emissions savings rather than using a reference loss rates of transformers, annual operating hours, power plant or average emissions for the entire grid. and the combined margin grid emission factor, as GHG Assessment Handbook: Baseline and project presented in UNFCCC (2009d). No-load losses are electricity generation required to meet demand are either calculated from annual power delivered divided by 1 minus technical losses (that is, before and after the minimum of (1) measured losses in the top project implementation). The change in energy gen- 20 percent of transformers and (2) the loss rate eration is divided by the reference power plant’s effi- specified in national regulations for transform- ciency to obtain fuel savings, which are multiplied ers, or by a fuel carbon emission factor to obtain carbon loss rates specified in national regulations, with- savings. This methodology implicitly assumes that out reference to measured losses. reducing technical losses reduces power generation in the grid, although that is represented by an indi- Project emissions are calculated similarly but vidual reference plant. using measured no-load loss rates of transformers 5. Generation Emissions Impacts of T&D Projects 40 installed by the project. This methodology does not tion would normally be reduced during an outage apply to load losses.1 period. The net impact of reduced power outages on GHG emissions is therefore the difference between Assessment of Available Methodologies the increased operation of connected power plants The available methodologies are similar to the stan- as a result of the T&D investment and the reduced dard economic analysis used for World Bank proj- use of backup power if any form of backup power is ects, in that they analyze technical loss reduction used. projects as reduced power generation. The reduction The critical question here is what sources of power, in generation is the difference in technical loss rates if any, are displaced when outages decline—in before and after the project multiplied by the total other words, what did consumers do during power electricity delivered (although it may be specified outages before the T&D reliability project was directly in some project documents rather than as a implemented? If consumers used captive power percentage of the total). The early methodology used for backup power during outages, these generation a weighted average emission factor, but the newer sources will be used less when outages decline. If CDM methodologies consider marginal changes and the emission factor of the captive power is higher thus use a marginal grid emission factor. than for grid power (which it usually is), net emis- sions will decline. If consumers do not have backup Increased Reliability power, however, and they simply use less electricity Not only do T&D upgrades and rehabilitation when there are power outages, net emissions would reduce technical losses, but they also increase the increase as outages decline. The concept of sup- reliability of the T&D system so that there are fewer pressed demand is useful here (Winkler and Thorne power outages for consumers (see, for example, 2002). Because there is a demand for this service World Bank 2004 and 2008b). These outages are that is constrained by technical factors (for example, costly for consumers not only because they may lose unreliable power supply), the emissions from the production (a factory that loses power) or inven- additional electricity supplied when outages decline tory (cold storage or supermarket), but also because could be compared to the alternatives for supplying they may purchase backup power supplies (for that power, even if the consumers do not actually example, diesel generators) to protect against out- own a backup power supply. This is not standard ages. Although these backup power supplies may practice in World Bank economic analysis of proj- only operate during power outages, they must be ects, which would compare the “with project” sce- maintained year round. nario to the current situation, even if the energy ser- vice levels were not the same in the two scenarios. In a typical financial and economic analysis of a T&D upgrade, reduced outages are treated as an Review of Existing Methodologies increase in sales. In other words, more electric- None of the GHG methodologies reviewed explicitly ity is delivered to the consumer after the project is addresses the GHG impacts of increased reliability implemented; therefore, more electricity is gener- and reduced outages. ated as well. This is also because grid power genera- Assessment of Available Methodologies 1 The absence of available methodologies could be a Load losses or coil losses are those losses caused by resis- tance in the electrical winding of the transformer; they include result of the challenges of estimating what on-site eddy current losses in the primary and secondary conductors of backup power source, if any, is displaced by the the transformer. Load losses of the transformer vary with load; therefore, crediting such reduction would require continuous increased sales of electricity to consumers and how monitoring of the load on the project activity transformer. to address the issue of suppressed demand. 5. Generation Emissions Impacts of T&D Projects 41 Distribution Capacity Expansion tricity use is zero. Another possibility is that the alternative energy source has a much higher unit Distribution projects that significantly increase the cost, so consumers cannot afford the alternative capacity of the power distribution system bring even though they can afford to use grid electricity. additional power generation to new or existing consumers. The rationale for increasing distribu- Review of Existing Methodologies tion capacity to existing consumers would be that their demand for power has increased, and there is None of the methodologies reviewed provide tools either surplus power capacity available elsewhere in or approaches to estimate the net impact on power the grid or new generation coming online that must generation of increased distribution capacity. be brought to these consumers. In other words, Assessment of Available Methodologies over the long term, distribution capacity expansion would almost always be accompanied by power gen- The absence of available methodologies means that eration capacity expansion. an approach for this project type must draw on experience with other project types, such as electrifi- Unlike technical loss reduction, distribution capac- cation and increased reliability. ity expansion could contribute to increased emis- sions from grid generation when compared to the baseline scenario. The project emissions for the type Electrification of project would be based on the additional electric- Electrification projects include additional distribu- ity being generated for distribution, which could tion (and, potentially, transmission) investments be from fossil fuel–fired power plants. Of course, if that connect new consumers to the electricity grid. the additional power generation is from renewable These may be consumers within existing electrified energy, project emissions would be much smaller, or areas that did not have access or entire communi- even nonexistent. ties that did not have grid electricity access. In either case, the electrification investment potentially dis- Baseline emissions depend on whether the addi- places existing or future energy sources with grid tional power supplied by the grid displaces other electricity. There must be additional power genera- off-grid alternatives for power supply. The addi- tion available to meet this increased demand, so tional power meets additional demand. How would generation from the grid-connected power plants this demand have been met if distribution capacity must increase (either through higher capacity utili- expansion had not been implemented? Would con- zation or new plant construction). sumers have used an alternative source of power, such as captive power or isolated grids? The reduc- In contrast to distribution capacity expansion to tion of captive or local grid power use must also be serve existing consumers, however, electrification considered in the analysis of impacts on emissions may displace energy sources other than electricity. from power generation. (This is why figure 2.4 In rural electrification, for example, new electricity includes captive power as well as grid power within supply to households could displace other energy the boundary of analysis.) There will also be situa- sources that were used for heating, lighting, and tions where end-use demand would not exist and cooking. the power would not have been supplied if the proj- ect were not implemented. An example would be a For new consumers connected within an electrified new factory that requires significantly more power area, the alternative could include captive power than the local grid can supply. Without an upgrade generation or a stand-alone minigrid. This could of T&D infrastructure, this new facility would not also be true for industrial and commercial con- be built, and so the baseline “without project” elec- sumers in an entirely new electrified community. 5. Generation Emissions Impacts of T&D Projects 42 For households in a newly electrified community, the literature on rural electrification to determine the baseline scenario could be a combination of whether there are consistent patterns of baseline off-grid power sources and nonelectrical energy energy use and shifts in patterns post-electrification sources. Alternatives to grid electricity will be used across different countries and regions. It will also by consumers as long as the financial resources are develop a new CDM methodology to address a sub- available and they are willing to pay the price of the set of rural electrification projects—the first such alternative. effort to address fuel displacement in the context of carbon financing or carbon accounting. Review of Existing Methodologies AM45: This methodology covers a subset of electri- Transmission Capacity Expansion fication projects that connect small isolated grids to Investments in new transmission capacity within a the interconnected national grid system. The meth- country may have several different purposes: odology applies where there was an existing isolated grid supplying a group of consumers, and the fossil To increase the capacity of an existing transmis- fuel–fired power plants in the isolated grid will no sion corridor by adding transmission lines longer be operated once the area is connected to the national grid (renewable power generation in the To connect a new power station that is far from isolated system must continue to operate). Baseline the major demand centers to a grid that serves emissions are initially the product of the weighted those demand centers average emission factor of the isolated grid and the To create new links between subnational grids amount of power supplied from the national grid that had not been previously connected after interconnection. To take into consideration the normal replacement of generation equipment in the The final case is similar to new interconnectors isolated system, the baseline emission factor declines between national grids, which is addressed in the over time toward the emission factor for the best next section, “Cross-Border Trade,” along with inter- available technology for isolated grid supply. Project national interconnectors. emissions are the product of electricity supplied For the first two cases, the capacity expansion proj- from the interconnected grid to the previously iso- ect is bringing additional power generation through lated area (adjusted for incremental technical losses) the transmission network, because this would be multiplied by the combined margin emission factor, the reason new lines were required. This additional calculated from UNFCCC (2009d). power generation may or may not be part of the Assessment of Available Methodologies same investment project as the new transmission line. Thus, the additional generation, and emis- AM45 provides a comprehensive and accurate sions from that generation, should be considered in approach for one type of electrification project, assessing the net impact on generation emissions. namely the displacement of isolated grid systems The project emissions could be based on the entire through connection to an integrated national grid. grid that transmits more power or on a single new Given the challenges of estimating the displacement plant, if the capacity expansion is to deliver that new of nonelectric energy sources—particularly in a plant’s production. rural electrification program—moving beyond this type of project will require substantial methodologi- As with distribution capacity expansion and cal work. The World Bank has recently commis- increased reliability, the selection of an appropriate sioned such work on a more comprehensive rural baseline scenario depends on whether consumers electrification methodology. That study will review would have used an alternative power source if the 5. Generation Emissions Impacts of T&D Projects 43 power generation and transmission capacity was not One final impact that should be considered is how built. The alternative to expanding the transmission increased transmission capacity along an existing system could be to generate power for end users on corridor affects technical losses over the entire cor- site or through a local minigrid, or possibly to build ridor. If the installation of a newer, more advanced a higher-cost power plant closer to the source of additional transmission line means that some of the demand. This holds true regardless of whether the power from the existing lines now flows through consumers are current grid customers or are off the the new line, technical losses for the entire corridor grid. Where there are existing sources of nongrid could be reduced. Projecting this impact would power being used in the areas served by a transmis- require a relatively sophisticated load-flow model- sion system expansion, they can be used to develop ing analysis as part of the project proposal. If such the baseline scenario. On the other hand, if consum- impacts and modeling exist, their GHG impacts can ers do not have the technical or financial means be analyzed as in any other loss reduction project. to generate their own power, which may be much more expensive per unit of energy, they will not use Review of Existing Methodologies additional power at all. Thus, the baseline is no addi- As with increased distribution capacity, none of the tional power production and consumption by those methodologies reviewed contain tools for estimat- consumers. ing the impact of increased transmission capacity on power generation beyond the impacts of reduced An important example of how a transmission losses and increased reliability. line would affect generation emissions is the link between transmission capacity and large-scale Assessment of Available Methodologies renewable power generation. Recent analysis on The absence of available methodologies means that large-scale roll-out of wind power has pointed out an approach for this project type must draw on that a lack of transmission capacity is often a major experience with other project types. constraint on construction of renewable power plants. Where renewable power sources are far from demand centers, and where current transmission Cross-Border Trade capacity is already constrained, transmission invest- One of the most important types of T&D invest- ments are an important component in achieving ments undertaken by the World Bank is new inter- emissions reductions from renewable power gen- connectors between separate existing grids, particu- eration investments. In this case, the increased grid larly connections between the grids of neighboring generation in the project scenario does not lead to countries (see, for example, World Bank 2003 and an increase in emissions; this may displace signifi- 2007). By connecting a national grid that has surplus cant fossil fuel alternatives in the baseline. power generation capacity with one that is capac- ity constrained, this new interconnector can pro- This example shows why, for both distribution and vide more electricity services to the region without transmission capacity expansion projects, there is increased investment in power generation. If the the potential for double counting between T&D exporting grid is a largely hydropower grid, and the projects and power generation projects (see “Double importing grid is largely fossil fuel–based genera- Counting,” page 19). The net impact of the T&D tion, this trade can reduce GHG emissions while project on power generation would be the same as increasing electricity supply to the connected system the net impact calculated from the new generation (Econ Analysis 2006) (see box 5.1). plant. Thus, it would not be accurate to add the net impacts of these projects together, since they over- What makes evaluating the impact of a new inter- lap. connector challenging is the fact that the connection 5. Generation Emissions Impacts of T&D Projects 44 Box 5.1: Cross-Border Trade and GHG Emissions Example: Cambodia-Vietnam The Cambodia Rural Electrification and Transmission Project—which formed the basis of a proposed CDM base- line methodology (NM0269)—includes the introduction of a 220 kV interconnection between Cambodia and Vietnam. This interconnection would facilitate the import of significant amounts of electricity (up to 200 MW capacity, 1,500 GWh per year) from Vietnam, which has larger, more efficient, and lower-emission-factor power plants, comprising a mix of hydropower (36 percent), gas-fueled, and coal-fired generation plants. This trade would meet Cambodia’s demand growth, which would otherwise have to be met by smaller, less efficient, higher-emission-factor sources, mainly diesel or heavy fuel oil–fired diesel engines and steam turbine–driven generators. The estimated GHG emissions impact was based on the difference in the grid emission factors of the two countries and the incremental amount of electricity imported into Cambodia. The emission factor for Vietnam is calculated as the average of the ex ante simple operating margin and the ex ante build margin. The emis- sion factor for Cambodia is calculated as the average of the ex post simple operating margin and the ex post build margin. These emission factors are defined in the “Tool to Calculate the Emission Factor for an Electricity System” approved by the CDM Executive Board. Fugitive emissions from SF6 from new T&D line components were also considered (although these are negligible, at only 230 tCO2/year). The calculated combined margin emission factors for Cambodia and Vietnam are 0.741 tCO2/MWh and 0.678 tCO2/MWh, respectively. The incremental traded electricity is projected to increase from 625 GWh in year 1 to 1,489 GWh in year 10. The emissions reductions over 10 years are therefore 536,158 tCO2. Source: UNFCCC 2008a. and free flow of power between the two previously ing local pollutants, such as oxides of nitrogen and separate grids may change the dispatch or merit sulfur. Dispatch models can also consider rules that order of many of the plants on both grids. Thus, the limit or minimize GHG emissions. Although these combined grid may not operate simply as the sum types of rules could be readily used to control GHG of the two grids (Econ Analysis 2006). A dispatch emissions, their application is limited given their model may be able to predict the operation of this implications for the cost of power generation. new system, but creating a simple ex ante estimate In terms of the impact of interconnectors on the in the absence of such models may be difficult. dispatch of power generation, one issue is which Assumptions about demand projections and gen- direction the electricity flows, because emissions erator unit parameters can be used for a simplified will only be reduced if power flows from the less estimate, but a dispatch model may more accurately carbon-intensive grid to the more carbon-intensive reflect how an interconnection will affect genera- grid. While power purchase contracts between the tion emissions, particularly if multiple dispatch utilities operating the grids will set the basic param- rules could be considered. Dispatch rules can sig- eters for trade, there is no guarantee that the flow is nificantly affect emissions from the power sector. always in one direction. High-voltage DC lines will Traditionally, least-cost (or price) rules are used to be easier to monitor for flow direction and current, dispatch power generation. In some power systems, but even in these systems power can potentially flow dispatch rules consider some constraints regard- in either direction—the utilities may want this to be 5. Generation Emissions Impacts of T&D Projects 45 so for when water supplies in the hydro-dominated from plants built explicitly for export capacity. The grid are low.2 underutilized plants could be higher marginal- cost plants, although hydropower capacity may be A second issue is whether the interconnected grid underutilized because of the seasonality of flows or has a combined or integrated dispatch center. If the distance from demand centers, even though this the two grids continue to operate largely indepen- is generally a lower marginal-cost power plant. New dently, it may be appropriate to simply consider the plants for export could include large hydropower transmission line as a single new low-carbon power plants, in which case the emission factors for these plant on the importing grid, and evaluate the emis- new plants should be used for project emissions sions impacts similarly to those of a new renew- rather than the overall grid. able power plant. However, if the connected grids operate as a single grid with a single or coordinated Aside from impacts on the operation of existing dispatch center, or with extensive communication plants, another major issue is whether the new between the dispatch centers, this approximation interconnector affects the construction of new will be less likely to reflect the actual impact of plants in each country. The impact of cross-border the new line on power generation in both grids. trade on plant construction is difficult to assess, There may be other unintended effects, based on because many countries or utilities have required the relative costs of generation across the grids, reserve margins and may choose to consider only the reliability of the generation sources, and the domestic supply as secure. Because of political con- mix of base load versus load following capacity in cerns about security of supply, governments and each grid. Ideally, the emissions impacts should be utilities may not necessarily delay the construc- evaluated based on ex ante and ex post monitoring tion of domestic power generation capacity even of generation in both grids. if additional imported power is available (Econ Analysis 2006). In fact, cross-border imports may A third issue is which plants are used for the be seen as a way to meet demand temporarily until exported power and which are displaced in the domestic supply construction can catch up with imported grid. While using average emission fac- demand. tors for both grids is the simplest way to analyze the change in emissions from increased trade, this Clearly, if a country already imports most of its elec- may not reflect actual grid operation. Power plants tricity from a neighboring country, or if the project are generally dispatched on merit order, which activity will supply a significant portion of con- reflects the marginal cost of power generation. In sumption in the neighboring country, cross-border other words, the best proxy for power generation imports have an impact on domestic capacity expan- displaced by imports would be the last plants dis- sion. The emission factor for surplus power supplied patched—those that have the highest marginal costs. by the exporting countries could be based on the These are generally fossil fuel plants, although the current marginal plants in the exporting country, highest-cost plants are not necessarily the most car- while the emission factor for additional power bon intensive. For example, a gas-fired plant might supplied (which requires the construction of new have a higher marginal cost, but lower emission fac- generation) should consider the new generation in tors than a coal plant. For exports, the power would the exporting grid. This is similar to the distinction come from the power plants that were underutilized made between trading surplus versus firm power, prior to the construction of the interconnector or where the cost of surplus power is the marginal cost of existing plants, and the cost of firm power is the 2 See the discussion by the Methodologies Panel on NM0269 full cost of new power generation in the exporting (UNFCCC 2009e). system. 5. Generation Emissions Impacts of T&D Projects 46 Review of Existing Methodologies there is surplus capacity in an exporting grid, or whether the importing country is historically a NM0272 and NM0269: Both NM0272 and NM0269 net importer), and limited the emissions reduc- were proposed as methodologies to estimate the net tions to net increases in traded electricity. This was impact of a transmission interconnector between accepted in the draft methodology proposed by the two national grids and the increased trade this Methodologies Panel. would allow. After several rounds of discussion between the project proponents and the CDM The choice of emission factors for the two grids Methodologies Panel on a draft approved methodol- has a critical impact on net emissions reductions. ogy incorporating both proposals, these methodolo- Originally, NM0269 proposed using the following gies were rejected. The major provisions of these emission factors for connected national grids. methodologies are discussed here, as well as the rea- sons why they were not accepted in the CDM. Baseline emissions: The combined margin emis- sion factor for the importing electricity system is Both methodologies begin from the premise that calculated using the latest version of the “Tool to when a transmission line connects a high-emission- Calculate the Emission Factor for an Electricity factor grid with a low-emission-factor grid (typically System” (UNFCCC 2009e) with the following hydropower dominated), power exports from the conditions: low-emission-factor grid will displace high-emis- sion-factor power generation in the importing grid. The operating margin is calculated as the In other words, the transmission line is similar to a simple operating margin, using ex post data. new power generation plant on the importing grid, The build margin is calculated using ex post with an emission factor reflecting the entire grid of data, updated annually (Option 2). the exporting country. Baseline emissions are there- fore the product of the importing grid emission fac- The combined margin uses 50-50 weightings tor and the amount of electricity imported. Project of the operating and build margins. emissions are the product of the exporting grid Project emissions: The emission factor for the emissions and the amount of electricity exported, exporting electricity system is calculated using the adjusted for technical losses if necessary. The meth- combined margin emission factor (EFgrid,CM) as in odologies only consider net increases in trade, to UNFCCC (2009d) with the following conditions: account for situations where there was some con- nectivity prior to the project. The operating margin is calculated as the simple operating margin, using three years of The two main issues raised during the subsequent historical data or the simple adjusted operat- revisions and discussions of these proposals are ing margin, if low-cost or must-run resources how to deal with a possible two-way flow of power are more than 50 percent of total grid genera- and which emission factors to use for each grid. tion. On the issue of flow, even if the exporting country currently has surplus power, there is still the pos- The build margin is calculated using ex ante sibility that a hydro-dominated exporter might have data (Option 1). to import power if water supplies were very low. The combined margin uses 50-50 weightings This reverse flow of power through transmission of the operating and build margins. therefore displaces low-emission-factor electricity with high-emission-factor electricity. To reduce this The approach in NM0272 was somewhat differ- likelihood, the methodology proponents included ent. For baseline emissions, the emission factor was various applicability conditions (that is, whether the combined margin (50-50 weightings) for the 5. Generation Emissions Impacts of T&D Projects 47 importing grid, where the operating margin was approach of the maximum of the operating or build calculated using current year dispatch data analysis, margin for the exporting system and the minimum and the project proponents choose the approach for of the operating or build margin for the importing the build margin. For project emissions, the emis- country, there would be no emissions reductions sion factor is the dispatch data operating margin for credited for this trade because the emission factor the exporting grid; no build margin is included. for the exporting grid is higher than for the import- ing grid. The draft approved methodology prepared by the Methodologies Panel modified the emission factors Because the draft approved methodology was never as follows: submitted to the CDM Executive Board, and both of the proposed methodologies were rejected, this dis- Baseline emissions: The operating margin is cussion is currently stalled within the CDM. calculated from either: (Option 1) dispatch data analysis or (Option 2) other recognized operat- ing margin approaches. The use of the build Summary of Impacts on Power margin was implicitly included, although the Generation: Direct and Indirect approach was not specified. The proposal speci- This discussion and review of available methodolo- fies that the minimum of the operating or build gies has several conclusions: margin must be used as the emission factor for the grid, rather than a weighted average. For the impacts of T&D projects on power gen- eration, some of the project categories have very Project emissions: The operating margin is limited coverage. There are no methodologies for calculated from either: (Option 1) dispatch impacts of distribution and transmission capac- data analysis or (Option 2) ex post calculations ity expansion. Within the wide scope of electrifi- that rank the plants of the operating margin cation projects, only the impact of grid electricity by decreasing emission factor and use the top- displacing isolated fossil fuel grids is covered. emitting ones to cover the annual net demand None of the proposed methodologies for cross- transferred over the new transmission lines. border trade has yet to gain acceptance in the The build margin approach was not discussed, CDM or other GHG accounting system. but the proposal specifies that the maximum of the operating or build margin must be used as Some of the CDM methodologies, such as the emission factor for the grid, rather than a AM67, have relatively narrow applicability con- weighted average. ditions, so they cannot cover the wider range of T&D projects within the World Bank portfolio. Assessment of Available Methodologies The CDM methodologies rely heavily on moni- The key differences in the draft approved methodol- tored data, since they must have a higher level ogy and the NM0269 and NM0272 proposals are of credibility and track actual project perfor- the choices of emission factors for the importing mance. This differs from the objective of this and exporting grids. The implications of using the study, which focuses on relatively simple ex ante highest-emission-factor plants instead of a simple approaches to estimating GHG impacts. operating margin or dispatch data are significant. The example in box 5.2 shows the calculation of the There is no methodology that addresses all of emission factors using the different approaches, in the multiple impacts of T&D investments (for a system where the exporting country is 90 percent example, technical loss reduction, increased hydropower and the importing country is 90 percent capacity, and primary effects of land clearing and fossil fuel. Using the draft approved methodology SF6 emissions). 5. Generation Emissions Impacts of T&D Projects 48 Box 5.2: Illustration of Sample Grid Emission Factor Calculations in the Draft Approved Methodology from NM0269/NM0272 The following calculations illustrate how the choice of grid emission factors (EFs) for exporting and importing grids can dramatically affect the emissions reductions credited to an interconnector project. If the proposal from the draft approved methodology prepared by the CDM Methodologies Panel had been accepted, there would be no emissions reductions credited to electricity trade—even in this example where the exporting country is 90 percent hydropower and the importing country is 90 percent fossil fuel. This is because the EF for the most carbon-intensive plant in the exporting grid is higher than the EF for the importing grid. Exporting Country Size Load Power generation Date EF Plant  (MW) factor (%) (MWh) Fuel  built tCO2/MWh OM  BM  1 100 80 700,880 Hydro 1970 0 k N 2 100 80 700,880 Hydro 1990 0 k N 3 100 80 700,880 Hydro 1990 0 k N 4 100 80 700,880 Hydro 1990 0 k N 5 100 80 700,880 Hydro 1991 0 k Y 6 100 80 700,880 Hydro 1995 0 k Y 7 100 70 613,270 Hydro 1960 0 k N 8 100 70 613,270 Hydro 1990 0 k N 9 100 60 525,660 Hydro 1990 0 k N 10 100 50 438,050 Coal 1980 1.1 j N Total 1,000   6,395,530 Estimated lambda 0.50 Assuming fossil plant only runs 50% of year Operating margin (OM) highest EF plant 1.10 Suggested in draft approved methodology OM simple adjusted 0.55 Suggested in NM0269 Build margin (BM) – EF (maximum BM & OM) 0.55 Using simple adjusted OM EF (maximum BM & OM) 1.10 Using top EF plant for OM Combined margin (CM) 0.28 Using simple adjusted OM and 50-50 OM/BM weighting Importing Country Size Load Power generation Date EF Plant  (MW) factor (%) (MWh) Fuel  built tCO2/MWh OM  BM 1 100 80 700,880 Coal 1970 1.1 Y N 2 100 80 700,880 Coal 1990 1.1 Y N 3 100 80 700,880 Coal 1990 1.1 Y N 4 100 80 700,880 Coal 1990 1.1 Y N 5 100 80 700,880 Oil 1991 0.9 Y N 6 100 80 700,880 Oil 1995 0.9 Y Y 7 100 70 613,270 Oil 1960 1.0 Y N 8 100 70 613,270 Oil 1990 0.9 Y N 9 100 60 525,660 Oil 1990 1.0 Y N 10 100 50 438,050 Hydro 1999 0.0 N Y Total 1,000   6,395,530         OM simple 1.01 BM 0.55 EF (minimum BM & OM) 0.55 As proposed in draft approved methodology CM 0.78 Using 50-50 OM/BM weighting Source: World Bank calculations. 5. Generation Emissions Impacts of T&D Projects 49 The relationship between T&D project categories power generation of each project category. As dis- and possible impacts on power generation is sum- cussed earlier, within the categories of T&D capacity marized in table 5.1. In this table, “Y” does not mean expansion, there are some additional distinctions by that the project definitely has an impact, but that project types, which are thus presented in more than many projects within that category are likely to have one category. that impact, which should be evaluated on a project- Box 5.3 illustrates the various impacts that have by-project basis. been discussed in this chapter, using the hypotheti- Table 5.2 summarizes conceptually the baseline and cal project originally presented in box 4.1. project scenarios for determining the impact on Table 5.1: Possible Impacts of Different T&D Project Categories on Power Generation Possible impacts on power generation Reduce Increase Change marginal marginal Displace Displace power power power alternative other build Project category generation generation power energy plan Direct generation effects Technical loss reduction Y N N N N Indirect generation effects Increased reliability N Y Y N N Distribution capacity expansion N Y Y N N Electrificationa N Y Y Y N Transmission capacity expansion—new N Y Y N Y? lines within a grid Transmission capacity expansion—connect Y (one grid) Y (one grid) N N Y grids Cross-border trade Y (one grid) Y (one grid) N N Y Source: Authors’ analysis. Note: Y = this project type may have this type of impact; N = this project type would not have this type of impact. a Electrification includes distribution capacity expansion that is directed at new customers. 5. Generation Emissions Impacts of T&D Projects 50 Table 5.2: Baseline and Project Scenarios for Impacts of T&D Investments on Power Generation Project category Project scenario Baseline scenario Direct generation effects Technical loss Generated electricity lost through technical Generated electricity lost through reduction losses after project implementation technical losses prior to project Indirect generation effects Increased reli- Additional power generation during longer Power source used during power ability supply hours outages or no emissions if alterna- tive is not available Distribution capac- Additional grid generation delivered to con- Alternative power source displaced ity expansion sumers, or generation from new plant by additional grid power or no emis- sions if alternative not available Electrificationa Additional grid generation delivered to con- Alternative power sources displaced sumers, or generation from new plant by additional grid power or no emis- sions if alternative not available Transmission Additional grid generation delivered to con- Alternative power sources displaced capacity expan- sumers, or generation from new plant by additional grid power or no emis- sion—new lines sions if alternative not available within grid Transmission Marginal or surplus power generation in Marginal power generation in capacity expan- exporting grid, or generation from new plants importing grid sion—connect built for export grids Cross-border trade Marginal or surplus power generation in Marginal power generation in exporting country, or generation from new importing country plants built for export Source: Authors’ analysis. a Electrification includes distribution capacity expansion that is directed at new customers 5. Generation Emissions Impacts of T&D Projects 51 Box 5.3: Example of Impact of Generation Emissions from a Typical Transmission Project For the same transmission line presented in box 4.1, consider the possible impacts on power generation emis- sions. If technical losses on this line were 15 percent, the project reduced these losses to 10 percent, and the grid had a relatively carbon-intensive emission factor of 700 kg CO2/MWh, the loss reduction project would save 4,861,435 tCO2 over the project life. This is equivalent to 35 kg CO2/MWh. If, on the other hand, the project was a capacity expansion project that did not have any alternative baseline supply of power (for example, no additional power would have been used if this capacity expansion was not built), transmitting an additional 138,898 GWh over 20 years leads to an increase of generation emissions of 97,228,702 tCO2 (700 kg CO2/MWh). If this new capacity displaced isolated diesel generators with an emission factor of 800 kg CO2/MWh, emissions would be reduced by 13,889,814 tCO2 or 100 kg CO2/MWh. If the transmission line was an interconnector that connected two grids, and the marginal emission factors of the exporting and importing grids were 100 kg CO2/MWh and 750 kg CO2/MWh, respectively, generation emis- sions would be reduced by 90,283,795 tCO2 or 650 kg CO2/MWh. If the transmission line was part of an electrification project that brought power from a new hydropower sta- tion (emission factor of 0 kg CO2/MWh) and displaced small diesel generators (emission factor of 1,200 kg CO2/ MWh), emissions would be reduced by 166,677,775 tCO2. If the grid supplying the new power was coal domi- nated (emission factor of 1,000 kg CO2/MWh), emissions reductions would only be 27,779,629 tCO2. For all these cases, it is clear that the impact on generation emissions is likely to be much larger than the direct nongeneration emissions. 5. Generation Emissions Impacts of T&D Projects 52 6. Recommended Approach In considering which elements of the available existing methodologies and the objectives of this methodologies to adapt for estimating net GHG study point to several key features that should be impacts of T&D projects, one of the key issues is considered for net emissions accounting method- feasibility. Many of these projects will not have ologies: extensive data available (either historic or projected) Modularity: Because a given investment may on power flows and individual plant-level power have multiple objectives and multiple compo- generation; any new methodology must acknowl- nents, emissions accounting methodologies edge this and not require extensive additional data should be modular. Each module should address collection. The teams conducting the economic, a specific impact, so these can be combined to financial, and technical analyses of new World Bank assess a wide range of T&D investments. projects should be able to apply the tools fairly eas- ily. The approaches should be standardized as much Tiered: In the IPCC “tiered” approach, more as possible, so they are easy to apply across different detailed methodologies are used where more countries. detailed data are available. Similarly, the emis- sions accounting methodologies should include Credibility is also important, so if the new meth- some default parameters, but require that more odology is going to deviate from the practice of detailed approaches be used where the data are other carbon methodologies, there should be a available. An example would be dispatch data or clear justification for this, and the risk of overesti- detailed system flow modeling, which should be mating net impacts should be relatively low. This used whenever available, but which may not be measure will also help with harmonization of the available for all projects. Simplified approaches new methodology with existing methodological are also required for the latter cases. approaches. Credibility also includes addressing the multiple impacts of T&D projects to ensure that Ex ante: As specified in the terms of reference the emissions impact analysis presents a compre- for this study, the net emissions impact should be hensive picture. Any risk of double counting should assessed ex ante, and not require monitoring, so be clearly highlighted (see “Double Counting,” page historical data should be used for all calculations. 19). Addressing the multiple impacts of T&D investments will require a modular approach to the Recommended Project Boundary analysis, which is discussed in more detail below. Based on the analysis of existing methodologies Comprehensive subsectoral coverage in this method- and case studies examined in chapter 4, the recom- ology, composed of a number of modules, will also mended project boundary is shown in figure 6.1. support the understanding of sectoral approaches to Project characteristics may mean that some of these estimating GHG emissions reductions. emissions sources may be zero, but they must still be In addition to the general principles for GHG explicitly assessed. Corona discharge is excluded for accounting described in chapter 2, the analysis of the reasons explained under “N2O Emissions from 53 Figure 6.1: Recommended Project Boundary for T&D Figure 6.2: Recommended Baseline and Project Projects Emissions Sources for Assessing the Impacts of Emissions on Generation Value chain Life-cycle phase Energy Baseline Project Power T&D source scenario scenario generation emissions emissions Manufacture Manufacture Materials b Combustionn b Combustionn of metal, of metal, production Grid power in grid d in grid and so on and so on wer power plant power plant Construction Energy use in Combustion Co b Co b Combustion construction Captive Construction of power in captive in captive power stations Land clearing wer power plant power plant SF6 fugitive iti Non- Combustion of Combustion of Combustion emissions electrical other energy other energy Operation in power energy sources sources plant Corona discharge Source: Authors’ analysis. Decommis- SF SF6 disposal sioning emissions example, for a coal-fired power plant, the emissions from upstream fuel production (for example, meth- Source: Authors’ analysis. ane emissions from, and energy use in, coal min- ing) could be between 0.4 percent and 11 percent, depending on the origin of the coal.1 Natural gas Corona Discharge” on page 34: namely, that there upstream emissions are estimated at 0.4 percent of are no methodologies for translating the numerous combustion emissions, and oil-fired power would site-specific drivers of this emissions source into be similar.2 Given that all the grid electricity CDM an accurate estimate. The inclusion of generation methodologies consider only combustion emissions emissions in the project boundary is also indicated. at the power plant, and not upstream, this study In keeping with the review of other methodologies proposes to limit the project boundary for assess- presented in chapter 5, fuel supply emissions and ing net impacts to only the power generation stage. emissions related to consumption of electricity are Because the electricity supplied by a new T&D not included. project could displace nongrid sources of energy Since many T&D projects may displace captive (for example, captive or backup power), baseline power, this is included in figure 6.2. Emissions from nonelectric energy sources are not included, because 1 Based on a combustion emission factor for other bitu- they only apply to electrification projects, and none minous coal of 0.0946 tCO2/gigajoule, net calorific value of of the methodologies reviewed provide any guid- 25.8 gigajoule/t (IPCC 2006b), and upstream emissions of 0.0134 t methane/t coal and 0.008 t methane/t coal for under- ance or data on this impact. ground and open-cast mining, respectively (UNFCCC 2009b). 2 Fuel supply stages are excluded based on their Based on combustion emissions of 0.0561 tCO2/gigajoule gas (IPCC 2006b) and upstream emissions of 0.296 kg methane/ very small contribution to life-cycle emissions. For gigajoule gas (also from ACM9). 6. Recommended Approach 54 and project emissions must be assessed for all these tory should include the quantities of various sources. The importance of combustion emissions metals, concrete, wood, and so on, that will be to power sector emissions and global energy sector used for the project. The country of origin for emissions makes it more meaningful to focus on these materials must be known so as to deter- these upstream impacts rather than the downstream mine whether the available databases of embod- impacts from energy consumption. ied emissions of construction materials are appli- cable. Direct and indirect emissions impacts are estimated for the same project life used in technical and eco- Are data available on energy consumption dur- nomic appraisal of a Bank-funded project to ensure ing the construction phase of the T&D project? the consistency and feasibility of the proposed This module will only be applied to projects for approach. which there are data on fuel consumption by construction vehicles or other energy sources used during the construction phase of the proj- Step 1. Determine Which Direct ect. Nongeneration Emissions Will Be Included Does the project involve clearing any land? Projects that only upgrade or rehabilitate existing Table 6.1 presents the questions to be asked, through lines and installations will not have an impact on review of the project preparation documentation, to land use, nor would projects that construct lines determine which modules for direct nongeneration along existing roads or rail lines. Only in cases impacts will be included in the project assessment. where there is clearing of land specifically for The modules are presented in detail in the next sec- new lines or equipment should this module be tion. Following is further explication of the ques- applied. tions. Does the project include new lines or capacity Are data available on materials consumption expansion that includes new SF6-containing by the T&D project and on the origin of those equipment? Projects that do not install entirely materials? This module will only be applied for new lines or substations are unlikely to cause a projects where a detailed materials inventory is net increase in SF6 emissions. Thus, this mod- available during project preparation or can be ule should only be applied to projects involving obtained from project proponents. This inven- capacity expansion, electrification, and cross- Table 6.1: Questions to Determine Which Direct Nongeneration Emissions Calculation Modules to Apply Question Module Are data available on materials consumption by the T&D Apply Module D1: Embodied Emissions project and on the origin of those materials? Are data available on energy consumption during the con- Apply Module D2: Construction Emissions struction phase of the T&D project? Does the T&D project involve clearing any land? Apply Module D3: Land Clearing Emissions Does the T&D project include new lines or capacity expan- Apply Module D4: SF6 Emissions sion that includes new SF6-containing equipment? Source: Authors’ analysis. 6. Recommended Approach 55 border trade connectors—and not to technical Module D2: Construction Emissions loss reduction or increased reliability projects— This module is only applied where the project prepa- that install entirely new SF6-containing equip- ration documentation estimates energy use by con- ment in order to account for direct nongenera- struction vehicles. tion emissions from SF6 leakage. Emissions are based on the fuel consumption in Step 2. Calculate Direct construction vehicles, the net calorific value of the fuel, and the emission factor of the fuel. Nongeneration Emissions for the T&D Projects PEconst = ∑(FCconst,i × NCVi × EFCO ,i) i 2 This section explains how to use the four modules to calculate direct nongeneration emissions. Where Module D1: Embodied Emissions PEconst = Project emissions from energy use in construction (tCO2) This module is only applied where the project prepa- ration documentation includes data on quantities of FCconst,i = Quantity of fossil fuel type i consumed materials used. The origin of the materials should also during construction (t) be known to identify the correct emission factor. NCVi = Net calorific value of fossil fuel type i Embodied emissions are the product of the mass of (GJ/t) materials used and the relevant embodied emission EFCO ,i = Carbon emission factor of fossil fuel 2 factor, summed across all significant materials. type i used in construction (tCO2e/GJ) PEEmb = ∑(Qp × EFEmb,p) p Parameter Source Where FCconst,i Project site records, feasibility studies by construction companies, or records from similar construction projects PEEmb = Project emissions from embodied emis- NCVi Local or national default factor, or IPCC sions in construction materials (tCO2) 2006 Guidelines Qp = Quantity of material p used in construc- EFCO ,i IPCC 2006 Guidelines 2 tion (t) EFEmb,p = Embodied emission factor of material p Module D3: Land Clearing Emissions (tCO2e/t) This module is based on AM45 and similar method- ologies. Parameter Source Qp Engineering studies in project documen- PELC = Adef × BD tation or feasibility studies Where EFEmb,p Table from the IFC CEET, or other similar databases (for example, Hammond and Jones 2008; Öko Institute for Applied PELC = Direct nongeneration emissions from land Ecology 2009). The embodied emission clearing (tCO2) factors should reflect the energy mix of the country of origin; for example, the Adef = Area of land deforested (ha) sources of construction materials from BD = Biomass density per unit area (above a country dominated by hydropower ground, below ground, soil carbon, litter, should not use emission factors from Europe. and dead biomass) (tCO2/ha) 6. Recommended Approach 56 Parameter Source Figure 6.3: Decision Tree for SF6 Calculation Approach Adef Project feasibility documents, or the product of default right of way and line length SF6 capacity BD IFC CEET table (which is taken from available? IPCC 2006 Guidelines; see annex A, table A.1 of this report) N Y Electrical capacity Option A Module D4: SF6 Emissions available? N Y This module does not apply to technical loss reduc- Equipment tion or increased reliability projects, unless the number and Option B project contains detailed information on what new type known? SF6-containing equipment would be installed. If new N Y equipment is specified, Option A may be used for that equipment only. Rehabilitation of existing SF6- Option D Option C containing equipment is not included, because that equipment was already in the electricity system, and Source: Authors’ analysis. any emissions from it are not incremental emissions from the project activity. Because this approach should rely on ex ante data, for the following equipment (IPCC 2006c, tables monitoring the changes in the SF6 inventory and 8.2–8.4): sealed-pressure electricity equipment using a mass balance (similar to AM35) is not pos- (medium-voltage switchgear), closed-pressure sible. Rather, the approach must use default emis- electrical equipment (high-voltage switchgear), sion factors for the various equipment that will be and gas-insulated transformers. The project would installed, considering the lifetime of that equipment have to provide data on nameplate capacity (kg and the maintenance that will be required. It is thus SF6) of all SF6-containing equipment and separate similar to the IPCC Tier 1 approach, except that this inventory into the relevant categories (that is, manufacturing emissions would not be included, sealed, closed pressure, gas-insulated transformer). since they are upstream of the T&D project. Where Because the use emission factors are annual leak- detailed equipment capacity data are not available, age, the economic life of the equipment would also a default factor based on a portion of national emis- be required to calculate lifetime direct nongenera- sions could be used. See the decision tree in fig- tion emissions. ure 6.3 to determine which option to follow. Annual project emissions would therefore be Option A: Nameplate SF6 Capacity for All PESF ,y = [(CapSP × EFSF ,Use,SP) + (CapCP × EFSF ,Use,CP)] Equipment Is Available 6 6 6 × GWPSF 6 Where the project documents provide an inven- Where tory of the SF6-containing equipment that will be installed as part of the project, and the nameplate PESF ,y = Annual project emissions of SF6 6 SF6 capacity of this equipment, standard emission (tCO2e/year) factors can be applied to the SF6 inventory. CapSP = Nameplate capacity of all sealed-pres- The default emission factors for installation, use, sure SF6-containing equipment used in and disposal are in the IPCC 2006 Guidelines the project (t SF6) 6. Recommended Approach 57 EFSF ,Use,SP = SF6 operational emission factor for Parameter Source 6 sealed-pressure electrical equipment CapSP Project preparation documentation (% SF6/year) CapCP Project preparation documentation EFSF ,Use,SP See table 6.2 CapCP = Nameplate capacity of all closed-pres- 6 sure SF6-containing equipment used in EFSF ,Use,CP See table 6.2 6 the project (t SF6) GWPSF IPCC 2006 Guidelines (23,900) 6 EFSF ,disp,SP Project or manufacturer guidelines for EFSF ,Use,CP = SF6 operational emission factor for 6 how SF6 will be disposed of at end of 6 closed-pressure electrical equipment project life (% SF6/year) EFSF ,disp,CP Project or manufacturer guidelines for 6 how SF6 will be disposed of at end of GWPSF = Global warming potential of SF6 project life 6 (23,900 tCO2e/t SF6) ELSF6 Manufacturer nameplate ratings of equipment life Project emissions at disposal would be PESF ,Disp = [(CapSP × EFSF ,disp,SP) + The emission factors for use should be as shown in 6 6 (CapCP × EFSF ,disp,CP)] × GWPSF table 6.2, based on the IPCC guidelines: 6 6 Where And total lifetime emissions would be as follows: ELSF6 PESF ,Disp = Project SF6 emissions at disposal PESF ,tot =y∑PESF ,y + PESF ,Disp =1 6 6 6 6 (tCO2e) CapSP = Nameplate capacity of all sealed-pres- Where sure SF6-containing equipment used in the project (t SF6) Table 6.2: Default Emission Factors for SF6 Losses in EFSF ,disp,SP = SF6 disposal emission factor for sealed- Operation 6 pressure electrical equipment (% SF6) Type of equipment %/year of name- CapCP = Nameplate capacity of all closed-pres- plate capacity lost sure SF6-containing equipment used in Sealed pressure SF6- 0.2 the project (t SF6) containing equipment EFSF ,disp,CP = SF6 disposal emission factor for closed- Closed pressure SF6- 2.6 6 pressure electrical equipment (% SF6) containing equipment GWPSF = Global warming potential of SF6 Source: IPCC 2006c, tables 8.2 and 8.3. 6 (23,900) PESF ,tot = Total project emissions from SF6- 6 containing equipment over project life (tCO2e) PESF ,y = Annual project emissions of SF6 6 (tCO2e/year) ELSF6 = Average economic life of all SF6- containing equipment (years) PESF ,Disp = Project SF6 emissions at disposal (tCO2e) 6 6. Recommended Approach 58 Option B: Electricity Capacity for All SF6- CapCP = Nameplate capacity of all closed-pressure Containing Equipment Is Available, But Not SF6 SF6-containing equipment used in the Capacity project (t SF6) For projects that have a detailed list of all SF6- CapCP,kV Nameplate electrical capacity of closed- containing equipment according to its rated power pressure SF6-containing equipment used capacity (for example, kV rating), but no actual data in the project (kV) on how much SF6 is in this equipment, the power SFCP = Scaling factor for closed-pressure equip- capacity may be converted to SF6 capacity using a ment (kg SF6/kV capacity) scaling factor. This factor is derived from a study by Wartmann and Harnisch (2005) on global SF6 emis- Parameter Source sions from the power sector. It converts the power CapSP,kV Project preparation documentation rating into SF6 capacity, assuming a linear relation- CapCP,kV Project preparation documentation ship between power and SF6 use. These estimated SFSP See table 6.3 SF6 capacity values are then used in the equations for Option A to estimate SF6 emissions. SFCP See table 6.3 CapSP = CapSP,kV × SFSP/1,000 Option C: Only Number and Type of Equipment Is Known, Not Capacity CapCP = CapCP,kV × SFCP/1,000 If only the number of pieces of SF6-containing equipment is known, and whether they are closed Where pressure or sealed pressure is specified, the simpli- CapSP = Nameplate capacity of all sealed pressure fied approach here uses an average SF6 capacity as SF6-containing equipment used in the shown in table 6.3. project (t SF6) PESF ,y = [(NSP × ACapSP × EFSF ,Use,SP) + 6 6 CapSP,kV = Nameplate electrical capacity of sealed- (NCP × ACapCP × EFSF ,Use,CP)] × GWPSF 6 6 pressure SF6-containing equipment used Where in the project (kV) SFSP = Scaling factor for sealed-pressure equip- PESF ,y = Annual project emissions of SF6 6 ment (kg SF6/kV capacity) (tCO2e/year) Table 6.3: Relationship between Power Rating and SF6 Capacity for T&D Equipment Default value if Scaling factor (kg power rating not Type of equipment Power rating SF6 capacity (kg) SF6/kV capacity) known (kg SF6) Sealed-pressure SF6- 1–52 kV 0.25–10 0.2 5 containing equipment Closed-pressure SF6- > 52 kV 3–200 0.5 100 containing equipment Source: Wartmann and Harnisch 2005. Note: Scaling factor for closed-pressure equipment assumes that 200 kg capacity would be up to 400 kV equipment. 6. Recommended Approach 59 Note that disposal emissions are not included NSP = Number of pieces of sealed-pressure because of high uncertainty and the fact that equipment (no units) these data are unlikely to be available if there is no ACapSP = Average capacity of sealed-pressure SF6- detailed SF6 capacity inventory. containing equipment (t SF6) Option D: No Inventory of SF6-Containing EFSF ,Use,SP = SF6 “use” emission factor for sealed- 6 Equipment Is Available pressure electrical equipment (% SF6/year) Where there is no detailed inventory of equipment NCP = Number of pieces of closed-pressure using SF6 during project preparation, average SF6 use equipment (no units) over the entire power sector for that country is used as the basis for determining a default emission fac- ACapCP = Average capacity of closed-pressure SF6- tor per unit of electricity (for example, kg SF6/kWh) containing equipment (t SF6) for high-, medium-, and low-voltage T&D systems. EFSF ,Use,CP = SF6 operational emission factor for This is then allocated, with 75 percent to high-volt- 6 closed-pressure electrical equipment age (> 100 kV) and 25 percent to medium-voltage (% SF6/year) (38–100 kV) equipment. If the project is entirely GWPSF = Global warming potential of SF6 below 38 kV, no SF6 emissions are estimated. 6 (23,900 tCO2e/t SF6) PESF ,y = [(ELECHV × EFSF6,z × 0.75) + 6 ELSF6 (ELECMV × EFSF ,z × 0.25)] × GWPSF /106 6 6 PESF ,tot =y∑PESF ,y =1 6 6 Where Where PESF ,y = Annual project emissions of SF6 6 PESF ,tot = Total project emissions from SF6- (tCO2e/year) 6 containing equipment over project life ELECHV = Electricity transmitted by the project (tCO2e) activity over new high-voltage lines PESF ,y = Annual project emissions of SF6 (> 100kV), measured at the exporting 6 (tCO2e/year) substation (MWh/year) ELSF6 = Average economic life of all SF6- EFSF ,z = Average SF6 emission factor for power 6 containing equipment (years) sector in country z (g SF6/MWh) ELECMV = Electricity transmitted by project activ- Parameter Source ity over new medium-voltage lines NSP Project preparation documentation (38–100 kV), measured at the exporting NCP Project preparation documentation substation (MWh/year) ACapSP See table 6.3 GWPSF = Global warming potential of SF6 6 ACapCP See table 6.3 (23,900 tCO2e/t SF6) EFSF ,Use,SP See table 6.2 6 EFSF ,Use,CP See table 6.2 ELSF6 6 PESF ,tot =y∑PESF ,y GWPSF IPCC 2006 Guidelines (23,900) 6 =1 6 6 ELSF Manufacturer nameplate ratings of 6 equipment life Where 6. Recommended Approach 60 term simulation models such as load flow and PESF ,to = Total project emissions from SF6- 6 t long-term economic dispatch simulations. containing equipment over project life (tCO2e) Identified alternative to additional electricity? PESF ,y = Annual project emissions of SF6 This question concerns whether the project tech- 6 (tCO2e/year) nical and economic analysis identifies a specific alternative source of power that would be used if ELSF6 = Average economic life of all SF6- the project were not implemented. This could be containing equipment (years) an existing captive power source, or a source that is likely to be constructed in the absence of the Parameter Source project. ELECHV Project preparation documentation ELECMV Project preparation documentation Identified source of incremental supply? This EFSF ,z Calculated from the U.S. EPA’s inventory question concerns whether the additional elec- 6 of SF6 fugitive emissions from the power tricity delivered by the T&D project is coming sector by country (U.S. EPA 2006) and the from a new power plant constructed to supply IEA’s reported electricity consumption by country (U.S. EIA 2010), or similar sources the T&D system. This could be the case, for ELSF Manufacturer nameplate ratings of example, with a new power plant built to export 6 equipment life power through a new interconnector. The project team would have to justify that the entire supply for the T&D project would come from the new Step 3. Determine How Baseline plant. This could be based on the timing of con- and Project Emissions for Power struction, similarity in capacity of the plant and Generation Effects Should Be the new T&D capacity, and contractual agree- Calculated ments between the grid operator and the owners of the new plant. To determine which modules to use for calculat- ing baseline and project emissions for the T&D project, the decision trees shown in figures 6.4–6.8 Step 4. Calculate Baseline Power should be used. There is a decision tree for each Generation Emissions for the T&D major project type, indicating which modules Projects should be applied to calculate baseline and project Both baseline and project emissions are always emissions using the designations “BE1,” “PE1,” calculated over the life of the project, since the and so on. Where a T&D investment package has more than one component (for example, capacity expansion and increased reliability), both modules Figure 6.4: Decision Tree for Technical Loss Reduction should be applied with relevant data from the proj- Projects ect preparation documentation. All the modules are described in detail below. BE1 = modeled w/o project Y PE1 = modeled w/ project The questions in the decision trees are as follows: Is a system model Is a system model available? Where there is a available? detailed power systems analysis model available, N BE2 = EFCM × TLBL PE2 = EFCM × TLPJ the most accurate approach for estimating GHG emissions is based on modeled power generation Source: Authors’ analysis. and/or fuel consumption using short- and mid- 6. Recommended Approach 61 Figure 6.5: Decision Tree for Increased Reliability Projects BE1 = modeled w/o project Y PE1 = modeled w/ project Is a system BE3 = EFAE × IE model Y PE4 = EFCM × IE available? N Identified alternative to additional electricity? N BE4 = 0 PE4 = EFCM × IE Figure 6.6: Decision Tree for T&D Capacity Expansion Projects BE1 = modeled w/o project BE3 = EFAE × IE Y PE3 = EFAS × IE Y PE1 = modeled w/ project Is a system Identified alternative model to additional Y electricity? available? BE4 = 0 N PE3 = EFAS × IE N Identified source of incremental supply? BE3 = EFAE × IE Y PE4 = EFCM × IE N Identified alternative to additional electricity? BE4 = 0 N PE4 = EFCM × IE Figure 6.7: Decision Tree for Electrification Projects BE1 = modeled w/o project BE3 = EFAE × IE Y PE3 = EFAS × IE Y PE1 = modeled w/ project Is a system Identified alternative model to additional Y electricity? available? BE4 = 0 N PE3 = EFAS × IE N Identified source of incremental supply? BE3 = EFAE × IE Y PE4 = EFCM × IE N Identified alternative to additional electricity? BE4 = 0 N PE4 = EFCM × IE 6. Recommended Approach 62 Figure 6.8: Decision Tree for Cross-Border Trade Projects BE1 = modeled w/o project Y PE1 = modeled w/ project Is a system BE5 = EFCM,M × IET model Y PE6 = (EFAS,X × IET)/(1 − TL) available? N Identified source of incremental supply? N BE5 = EFCM,M x IET PE5 = (EFCM,M × IET)/(1 − TL) total electricity flowing through the project can ηk = Conversion efficiency of grid-connected change over time. Thus, the following questions are power plant k (%) summed over all years y of the project life. The eco- nomic life should be the same one used in the feasi- 3.6 = Unit conversion factor (GJ/MWh) bility study for the project. EFCO ,i = Carbon emission factor of fuel type i 2 (tCO2/GJ) Module BE1 and BE1A: Modeled Baseline n = Economic life of project (years) Emissions This module is used for any project where a detailed Option B: Using Plant-Level Fuel Consumption power system model is available that can estimate If fuel consumption for each power plant is provided power generation by each plant with and without by the power system model, baseline emissions are the project. given as follows: Option A: Using Conversion Efficiencies n BE1 =y ∑1∑(FCBL,i,k,y × NCVi × EFCO2,i) = i,k If the model only reports power generation by plant, and not fuel consumption, baseline emissions are Where calculated as follows: n BE1 = Baseline emissions modeled over project BE1 =y ∑1∑[(EGBL,k,y /ηk) × 3.6 × EFCO2,i] = k life (tCO2) FCBL,i,k,y = Consumption of fuel type i in grid-con- Where nected plant k in the “without project” scenario in modeling year y (t) BE1 = Baseline emissions modeled over project life (tCO2) NCVi = Net calorific value of fuel type i (GJ/t) EGBL,k,y = Electricity generated by grid-connected EFCO ,i = Carbon emission factor of fuel type i 2 power plant k in the “without project” (tCO2/GJ) scenario in modeled year y (MWh) n = Economic life of project (years) 6. Recommended Approach 63 Parameter Source 3.6 = Unit conversion factor (GJ/MWh) EGBL,k,y Power system model EFCO ,i = Carbon emission factor of fuel type i 2 ηk Conversion efficiency should come from (tCO2/GJ) one of the following sources, in order of preference: EGBL,m,y = Electricity generated by grid-connected Source 1: Utility data on actual effi- power plant m in the importing grid in ciency of existing plants Source 2: Relevant national or regional the “without project” scenario in mod- studies on power plant efficiency eled year y (MWh) Source 3: Default efficiency from UNFCCC (2009d) (see annex A, ηm = Conversion efficiency of importing grid- table A.3) connected power plant m (%) EFCO ,i IPCC 2006 Guidelines n = Economic life of project (years) 2 n Project preparation documentation FCBL,i,k,y Power system model If fuel consumption for each power plant is provided NCVi Local or national default factor, or IPCC by the power system model, baseline emissions for 2006 Guidelines the two systems are given as follows: n Baseline emissions Module 1A is almost the same, BE1A =y∑ 1[∑(FCBL,i,x,y × NCVi × EFCO2,i) + = i,x except that it is the sum of modeled emissions for ∑(FCBL,i,m,y × NCVi × EFCO2,i)] two separate grids that are connected by the cross- i,m border trade project. Where n BE1A =y∑ {∑[(EGBL,x,y/ηx) × 3.6 × EFCO2,i] + =1 x BE1A = Baseline emissions modeled over project ∑[(EGBL,m,y/ηm) × 3.6 × EFCO2,i]} life (tCO2) m FCBL,i,x,y = Consumption of fuel type i in exporting Where grid-connected plant x in the “without project” scenario in modeling year y (t) BE1A = Baseline emissions modeled over project life (tCO2) NCVi = Net calorific value of fuel type i (GJ/t) EGBL,x,y = Electricity generated by grid-connected EFCO ,i = Carbon emission factor of fuel type i 2 power plant x in the exporting grid in (tCO2/GJ) the “without project” scenario in mod- FCBL,i,m,y = Consumption of fuel type i in importing eled year y (MWh) grid-connected plant m in the “without ηx = Conversion efficiency of exporting grid- project” scenario in modeling year y (t) connected power plant x (%) n = Economic life of project (years) 6. Recommended Approach 64 Parameter Source Parameter Source EGBL,x,y Power system model TLBL,y Power system model EGBL,m,y Power system model EFCM Calculated using UNFCCC (2009d) ηx, ηm Conversion efficiency should come from with ex ante options for operating and one of the following sources, in order of build margins. The operating margin preference: should be calculated as the simple Source 1: Utility data on actual effi- operating margin if low-cost/must-run ciency of existing plants resources are less than 50% of total Source 2: Relevant national or regional power generation, or as the weighted studies on power plant efficiency average operating margin if low-cost/ Source 3: Default efficiency from must-run resources are more than 50% UNFCCC (2009d) (see annex A, of total generation. New plants that are table A.3) committed to new capacity should be included in the margin calculations. EFCO ,i IPCC 2006 Guidelines 2 n Project preparation documentation n Project preparation documentation FCBL,i,k,y Power system model Module BE3: Emissions from Alternative NCVi Local or national default factor, or IPCC 2006 Guidelines Baseline Energy Source This module is used where more electricity is deliv- ered to the system by the T&D project, and there is a Module BE2: Emissions from Existing clearly identified alternative baseline energy source Technical Loss Rates in the project preparation documentation. In other This module is used for a project that reduces power words, the project documents either specify the generation but where there is no power system current energy sources that will be displaced or the model available to estimate change in plant-level alternatives that would have been built or used if the generation. The module is thus for technical loss project had not been implemented. reduction projects. n n BE3 =y∑(IEy × EFAE) =1 BE2 =y∑(TLBL,y × EFCM) =1 Where Where BE3 = Baseline emissions for project with identi- BE2 = Baseline emissions from losses (tCO2) fied alternative energy source (tCO2) TLBL,y = Estimated technical losses in year y with- IEy = Incremental electricity transmitted and dis- out the project (MWh) tributed as a result of the project in year y EFCM = Combined margin emission factor for the (MWh) interconnected grid, based on UNFCCC EFAE = Emission factor for the alternative baseline (2009d) (tCO2/MWh) energy supply source (tCO2/MWh) n = Economic life of project (years) n = Economic life of project (years) 6. Recommended Approach 65 Parameter Source EGBL,k = Electricity generated by minigrid power IEy Project preparation documentation source k in most recent three years EFAE If the alternative energy supply source is (MWh) an on-site/captive power supply plant, the emission factor for the alternative energy supply source should come from Parameter Source one of the following, in order of prefer- FCBLk,i Local utility records or minigrid opera- ence: tor records Source 1: Historical measurements if NCVi Local or national default factor, or IPCC the alternative energy supply source 2006 Guidelines already exists or can be identified. EFCO ,i IPCC 2006 Guidelines Source 2: Survey of local industry and 2 commercial facilities to show the type EGBL,k Local utility records or minigrid opera- of backup power used (for example, tor records fuel type, generation capacity), effi- ciency of that power supply, and/or fuel consumption Module BE4: No Emissions in the Baseline Source 3: Relevant national or regional studies on backup power supplies to For many of the project types where there is no determine the typical mix of capacity, alternative baseline energy source specified in the fuel type, and efficiency/fuel con- project documentation, it is assumed that there sumption would not be power consumption in the absence of Source 4: Default emission factor for diesel power generation from AMS I.D. the project. In other words, the new or expanded “Grid Connected Renewable Electricity end uses supplied by the project (for example, new Generation” (that is, 0.8 tCO2/MWh for commercial and industrial facilities, or new housing units over 200 kW capacity) areas) would not have been constructed or would If the alternative supply source is a mini- grid, refer to equation below not have had access to electricity. In the majority of cases, the T&D project preparation process will n Project preparation documentation identify and document the energy alternatives to If the alternative supply source is an isolated mini- increased grid power supply. However, there will be grid, the emission factor is calculated as follows: cases where the alternatives are unknown, or where it is unlikely that any supply would have existed EFAE = (∑ FCBLk,i × NCVi × EFCO ,i)/∑EGBL,k k,i k without the project. This baseline alternative accom- 2 modates that situation. Where Module BE5: Emissions from Importing Grid EFAE = Emission factor for the alternative base- This module is used for cross-border trade projects line energy supply source (tCO2/MWh) where this is no power system model available to FCBLk,i = Quantity of fuel type i consumed by project plant-level power generation with and with- minigrid power source k in the most out the project. Although in practice there may be recent three years (t or liters) some two-way flow of power on the transmission line, and there may also have been some historical NCVi = Net calorific value of fuel type i (GJ/t or electricity trade, only the incremental flow of power liter) from the exporting to importing country should be EFCO ,i = Carbon emission factor of fuel type i considered in estimating the net emissions impact of 2 (tCO2/GJ) the transmission investment. 6. Recommended Approach 66 n power generation by each plant with and without BE5 =y∑(IETm,y × EFCM,m) =1 the project. If the model only reports power gen- eration by plant and not fuel consumption, project Where emissions are calculated as follows: n BE5 = Baseline emissions for cross-border trade PE1 =y ∑1∑[(EGPJ,k,y /ηk) × 3.6 × EFCO2,i] = k project (tCO2) IETm,y = Projected incremental electricity received Where in the importing country because of the project in year y, measured at receiving PE1 = Project emissions modeled over project substation (MWh) life (tCO2) EFCM,m = Combined margin emission factor for the EGPJ,k,y = Electricity generated by grid-connected importing grid (tCO2/MWh) power plant k in the “with project” sce- n = Economic life of project (years) nario in modeled year y (MWh) ηk = Conversion efficiency of grid-connected Parameter Source power plant k (%) IETm.y Project preparation documentation 3.6 = Unit conversion factor (GJ/MWh) EFCM,m Calculated for the importing grid using UNFCCC (2009d) with ex ante options EFCO ,i = Carbon emission factor of fuel type i 2 for operating and build margin. The (tCO2/GJ) operating margin should be calcu- lated as the simple operating margin if low-cost/must-run resources are less If fuel consumption for each power plant is provided than 50% of total power generation, by the power system model, project emissions are or as the weighted average operating margin if low-cost/must-run resources given as follows: are more than 50% of total generation. n New plants that are committed to new PE1 =y ∑1∑(FCBL,i,k,y × NCVi × EFCO2,i) = i,k capacity should be included in the mar- gin calculations. n Project preparation documentation Where PE1 = Project emissions modeled over project Step 5. Calculate Project Power life (tCO2) Generation Emissions for the T&D FCBL,i,k,y = Consumption of fuel type i in grid con- Projects nected plant k in the “with project” sce- nario in modeled year y (t) Module PE1 and PE1A: Modeled Project Emissions NCVi = Net calorific value of fuel type i (GJ/t) This module is used for any project where a detailed EFCO ,i = Carbon emission factor of fuel type i 2 power system model is available that can estimate (tCO2/GJ) 6. Recommended Approach 67 Parameter Source n = Economic life of project (years) EGPJ,k,y Power system model ηk Conversion efficiency should come from If fuel consumption for each power plant is provided one of the following sources, in order of preference: by the power system model, project emissions for Source 1: Utility data on actual effi- the two systems are given as follows: ciency of existing plants Source 2: Relevant national or regional n studies on power plant efficiency PE1A =y∑ 1[∑(FCPJ,i,x,y × NCVi × EFCO2,i) + = i,x Source 3: Default efficiency from UNFCCC (2009d) (see annex A, table A.3) ∑(FCPJ,i,m,y × NCVi × EFCO2,i)] i,m EFCO ,i IPCC 2006 Guidelines 2 n Project preparation documentation Where FCPJ,i,k,y Power system model NCVi Local or national default factor, or IPCC PE1A = Baseline emissions modeled over project 2006 Guidelines life (tCO2) FCPJ,i,x,y = Consumption of fuel type i in exporting Project emissions Module 1A is almost the same, grid-connected plant x in the “with proj- except that it is the sum of modeled emissions for ect” scenario in modeling year y (t) two separate grids that are connected by the cross- border trade project. NCVi = Net calorific value of fuel type i (GJ/t) n EFCO2,i = Carbon emission factor of fuel type i {∑[(EGPJ,x,y/ηx) × 3.6 × EFCO2,i] + PE1A = y∑ 1 x = (tCO2/GJ) ∑[(EGPJ,m,y/ηm) × 3.6 × EFCO2,i]} m FCPJ,i,m,y = Consumption of fuel type i in importing grid-connected plant m in the “with proj- Where ect” scenario in modeling year y (t) PE1A = Project emissions modeled over project n = Economic life of project (years) life (tCO2) EGPJ,x,y = Electricity generated by grid-connected Parameter Source power plant x in the exporting grid in the EGPJ,x,y Power system model “with project” scenario in modeled year y EGPJ,m,y Power system model (MWh) ηx, ηm Conversion efficiency should come from ηx = Conversion efficiency of exporting grid- one of the following sources, in order of preference: connected power plant x (%) Source 1: Utility data on actual effi- 3.6 = Unit conversion factor (GJ/MWh) ciency of existing plants Source 2: Relevant national or regional EFCO2,i = Carbon emission factor of fuel type i studies on power plant efficiency (tCO2/GJ) Source 3: Default efficiency from UNFCCC (2009d) (see annex A, EGPJ,m,y = Electricity generated by grid-connected table A.3) power plant m in the importing grid in EFCO ,i IPCC 2006 Guidelines 2 the “with project” scenario in modeled n Project preparation documentation year y (MWh) FCBL,i,k,y Power system model ηm = Conversion efficiency of importing grid- NCVi Local or national default factor, or IPCC connected power plant m (%) 2006 Guidelines 6. Recommended Approach 68 Module PE2: Emissions from Expected Where Project Technical Loss Rates PE3 = Project emissions for project with identified This module is used for a project that reduces power new source of supply (tCO2) generation but where there is no power system model available to estimate changes in plant-level IEy = Incremental electricity transmitted and dis- generation. This module is thus for technical loss tributed as a result of the project in year y reduction projects. (MWh) n EFAS = Emission factor for the new source of sup- PE2 =y∑(TLPJ,y × EFCM) =1 ply (tCO2/MWh) Where n = Economic life of project (years) PE2 = Project emissions from losses (tCO2) Parameter Source TLPJ,y = Estimated technical losses in year y with IEy Project preparation documentation the project (MWh) EFAS The emission factor for the new source EFCM = Combined margin emission factor for the of supply may be determined in several ways, in order of preference: interconnected grid, based on UNFCCC Source 1: Estimated project-specific (2009d) (tCO2/MWh) annual fuel consumption and power generation, calculated according the n = Economic life of project (years) equation below Source 2: Based on manufacturer Parameter Source nameplate efficiency rating, calculated TLPJ,y Power system model according to the equation below Source 3: Feasibility studies for the EFCM Calculated using UNFCCC (2009d) with new source of supply ex ante options for operating and build Source 4: Default efficiencies from margin. The operating margin should be UNFCCC (2009d) (see annex A, calculated as the simple operating mar- table A.3) gin if low-cost/must-run resources are n Project preparation documentation less than 50% of total power generation, or as the weighted average operating margin if low-cost/must-run resources Equation for Source 1: Estimated Project-Specific are more than 50% of total generation. Annual Fuel Consumption and Power Generation New plants that are committed to new capacity should be included in the mar- EFAS = (∑FCAS,i × NCVi × EFCO ,i)/EGAS i 2 gin calculations. n Project preparation documentation Where Module PE3: Emissions from Identified New EFAS = Emission factor for the new source of sup- Source of Supply ply (tCO2/MWh) Where there is a clearly identified source of new FCAS,i = Estimated annual fossil fuel type i con- supply that will provide the power in the T&D proj- sumed by new power unit (mass or vol- ect, the emission factor for this source of supply is ume unit) used rather than a combined margin grid emission NCVi = Net calorific value (energy content) of fos- factor. sil fuel type i (GJ/mass or volume unit) n EFCO ,I = Carbon emission factor of fossil fuel type i 2 PE3 =y∑(IEy × EFAS) =1 (tCO2/GJ) 6. Recommended Approach 69 = Estimated annual net generation by new Where EGAS power unit (MWh) PE4 = Project emissions for project with identi- fied new source of supply (tCO2) Parameter Source FCAS,i Project preparation documentation, IEy = Incremental electricity transmitted and feasibility studies for new power plant, distributed as a result of the project in or utility data year y (MWh) EGAS Project preparation documentation, feasibility studies for new power plant, EFCM = Combine margin emission factor of elec- or utility data tricity grid (tCO2/MWh) NCVi Local or national default factor, or IPCC n = Economic life of project (years) 2006 Guidelines EFCO ,i IPCC 2006 Guidelines IL = Incremental technical losses caused by the 2 project (%) Equation for Source 2: Based on Manufacturer Parameter Source Nameplate Efficiency Rating IEy Project preparation documentation EFAS = (EFCO ,i × 3.6)/ ηAS,y EFCM Calculated using UNFCCC (2009d) with 2 ex ante options for operating and build Where margin. The operating margin should be calculated as the simple operating mar- EFAS = Emission factor for the new source of sup- gin if low-cost/must-run resources are ply (tCO2/MWh) less than 50% of total power generation, or as the weighted average operating EFCO ,i = Carbon emission factor of fossil fuel type i margin if low-cost/must-run resources 2 (tCO2/GJ) are more than 50% of total generation. New plants that are committed to new ηAS = Manufacturer nameplate efficiency rating capacity should be included in the mar- for new power unit (%) gin calculations. n Project preparation documentation Parameter Source IL This would be zero for most projects, unless the project preparation docu- EFCO ,i IPCC 2006 Guidelines 2 mentation specifies that the project ηk Manufacturer nameplate efficiency rat- activity involves a major line extension ing that would have higher technical losses than the local grids or power generation systems it is displacing. In the latter case, Module PE4: Emissions from Project Grid the project preparation documentation would be the source for the incremental Project emissions are from increased grid supply, technical losses. and are calculated as shown below. This assumes that the project does not lead to an increase in Module PE5: Emissions from Exporting Grid overall technical losses in the T&D system. If there is an increase in the technical losses, such as in an This module is used for cross-border trade projects electrification project involving a long-line exten- where this is no power system model available to sion to a remote village, the incremental losses are estimate plant-level power generation with and considered. without the project. Although in practice there may be some two-way flow of power on the transmission n line and there may also have been some historical PE4 =y∑(IEy × EFCM)/(1 − IL) =1 electricity trade, only the incremental flow of power 6. Recommended Approach 70 from the exporting to importing country should be mission line and there may also have been some considered in estimating the net emissions impact of historical electricity trade, to estimate the net emis- the transmission investment. sions impact of the transmission investment, only the incremental flow of power from the exporting to n PE5 =y∑(IETm,y × EFCM,x)/(1 − TLIC) importing country should be considered. =1 n Where PE6 =y∑(IETm,y × EFAS)/(1 − TLIC) =1 PE5 = Project emissions for cross-border trade Where project (tCO2) PE6 = Project emissions for cross-border trade IETm,y = Projected incremental electricity received project (tCO2) in the importing country because of the project in year y, measured at receiving IETm,y = Projected incremental electricity received substation (MWh) in the importing country because of the project in year y, measured at receiving EFCM,x = Combined margin emission factor for the substation (MWh) exporting grid (tCO2/MWh) EFAS = Emission factor for the new source of sup- TLIC = Technical losses on new interconnector (%) ply (tCO2/MWh) n = Economic life of project (years) TLIC = Technical losses on new interconnector (%) Parameter Source n = Economic life of project (years) IETm.y Project preparation documentation Parameter Source EFCM,x Calculated for the exporting grid using UNFCCC (2009d) with ex ante options for IETm.y Project preparation documentation operating and build margin. The operat- EFAS The emission factor for the new source ing margin should be calculated as the of supply may be determined in several simple operating margin if low-cost/must- ways, in order of preference: run resources are less than 50% of total Source 1: Estimated project-specific power generation, or as the weighted annual fuel consumption and power average operating margin if low-cost/ generation, calculated according the must-run resources are more than 50% equation below. of total generation. New plants that are Source 2: Based on manufacturer committed to new capacity should be nameplate efficiency rating, calculated included in the margin calculations. according to the equation below TLIC Project preparation documentation Source 3: Feasibility studies for the new source of supply n Project preparation documentation Source 4: Default efficiencies from UNFCCC (2009d) (see annex A, Module PE6: Emissions from Identified New table A.3) Source of Supply for Export TLIC Project preparation documentation This module is used for cross-border trade proj- n Project preparation documentation ects where this is no power system model available Equation for Source 1: Estimated Project-Specific to project long-term plant-level power generation Annual Fuel Consumption and Power Generation with and without the project, and where there is an identified new power plant that will produce the PE2 = (∑FCAS,i × NCVi × EFCO ,i)/EGAS i 2 electricity for export. Although in practice there may be some two-way flow of power on the trans- Where 6. Recommended Approach 71 EFAS = Emission factor for the new source of Note on Emission Factors for Cross-Border supply (tCO2/MWh) Trade Projects FCAS,i = Estimated annual fossil fuel type i con- Most large high-voltage transmission intercon- sumed by new power unit (mass or vol- nection projects financed by the World Bank are ume unit) expected to conduct short- and long-term power system simulation studies, which directly provide NCVi = Net calorific value (energy content) of emission factors and emissions reductions for dif- fossil fuel type i (GJ/mass or volume ferent integration scenarios and different dispatch unit) rules. However, in the absence of these studies, some EFCO ,I = Carbon emission factor of fossil fuel type assumptions will have to be made on emission fac- 2 i (tCO2/GJ) tors. These assumptions include what type of oper- EGAS = Estimated annual net power generation ating margin should be used and whether a build by new power unit (MWh) margin should also be included in the grid emission factor. For the importing grid, most of the propos- als have included both operating and build margins, Parameter Source on the grounds that imported power is being used FCAS,i Project preparation documentation, fea- in many countries to substitute or delay new con- sibility studies for new power plant, or utility data struction of power plants. As discussed earlier, the EGAS Project preparation documentation, fea- Methodologies Panel suggested using the minimum sibility studies for new power plant, or of operating and build margins for the grid emis- utility data sion factor, and basing the operating margin on NCVi Local or national default factor, or IPCC ex post dispatch data, if the data were available, or 2006 Guidelines other accepted approaches. No other methodology EFCO ,i IPCC 2006 Guidelines guideline or approved baseline methodology speci- 2 fies using the minimum of operating and build mar- Equation for Source 2: Based on Manufacturer gins.3 The approach suggested here is therefore to Nameplate Efficiency Rating use the combined margin, with ex ante simple oper- ating margin and ex ante build margin with 50-50 EFAS = (EFCO ,i × 3.6)/ηAS,y weighting. If historical dispatch data are available 2 to construct a more detailed operating margin, this Where may also be used. An alternative would be to decide whether to EFAS = Emission factor for the new source of sup- include the build margin for the importing grid ply (tCO2/MWh) based on two factors: (1) the amount of imports EFCO ,I = Carbon emission factor of fossil fuel type i prior to the project activity relative to the total con- 2 (tCO2/GJ) sumption of electricity in the importing country;4 ηAS = Manufacturer nameplate efficiency rating for new power unit (%) 3 The only exception among all the approved CDM method- ologies and tools is AM29 for grid-connected gas-fired power Parameter Source plants. EFCO ,i IPCC 2006 Guidelines 4 Note that this is different from the question of whether 2 absolute project size should influence the weightings of the ηk Manufacturer nameplate efficiency rat- operating and build margins, as discussed by the UNFCCC ing (2005, annex 2). 6. Recommended Approach 72 Table 6.4: Decision Matrix for Whether to Use Build Margin as Part of Baseline (Importing Country) Electricity Emission Factor Imports relative to total electricity consumption in importing country prior to project activity Project projected imports relative to total electricity consumption in Below threshold (for Above threshold (for example, > importing country example, < 50%) 50%) Above threshold (for example, > 10%) Yes Yes Below threshold (for example, < 10%) No Yes Source: Authors’ analysis. and (2) the projected amount of imports from the project activity relative to total consumption of elec- Table 6.5: Example of Summary Table for T&D Project tricity in the importing country (both average of last GHG Emissions (all tCO2 over project life) three years). Table 6.4 shows how these two vari- Direct nongeneration impacts ables would affect the use of the build margin. The exact thresholds would require further research. Embodied 5,000 emissions For the exporting grid, the proposals reviewed Energy in con- 12,000 take different approaches, from dispatch data, to struction combined margin, to using the emission factor of Land clearing 33,000 the most carbon-intensive plant on the exporting SF6 1,500 grid. Given that the approved methodology for grid Baseline Project Net extension (AM45) uses the combined margin, and that this can be used for relatively large flows of Direct generation impacts existing and new power generation, it is suggested Technical loss 30,000 10,000 −20,000 that the combined margin for the exporting coun- reduction try be used for this type of transmission project. Indirect generation impacts Alternatively, a matrix similar to that in table 6.4 Increased could be used to determine the extent to which reliability the build margin should be included. As discussed Capacity 25,000 30,000 5,000 above, where a single new plant can be identified as expansion the source for practically all of the export power, the Electrification emission factor for this plant should be used rather Cross-border than a grid emission factor. trade Source: Authors’ analysis. Step 6. Summarize GHG Emissions Impacts The GHG emissions impacts of the T&D proj- ent in terms of their effect on the overall power sec- ect should be summarized as shown in table 6.5. tor, they should be reported separately rather than Because the various impacts are qualitatively differ- summed. 6. Recommended Approach 73 7. Case Studies Three case studies from actual operations in the Case Study 1: Ethiopia-Kenya Power early stages of the project pipeline were selected Systems Interconnection Project to test the proposed approach. The type of invest- ments considered are a good representation of the The project is an interconnector for the Ethiopia different types of interventions typically present in and Kenya power systems over a high-voltage trans- mission line starting from Wolayta/Sodo on the loans that contain T&D components. The case stud- Ethiopian side and ending in the Nairobi area on the ies were developed using the approach and mod- Kenya side. Depending on the location of the land- ules described in the previous chapters. The aim in ing point on the Kenya side, the transmission line undertaking these case studies was to evaluate the will cover approximately 1,200 km. proposed methodology, in particular with respect to its ease (feasibility) of implementation and its reli- Under various assumptions on energy exchanges ability in determining the GHG impacts of interven- with Sudan, Egypt, and Djibouti, and the hydrologi- tions. In most cases, the project data reviewed were cal risks in Ethiopia, a two-stage development of the part of the feasibility studies from project prepara- interconnection capacity to Kenya has been defined: tion. Feasibility studies were prepared in most of the cases by external consultants working with the Phase 1: 1,000 MW transfer capacity by 2012, World Bank project teams. Besides the data con- the year of the targeted availability of hydro- tained in feasibility studies, the environmental and power from Gilbel Gibe III in Ethiopia social assessments were consulted, as well as other Phase 2: 2,000 MW transfer capacity beyond information from project preparation described in 2020 up to the planning horizon of 2030 the project appraisal documents. The very long interconnection and the high transfer The three case studies chosen were from the follow- capacity allow for the use of either high-voltage AC ing operations: or DC technologies (or combinations thereof) to Case study 1: Ethiopia-Kenya Power Systems ensure acceptable technical and economical perfor- Interconnection Project mance. The relevant data for this project were from the feasibility study prepared by Fichtner (2008). Case study 2: Energy Access Scale-Up Program, Kenya Description of Modules and Data Availability Case study 3: Eletrobras Distribution Step 1: Determine Which Direct Nongeneration Rehabilitation Project, Brazil Emissions Will Be Included The first project corresponds to preliminary feasibil- Based on the project type and data availability, the ity studies, but not a pipeline project. The second following table determines which direct nongener- and third case studies are already at advanced stages ation emissions calculation modules apply to this of approval. project. 75 Question Answer Module D4: SF6 Emissions. Since no information Are data available on materials No was available on the nameplate capacity of the SF6- consumption by the T&D project containing equipment, nor on the electrical capacity of and on the origin of those mate- this equipment, the option chosen was to use a default rials? value for SF6 emissions from electrical power systems, Are data available on energy No and multiply that figure by the electricity transmitted consumption during the con- struction phase of the T&D over the new line on an annual basis. Since no figure project? was available for SF6 emissions from electrical power Does the T&D project involve Yes (Apply Module systems in either Ethiopia or Kenya, the default emis- clearing any land? D3: Land Clearing sion factor for Africa as a whole was chosen (0.13 g Emissions) SF6/MWh). The results of this assessment are pre- Does the T&D project include Yes (Apply Module sented below. As the totality of the line is considered to new lines or capacity expan- D4: SF6 Emissions) be a high-voltage line (500 kV and 400 kV), 75 percent sion that includes new SF6- containing equipment? of the default emission factor is used. PESF6,y = [(ELECHV,y × EFSF6,z × 0.75) + Step 2: Calculate Direct Nongeneration Emissions (ELECMV,y × EFSF6,z × 0.25)] × GWPSF6/106 for the T&D Projects Parameter Unit Value Module D3: Land Clearing Emissions. The exec- ELECHV MWh 111,118,518 utive summary mentions that the right of way EFSF ,z g SF6/MWh 0.13 for the high-voltage AC 400 kV double circuit 6 line is 60 m. The right of way for a high-voltage ELECMV MWh 0 DC bipolar 500 kV line is 50 m, except in popu- GWPSF tCO2e/t SF6 23,900 6 lated areas where the right of way increases up PESF ,y 249,971 6 to 70 m. A figure of 60 m has therefore been taken as an appropriate compromise for this Step 3: Determine How Baseline and Project project. With respect to the potential emissions Emissions from Power Generation Should Be from land clearing, an assumption has been Calculated made from the limited information available in the FSR that the land can be described as The decision tree for cross-border trade projects is “Cropland—Tropical (moist region), perennial shown in figure 6.8. According to this flow chart, woody biomass” according to the IFC CEET baseline power generation emissions should be table; the relevant biomass density has therefore calculated using Module BE5, and project power been applied. generation emissions should be calculated using Module PE6. The results for this module are summarized below: Step 4: Calculate Baseline Power Generation Emissions for the T&D Projects PELC = Adef × BD Module BE5: Emissions from Importing Grid. The Parameter Unit Value baseline emissions from the importing grid are cal- def ha 7,200 culated as follows: BD tCO2/ha 77 n PELC 554,400 BE5 =y∑(IETm,y × EFCM,m) =1 7. Case Studies 76 Parameter Unit Value Table 7.1: Summary of GHG Impacts for Ethiopia-Kenya EFCM,m tCO2/MWh 0.6545 Power Systems Interconnection Project (tCO2) IETm MWh 106,672,377 BE5 69,817,071 Direct nongeneration impacts Embodied n.a. emissions Note that IETm is the sum of all of the years of the Energy in con- n.a. project life, and is the value that would be measured struction at the receiving substation, net of technical losses on Land clearing 554,400 the transmission line. SF6 249,971 Step 5: Calculate Project Power Generation Baseline Project Net Emissions for the T&D Projects Direct generation impacts Module PE6: Emissions from Identified New Technical loss n.a. n.a. Source of Supply for Export. The new cross-border reduction transmission line is linked to the construction of a Indirect generation impacts new hydropower plant in Ethiopia (Gilbel Gibe III). Increased reli- n.a. n.a. Therefore, the source of incremental supply is iden- ability tified as being this new plant with an emission factor Capacity n.a. n.a. of 0 tCO2/MWh. expansion The emissions from the exporting grid are calculated Electrification n.a. n.a. as follows: Cross-border 69,817,071 0 −69,817,071 trade n PE6 =y∑(IETm,y × EFAS)/(1 − TLIC) =1 Source: Authors’ analysis. Note: n.a. = not applicable. Parameter Unit Value EFAS,x tCO2/MWh 0 IETm MWh 106,672,377 that although direct nongeneration emissions are PE6 0 significant—the highest of all three case studies— they are by far outweighed by the impacts on gen- eration emissions. For this project, direct nongener- Note that IETm is the sum of all the years of the proj- ation emissions represent approximately 1 percent of ect life, and is the value that would be measured at the impact on generation emissions. the receiving substation, net of technical losses on the transmission line. Case Study 2: Energy Access Scale-Up Step 6: Summarize GHG Emissions Impacts Program, Kenya The Energy Access Scale-Up Program is an opera- Table 7.1 presents a summary of the estimated GHG tion that contributes to Kenya’s effort to improve impacts from this project. and expand electricity services to the country. The The table indicates that the project results in a very loan includes components that support investments significant reduction in power generation emissions in all segments of the electricity chain: generation, over the period 2012–27 because of indirect genera- T&D, and increased access. The GHG account- tion impacts (−69.8 MtCO2). The results also show ing for this case study focused on the transmission 7. Case Studies 77 component. This component supports the govern- The table below determines which direct nongener- ment’s plan to expand transmission capacity to serve ation emissions calculation modules apply to this growing demand from the distribution sector and project. to improve the reliability of the electricity network. Question Answer This network is highly radial, which is characteristic of countries with low electrification rates. Are data available on materials No consumption by the T&D proj- The government plan consists of 13 subprojects ect and on the origin of those materials? on the 132/33 kV transmission Kenya Power and Are data available on energy No Light network. The projects in the plan have been consumption during the con- developed by the network with the help of an exter- struction phase of the T&D nal consultant and put forward for Bank financ- project? ing. During the preparation of this report, two of Does the T&D project involve Yes (Apply Module the projects completed their technical, economic, clearing any land? D3: Land Clearing Emissions) financial, and environmental and social feasibility Does the T&D project include No analyses. new lines or capacity expan- sion that includes new SF6- Kisii-Awendo Line: This project will involve con- containing equipment? struction of a 44 km 132 kV transmission line between the proposed Kisii 132/33 kV substation and a 132/33 kV, 1x23 MVA substation to be built Step 2: Calculate Direct Nongeneration Emissions in the vicinity of Awendo, and the construction of a for the T&D Projects 132 kV line bay at Kisii. Module D3: Land Clearing Emissions. The FSR Eldoret-Kitale Line: This project will involve con- makes no mention of the right of way for the new struction of approximately 60 km of 132 kV single 132 kV/33 kV line, so the default figure of 30 m, circuit transmission line, including establishment of based on discussions and feedback from World a 132/33 kV, 23 MVA substation at Kitale. The 33 kV Bank staff, is applied, which, combined with the dis- network that will be influenced by this project is tance of 44 km, gives an area of 132 ha. With respect the 33 kV radial from Eldoret 132/33 kV substation to the potential emissions from land clearing, an supplying the 33/11 kV substations Moi Barracks, assumption has been made from information avail- Moi’s Bridge, Cheranguria, Kitale, and Kapenguria. able in the environmental and social impact assess- Several 33/0.4 kV distribution transformers are also ment that the land can be described as “Cropland— connected to this 33 kV radial, most concentrated Annual crops” according to the IFC CEET table, between Moi Barracks and Moi’s Bridge. and the relevant biomass density has therefore been applied. Description of Projects and Data Availability The results for this module are summarized Project I: Kisii-Awendo Line below: The relevant data for this subproject were sourced PELC = Adef × BD from the feasibility study prepared by Snowy Mountains Engineering Corporation and dated Parameter Unit Value April 2009. Adef ha 132 BD tCO2/ha 17 Step 1: Determine Which Direct Nongeneration PELC 2,244 Emissions Will Be Included 7. Case Studies 78 Module D4: SF6 Emissions. Because the new equip- n BE2 =y∑(TLBL,y × EFCM) ment will displace old equipment that is being =1 retired, there should be no net increase in SF6 emis- sions. Parameter Unit Value EFCM tCO2/MWh 0.6545 Step 3: Determine How Baseline and Project Year TLBL,y BEy Emissions from Power Generation Should Be   MWh tCO2 Calculated 2012 2,312 1,513 This project leads to technical loss reductions, 2013 2,471 1,617 increased reliability, and T&D capacity expansion. 2014 2,562 1,677 The decision trees for these three project types 2015 2,777 1,818 are presented in figures 6.4–6.6, respectively. For 2016 2,978 1,949 technical loss reduction baseline emissions should 2017 3,107 2,033 be calculated using Module BE2, and project emis- 2018 3,349 2,192 sions should be calculated using Module PE2. For 2019 3,493 2,286 increased reliability and T&D capacity expansion, 2020 3,780 2,474 baseline emissions should be calculated using 2021 4,068 2,662 Module BE4, and project emissions should be calcu- 2022 4,257 2,786 lated using Module PE4. 2023 4,753 3,111 2024 5,653 3,700 Note that because both increased reliability and 2025 6,615 4,329 capacity expansion use Module PE4 (emissions from 2026 7,705 5,043 the project grid) but have different quantities of incremental electricity supplied, Module PE4 must 2027 8,753 5,729 be applied separately to each project objective. The 2028 9,786 6,405 same equations are thus used for each of these three 2029 10,805 7,072 impacts, but with different input parameters for total 2030 11,815 7,733 incremental electricity. 2031 12,818 8,390 2032 13,784 9,021 Step 4: Calculate Baseline Power Generation BE2   83,541 Emissions for the T&D Projects Increased Reliability Technical Loss Reduction Module BE4: No Emissions in the Baseline. Module BE2: Emissions from Existing Technical Because there is no source of alternative energy Loss Rate. The feasibility study does not provide specified in the project documentation, the assump- figures on technical losses for each year, but rather tion is made that there would be no power con- provides figures on annual technical loss reductions sumption in the absence of the project (BE4 = 0). in MWh for the period 2012–32. It is therefore not possible to resolve this module as described, but T&D Capacity Expansion the same result will be achieved by using the figures Module BE4: No Emissions in the Baseline. provided for technical loss as technical losses for the Because there is no source of alternative energy baseline, and assuming technical losses in the proj- specified in the project documentation, the assump- ect scenario to be zero. tion is made that there would be no power con- The results for this module are summarized here: sumption in the absence of the project (BE4 = 0). 7. Case Studies 79 Electrification Year IEy PEy Although the FSR mentions electrification, there are MWh tCO2 no data in the feasibility studies on the incremental 2012 2,149 1,407 energy supplied to new customers by electrifica- 2013 2,241 1,467 tion. Thus, the technical and economic assessment 2014 2,310 1,512 assumed that all additional capacity would be used 2015 2,407 1,576 to supply the incremental demand of existing con- 2016 2,476 1,621 sumers. For this reason, this module is not applied. 2017 2,548 1,668 2018 2,649 1,733 Step 5: Calculate Project Power Generation 2019 2,723 1,782 Emissions for the T&D Projects 2020 2,794 1,828 Technical Loss Reduction 2021 2,898 1,897 2022 2,976 1,948 Module PE2: Emissions from Expected Project 2023 3,218 2,106 Loss Rates. Because the FSR only provides data on 2024 3,450 2,258 total loss reduction, this module is not necessary. 2025 3,674 2,404 Project losses are zero and baseline losses represent 2026 3,849 2,519 the full benefit of the loss reduction. 2027 4,024 2,633 Increased Reliability 2028 4,197 2,747 2029 4,371 2,861 Module PE4: Emissions from Project Grid. 2030 4,546 2,975 Annex D4 of the FSR provides tables that summa- 2031 4,719 3,089 rize the project’s economic and financial reliability. PE4 42,032 The figures for financial reliability (which is defined as the net benefit resulting from the reconfigured system in net kWh added energy sales) provide an T&D Capacity Expansion estimate of the additional power that can be sold to customers because of improved reliability. The fig- Module PE4: Emissions from Project Grid. ure presented in the FSR assumes 30 percent of the Annex D4 of the FSR provides tables that summa- power provided through increased reliability is non- rize the project’s economic and financial reliability. recoverable; thus, this 30 percent must be taken into The table provides figures for customer consump- account (emissions reductions are independent of tion growth (also called financial growth). These whether costs for provided electricity are recovered figures are required to estimate additional project or not). The emission factor for the grid is from the emissions caused by expansion of T&D capacity. The Institute for Global Environmental Strategies (IGES) emission factor for the Kenya grid is taken from the database for the grid emission factor (combined IGES database. The data and results for this module margin) for the Kenya grid. are summarized below: n n PE4 =y∑(IEy × EFCM)/(1 − IL) PE4 =y∑(IEy × EFCM)/(1 − IL) =1 =1 Parameter Unit Value Parameter Unit Value EFCM tCO2/MWh 0.6545 EFCM tCO2/MWh 0.6545 IL % 0 IL % 0 7. Case Studies 80 Year IEy PEy Table 7.2: Summary of GHG Impacts for Kisii-Awendo MWh tCO2 Line (tCO2) 2012 266 174 2013 538 352 Direct nongeneration impacts 2014 1,097 718 Embodied n.a. 2015 1,678 1,098 emissions 2016 2,283 1,494 Energy in con- n.a. 2017 2,912 1,906 struction 2018 3,566 2,334 Land clearing 2,244 2019 4,246 2,779 SF6 n.a. 2020 4,954 3,242 Baseline Project Net 2021 5,690 3,724 2022 6,455 4,225 Direct generation impacts 2023 6,649 4,352 Technical loss 83,541 0 −83,541 2024 6,848 4,482 reduction 2025 7,054 4,617 Indirect generation impacts 2026 7,265 4,755 Increased reli- 0 42,032 42,032 2027 7,483 4,898 ability 2028 7,708 5,045 Capacity 0 66,255 66,255 2029 7,939 5,196 expansion 2030 8,177 5,352 Electrification n.a. n.a. 2031 8,422 5,512 Cross-border n.a. n.a. PE4 66,255 trade Source: Authors’ analysis. Note: n.a. = not applicable. Step 6: Summarize GHG Emissions Impacts Table 7.2 presents a summary of the estimated GHG impacts from this project. Question Answer Are data available on materials con- No Project II: Eldoret-Kitale Line sumption by the T&D project and The relevant data for this project were sourced from on the origin of those materials? the feasibility study prepared by Norconsult and Are data available on energy con- No sumption during the construction dated September 2009 as part of the project prepara- phase of the T&D project? tion technical and economic analysis. Does the T&D project involve clear- Yes (Apply ing any land? Module D3: Step 1: Determine Which Direct Nongeneration Land Clearing Emissions Will Be Included Emissions) Does the T&D project include Yes (Apply The following table determines which direct non- new lines or capacity expansion Module D4: SF6 generation emissions calculation modules apply to that includes new SF6-containing Emissions) this project. equipment? 7. Case Studies 81 Step 2: Calculate Direct Nongeneration Emissions Parameter Unit Value for the T&D Projects NSP no units 7 Module D3: Land Clearing Emissions. The FSR NCP no units 6 makes no mention of the right of way for the new ACapSP t SF6 0.005 132 kV line, so the default figure of 30 m, based on ACapCP t SF6 0.1 discussions and feedback from World Bank staff, is EFSF ,Use,SP % 0.2% 6 applied. Combined with the distance of 60 km, this EFSF ,Use,CP % 2.6% gives an area of 180 ha. With respect to the poten- 6 GWPSF tCO2e/t SF6 23,900 tial emissions from land clearing, an assumption 6 has been made based on the limited information ELSF years  20 6 available in the FSR that the land can be described PESF ,y tCO2e 375 6 as “Cropland—Tropical (moist region), perennial PESF ,tot tCO2e 7,490 6 woody biomass” according to the IFC CEET table; the relevant biomass density has therefore been Because the new equipment will displace old equip- applied. The results for this module are summa- ment that is being retired, there should be no net rized below: increase in SF6 emissions. For this reason, SF6 emis- sions have not been included in the summary. PELC = Adef × BD Step 3: Determine How Baseline and Project Emissions from Power Generation Should Be Parameter Unit Value Calculated Adef ha 180 This project leads to technical loss reductions, BD tCO2/ha 77 increased reliability, T&D capacity expansion, and PELC 13,860 electrification. The decision trees for these four proj- ect types are presented in figures 6.4–6.7, respec- Module D4: SF6 Emissions. This case study was tively. For technical loss reduction, baseline emis- the only one for which data were available, which sions should be calculated using Module BE2 and provided an indication of the number of SF6- project emissions using Module PE2. For increased containing equipment that would be installed dur- reliability and T&D capacity expansion, baseline and ing project implementation and their respective project emissions should use Modules BE4 and PE4, capacities. Therefore, it was possible to use Option respectively. For electrification, baseline and project C to estimate GHG emissions from SF6-containing emissions should use Modules BE3 and PE4, respec- equipment use. The project documentation stated tively. that seven units would be installed that could be Because increased reliability, capacity expansion, considered sealed-pressure SF6-containing equip- and electrification all use Module PE4 (emissions ment and six units would be installed that could be from the project grid) but have different quantities considered closed-pressure SF6-containing equip- of incremental electricity supplied, Module PE4 ment. For disposal emissions, it has been assumed must be applied separately to each project objec- that all of the SF6 will be recovered, because World tive. In other words, the same equations are used for Bank projects must follow strict environmental each of these three impacts, but with different input guidelines. parameters for total incremental electricity. 7. Case Studies 82 Step 4: Calculate Baseline Power Generation Increased Reliability Emissions for the T&D Projects Module BE4: No Emissions in the Baseline. Technical Loss Reduction Because there is no source of alternative energy Module BE2: Emissions from Existing Technical specified in the project documentation, the assump- Loss Rate. The feasibility study does not provide tion is made that there would be no power con- figures on technical losses for each year, but rather sumption in the absence of the project (BE4 = 0). on annual technical loss reductions in MWh for a T&D Capacity Expansion 20-year period (2,900 MWh/year). This is incorpo- rated into the model by designating this amount as Module BE4: No Emissions in the Baseline. the baseline losses and setting project losses to zero. Because there is no source of alternative energy The results for this module are summarized below: specified in the project documentation, the assump- tion is made that there would be no power con- n BE2 =y∑(TLBL,y × EFCM) =1 sumption in the absence of the project (BE4 = 0). Parameter Unit Value Electrification EFCM tCO2/MWh 0.6545 Module BE3: Diesel Generator Emissions. The FSR identifies the cost of small-scale diesel as the Year TLBL,y BEy alternative to the electrification project, so the MWh tCO2 emission factor for a diesel generator has been 2012 2,900 1,898 used. 2013 2,900 1,898 2014 2,900 1,898 The results for this module are summarized 2015 2,900 1,898 below: 2016 2,900 1,898 n 2017 2,900 1,898 BE3 =y∑(IEy × EFAE) =1 2018 2,900 1,898 Parameter Unit Value 2019 2,900 1,898 EFAE tCO2/MWh 0.8 2020 2,900 1,898 2021 2,900 1,898 2022 2,900 1,898 2023 2,900 1,898 2024 2,900 1,898 2025 2,900 1,898 2026 2,900 1,898 2027 2,900 1,898 2028 2,900 1,898 2029 2,900 1,898 2030 2,900 1,898 2031 2,900 1,898 BE2   37,961 7. Case Studies 83 Year IEy BEy used is from the IGES database for the grid emission   MWh tCO2 factor (combined margin) for the Kenya grid. The 2012 3,350 2,680 results for this module are summarized below: 2013 5,200 4,160 Parameter Unit Value 2014 7,250 5,800 EFCM tCO2/MWh 0.6545 2015 9,500 7,600 IL % 0 2016 11,950 9,560 2017 14,600 11,680 Year IEy PEy 2018 17,450 13,960 MWh tCO2 2019 20,600 16,480 2012 1,900 1,244 2020 24,050 19,240 2013 1,900 1,244 2021 27,800 22,240 2014 1,900 1,244 2022 31,850 25,480 2015 1,900 1,244 2023 36,300 29,040 2016 1,900 1,244 2024 41,150 32,920 2017 1,900 1,244 2025 46,450 37,160 2018 1,900 1,244 2026 52,250 41,800 2019 1,900 1,244 2027 58,550 46,840 2020 1,900 1,244 2028 65,450 52,360 2021 1,900 1,244 2029 73,000 58,400 2022 1,900 1,244 2030 81,200 64,960 2023 1,900 1,244 2031 90,200 72,160 2024 1,900 1,244 BE3   574,520 2025 1,900 1,244 2026 1,900 1,244 Step 5: Calculate Project Power Generation 2027 1,900 1,244 Emissions for the T&D Projects 2028 1,900 1,244 2029 1,900 1,244 Technical Loss Reduction 2030 1,900 1,244 2031 1,900 1,244 Module PE2: Emissions from Expected Project Loss Rates. Because the FSR only provides data on PE4 26,115 total loss reduction, this module is not necessary. T&D Capacity Expansion Project losses are zero, and baseline losses represent the full benefit of the loss reduction. Module PE4: Emissions from Project Grid. The Increased Reliability feasibility study includes a table that summarizes the main operating parameters for the project’s Module PE4: Emissions from Project Grid. The economic analysis. The table provides figures on feasibility study includes a table that summarizes incremental energy supplied to customers after grid the main operating parameters for the project’s extension, which increases from 6.7 GWh in 2012 economic analysis. From this table, a figure of to 180 GWh in 2031. However, the study does not 1,900 MWh annual loss reduction is derived. clarify whether incremental energy supply is for “Annual loss reduction” in this context means reduc- connection of new consumers to the electricity grid tion of financial losses in the form of lost MWh (that is, electrification) or for supplying additional because of improved reliability. The emission factor power generation to existing consumers where there 7. Case Studies 84 is no alternative source of supply for these consum- Electrification ers. To test the modeling tool, it has been assumed Module PE4: Emissions from Project Grid. As that 50 percent of the incremental energy will be for explained above, 50 percent of the incremental capacity expansion and 50 percent for electrifica- energy supplied is assumed to be for electrification. tion. The critical difference here is not in project emissions, since Module PE4 would be the same for Parameter Unit Value either project type. Rather, the difference is that the EFCM tCO2/MWh 0.6545 baseline alternative for electrification is diesel gen- IL % 0 erators. Year IEy PEy Parameter Unit Value MWh tCO2 EFCM tCO2/MWh 0.6545 2012 3,350 2,193 IL % 0 2013 5,200 3,403 2014 7,250 4,745 Year IEy PEy 2015 9,500 6,218 MWh tCO2 2016 11,950 7,821 2012 3,350 2,193 2017 14,600 9,556 2013 5,200 3,403 2018 17,450 11,421 2014 7,250 4,745 2019 20,600 13,483 2015 9,500 6,218 2020 24,050 15,741 2016 11,950 7,821 2021 27,800 18,195 2017 14,600 9,556 2022 31,850 20,846 2018 17,450 11,421 2023 36,300 23,758 2019 20,600 13,483 2024 41,150 26,933 2020 24,050 15,741 2025 46,450 30,402 2021 27,800 18,195 2026 52,250 34,198 2022 31,850 20,846 2027 58,550 38,321 2023 36,300 23,758 2028 65,450 42,837 2024 41,150 26,933 2029 73,000 47,779 2025 46,450 30,402 2030 81,200 53,145 2026 52,250 34,198 2031 90,200 59,036 2027 58,550 38,321 PE4 470,029 2028 65,450 42,837 2029 73,000 47,779 2030 81,200 53,145 Step 6: Summarize GHG Emissions Impacts 2031 90,200 59,036 Table 7.3 presents a summary of the estimated GHG PE4 470,029 impacts from this project. 7. Case Studies 85 Advance metering infrastructure Table 7.3: Summary of GHG impacts for Eldoret-Kitale Line (tCO2) Modernization of distribution company man- agement information system Direct nongeneration impacts Component 2: Institutional Strengthening Embodied n.a. emissions The subtransmission and distribution network Energy in con- n.a. reinforcement subcomponent aims to strengthen struction and rehabilitate the subtransmission and distribu- Land clearing 13,860 tion grid, including strengthening and rehabilitat- SF6 7,490 ing substations, which would entail the acquisition Baseline Project Net and installation of cables, transformers, switches, breakers, posts, automatic meters in feeders, pro- Direct generation impacts tection systems, ancillary equipment, and so on. Technical loss 37,961 0 −37,961 reduction Other equipment to be acquired and installed include distribution equipment for the supervisory Indirect generation impacts control, voltage control, and switching needed to Increased reli- 0 26,115 26,115 improve the reliability and quality of the electric- ability ity supply. This subcomponent, which would rep- Capacity 0 470,029 470,029 resent the bulk of the project investment, would expansion help reduce service interruptions, reduce technical Electrification 574,520 470,029 −104,491 losses, and improve the ability of the distribution Cross-border n.a. n.a. companies to manage the grid effectively (includ- trade ing reducing nontechnical and billing losses). It is Source: Authors’ analysis. this subcomponent that will lead to the impacts on Note: n.a. = not applicable. emissions evaluated here. The relevant data for this project was sourced from Case Study 3: Eletrobras Distribution a number of World Bank documents, including the project concept note dated August 2009, the invest- Rehabilitation Project, Brazil ment analysis spreadsheets dated September 2009, The proposed project would strengthen the manage- and the World Bank project appraisal document ment, operations, and corporate governance of the dated January 2009. No consultant feasibility study six distribution companies managed by Eletrobras was made available for this project. (Amazonas Energia, Eletroacre, Ceron, Boa Vista, Cepisa, and Ceal), through the following compo- Description of Modules and Data Availability nents and subcomponents. Step 1: Determine Which Direct Nongeneration Component 1: Service Quality Improvement Emissions Will Be Included and Loss Reduction Program The following table determines which direct non- Subtransmission and distribution network generation emissions calculation modules apply to reinforcement this project. 7. Case Studies 86 Question Answer Step 4: Calculate Baseline Power Generation Are data available on materials con- No Emissions for the T&D Projects sumption by the T&D project and on the origin of those materials? Technical Loss Reduction Are data available on energy consump- No Module BE2: Emissions from Existing Technical tion during the construction phase of the T&D project? Loss Rate. The investment analysis spreadsheets Does the T&D project involve clearing No provide data on transmission losses both before and any land? after project implementation for each of the six dis- Does the T&D project include new lines No tribution companies. The emission factor (combined or capacity expansion that includes margin) for the Brazilian grid is also available from new SF6-containing equipment? the IGES database (0.1045 tCO2e/MWh). None of the equipment to be installed will require The feasibility study does not provide figures on additional right of ways, since the project will reha- technical losses for each year, but rather provides bilitate or strengthen existing distribution infra- figures on technical loss reductions in MWh for structure only, and Module D3 is therefore not used. a 10-year period (5,464 GWh over 10 years). This As this project involves only technical loss reduction information is incorporated into the model by des- and increased reliability, Module D4 (SF6 Emissions) ignating this amount as the baseline losses and set- also does not apply because of the low-voltage level ting project losses to zero. of the system. The project team confirmed with the distribution companies that SF6 is not installed in The results for this module are summarized low-voltage distribution lines, which is the main below: focus of the project. n BE2 =y∑(TLBL,y × EFCM) Step 2: Calculate Direct Nongeneration Emissions =1 for the T&D Projects Parameter Unit Value This step is not applicable, since none of the direct EFCM tCO2/MWh 0.1045 nongeneration emissions calculation modules apply Year TLBL,y BEy to this project.   MWh tCO2 Step 3: Determine How Baseline and Project 2012 546,400 57,099 Emissions from Power Generation Should Be 2013 546,400 57,099 Calculated 2014 546,400 57,099 2015 546,400 57,099 This project leads to technical loss reduction and 2016 546,400 57,099 increased reliability. The decision trees for these 2017 546,400 57,099 project types are presented in figures 6.4 and 6.5, 2018 546,400 57,099 respectively. For technical loss reduction, baseline 2019 546,400 57,099 emissions should be calculated using Module BE2 and project emissions using Module PE2. For 2020 546,400 57,099 increased reliability, baseline and project emissions 2021 546,400 57,099 should use Modules BE4 and PE4, respectively. BE2   570,988 7. Case Studies 87 Increased Reliability Year IEy BEy Module BE3: Emissions from Alternative Baseline MWh tCO2 Energy Source. The reduction in frequency and 2012 26,092 20,874 duration of interruptions in electricity supply 2013 26,092 20,874 because of improved service quality would result in 2014 26,092 20,874 an increase of electricity supply of 606.8 GWh over 2015 26,092 20,874 a 10-year period. Distribution company customers 2016 26,092 20,874 would be able to consume electricity during periods 2017 26,092 20,874 in which they currently experience interruptions in 2018 26,092 20,874 supply and abnormal voltage drops, thus resorting 2019 26,092 20,874 to reduced consumption or the use of alternative 2020 26,092 20,874 energy sources. The emission factors used are the 2021 26,092 20,874 default value for diesel generators taken from the BE3   208,736 CDM methodology AMS I.D. and from the IGES database. In this case study, it is assumed from Step 5: Calculate Project Power Generation country experience that only the medium- and Emissions for the T&D Projects high-voltage customers—which have been calcu- lated to account for 43 percent of total consumption Technical Loss Reduction in the six distribution companies—will have access Module PE2: Emissions from Expected Project to alternative energy sources (diesel generator sets). Thus, for the baseline, 260.9 GWh (43 percent of Loss Rates. As described above, technical losses in 606.8 GWh) of the electricity supplied originates the project scenario are set to zero because all loss from the use of diesel generators (default value for reductions are captured in the baseline. diesel generators used). For the remainder of the Increased Reliability electricity supplied (57 percent of 606.8 GWh), it is assumed that there would be no power consumption Module PE4: Emissions from Project Grid. The in the absence of the project (baseline BE4 = 0). The results for this module are summarized below: results for this module are summarized below: n n PE4 =y∑(IEy × EFCM)/(1 − IL) BE3 =y∑(IEy × EFAE) =1 =1 Parameter Unit Value Parameter Unit Value EFCM tCO2/MWh 0.1045 EFAE tCO2/MWh 0.8 IL % 0 7. Case Studies 88 Year IEy PEy Table 7.4: Summary of GHG Impacts for Eletrobras   MWh tCO2 Distribution Rehabilitation Project (tCO2) 2012 60,680 6,341 2013 60,680 6,341 Direct nongeneration impacts 2014 60,680 6,341 Embodied n.a. 2015 60,680 6,341 emissions 2016 60,680 6,341 Energy in con- n.a. 2017 60,680 6,341 struction 2018 60,680 6,341 Land clearing n.a. 2019 60,680 6,341 SF6 n.a. 2020 60,680 6,341 Baseline Project Net 2021 60,680 6,341 PE4   63,411 Direct generation impacts Technical loss 570,988 0 −570,988 Step 6: Summarize GHG Emissions Impacts reduction Indirect generation impacts Table 7.4 presents a summary of the estimated GHG impacts from this project. Increased reli- 208,736 63,411 −145,325 ability Capacity n.a. n.a. Summary of Results and Conclusions expansion from the Three Case Studies Electrification n.a. n.a. Table 7.5 summarizes the results from the three case Cross-border n.a. n.a. studies presented above: trade Source: Authors’ analysis. Some important conclusions on the feasibility Note: n.a. = not applicable. of implementing the proposed approach can be gleaned from the three case studies. While the approach requires a certain amount of data analysis and processing, the additional effort will not drasti- evaluations of the projects—will be required in cally increase the effort required to perform eco- order to gather the data. Referring to this document, nomic and technical analysis. For a typical project those team members performing the technical and component where two or three GHG estimation economic evaluation should be able to conduct the modules may be required, approximately three days analysis by themselves in the same number of days. are needed for data collection and setup plus a day of analysis. Some knowledge of GHG accounting While current project preparation procedures will be required on the part of the analyst, especially already provide most of the data that will be impor- to interpret the modules and understand the data tant to estimate net impacts, some improved data requirements. For a typical Bank project, which may collection will be needed, especially for the direct include one component in transmission and one nongeneration emissions modules. Determining in distribution, the total cost would be four days the type of incremental demand being served by the of work by a research analyst. Strong collaboration T&D project is also important for the correct appli- between this analyst and the project team—espe- cation of some modules, such as electrification and cially with those team members or consultants deal- capacity expansion. Some of the issues regarding ing with the economic and social and environmental data availability are further developed below. 7. Case Studies 89 Table 7.5: Summary Results for Three Case Studies (tCO2) Case 2 Case 1 Project I Project II Case 3 Direct non- Embodied emissions n.a. n.a. n.a. n.a. generation impacts Energy in construction n.a. n.a. n.a. n.a. Land clearing 554,400 2,244 13,860 n.a. SF6 249,971 n.a. n.a. n.a. Direct genera- Technical loss reduction n.a. −83,541 −37,961 −570,988 tion impacts Indirect Increased reliability n.a. 42,032 26,115 −145,325 generation impacts Capacity expansion n.a. 66,255 470,029 n.a. Electrification n.a. n.a. −104,491 n.a. Cross-border trade −69,817,071 n.a. n.a. n.a. Source: Authors’ analysis. Note: A negative total represents a reduction in GHG emissions. Project I refers to the Kisii-Awendo transmission project; Project II refers to Eldoret-Kitale. n.a. = not applicable. Embodied Emissions in Materials always available in the documentation provided, and was not described using the same classification This impact can only be evaluated where the project as the IFC CEET table or other sources that provide preparation documentation provides the necessary emission factors for different vegetation types (for data. Data on materials that would be used during example, annex A, table A.1). construction may be available only for large projects (for example, large transmission interconnectors). SF6 Emissions Such data are not usually collected for smaller proj- ects (for example, distribution rehabilitation). The Data were not available on the nameplate capacity origin of materials is generally not known during of the SF6-containing equipment to be installed in the project preparation phase, but rather only after most cases. This may be because such equipment is the construction contracts are awarded, according to not installed for all T&D projects, especially for low- Bank procurement rules. Emissions from energy use voltage distribution lines. When such equipment is in construction have similar data collection needs. installed, data on SF6 capacity and leakage rates are not traditionally collected. A systemwide average SF6 Land Clearing emission factor can be applied where project-spe- cific equipment data are not available. This emission The impacts of land clearing can be estimated more easily. Information is usually available as part of the factor should reflect the fact that most SF6 emissions project’s environmental and social analysis. If this will be from high-voltage equipment. information is not available, standard right-of-way Technical Loss Reduction widths can be used for each voltage level. The length of the line would almost always be available dur- Figures for technical loss reduction rates were avail- ing project preparation. Vegetation type was not able for most case studies, although the information 7. Case Studies 90 was not always provided in the format used in the narios that can provide most of the information relevant module. This will not affect the accuracy required. of the results, as long as total reduction in losses is Project lifetimes of between 10 and 21 years were available in the project documentation. used in the economic analysis of the projects based Increased Reliability on the information available in the feasibility stud- ies. This lifetime may be appropriate, given the dif- In the two case studies where increased reliabil- ferent technologies and equipment installed. For ity is an important project component, relevant the proposed approach, the time frame for assessing data on increased energy supply was available, so GHG accounting should be consistent with that this module could be applied successfully. Energy used for the economic analysis. demand not served because of reliability problems is a parameter input for most economic evaluations of Nongeneration versus Generation Emissions T&D projects. The three cases explored indicate that direct non- T&D Capacity Expansion and Electrification generation emissions are relatively small compared to direct and indirect impacts on power generation. This The most challenging task in these two modules is is supported by evidence from the literature review. to identify the type of incremental demand that will In all cases, direct nongeneration emissions range be served by the project. Project teams will need from 0 to 6 percent of generation impacts. The direct to differentiate between suppressed demand and nongeneration emissions for the interconnection demand that, in the absence of the project, will be between Ethiopia and Kenya are estimated at +804 supplied by alternative on-site electricity supply. Not ktCO2, largely from land clearing, while the indirect all demand forecasts used in economic evaluation impact on power generation is estimated at −69,812 of distribution projects will make this distinction, ktCO2 because of the displacement of power from a which has an important impact on emissions as pre- higher emissions grid. For one of the transmission sented in the examples. projects in Kenya, direct nongeneration emissions are estimated at +14 ktCO2, while the direct genera- Cross-Border Trade tion impact is −38 ktCO2 and the indirect generation This module was only used for the first case impact is +392 ktCO2. The T&D rehabilitation project study and did not pose any particular difficulties. in Brazil results in a direct generation impact of −571 International interconnector projects are likely to ktCO2 and an indirect generation impact of −145 have power system and generation simulation sce- ktCO2, and has no direct nongeneration emissions. 7. Case Studies 91 8. Conclusions The objective of this study was to review existing no accepted methodologies at all in the context of methodologies and to recommend feasible ones climate financing mechanisms such as the CDM, that capture the most relevant GHG impacts of which underscores the importance of this study. The T&D projects in the context of the World Bank direct generation impacts of technical loss reduc- project preparation cycle. The diversity and quantity tion and the indirect generation impacts of elec- of T&D interventions, their varied technical and trification are noted in several methodologies and economic impacts, and data availability at the time international studies, including the World Bank’s of project preparation emphasize the necessity of GHG Assessment Handbook. However, impacts such a flexible, modular approach. The study approach as increased reliability and T&D capacity expan- and conclusions are not intended to be the final sion have not been analyzed for their GHG impacts. word on T&D project GHG emissions accounting, Cross-border trade, although discussed by several but instead viewed as a starting point for accurately proposed CDM methodologies, also does not have understanding the most important implications an accepted standard of analysis. of T&D interventions using a framework that can One of the most important conclusions of this work be implemented credibly in the context of project is that the impacts of T&D projects on power gen- preparation. eration emissions are likely to be much greater than direct nongeneration emissions. For some projects, Importance of Net Emissions the net emissions impacts could be negative; that Accounting and Including Power is, the project contributes to reduced overall power Generation Emissions Impacts system emissions even though direct nongenera- tion emissions are positive. Although this increases The survey of methodologies and case studies the level of effort required to assess the impacts on indicate that direct nongeneration emissions for power generation, not analyzing the impacts on gen- T&D projects are well covered by many existing eration emissions could significantly underestimate approaches. There is broad consistency on the type the impact of T&D projects on GHG emissions. of emissions that are relevant and how they can be estimated. In addition to data now being collected for technical and economic assessment of projects, Implementation Issues: Level some additional data are needed to estimate emis- of Effort, Data Collection, and sions. Direct nongeneration emissions from T&D Uncertainty projects are small when compared to the impacts of While the proposed approach is relatively simple T&D projects on power generation emissions. and robust for estimating the most important GHG There is very little experience with the analysis impacts of World Bank T&D interventions, this of the effects of T&D projects on emissions from analysis will require some additional effort from power generation. Several key project types have project teams. For some projects, this effort will 93 involve additional data collection; mainly, it will integrate the data collection process into the project entail additional time in applying the modules. preparation cycle. This assessment is based on the work entailed in Some limited additional data are needed to assess screening the project appraisal documents listed in the impact on power generation emissions. The annex B and for the three case studies presented in majority of these modules require some of the mar- chapter 7 and the projects under preparation used ginal grid emission factors. Although emission fac- in this report to perform GHG accounting with the tors from the IGES CDM database or a registered proposed approach. For a typical two-component CDM project can be used, project teams could project, the research and analysis required to per- consider collecting primary data from a national form GHG accounting using the proposed approach utility or similar source during project preparation. should take a research analyst a total of 8 days, Projects that conduct a power generation simulation working in coordination with the team members for their economic analysis will have this informa- performing the technical and economic evaluation tion. However, distribution and electrification proj- of the project. ects generally will not collect such information. Some data collection issues that project teams need An important challenge in assessing net impacts of to be aware of include the following: increased reliability, technical loss reduction, and For direct nongeneration emissions accounting, capacity expansion projects is the clear separation the quantity of construction materials required of the impacts of these objectives, both theoretically for different projects is not usually known with and practically. While load flow and long-term eco- certainty at the time of project preparation nomic dispatch simulations could provide reliable because the detailed feasibility studies have not information to supplement all the modules, they are yet been completed. The relatively small size of not conducted for all types of projects. If the impacts this impact would not merit additional effort by on losses and reliability are determined separately, project teams. it is essential that the teams use consistent baselines and project scenarios. For instance, if the impact in While land clearing is generally covered in the losses of the project is estimated for an entire net- environmental and social impact assessments, work, then the impact of the project on increased the project documentation should clarify the transmission capacity should also be analyzed for IPCC-defined vegetation types so the correct the entire network. emission factors can be used. For capacity expansion projects and, to a lesser Detailed data on equipment containing SF6 is extent, electrification projects, an additional a gap that must be addressed, particularly for source of uncertainty is in the manner in which high-voltage equipment. Existing environmental the baseline captures alternatives to the grid. In and social safeguards require regulated handling other words, if a capacity expansion project were of SF6, but there is no requirement to quantify not implemented, would the customers find other fugitive emissions or specify the characteristics equivalent power sources? This is both a question of of all equipment being installed. principle and of practice. The principle issue is that The review of existing studies and the case studies economic development will drive the need for more indicate that direct nongeneration emissions from power, and must be provided by the grid or by other T&D projects are small relative to power genera- sources. Even if those alternatives are not currently tion emissions impacts. Erring on the high side of in place, to exclude them from the baseline would estimating direct nongeneration emissions is prefer- be in essence to assume that the demand for power able to underestimating them. The best solution is to is not growing. At the same time, the reality is that 8. Conclusions 94 the lack of power is a major constraint to develop- Lessons for the Bank’s Overall Effort ment, and there are many large industrial projects on GHG Accounting under the SFDCC that would not be implemented without significant T&D capacity expansion. The practical issue is This study has shown that simple, feasible, and cred- whether the project’s technical and economic analy- ible methodologies for direct nongeneration impacts sis provides information on how electricity demand and generation emissions impacts of T&D interven- will be supplied if the project is not implemented. tions can be applied to World Bank projects. While Uncertainty is always present in project evaluation the Bank’s interventions differ significantly from and will affect the credibility of the baseline and private sector transactions and traditional CDM project scenarios. The approach should be consis- projects, the existing project preparation cycle and tent and applicable to current project preparation formal requirements for technical, economic, and practices. Thus, if alternative power sources have environmental analysis of projects provide a good been identified in a project’s technical and economic starting platform for GHG accounting in the T&D assessment, they should be used to define the base- sector. Implementing GHG accounting with the pro- line emissions. A zero emissions baseline should be posed approach does not present a major burden to assumed if no alternative source is identified. project teams, although certain data collection issues do need to be addressed. For cross-border trade projects where load flow and long-term dispatch modeling data exist, estimating The review and results obtained emphasize the emissions impacts is straightforward. This is likely to importance of the T&D infrastructure sector in be the case for some large interconnector projects, achieving lower-carbon development paths in the but certainly not for all. Where these data are not power sector, which has historically been largely available, the challenge is to determine whether the overlooked. The T&D sector is particularly signifi- use of marginal emission factors for the grid accu- cant in World Bank operations in client countries rately represents the impacts on dispatch caused where losses are high or electricity systems are weak by the project. The answer to this will depend on and relatively small. An efficient and integrated grid the dispatch systems used for both grids, their level can enable large-scale investment in clean technolo- of integration, and the grid characteristics. Using gies and increase the operational efficiency of exist- marginal grid emission factors is feasible and in line ing power generation sources, potentially reducing with some of the proposed CDM methodologies, emissions. In electricity systems where reliability is but sacrifices a certain level of accuracy. low and technical losses in the T&D sector are high, T&D investments can have major impacts on low- For electrification, the main challenge lies in carbon growth. addressing the displacement of fuels other than electricity. All the existing methodologies only look As some of the case studies show, a reliable and at the displacement of alternative forms of electric- efficient transmission system can contribute to the ity by grid power installed. This is even true of the fulfillment of the twin objectives of efficient and reli- small-scale CDM methodologies applied to renew- able energy supply and a contribution to reduced able energy systems for individual households. emissions. This is especially true in situations where While electricity will clearly displace some other low reliability and losses lead to wasteful use of energy sources, such as kerosene use for lighting, power generation sources and therefore to increased the quantity displaced, the time frame, and the dis- emissions. It is also true where transmission sys- placement conditions are complex issues that need tems in developing and developed countries are significant attention. The World Bank is undertaking challenged by the need to connect more renewable additional work on this important subject. energy sources. The transmission system will be an 8. Conclusions 95 important enabler to ensure that the power sector adoption of GHG accounting procedures for Bank can move toward lower-carbon power generation operations may require some uniformity and con- options. This work provides a platform to estimate sistency across all sectors. As the work on piloting the GHG impacts of T&D projects and contributes GHG accounting in other sectors moves forward, to the ongoing work on low-carbon development a Bank-wide proposal on GHG analysis should be planning being undertaken at the World Bank. proposed to the Board as envisaged by the SFDCC. The SFDCC clearly states that GHG accounting is The proposed approach has been designed to suit an analytical exercise and should not be used as a the structure of Bank projects that contain T&D decision-making tool for Bank-financed projects. components that can be categorized as subsectoral The purpose of this effort is to increase knowledge programs and not discrete projects, such as many and capacity building, understand the implications carbon financing transactions. The move toward of new approaches on GHG accounting, and facili- more comprehensive or sector-based approaches by tate the use of emerging climate financing funds. climate financing mechanisms is increasingly being This work has also aimed at increasing knowledge recognized as a possible solution to the drawbacks and understanding of the implications of new of project-based climate financing mechanisms approaches and Bank interventions. The formal (Bodansky 2007; CCAP 2008). 8. Conclusions 96 Annex A: Data Tables for Methodology Proposals Table A.1: Carbon Density in Biomass Types (C) Above-ground bio- (B) Above-ground bio- mass (t dry matter/ha) (F) Below-ground bio- ground ratio (t d.m.) mass (t C/ha = AxB) mass (t C/ha = CxE) (t C/t dry matter) (E) Below/above Carbon fraction Type of tree (Please select from drop-down list)  Natural Forest—Tropical (avg) 164.0 77.1 0.34 25.9 0.47 Natural Forest—Tropical rainforest 300 141.0 0.37 52.2 0.47 Natural Forest—Tropical moist deciduous 180 84.6 0.22 18.6 0.47 Natural Forest—Tropical dry 130 61.1 0.42 25.7 0.47 Natural Forest—Tropical shrubland 70 32.9 0.40 13.2 0.47 Natural Forest—Tropical mountain systems 140 65.8 0.27 17.8 0.47 Natural Forest—Subtropical (avg) 140 65.8 0.32 21.1 0.47 Natural Forest—Subtropical humid 220 103.4 0.22 22.7 0.47 Natural Forest—Subtropical dry 130 61.1 0.42 25.7 0.47 Natural Forest—Subtropical steppe 70 32.9 0.32 10.5 0.47 Natural Forest—Subtropical mountain systems 140 65.8  n.a.  n.a. 0.47 Natural Forest—Temperate (avg) 133.3 62.7 0.25 15.5 0.47 Natural Forest—Temperate oceanic 180 84.6 0.22 18.6 0.47 Natural Forest—Temperate continental 120 56.4 0.26 14.7 0.47 Natural Forest—Temperate mountain systems 100 47.0 0.26 12.2 0.47 Natural Forest—Boreal (avg) 31.7 14.9 0.39 5.8 0.47 Natural Forest—Boreal coniferous 50 23.5 0.39 9.2 0.47 Natural Forest—Boreal tundra woodland 15 7.1 0.39 2.7 0.47 Natural Forest—Boreal mountain systems 30 14.1 0.39 5.5 0.47 97 (C) Above-ground bio- (B) Above-ground bio- mass (t dry matter/ha) (F) Below-ground bio- ground ratio (t d.m.) mass (t C/ha = AxB) mass (t C/ha = CxE) (t C/t dry matter) (E) Below/above Carbon fraction Type of tree (Please select from drop-down list)  Plantation Forest—Tropical (avg) 90.0 42.3 0.34 14.2 0.47 Plantation Forest—Tropical rain forest 150 70.5 0.37 26.1 0.47 Plantation Forest—Tropical moist deciduous forest 120 56.4 0.22 12.4 0.47 Plantation Forest—Tropical dry forest 60 28.2 0.42 11.8 0.47 Plantation Forest—Tropical shrubland 30 14.1 0.40 5.6 0.47 Plantation Forest—Tropical mountain systems 90 42.3 0.27 11.4 0.47 Plantation Forest—Subtropical (avg) 80.0 37.6 0.32 12.0 0.47 Plantation Forest—Subtropical humid forest 140 65.8 0.22 14.5 0.47 Plantation Forest—Subtropical dry forest 60 28.2 0.42 11.8 0.47 Plantation Forest—Subtropical steppe 30 14.1 0.32 4.5 0.47 Plantation Forest—Subtropical mountain systems 90 42.3 0.00  n.a. 0.47 Plantation Forest—Temperate (avg) 120.0 56.4 0.25 13.9 0.47 Plantation Forest—Temperate oceanic forest 160 75.2 0.22 16.5 0.47 Plantation Forest—Temperate continental forest 100 47.0 0.26 12.2 0.47 Plantation Forest—Temperate mountain systems 100 47.0 0.26 12.2 0.47 Plantation Forest—Boreal (avg) 28.3 13.3 0.39 5.2 0.47 Plantation Forest—Boreal coniferous forest 40 18.8 0.39 7.3 0.47 Plantation Forest—Boreal tundra woodland 15 7.1 0.39 2.7 0.47 Plantation Forest—Boreal mountain systems 30 14.1 0.39 5.5 0.47 Cropland—Temperate (all regions), woody biomass  n.a. 63  n.a.  n.a. 0.47 Cropland—Tropical (dry region), perennial woody biomass  n.a. 9  n.a.  n.a. 0.47 Cropland—Tropical (moist region), perennial woody biomass  n.a. 21  n.a.  n.a. 0.47 Cropland—Tropical (wet region), perennial woody biomass  n.a. 50  n.a.  n.a. 0.47 Cropland—Annual crops (all) 10 4.7  n.a.  n.a. 0.5 Grassland—Boreal (dry and wet) 1.7 0.68 4.0 1.6 0.4 Grassland—Cold Temperate (dry) 1.7 0.68 2.8 1.1 0.4 Annex A: Data Tables for Methodology Proposals 98 (C) Above-ground bio- (B) Above-ground bio- mass (t dry matter/ha) (F) Below-ground bio- ground ratio (t d.m.) mass (t C/ha = AxB) mass (t C/ha = CxE) (t C/t dry matter) (E) Below/above Carbon fraction Type of tree (Please select from drop-down list)  Grassland—Cold Temperate (wet) 2.4 0.96 4.7 1.9 0.4 Grassland—Warm Temperate (dry) 1.6 0.64 2.8 1.1 0.4 Grassland—Warm Temperate (wet) 2.7 1.08 4.0 1.6 0.4 Grassland—Tropical (dry) 2.3 0.92 2.8 1.1 0.4 Grassland—Tropical (moist and wet) 6.2 2.48 1.6 0.6 0.4 Settlement—Construction  n.a.  n.a.  n.a.  n.a.  n.a. Source: IPCC 2006. Note: Entries in blue are an average of IPCC values. n.a. = not applicable. Table A.2: Default Emission Factors for Generator Systems in Small-Scale Diesel Power Plants for Three Load Factor Levels (kg CO2e/kWh) Load factor (%) Minigrid with tempo- rary service (4–6 hr/day; Minigrid with 24-hour productive applications; service water pumps) Minigrid with storage Case (25%) (50%) (100%) < 15 kW 2.4 1.4 1.2 ≥ 15 < 35 kW 1.9 1.3 1.1 ≥ 35 < 135 kW 1.3 1.0 1.0 ≥ 135 < 200 kW 0.9 0.8 0.8 > 200 kWa 0.8 0.8 0.8 Source: UNFCCC 2009a. Note: A conversion factor of 3.2 kg CO2 per kg of diesel has been used (following revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories). Values are derived from fuel curves in the online manual of RETScreen lnternational’s PV 2000 model, downloadable from http://retscreen.net/. a. Default values. Annex A: Data Tables for Methodology Proposals 99 Table A.3: Default Energy Efficiencies of Different Power Plant Types (%) Power plant type Old (before 2000) New (after 2000) Coal Subcritical 37 39 Supercritical 45 Ultrasupercritical 50 IGCC 50 FBS 35.5 CFBS 36.5 40 PFBS 41.5 Oil Steam turbine 37.5 39 Open cycle 30 39.5 Combined cycle 46 46 Natural gas Steam turbine 37.5 37.5 Open cycle 30 39.5 Combined cycle 46 60 Source: UNFCCC 2009e. Annex A: Data Tables for Methodology Proposals 100 Annex B: World Bank T&D Projects $ mil- FY Project ID Project Name Region Country Product line lions URL 2003 P063913 ID-Java-Bali Pwr EAP Indonesia IDA 99.09 P063913 Sector & Strength 2003 P043311 Power Development SAR Nepal IDA 26.55 P043311 Project 2004 P083908 Emergency Power SAR Afghanistan IDA 84.75 P083908 Rehabilitation Project 2004 P064844 KH-Rural Electrif. & EAP Cambodia IDA 16.97 P064844 Transmn 2004 P069183 MZ—Energy Reform AFR Mozambique IDA 30.60 P069183 and Access SiL (FY2004) 2004 P066532 PH-GEF-Electric EAP Philippines GEF 12.00 P066532 Cooprtv System Loss Redu 2005 P094735 Emerg National SAR Afghanistan IDA 5.60 P094735 Solidarity— Supplemental 2005 P075994 3A-WAPP Phase 1 AFR AFR Region IDA 40.00 P075994 APL 1 (FY2005) 2005 P090656 ECSEE APL2 (Albania) ECA Albania IDA 27.00 P090656 2005 P083341 Power Transmission ECA Azerbaijan IDA 48.00 P083341 2005 P079633 BJ-Energy Srvc AFR Benin IDA 28.80 P079633 Delivery APL (FY2005) 2005 P076807 CL-Infrastructure for LCR Chile IBRD 4.52 P076807 Territorial Dvlpmt 2005 P088619 CD-Emergen Living AFR Congo, Dem. Rep. IDA 12.30 P088619 Conditions Impr (FY2005) 2005 P082712 DO Power Sector LCR Dominican IBRD 150.00 P082712 Program Loan Republic 101 $ mil- FY Project ID Project Name Region Country Product line lions URL 2005 P057929 ER-Power AFR Eritrea IDA 45.00 P057929 Distribution SIL (FY2005) 2005 P083131 KE-Energy Sec AFR Kenya IDA 68.00 P083131 Recovery Prj (FY2005) 2005 P090194 RW-Urgent Electricity AFR Rwanda IDA 3.00 P090194 Rehab SIL (FY2005) 2005 P073477 SN-Elec Sec Effi. AFR Senegal IDA 8.60 P073477 Enhanc. Phase 1 APL-1 2005 P085708 SN-Elec. Serv. for AFR Senegal IDA 17.04 P085708 Rural Areas (FY2005) 2005 P088867 ECSEE APL #2 (Serbia) ECA Serbia IBRD 21.00 P088867 2005 P087203 SL-Power & Water SIL AFR Sierra Leone IDA 15.40 P087203 (FY2005) 2005 P094176 ECSEE APL #2 ECA Turkey IBRD 66.00 P094176 (Turkey) (CRL) 2005 P074688 VN-Rural Energy 2 EAP Vietnam IDA 220.00 P074688 2006 P094917 3A-WAPP APL 1 (CTB AFR AFR Region IDA 3.00 P094917 Phase 2) Project 2006 P094917 3A-WAPP APL 1 (CTB AFR AFR Region IDA 27.00 P094917 Phase 2) Project 2006 P094917 3A-WAPP APL 1 (CTB AFR AFR Region IDA 27.00 P094917 Phase 2) Project 2006 P090666 ECSEE APL3-BiH ECA Bosnia and IDA 36.00 P090666 Herzegovina 2006 P093787 BR Bahia State Integ LCR Brazil IBRD 8.70 P093787 Proj Rur Pov 2006 P052256 BR-MG Rural Poverty LCR Brazil IBRD 8.75 P052256 Reduction 2006 P096305 CD-Emerg MS Rehab AFR Congo, Dem. Rep. IDA 10.00 P096305 & Recov ERL Sup (FY2006) 2006 P086379 DJ-Power Access And MNA Djibouti IDA 1.54 P086379 Diversification 2006 P097271 ET-Electricity Access AFR Ethiopia IDA 130.73 P097271 (Rural) Expansion 2006 P097975 GW-MS Infrastructure AFR Guinea-Bissau IDA 6.00 P097975 Rehab SIM (FY2006) 2006 P086775 HN (CRL1) Rural LCR Honduras IDA 3.76 P086775 Infrastructure Project Annex B: World Bank T&D Projects 102 $ mil- FY Project ID Project Name Region Country Product line lions URL 2006 P086414 Power System SAR India IDA 400.00 P086414 Development Project III 2006 P091299 JM Inner City Basic LCR Jamaica IBRD 1.47 P091299 Services Project 2006 P095155 N-S Elec Transm ECA Kazakhstan IBRD 100.00 P095155 2006 P100160 LR-Emergency AFR Liberia IBRD 2.70 P100160 Infrastructure ERL (FY2006) 2006 P082337 ECSEE APL #3 ECA Macedonia, FYR IBRD 25.00 P082337 (Macedonia, FYR) 2006 P057761 MW-Infrastr Srvcs SIM AFR Malawi IDA 5.20 P057761 2006 P096598 ECSEE APL #3— ECA Montenegro IBRD 1.71 P096598 Montenegro 2006 P090104 NG-Natl Energy Dev AFR Nigeria IDA 146.20 P090104 SIL (FY2006) 2006 P088181 TP Consolidation EAP Timor-Leste IDA 0.07 P088181 Support Program (CSP) 1 2006 P096400 ECSEE APL #3 ECA Turkey IBRD 150.00 P096400 (Turkey) 2006 P084871 VN-Trans & Distrib 2 EAP Vietnam IDA 200.00 P084871 2006 P086865 RY-Power Sector MNA Yemen, Rep. IDA 44.00 P086865 2006 P086865 RY-Power Sector MNA Yemen, Rep. IDA 6.00 P086865 2007 P090928 AF PSD Support SAR Afghanistan IDA 7.50 P090928 Project 2007 P095229 AO-MS ERL 2 AFR Angola IDA 25.50 P095229 2007 P105329 KH-GMS Power Trade EAP Cambodia IDA 18.50 P105329 Project 2007 P094306 JO-Amman East MNA Jordan Guarantees 45.00 P094306 Power Plant 2007 P098949 VIP 2 ECA Kyrgyz Republic IDA 1.20 P098949 2007 P105331 LA-GMS Power Trade EAP Lao PDR IDA 12.90 P105331 Project 2007 P104774 LB-Emergency Pwr MNA Lebanon SF 2.50 P104774 Reform Capacity Reinfor 2007 P095240 MG—Pwr/Wtr AFR Madagascar IDA 6.30 P095240 Sect. Recovery and Restruct. Annex B: World Bank T&D Projects 103 $ mil- FY Project ID Project Name Region Country Product line lions URL 2007 P096801 Elect Distrib Rehab ECA Turkey IBRD 269.40 P096801 2007 P069208 UG-Power Sector AFR Uganda IDA 288.00 P069208 Dev. Project (FY2007) 2007 P074594 GZ-Emergency MNA West Bank and SF 2.30 P074594 Municipal Service Gaza Rehab II 2008 P106654 ARTF Kabul-Aybak SAR Afghanistan RE 52.44 P106654 MazareSharif Power Proj 2008 P084404 3A- MZ-MW AFR AFR Region IDA 93.00 P084404 Transmission Interconnection 2008 P109885 Rural Investment ECA Azerbaijan IDA 1.40 P109885 (AZRIP) Additional Financing 2008 P108843 Bangladesh DSC SAR Bangladesh IDA 19.50 P108843 IV-Supplemental Financing 2008 P110110 BD DSC SAR Bangladesh IDA 25.00 P110110 IV-Supplemental Financing II 2008 P111019 Additional Financing AFR Benin IDA 7.00 P111019 For The Benin Energy Services Delivery Project 2008 P078091 BF-Energy Access SIL AFR Burkina Faso IDA 17.46 P078091 2008 P097974 BI-Multisectoral AFR Burundi IDA 16.80 P097974 Water & Electricity Inf 2008 P108905 ZR-EMRRP Supp 2 AFR Congo, Dem. Rep. IDA 7.90 P108905 ERL (FY2008) 2008 P109932 DO Emergency LCR Dominican IBRD 20.80 P109932 Recovery & Disaster Republic Mgmt 2008 P110202 ER-Add-Fin Power AFR Eritrea IDA 15.80 P110202 distr & rural electric 2008 P074011 ET/Nile Basin AFR Ethiopia IDA 41.05 P074011 Initiative:ET-SU Interconn 2008 P101556 ET-Elect. Access Rural AFR Ethiopia IDA 122.20 P101556 II SIL (FY2007) 2008 P074191 GH-Energy Dev & AFR Ghana IDA 77.40 P074191 Access SIL (FY2008) Annex B: World Bank T&D Projects 104 $ mil- FY Project ID Project Name Region Country Product line lions URL 2008 P101653 Power System SAR India IBRD 600.00 P101653 Development Project IV 2008 P106899 ECSEE APL #3— ECA Montenegro IDA 8.20 P106899 Montenegro 2008 P104265 MA–One Support MNA Morocco IBRD 123.00 P104265 Project 2008 P095982 Electricity SAR Pakistan IDA 76.45 P095982 Distribution and Transmission 2008 P095982 Electricity SAR Pakistan IBRD 159.71 P095982 Distribution and Transmission 2008 P106262 PH- Bicol Power EAP Philippines IBRD 12.94 P106262 Restoration Project 2008 P101645 TZ-Energy AFR Tanzania IDA 90.30 P101645 Development & Access Expansion 2008 P096207 Power Transmission ECA Ukraine IBRD 200.00 P096207 Project 2008 P099211 VN-Rural Distribution EAP Vietnam IDA 150.00 P099211 Project 2008 P084461 GZ–Electric Utility MNA West Bank and SF 7.40 P084461 Management Gaza 2008 P077452 ZM–Incr. Eff. & Access AFR Zambia IDA 5.12 P077452 to Elec SIL (FY2008) 2009 P111943 ATRF–Power System SAR Afghanistan RE 35.00 P111943 Development 2009 P105654 3A–S. Afr Power AFR AFR Region IDA 180.62 P105654 Market—Add.Fin. APL1 2009 P112242 Power Distribution ECA Albania Guarantees 78.00 P112242 Privatization PRG 2009 P095965 Siddhirganj Peaking SAR Bangladesh IDA 43.30 P095965 Power Project 2009 P110614 BR–Sergipe State Int. LCR Brazil IBRD 3.12 P110614 Project: Rural Pov 2009 P105651 GPOBA W3–Ethiopia AFR Ethiopia RE 7.00 P105651 Rural Elect Expn, Ph2 2009 P114167 Supplemental Credit ECA Georgia IDA 8.00 P114167 for PRSO IV 2009 P112798 Power Sys Dev IV SAR India IBRD 400.00 P112798 Addl Financing Annex B: World Bank T&D Projects 105 $ mil- FY Project ID Project Name Region Country Product line lions URL 2009 P110173 KE–ESRP Additional AFR Kenya IDA 71.40 P110173 Financing SIL 2009 P096648 NG–Commercial AFR Nigeria IDA 27.00 P096648 Agriculture Development 2009 P113159 PH–Additional EAP Philippines IBRD 20.00 P113159 Financing for RPP 2009 P112334 UG–Energy for Rural AFR Uganda IDA 26.60 P112334 Transformation APL2 2009 P113495 Rural Energy II– EAP Vietnam IDA 200.00 P113495 Additional Financing 2009 P116854 GZ–Electric Utility MNA West Bank and SF 2.50 P116854 Management Add. Gaza Fin. 2009 P092211 RY–Rural Energy MNA Yemen, Rep. IDA 16.02 P092211 Access Source: World Bank. Note: IBRD = International Bank for Reconstruction and Development; IDA = International Development Agency. Annex B: World Bank T&D Projects 106 Glossary Additionality. A criterion often applied to GHG emissions reductions from climate change mitiga- projects, stipulating that project-based GHG reduc- tion projects undertaken in non-Annex I countries. tions should only be quantified if the project activity Combined margin. The weighted average of the would not have happened anyway—that is, that the operating and build margins. project activity (or the same technologies or prac- tices it employs) would not have been implemented Emission factor. A factor relating GHG emissions in its baseline scenario and/or that project activity to a level of activity or a certain quantity of inputs emissions are lower than baseline emissions. or products or services (for example, tonnes of fuel consumed, or units of a product). For example, an Baseline emissions. An estimate of GHG emissions, electricity emission factor is commonly expressed as removals, or storage associated with a baseline sce- tCO2eq/MWh. nario. Fugitive emissions. Emissions that are not physi- Baseline scenario. A hypothetical description cally controlled but rather result from the inten- of what would most likely have occurred in the tional or unintentional releases of GHGs. They absence of any considerations about climate change commonly arise from the production, processing mitigation. transmission storage, and use of fuels and other Build margin. The grid electricity emission factor chemicals, often through joints, seals, packing, gas- that reflects how a new power generation or sav- kets, and so on. ing project activity affects the construction of new GHG accounting. The process of quantifying the power plants. impacts on GHG emissions from an activity or orga- Carbon dioxide equivalent. The universal unit of nization/institution. measurement used to indicate the global warming GHG emissions. GHGs released into the atmo- potential of GHGs. It is used to evaluate the impacts sphere. of releasing (or avoiding the release of) different GHGs. GHG Protocol. A multistakeholder partnership of businesses, nongovernmental organizations, Clean Development Mechanism (CDM). A mecha- governments, academics, and others convened nism established by Article 12 of the Kyoto Protocol by the World Business Council for Sustainable for project-based emissions reduction activities in Development and the World Resources Institute to developing countries. The CDM is designed to meet design and develop internationally accepted GHG two main objectives: to address the sustainability accounting and reporting standards and/or proto- needs of the host country and to increase the oppor- cols, and to promote their broad adoption. tunities available to Annex I parties to meet their GHG reduction commitments. The CDM allows for GHG source. Any physical unit or process that the creation, acquisition, and transfer of certified releases GHGs into the atmosphere. 107 Global warming potential. A factor describing tion of the project and the GHG emissions that the radiative forcing impact (degree of harm to the would have occurred in a “without project” baseline atmosphere) of one unit of a given GHG relative to scenario. one unit of CO2. Operating margin. The grid electricity emission Greenhouse gases. Gases that absorb and emit radi- factor that reflects how a new power-generation or ation at specific wavelengths within the spectrum of -saving project activity affects the operation of exist- infrared radiation emitted by the Earth’s surface, the ing power plants. atmosphere, and clouds. The six main GHGs whose Project boundary. The physical location of the emissions are human-caused are carbon dioxide, activities that are evaluated for their GHG impacts methane, nitrous oxide, hydrofluorocarbons, per- and the list of GHG sources that are included in fluorocarbons, and sulfur hexafluoride. a GHG accounting exercise. Under the CDM, the Intergovernmental Panel on Climate Change project boundary is “all anthropogenic emissions (IPCC). International body of climate change scien- by sources of GHGs under the control of the proj- tists. The role of the IPCC is to assess the scientific, ect participants that are significant and reasonably technical, and socioeconomic information relevant attributable to the CDM project activity.” to the understanding of the risk of human-induced Project emissions. An estimate of GHG emissions, climate change (www.ipcc.ch). removals, or storage associated with a project sce- Inventory. A quantified list of a project’s, organiza- nario. tion’s, or country’s GHG emissions and sources. Project scenario. A description of the technology Life-cycle analysis. Assessment using a “cradle-to- and operational characteristics of the project activity grave” approach of the sum of a product’s effects implemented. (for example, GHG emissions) at each step in its life Value chain. All the upstream and downstream cycle, including resource extraction, production of activities associated with the production of goods or material, use, and waste disposal. services. Net GHG accounting. Quantification of the differ- ence between all GHG emissions from GHG sources within the project boundary after the implementa- Sources: Adapted from GHG 2005a and UNFCCC 2009c. Glossary 108 References Ackermann, T., G. Ancell, L. D. Borup, P. B. Eriksen, B. CCAP (Center for Clean Air Policy). 2008. Sectoral Ernst, F. Groome, M. Lange, C. Mohrlen, A. Orths, Approaches: A Pathway to Nationally Appropriate J. O’Sullivan, and M. de la Torre. 2009. “Where the Mitigation Actions. Wind Blows.” IEEE Power and Energy Magazine 7(6): CEPOS (Center for Politiske Studier). 2009. “Wind 65–75. Energy: The Case of Denmark.” www.cepos.dk/ fileadmin/user_upload/Arkiv/PDF/Wind_energy_-_ Barroso, L., F. Porrua, F., L. Thome, and M. Pereira. 2007. the_case_of_Denmark.pdf. “Planning for Big Things in Brazil: Technical and Regulatory Challenges of Large-Scale Transmission CPUC (California Public Utility Commission). Networks in Competitive Hydrothermal Systems.” 2005. PG&E Delta Distribution Planning Area IEEE Power and Energy Magazine 5(5): 54–63. Capacity Increase Substation Project. Proponent’s Environmental Assessment. 16. Corona and Induced Bauer, Christian, Roberto Dones, Thomas Heck, Current Effects Sacramento, CA. and Stefan Hirschberg. 2008. “Environmental Assessment of Current and Future Swiss Electricity DeLuchi, M. A. 1991. Emissions of Greenhouse Gases Supply Options.” In Proceedings of the PHYSOR from the Use of Transportation Fuels and Electricity. Volume 1: Summary. Argonne, IL: Argonne National 08 Conference, Interlaken, Switzerland, September Laboratory. 14–19, 2008. Dones, R., C. Bauer, R. Bolliger, B. Burger, T. Heck, Baumert, K. 1999. “Understanding Additionality.” In A. Roder, M. F. Emmenegger, R. Frischknecht, Promoting Development While Limiting Greenhouse N. Jungbluth, and M. Tuchschmid. 2007. Life Cycle Gas Emissions: Trends & Baselines, pp. 135–45. New Inventories of Energy Systems: Results for Current York: World Resources Institute and United Nations Systems in Switzerland and Other UTCE Countries. Development Programme. Final report EcoInvent data v2.0, No. 5. Dübendorf: EcoInvent Swiss Centre for Life Cycle Inventories. Bodansky, D. 2007. International Sectoral Agreements in a www.ecoinvent.ch. Post-2012 Climate Framework. Washington, DC: Pew Center on Global Climate Change. www.pewclimate. Econ Analysis. 2006. Recommendations for CDM org/publications/workingpaper/international- Power Projects That Involve Imports and Exports sectoral-agreements-post-2012-climate-framework. of Power between Different National Grids. Report commissioned by World Bank Carbon Finance Unit. Bode, S., and A. Michaelowa. 2003. “Avoiding Perverse Report 2006-045. Oslo: Econ Analysis A.S. Effects of Baseline and Investment Additionality Ellis, J., J. Corfee-Morlot, and H. Winkler. 2007. “CDM: Determination in the Case of Renewable Energy Taking Stock and Looking Forward.” Energy Policy Projects.” Energy Policy 31: 505–17. 35: 15–28. Bosi, M., and A. Laurence. 2002. Road-Testing Baselines Fichtner. 2008. “Ethiopia-Kenya Power Systems for Greenhouse Gas Mitigation Projects in the Electric Interconnection Project. Consultancy Services for Power Sector. Paris: International Energy Agency Feasibility Study. Draft Final Report.” Ethiopian Electric and Organisation for Economic Co-operation and Power Corporation and Kenyan Ministry of Energy. Development. Gagnon, L., C. Belanger, and Y. Uchiyama. 2002. “Life- Brown, M. H., and R. P. Sedano. 2004. Electricity Cycle Assessment of Electricity Generation Options: Transmission: A Primer. Washington, DC: National The Status of Research in Year 2001.” Energy Policy Council on Electricity Policy. 30: 1267–78. 109 GEF (Global Environment Facility). 2008. Manual IFC (International Finance Corporation). 2009. IFC Carbon for Calculating GHG Benefits of GEF Projects: Emissions Estimator Tool (CEET). Updated October 9, Energy Efficiency and Renewable Energy Projects. 2009. Washington, DC: World Bank Group. Washington, DC: GEF. www.thegef.org/gef/sites/ thegef.org/files/documents/C.33.Inf_.18%20 IPCC (Intergovernmental Panel on Climate Change). Climate%20Manual.pdf. 2006a. 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Geneva: IPCC. www.ipcc-nggip.iges. GHG Protocol. 2004. A Corporate Accounting and or.jp/public/2006gl/index.html. Reporting Standard. Revised edition. Washington, DC: World Business Council for Sustainable —. 2006b. 2006 IPCC Guidelines for National Development and World Resources Institute. Greenhouse Gas Inventories. Volume 2: Energy. www.ghgprotocol.org. Geneva: IPCC. www.ipcc-nggip.iges.or.jp/ public/2006gl/vol2.htm. —. 2005a. GHG Protocol for Project Accounting. Washington, DC: World Business Council for —. 2006c. 2006 IPCC Guidelines for National Sustainable Development and World Resources Greenhouse Gas Inventories. Volume 3: Industrial Institute. www.ghgprotocol.org. Processes and Product Use. Geneva: IPCC. www.ipcc- nggip.iges.or.jp/public/2006gl/vol3.htm. —. 2005b. Guidelines for Quantifying GHG Reductions from Grid-Connected Electricity Projects. —. 2006d. 2006 IPCC Guidelines for National Washington, DC: World Business Council for Greenhouse Gas Inventories. Volume 4: Agriculture, Sustainable Development and World Resources Forestry and Other Land Use Data. Geneva: IPCC. Institute. www.ghgprotocol.org. www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html. —. 2007. Guidelines for Quantifying GHG Reductions Kartha, S., M. Lazarus, and M. Bosi. 2004. “Baseline from Grid-Connected Electricity Projects. Washington, Recommendations for Greenhouse Gas Mitigation DC: World Business Council for Sustainable Projects in the Electric Power Sector.” Energy Policy Development and World Resources Institute. 32: 545–66. www.ghgprotocol.org. Knapp, K., and T. Jester. 2001. “Empirical Investigation of Greiner, S., and A. Michaelowa. 2001. “Defining the Energy Payback Time for Photovoltaic Modules.” Investment Additionality for CDM Projects—Practical Solar Energy 71 (3): 165–72. Approaches.” Energy Policy 31(10): 1007–15. Lee, M.-K., R. M. Shrestha, S. Sharma, G. R. Timilsina, Hammond, G., and C. Jones. 2006. Inventory of Carbon & and S. Kumar. 2005. Baseline Methodologies for Energy (ICE). Version 1.5 Beta. Bath, UK: University Clean Development Mechanism Projects. Roskildge, of Bath, Department of Mechanical Engineering. Denmark: United Nations Environment Programme, www.bath.ac.uk/mech-eng/sert/embodied/. Riso Centre for Energy, Climate and Sustainable Development. —. 2008. Inventory of Carbon & Energy (ICE). Version 1.6a. Bath, UK: University of Bath, May, H. 2009. “Rewiring for Renewable: More Clean Department of Mechanical Engineering. www.bath. Power Coming Down the Line.” New Energy ac.uk/mech-eng/sert/embodied/. Magazine 2 (April). www.newenergy.info/index. php?id=878&tx_ttnews[tt_news]=3524&tx_ttnews[b Herzog, T. 2009. World Greenhouse Gas Emissions in ackPid]=882&cHash=3c10cdc3b1. 2005. WRI Working Paper. Washington, DC: World Resources Institute. www.wri.org/chart/world- May, N. 2005. “Eco-Balance of a Solar Electricity greenhouse-gas-emissions-2005. Transmission from North Africa to Europe.” Diploma Thesis. Braunschweig: Technical University IEA (International Energy Agency). 2007. IEA Statistics: of Braunschweig, Faculty for Physics and Geological Electricity/Heat. Paris: IEA. www.iea.org/stats/ Sciences. prodresult.asp?PRODUCT=Electricity/Heat. Accessed March 22, 2010. MME (Brazilian Federal Government Ministry of Mines and Energy) and EPE (Empresa de Pesquisa —. 2009. Energy Statistics of Non-OECD Countries— Energética). 2006. Brazilian Energy Balance 2006: Basic Energy Statistics. Paris: IEA. Year 2005. Rio de Janeiro: EPE. www.lib.utexas.edu/ benson/lagovdocs/brazil/federal/minasenergia/ BEN(English)2006.pdf. References 110 National Grid. 2008. Corporate Responsibility Reporting Tanwar, N. 2007. “Clean Development Mechanism Procedures. Annex II. London: National Grid. and Off-Grid Small-Scale Hydropower Projects: www.nationalgrid.com/NR/rdonlyres/BBE34B8A- Evaluation of Additionality.” Energy Policy 35: 714– 50BD-4EB8-9922-5B1324E5C744/26456/ 21. CRreportingprinciples2008.pdf. Transpower. 2009. Transpower’s Carbon Footprint Report Öko Institute for Applied Ecology. 2009. Global Emission 2008–09. Wellington, New Zealand: Transpower New Model for Integrated Systems. Berlin: Öko Institute for Zealand Limited. Applied Ecology. www.gemis.de/en/index.htm. UNFCCC (United Nations Framework Convention on Penman, J., M. Gytarsky, T. Hiraishi, T. Krug, D. Kruger, Climate Change). 2001. “Modalities and Procedures R. Pipatti, L. Buendia, K. Miwa, T. Ngara, K. Tanabe, for a Clean Development Mechanism, as Defined in and F. Wagner, eds. 2003. Good Practice Guidance for Article 12 of the Kyoto Protocol.” FCCC/CP/2001/13/ Land Use, Land-Use Change and Forestry. Kanagawa, Add.2. Decision 17/CP.7. Japan: Intergovernmental Panel on Climate Change National Greenhouse Gas Inventories Programme —. 2005. “Guidance Regarding Methodological Technical Support Unit. Issues.” EB 22 Report. Clean Development Mechanism Executive Board. Bonn: UNFCCC. Pinto, Neil (CEO PPA Energy). 2010. Calculating Electricity Losses in Distribution Utilities. World —. 2008a. “Methodological Tool: Tool for the Bank. Demonstration and Assessment of Additionality.” EB 39 Report, Annex 10, Version 05.2, CDM Executive Schmidt, J. H., and M. Thrane. 2009. Life Cycle Assessment Board: http://cdm.unfccc.int/methodologies/ of Aluminium Production in New Alcoa Smelter PAmethodologies/tools/am-tool-01-v5.2.pdf. in Greenland. Nuuk, Greenland: Government of Greenland. —. 2008b. “NM0269 Draft Project Design Document.” Bonn: UNFCCC. http://cdm.unfccc. Schneider, L. 2007. Is the CDM Fulfilling Its int/methodologies/PAmethodologies/publicview. Environmental and Sustainable Development html?meth_ref=NM0269. Objectives? An Evaluation of the CDM and Options for Improvement. Berlin: Öko-Institut. —. 2009a. “AMS I.D. Grid Connected Renewable Electricity Generation.” Version 15. Bonn: UNFCCC. Sharma, S., and R. M. Shrestha. 2006. “Baseline for http://cdm.unfccc.int/UserManagement/FileStorage/ Electricity Sector CDM Projects: Simplifying 7QXAZ5036WN8BEYKUDFRPJGL21V4I9. Estimation of Operating Margin Emission Factor.” Energy Policy 34: 4093–4102. —. 2009b. “Consolidated Baseline and Monitoring Methodology for Fuel Switching from Coal or Shrestha, R., and G. R. Timilsina. 2002. “The Petroleum Fuel to Natural Gas”). CDM methodology Additionality Criterion for Identifying Clean ACM9, Version 03.2. Development Mechanism Projects under the Kyoto Protocol.” Energy Policy 30(1): 73–79. —.2009c. “Glossary of CDM Terms.” Version 05. South Asia Sustainable Development Department. 2009. http://cdm.unfccc.int/Reference/Guidclarif/glos_ India: Options for Low Carbon Development. Synopsis CDM.pdf. of a Study by the World Bank for Government of India. —. 2009d. “Proposed New Baseline and Monitoring Washington, DC: World Bank. Methodologies.” CDM-NM, Version 03.1. Bonn: Spalding-Fecher, R., ed. 2002. The CDM Guidebook: UNFCCC. The Clean Development Mechanism of the Kyoto —. 2009e. “Tool to Calculate the Emission Factor Protocol—A Guide for Project Developers in Southern for an Electricity System.” Version 2. Bonn: Africa. Cape Town: University of Cape Town, Energy UNFCCC. https://cdm.unfccc.int/methodologies/ and Development Research Centre. PAmethodologies/tools/am-tool-07-v2.pdf. Sural. 2010. Product Catalogue—ACSR (Aluminum Conductor, Steel Reinforced). Caracas, Venezuela. —. 2010. “Guidelines on the Assessment of Investment Analysis.” EB 51 Report Annex 58, www.sural.com. Version 03.1, CDM Executive Board: http://cdm. unfccc.int/Reference/Guidclarif/reg/reg_guid03.pdf. References 111 U.S. DOE (U.S. Department of Energy). 2008. “20% —. 2003. Cambodia Rural Electrification and Wind Energy by 2030, Increasing Wind Energy’s Transmission Project. Project Appraisal Document. Contribution to U.S. Electricity Supply.” Washington, Report Number 27015-KH. Washington, DC: World DC: U.S. DOE. Bank. U.S. EIA (U.S. Energy Information Administration). 2010. —. 2004. Kenya Energy Sector Recovery Project. International Energy Statistics. http://tonto.eia.doe. Project Appraisal Document. Report Number 28314- gov/cfapps/ipdbproject/IEDIndex3.cfm. Accessed KE. Washington, DC: World Bank. February 11, 2010. —. 2007. Ethiopia/Nile Basin Initiative Power Export U.S. EPA (Environmental Protection Agency). 2006. Project: Ethiopia-Sudan Interconnector. Project Global Anthropogenic Non-CO2 Greenhouse Gas Appraisal Document. Report Number 41425-ET. Emissions: 1990–2020. Washington, DC: U.S. EPA. Washington, DC: World Bank. Wartmann, S., and J. Harnisch. 2005. “Reductions of SF6 —. 2008a. “Development and Climate Change: A Emissions from High and Medium Voltage Electrical Strategic Framework for the World Bank Group.” Equipment in Europe.” Project No. dm70047.2. Report to the Board, Sustainable Development Nürnberg, Germany: Ecofys GmbH. Sector, World Bank. http://beta.worldbank.org/ climatechange/overview/strategic-framework- Winkler, H., and S. Thorne. 2002. “Baselines for documents. Suppressed Demand: CDM Projects Contribution to Poverty Alleviation.” South African Journal of —. 2008b. Honduras Power Sector Efficiency Economic and Management Sciences 5(2): 413–29. Enhancement Project (PROMEF). Project Appraisal Document. Report Number 45791-HN. Washington, World Bank. 1996. Handbook on Economic Analysis of DC: World Bank. Investment Operations. Washington, DC: World Bank. —. 2010. World Development Report 2010. Washington, DC: World Bank Group. —. 1998. Greenhouse Gas Assessment Handbook: A Practical Guidance Document for the Assessment of WRI (World Resources Institute). 2006. Climate Analysis Project-Level Greenhouse Gas Emissions. Washington, Indicators Tool. Washington, DC: WRI. http://cait. DC: World Bank. wri.org/. References 112 The Energy and Mining Sector Board The World Bank 1818 H Street N.W. Washington, D.C., 20433 USA