A WORLD BANK STUDY 55842 Wind Energy in Colombia A FRAMEWORK FOR MARKET ENTRY Walter Vergara, Alejandro Deeb, Natsuko Toba, Peter Cramton, Irene Leino W O R L D B A N K S T U D Y Wind Energy in Colombia A Framework for Market Entry Walter Vergara Alejandro Deeb Natsuko Toba Peter Cramton Irene Leino Copyright © 2010 The International Bank for Reconstruction and Development/The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing: July 2010 Printed on recycled paper 1234 13 12 11 10 World Bank Studies are published to communicate the results of the Bank's work to the development community with the least possible delay. The manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank of the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission promptly to reproduce portions of the work. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, Tel: 978 750 8400, Fax: 978 750 4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Fax: 202 522 2422, email: pubrights@worldbank.org. ISBN: 978 0 8213 8504 3 eISBN: 978 0 8213 8508 1 ISSN: 1726 5878 DOI: 10.1596/978 0 8213 8504 3 Library of Congress Cataloging in Publication Data has been requested. Contents Preface ..................................................................................................................................... viii Acknowledgments ................................................................................................................... ix Acronyms and Abbreviations ................................................................................................. x Executive Summary............................................................................................................... xiii Objective............................................................................................................................xiii General Context ...............................................................................................................xiii Alternative Options for Colombia's Power Mix ..........................................................xiv Wind Energy Capital Costs Are Expected to Decrease ..............................................xiv Wind and Hydro Energy Resources Are Complementary.......................................... xv Options to Address Barriers to Entry............................................................................. xv Impact of Policy Options ...............................................................................................xvii Lessons Learned.............................................................................................................xviii 1. Introduction............................................................................................................................ 1 Context ................................................................................................................................. 1 Structure of the Report ....................................................................................................... 3 2. Summary of Findings from First Stage Report: Nonconventional Renewable Energy Barrier Analysis .................................................................................................... 4 3. Cost Comparison of Alternative Power Sources Based on the Expansion Plan for 2008­2025 ....................................................................................................................... 7 Methodology for Technology Cost Comparison ............................................................ 7 Least (Levelized) Cost Comparison ................................................................................. 9 4. Wind Power Costs Outlook ............................................................................................... 13 Technical Viability of Wind Power................................................................................. 13 Efficiency Gains over Time.............................................................................................. 14 Capital Cost Evolution ..................................................................................................... 14 Operation and Maintenance Costs Are Decreasing ..................................................... 15 Wind Power Grid Integration ......................................................................................... 16 Outlook............................................................................................................................... 16 5. Wind and Hydro in Colombia: Complementarity Analysis........................................ 18 Complementarity of the Wind and Hydro Regimes .................................................... 18 Firm Energy and Joint Operation of Wind and Hydroelectric Projects..................... 25 6. Options to Aid Market Entry of Wind Energy in the Country's Power Mix ............ 29 Introduction ....................................................................................................................... 29 Options to Facilitate Market Entry of Wind Energy .................................................... 29 Proposal to Address the Reliability Issue for Wind Energy........................................ 32 iii iv Contents 7. Assessing the Effectiveness of Policy Instruments and Policy Options: Impact on a 300 MW Wind Powered Power Plant Operating in the Wholesale Energy Market .................................................................................................................. 36 Baseline Information ........................................................................................................ 37 Baseline Results................................................................................................................. 39 Impact of Selected Policy Options .................................................................................. 40 Key Findings: Options to Foster Investment in Wind Power ..................................... 47 Conclusions of the Estimated Impact of Alternative Policy Options for a 300 MW Wind Energy Power Plant in the MEM .................................................. 48 8. Conclusions .......................................................................................................................... 50 Wind Energy Resources Could Become an Important Energy Option in Colombia .................................................................................................................... 50 Policy Instruments ............................................................................................................ 51 Policy Options ................................................................................................................... 51 Other Findings .................................................................................................................. 52 Applicability of the Analysis Conducted ...................................................................... 52 References................................................................................................................................. 53 APPENDIXES........................................................................................................................... 55 Appendix 1. Technology Cost Comparison........................................................................ 56 Appendix 2. Use of Earth Simulator to Estimate the Likelihood of Extreme Weather Events................................................................................................................. 58 Appendix 3. Pool Prices under Various Scenarios ............................................................ 60 Appendix 4. Results of the Expected Returns on Investments with the Individual Application of the Policy Instruments for Different Market Scenarios............................................................................................................................ 62 Appendix 5. Exempting CERE Payments by 50 or 100 Percent ....................................... 67 Appendix 6. Complementarity between Wind Power and Hydroelectric Resources........................................................................................................................... 68 Chapter 1: Introduction.................................................................................................... 68 Chapter 2: Methodology .................................................................................................. 69 Chapter 3: Data Base......................................................................................................... 70 Chapter 4: Extension of Jepírachi Information.............................................................. 77 Chapter 5: Case Studies for Complementarity Analysis ............................................. 80 Tables Table 1. Actions Required to Reach a Financial Threshold for a 300 MW Wind Power Plant on the Northern Coast ............................................................................. xix Table 3.1. Power Generation Options Included in the Screening Curve Analyses .......... 8 Table 3.2. Least Cost Capacity Expansion Mix (without CO2e revenue) ........................... 9 Table 3.3. Suggested Capacity Expansion Mix at US$18 per Ton CO2e ........................... 10 Contents v Table 5.1. Jepírachi Monthly Power Generation ................................................................. 19 Table 5.2. Wind Speed as a Fraction of Mean Yearly Wind Speeds.................................. 21 Table 5.3. El Niño Periods ...................................................................................................... 23 Table 5.4. Wind and Hydro Complementary during El Niño........................................... 24 Table 5.5. Complementarity of Joint Operation of Hydro Plant and Wind Farm; the Case of the Nare River .............................................................................................. 24 Table 5.6. Firm Energy Results for Guavio River Analyzed in Isolated and Joint Operation .......................................................................................................................... 25 Table 7.1. Demand Scenarios for the Interconnected Grid and Resulting Indicative Prices.................................................................................................................................. 39 Table 7.2. Expected Returns on Equity before Taxes for a 300 MW Wind Farm in Colombia--Business as Usual Results (no government intervention)..................... 39 Table 7.3. Policy Options, Allocation of Responsibilities and Associated Costs............. 41 Table 7.4a. Financial Results for a 300 MW Wind Farm In Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 20 Percent Firm Energy Factor ............................................................................................ 44 Table 7.4b. Financial Results for a 300 MW Wind Farm in Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 30 Percent Firm Energy Factor ............................................................................................ 45 Table 7.4c. Financial Results for a 300 MW Wind Farm in Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 36 Percent Firm Energy Factor ............................................................................................ 46 Table 7.5. Key Findings: Combination of Policy Instruments to Reach a Financial Threshold .......................................................................................................................... 47 Table A1.1. Least Levelized Cost Ranking of Electricity Generation Plant by Capacity Factor (%) without the Cost of CO2 Emissions ............................................ 56 Table A1.2. Least Levelized Cost Ranking of Electricity Generation Plant by Capacity Factor (%) with US$18/Ton CO2 Emissions.................................................. 57 Table A3.1 MEM Scenarios .................................................................................................... 60 Table A4.1. Effectiveness Analysis of Individual Policy Instruments.............................. 63 Table A4.2. Effectiveness Analysis of Policy Options: Use of Financial Instruments .... 64 Table A4.3. Effectiveness Analysis of Policy Options: Use of Government Fees and Payments........................................................................................................................... 65 Table A4.4. Effectiveness Analysis of Policy Options: Use of Regulatory Instruments....................................................................................................................... 66 Table A5.1. Financing Necessary if CERE Is Returned 50 Percent or 100 Percent, Depending on Investment Costs.................................................................................... 67 Table A6.1. Mean Monthly Values for the Guavio, Nare, Cauca, and Magdalena Rivers................................................................................................................................. 72 Table A6.2. Jepírachi Monthly Hour Generation kWh (1 to 12)........................................ 74 Table A6.3. Jepírachi Monthly Hour Generation kWh (13 to 24)...................................... 75 Table A6.4. Extended Monthly Generation for Jepírachi (January to June) .................... 78 Table A6.5. Extended Monthly Generation for Jepírachi (July to December) ................. 79 Table A6.6. El Niño Periods since 1950 ................................................................................ 82 Table A6.7. Analysis of El Niño Occurrences in Guavio River Discharges (1986­ 1995)................................................................................................................................... 83 vi Contents Table A6.8. Analysis of El Niño Occurrences in Guavio River Discharges (1997­ 2007)................................................................................................................................... 84 Table A6.9. Analysis of El Niño Occurrences in Nare River Discharges (1986­1995).... 85 Table A6.10. Analysis of El Niño Occurrences in Nare River Discharges (1997­ 2007)................................................................................................................................... 86 Table A6.11. Analysis of El Niño Occurrences in Cauca River Discharges (1986­ 1995)................................................................................................................................... 87 Table A6.12. Analysis of El Niño" Occurrences in Cauca River Flows (1997­2007) ...... 88 Table A6.13. Analysis of El Niño occurrences in Magdalena River discharges (1986­1995)........................................................................................................................ 89 Table A6.14. Analysis of El Niño Occurrences in Magdalena River Discharges (1997­2007)........................................................................................................................ 90 Table A6.15. Analysis of El Niño Occurrences at Jepírachi Power Plant (1986­ 1995)................................................................................................................................... 91 Table A6.16. Analysis of El Niño Occurrences at Jepírachi Power Plant (1997­ 2007)................................................................................................................................... 91 Table A6.17. Summary of El Niño occurrences, 1986­2007 ............................................... 92 Table A6.18. Firm Energy for Guavio and Jepírachi in Isolated and Joint Operation.... 93 Table A6.19. Firm Energy for Nare and Jepírachi in Isolated and Joint Operation ........ 95 Table A6.20. Firm Energy for Cauca and Jepírachi in Isolated and Joint Operation...... 96 Table A6.21. Firm Energy for Magdalena and Jepírachi in Isolated and Joint Operation .......................................................................................................................... 98 Figures Figure 1.1 Installed Capacity per Technology Type ............................................................. 2 Figure 3.1. Screening Curve for Levelized Total Costs Measured in Cost of Capacity of a Plant per Year (US$/kW yr) at Different Capacity Factors................. 10 Figure 3.2. Screening Curve for Levelized Total Costs at Different Capacity Factors Measured in Terms of Generation Costs (US cents/kWh) ............................ 11 Figure 4.1. World Total Wind Power Installed Capacity (MW)........................................ 13 Figure 4.2. Project Capacity Factors by Commercial Operation Date .............................. 14 Figure 4.3. Reported US Wind Turbine Transaction Prices over Time ............................ 15 Figure 4.4. Average Operation and Maintenance Costs for Available Data Years from 2000 to 2007, by Last Year of Equipment Installation........................................ 15 Figure 5.1. Stations Used to Characterize Wind Power in Colombia ............................... 20 Figure 5.2. Almirante Padilla Airport, Guajira .................................................................... 21 Figure 5.3. Graphic Representation of Wind Conditions in Northern Colombia ........... 22 Figure 5.4. Firm Energy for Guavio River as a Result of Isolated and Joint Operation .......................................................................................................................... 26 Figure 5.5. Guavio River Reservoir Operation with a Reservoir Size of 0.2 in Isolated and Joint Operation .......................................................................................... 27 Figure 5.6. Guavio River Reservoir Operation with a Reservoir Size of 0.5 in Isolated and Joint Operation .......................................................................................... 27 Figure 7.1. Colombia NIS Demand Forecasts, 2007­2028 .................................................. 37 Figure 7.2. Wind Project Generation Estimates 2012­2025 ................................................ 37 Figure 7.3. Pool Prices, Base Scenario ................................................................................... 38 Contents vii Figure 7.4. Comparison of Pool Prices for Base, High, and Low Scenarios..................... 38 Figure A2.1. Changes in Maximum Five Day Precipitation Total (mm) between the Present and the End of the 21st Century for (a) 60 km and (b) 20 km, Respectively...................................................................................................................... 58 Figure A2.2. The Same as In Figure A.2.1 Except for Consecutive Dry Days (day) ....... 59 Figure A3.1 Pool Prices, Base High Hydro Scenario .......................................................... 60 Figure A3.2. Pool Prices, High Scenario ............................................................................... 60 Figure A3.3. Comparison of Pool Prices for Base and Base High Hydro Scenarios ....... 61 Figure A6.1. Hourly Wind Velocity: Puerto Bolívar........................................................... 70 Figure A6.2. Seasonal Behavior of Mean Wind Velocity.................................................... 71 Figure A6.3. Hourly Mean Velocity: Barranquilla Airport ................................................ 71 Figure A6.4. Mean Wind Velocity: Barranquilla Airport ................................................... 72 Figure A6.5. Normalized Monthly Discharges of the Four Rivers ................................... 73 Figure A6.6. Power Curve for Each Unit.............................................................................. 73 Figure A6.7. Jepírachi: Hourly Generation .......................................................................... 76 Figure A6.8. Jepírachi: Monthly Mean Generation ............................................................. 76 Figure A6.9. Mean Monthly Values at the Guavio River Dam Site .................................. 80 Figure A6.10. Mean Monthly Values at the Santa Rita Dam Site on the Nare River...... 80 Figure A6.11. Mean Monthly Values at the Salvajina Dam Site on the Cauca River ..... 81 Figure A6.12. Mean Monthly Values at the Salvajina Dam Site on the Magdalena River................................................................................................................................... 81 Figure A6.13. Firm Energy for Guavio and Jepírachi in Isolated and Joint Operation .......................................................................................................................... 93 Figure A6.14. Guavio River Reservoir Operation with Reservoir Size 0.2 ...................... 94 Figure A6.15. Guavio River Reservoir Operation with Reservoir Size 0.5 ...................... 94 Figure A6.16. Firm Energy for Nare and Jepírachi in Isolated and Joint Operation ...... 95 Figure A6.17. Nare River Reservoir Operation with Reservoir Size 0.2 .......................... 95 Figure A6.18. Nare River Reservoir Operation with Reservoir Size 0.5 .......................... 96 Figure A6.19. Firm Energy for Cauca and Jepírachi in Isolated and Joint Operation .... 97 Figure A6.20. Cauca River Reservoir Operation with Reservoir Size 0.2 ........................ 97 Figure A6.21. Cauca River Reservoir Operation with the Reservoir Size 0.5.................. 98 Figure A6.22. Firm Energy for Magdalena and Jepírachi in Isolated and Joint Operation........................................................................................................................... 98 Figure A6.23. Magdalena River Reservoir Operation with Reservoir Size 0.2................ 99 Figure A6.24. Magdalena River Reservoir Operation with Reservoir Size 0.5................ 99 Preface T he urgent need to reduce the carbon footprint of human activities and the increased awareness of the consequences of climate destabilization have rekindled interest in renewable energy sources as important elements to consider in the expansion or retrofitting of power systems. This report, the second in a series aimed at assessing and addressing barriers to the market entry of wind energy in Colombia's power sector, is but one example of the renewed attention that is rightly being conferred to the potential for wind to become a forceful player in low carbon futures in Latin America. The role of wind will not only be a function of cost effectiveness and/or technology advances but also of the ability to address policy and regulatory barriers that in the past have hampered their entry into developing markets. Although the report refers to the specifics of Colombia, its approach and conclusions may be valuable to a wider audience in the region and worldwide. If these barriers are successfully addressed, wind energy may contribute substantially to maintain the current, relatively low carbon footprint of Colombia s power sector, aided by a strong hydro contribution. Furthermore, as the report suggests, the wind option may also contribute to the diversification of power sources without increasing their carbon footprint, while also addressing concerns related to the vulnerability of hydropower to increased climate variability. Walter Vergara Team Leader Global Expert Team on Climate Change Adaptation viii Acknowledgments M any people contributed to this study, and the authors would like to thank them all. The authors would like to express their gratitude for the support and inputs provided by J. Mejía (energy specialist), A. Brugman (power engineer), and A. Valencia (renewable energy specialist) in the preparation of this study. The authors want to thank the technical staff at UPME led by C. A. Flórez, in particular, J.V. Dulce, as well as the technical staff at ISAGEN led by L.A. Posado, and Las Empresas Públicas Municipales (EPM), led by L.F. Rodríguez Arbelaez, for their valuable comments. The authors are also most grateful to J. Nash, P. Benoit, G. Grandolini, C. Feinstein, and D. Reinstein for their comments and suggestions. This study is a product of the Energy Unit of the Sustainable Development Department of the Latin America and Caribbean Region of the World Bank and funded through the Energy Sector Management Assistance Program. ix Acronyms and Abbreviations AGC Automatic Generation Control ANH Agencia Nacional de Hidrocarburos (National Hydrocarbon Agency) CCGT Combined Cycle Gas Turbine CCS Carbon Capture and Storage CDM Clean Development Mechanism CER Certified Emission Reductions CERE Real Equivalent Cost of the Capacity Charge CFB Circulating Fluidized Bed CNG Compressed Natural Gas COLCIENCIAS Departamento Administrativo de Ciencia, Tecnología e Innovación (Colombian Institute for the Development of Science, Technology and Innovation) CREG Comisión de Regulación de Energía y Gas (Regulatory Commission for Electricity and Gas) CTF Clean Technology Fund DNP Departamento Nacional de Planeación (National Planning Department) EPM Empresas Públicas de Medellín ESP (Public Companies of Medellín), one of Colombia's largest energy producers ENSO El Niño Southern Oscillation ESMAP World Bank Energy Sector Management Assistance Program ESP Electrostatic Precipitator FAZNI Fondo de Apoyo Financiero para la Energización de las Zonas No Interconectadas (Fund for the Electrification of Off grid Regions) FDG Flue Gas Desulfurization Gypsum FGD Flue Gas Desulfurization GCM General Circulation Model GDP Gross Domestic Product GHG Greenhouse Gas GOC Government of Colombia IDEAM Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia (Institute of Hydrology, Meteorology and Environmental Studies of Colombia) IEA International Energy Agency IGCC Integrated Gasification Combined Cycle IPCC Intergovernmental Panel on Climate Change IPP Independent Power Producer IRR Internal Rate of Return ISA Interconexión Eléctrica S.A. ISAGEN A major power producer and commercialization company in Colombia JMA Japan Meteorological Agency LVRT Low Voltage Run Through x Acronyms and Abbreviations xi MDB Multilateral Development Bank MEM Wholesale Energy Market MRI Meteorological Research Institute of Japan NIS National Interconnected System NREL National Renewable Energy Laboratory O&M Operation and Maintenance PM Particulate Matter PPP Purchasing Power Parity R&D Research and Development RE Renewable Energy RET Renewable Energy Technologies SC Subcritical SCR Selective Catalytic Reduction SDL Sistema de Distribución Local (Local Distribution System) SOPAC Pacific Islands Applied Geoscience Commission SOx Sulfur Oxide Gases SPC Super Critical SSPD Superintendencia de Servicios Públicos Domiciliarios (Superintendency for Residential Public Services) STN Sistema de Transmisión Nacional (National Transmission System) STR Sistema de Transmisión Regional (Regional Transmission System) TPC Total Plant Cost UNFCCC United Nations Framework Convention on Climate Change UPME Unidad de Planeamiento Minero Energética (Colombia's Energy and Mining Planning Unit) URE Uso Racional de Energía (Rational and Efficient Use of Energy) USPC Ultra Supercritical List of Units BTU British Thermal Units Cal Calories Cz Ash GJ Gigajoule GW Gigawatt GWh Gigawatt hour Kbpd Thousand barrels per day Kcal Kilo calories Kg Kilogram KTOE Thousand tons of oil equivalent KWh Kilowatt hour (103) Lb Pounds M/s Meters per second MBTU Million British Thermal Units Mbbl Million barrels MJ Megajoules xii Acronyms and Abbreviations MMT Million metric tons MTOE Million tons of oil equivalent MWa Megawatt average MWh Megawatt hour (106) QUADS Quadrillion BTU Tcf Trillion cubic feet TOE Tons of oil equivalent TWh Terawatt hour US$PPP Purchasing power parity Executive Summary Objective The purpose of this report is to provide decision makers in Colombia (and by extension other countries or regions), who are considering the deployment or consolidation of wind power, with a set of options to promote its use. The options presented are the result of an analysis of the Colombian market; this analysis included simulations and modeling of the country's power sector, and extensive consultations with operators, managers, and agents. More information on the analysis and simulations is presented in the appendixes. Wind was chosen to exemplify the range of renewable energy alternatives available to complement traditional power sector technologies on the basis of its technical maturity, its relatively low cost compared to other options, the country's experience, and its wind power potential. This report constitutes the second phase of a barrier analysis to wind energy in Colombia (Vergara et al. 2008). General Context Colombia has a rich endowment of energy sources. The natural gas reserves in 2008 were 7.3 tera cubic feet (of which 60 percent were proven reserves). At the current rate of utilization these reserves would last 23 years.1 Likewise, Colombia's coal reserves are rated at seven billion tons (or about 100 years of production at the present mining rate). Most coal mined is anthracite, with very low ash and sulfur content, ideal for exports to the European market. Oil reserves are more limited but recent discoveries have expanded reserves in number of years of supply, which until recently had been estimated at eight years (Ministry of Mines and Energy 2008). The country has also a substantial, relatively low cost hydropower potential resulting from its location in the tropical inter convergence zone and its mountain ranges. Within this context, the country has developed a power sector that relies heavily on installed, large capacity hydropower units that provide cost effective electricity. In 2008 the installed power mix in Colombia (13.5 GW) was 67 percent hydro, 27 percent natural gas, 5 percent coal, and 0.3 percent wind and cogeneration. The total power demand that same year was 54 TWh (UPME 2009), met with about 9 GW of installed capacity.2 This structure also results in a low carbon footprint, among the lowest in the region, with 87 percent of power generated and delivered to the grid by hydropower plants, resulting in an estimated 350 tons of CO2 per GWh generated (about half that of Mexico). From a management perspective, Colombia's power sector is maturing quickly, with relative stability in its regulations, an unbundled system, and a dispatch mechanism that closely resembles a well functioning competitive market. Competition is promoted and tools have been designed to attract cost effective capacity expansions that would promote reliability3 of service (a fuller description of the system and its dispatch mechanism was included in the phase one report). The wind regime in Colombia has been rated among the best in South America. Offshore regions of the northern part of Colombia have been classified with class seven xiii xiv Executive Summary winds (winds over nine meters per second [m/s] at heights of 50 meters). The only other region in South America with similar wind intensity is the Patagonia region of Chile and Argentina. Colombia has an estimated wind power potential of 18 GW in the La Guajira region--enough to generate power to meet the national power demand twice over4 (Pérez and Osorio 2002). However, the country has an installed capacity of only 19.5 MW of wind energy (Jepírachi Project) and several projects under consideration, including a 200 MW project in Ipapure, northern Colombia. Under the current circumstances, and on its own, the interconnected system would not likely promote nonconventional renewable energy resources (for example, other than hydropower), such as wind, but would instead maintain its high capacity share of hydro. Alternatively, the system may move toward a more carbon intensive energy resource mix (likely reliant on abundant coal reserves) to meet any additional demand that cannot be met through hydropower and/or to strengthen the system's resilience to deal with the effects of droughts and El Niño years. Expanding the coal based power generation capacity would result in an increase in the carbon footprint of the economy from its current relatively low level of greenhouse gas (GHG) emissions.5 Alternative Options for Colombia's Power Mix A cost comparison of 37 alternative technology options for power generation in Colombia, using a levelized curve/netback analysis, indicates that, as expected, large hydropower is the least cost power option with or without CO2e emission reduction revenues over a wide range of capacity factors. After hydropower, the rehabilitation of existing (subcritical) coal power plants and the fuel switch from oil or natural gas to coal fired power plants present some of the lowest levelized costs at any capacity factor; these options are not currently used in the country. Allowing for CO2 revenues does not significantly change the least cost capacity expansion ranking. For 2007 investment costs (based on which the analysis was made) even at a CO2e price of US$50, wind power is still not the least cost option. Within this range of revenues, carbon credits fail to effectively affect the ranking of options, proving that the Clean Development Mechanism (CDM) alone at the 2007 price level is not enough to promote alternative zero carbon energy under existing conditions in Colombia. Therefore, other policy options are required to facilitate market entry for wind power. Wind Energy Capital Costs Are Expected to Decrease Primarily because of the increased interest caused by climate concerns, wind power installations are experiencing rapid change and improvements. For example, the energy produced per unit of installed capacity (measured as the weighted average of capacity factors) went from 22 percent for wind power projects installed before 1998 to 30­32 percent for projects installed from 1998 to 2003 and to 33­35 percent for projects installed during 2004­2006 (LBNL 2008). Investment costs have decreased in the last year after peaking late in 2008. Investment costs for wind energy projects experienced a decreasing trend, which was interrupted between 2004 and 2008 as consequence of high demand, limited production capacity, and the global high demand for raw materials. Recent Executive Summary xv information indicates that investment costs have continued the long term downward trend, with mid 2009 average costs at around $1,800/kW. Annual average operation and maintenance costs of wind power production have also continuously declined6 since 1980. Most importantly, the capacity weighted average of 2000­2007 Operation and Maintenance (O&M) costs for projects constructed in the 1980s was equal to US$30/MWh, but dropped to US$20/MWh for projects installed in the 1990s and to US$9/MWh for projects installed in the 2000s. These trends are expected to continue in the foreseeable future, gradually improving the relative competitiveness of wind power. Wind and Hydro Energy Resources Are Complementary The report examines the extent to which the wind resource is complementary to the hydro regime in Colombia.7 Wind power appears to be available when its contribution to the national grid is most needed, that is, during the dry periods and to an extent during the early evening when demand peaks. Large scale droughts could affect Colombia's interconnected power system due to its high reliance on hydropower. Historically, critical drought conditions are linked to El Niño events, such as those of 1991­1992 and 2002­2003. Existing power generation data from Jepírachi (for the period from February 2004 to March 2009) and wind velocity records data from Puerto Bolívar were extended to cover the period from 1985 to 2008 to assess wind generation capacity during drought periods. The analysis considered four rivers with substantial hydropower development: Guavio, Nare, Cauca, and Magdalena. The most severe droughts in these basins correspond to the El Niño period from April 1991 to July 1992 when strict energy rationing occurred, and from April 1997 to May 1998 when pool prices reached very high spot prices, forcing regulatory changes in the market. During these periods the estimated generation from wind was well above the mean value. That is, during periods of extreme drought associated with El Niño, wind energy from northern Colombia was above average. This analysis is described in detail in Appendix 6. Complementarity was also explored by analyzing the joint operation of a simple system consisting of a wind farm operating in tandem with a hydropower plant of similar size for each of the rivers studied and for a range of reservoir sizes. The analysis is summarized for each of the rivers and is also described in Appendix 6. Results suggest that firm energy from the joint operation of wind and hydropower plants surpasses the isolated operation of the hydropower plant and of the wind farm. This result holds for a wide range of possible reservoir sizes studied. The strong complementarity that the joint operation of wind and hydropower plants exhibits has not been recognized by the current regulatory system adopted by Colombia. Options to Address Barriers to Entry Despite the resource endowment and strategic advantages, under current circumstances wind based generation faces considerable obstacles to participate in the nation's power mix. Key obstacles (described in the first stage report8) include the current relatively high capital intensity and the structure of the regulatory system, which does not acknowledge wind's potential firm capacity.9 Specifically, there is a xvi Executive Summary mechanism in place that remunerates firm energy10 (through auctions), in which wind power currently cannot participate. The first stage report identifies barriers that nonconventional renewable energy sources face in the country and proposes various sets of policy options that may lead to a wide market entry. There is a wide range of potential instruments through which governments can guide the functioning of power markets. Many of these instruments would be applicable to the energy sector in Colombia. However, only a subset of options was explored in detail (those that are in agreement with the existing regulatory system in Colombia and have the effect of changing the financial results for a potential investor): Access international financial instruments to internalize global externalities in national and private decisions. The government can play an active role in promoting access to financial instruments aimed at reducing GHG emissions through: o Active participation in the CDM by engaging in the global carbon market. This is already mainstreamed into the environmental policy in Colombia, but it could be further strengthened within the energy policy; and o Access to multilateral soft loans earmarked for alternative energies or other concessionary funding sources for low carbon investments such as the Clean Technology Fund (CTF). Target subsidies through government fiscal mechanisms. The government could utilize fiscal measures for the benefit of potential investors. Specifically, the mechanisms identified are: o Reduction in income tax. As previously indicated, tax exemptions or reductions are policy mechanisms to guide investment toward areas of policy interest. From the investor's point of view, such policies are tools to improve the after taxes returns; and o Exemptions from system charges. The government could use the regulatory system to reduce or eliminate charges paid for automatic generation control, environmental charges, and/or contributions to the Fund for the Electrification of Off grid Regions (FAZNI). Reform the regulatory system. The regulatory system should be adjusted to promote a level playing field for wind power, and to guide the country toward low carbon intensity development. The existing regulatory system has developed mechanisms to steer the market in order to provide a more resilient interconnected system (measured by its capacity to deliver the demand even during the most difficult hydrological conditions). In doing so, Renewable Energy Technologies (RETs) have not received adequate compensation for their contribution. This situation could be remedied by: o Adjustment of the reliability charge. Colombia has developed a financial mechanism to produce an economic signal to investors as a price premium on reliable installed power capacity. Unfortunately, the existing regulation does not have clear rules to assess the potential contribution of wind energy to the overall reliability of the interconnected system and Executive Summary xvii thus favors conventional power plants. In practice this discriminatory treatment has been identified as a major barrier to further investments in the wind sector; o In relation to the above, an alternative policy option analyzed is the possibility of reducing or eliminating Real Equivalent Cost of Capacity Charge (CERE) payment obligations for certain RETs, as an extension of the existing option for small scale investments;11 and o The regulatory system could also be adjusted to correct market failures by creating charges and payments to adjust for externalities. To correct the economic signal for environmental externalities with impacts on local communities, ecosystems and economic sectors, a sustainability charge (green charge) has been proposed. Highly polluting technologies would be charged while clean technologies would receive a payment, making the system cost neutral to the government. As found in discussions with decision makers and high level policy advisors, the selected options are consistent with the existing regulatory system in Colombia and agreeable to the key stakeholders for further analysis. This analysis could likely take place when the government further fine tunes its decision on policy instruments and policy options to guide the power sector in the future. Impact of Policy Options The assessment focuses on the identification of policy options (government intervention) that would enable a wind power plant to reach a 14 percent rate of financial return (independent investor decision). The main results of the assessment can be found in table 1. The table also summarizes the results of applying different options to a 300 MW wind power project, assuming three investment costs. For each investment cost, three scenarios are described, depending on the reliability factor used to recognize the project's contribution to firm energy during dry periods. The values include a worst case assessment of firm energy contribution (reliability factor of 0.20), an intermediate value (reliability factor of 0.30), and a moderate estimate of the reliable firm energy (0.36). Main results of the impact assessment of the policy instruments are: The single most effective policy instrument to promote wind power in Colombia is the granting of access to reliability payments, recognizing the firm energy and complementarity offered by wind. The implementation of this policy option is relatively easy to incorporate into the existing regulatory system. For new wind power plants with costs in the range of $1,800/kW installed, the adoption of the reliability payments is enough to attract investors operating in wind fields with similar characteristics to that found in Northern Guajira. Higher capital costs require access to concessionary financial conditions, such as those provided under the CTF or fiscal incentives. xviii Executive Summary Lessons Learned The principal lessons learned from this study are as follows: Wind powered power plants are experiencing improvements in efficiency and reductions in operation and maintenance costs. Moreover, since 2008 investment costs have decreased, returning to the expected technology maturing behavior of cost reductions with time, a trend that is expected to continue. In certain locations, such as northern Colombia, wind resources are plentiful and could provide substantial complementarity to hydro based power systems. Under existing conditions wind is not a competitive technology option in Colombia. Of the several barriers found, the most relevant is the difficulty in accessing payments for wind's contribution to firm energy. Governments have a wide range of policy instruments and policy options available to promote RET. To foster wind resources, governments should strengthen wind data collection as a public service, improve access to research and technology developments, and modernize grid access to wind power. Although the analysis has centered on Colombia and its energy sector, the approach and main results are applicable to other countries relying on hydropower. In summary, under existing conditions wind farms are not financially attractive in Colombia even considering the drop in investment costs recorded during 2009. However, wind investments would become financially attractive if the benefits of reliability payments are extended to wind power, even under current investment costs. The government has other multiple policy instruments to steer independent investors toward RETs. Adopting several of these options, as detailed in the report, seems relatively simple and will not distort the market. Improving the conditions for market entry of the wind option will serve to prepare the sector for the anticipated improvement of conditions as investment costs for wind decrease over time. Finally, deployment of the wind option would help the sector to strengthen its climate resilience and be better prepared to face climate variability, without increasing its carbon footprint. Executive Summary xix Table 1. Actions Required to Reach a Financial Threshold for a 300 MW Wind Power Plant on the Northern Coast If reliability Investment payment cost/kW (US$) considered at % Required actions to reach a 14% Internal Rate of Return (IRR) Elimination of sector fees (AGC, FAZNI, CERE) and considerable financial Nonea support: i.e., 10% CTF financing and access to 60% soft loansb Requires considerable financial support: i.e., 40% CTFc financing and access to 20% 20% in soft loans $2,400 Requires considerable financial support: i.e., 30% CTF financing and access to 30% 30% in soft loans Requires considerable financial support: i.e., 20% CTF financing and access to 36% 50% in soft loans Elimination of sector fees (AGC, FAZNI, CERE) and special financial support: i.e., None access to 30% soft loans Requires considerable financial support: i.e., 15% CTF financing and access to 20% 55% in soft loans $2,100 Requires considerable financial support: i.e., 5% CTF financing and access to 30% 65% in soft loans 36% Requires financing support: i.e., 60% access to soft loans None Elimination of sector fees (AGC, FAZNI, CERE) 20% Requires financing support: i.e., 40% access to soft loans $1,800 30% No additional interventions required 36% No additional interventions required Source: Authors' data. Notes: a. Waiving a project's obligation to make CERE contributions is financially equivalent to remunerating the project with a reliability factor of around 0.4, as is shown later in this analysis. b. Soft loans here mean those with conditions typical of IBRD loans in Colombia: currently, a 17 year repayment period, interest rate LIBOR + 1.05%, front end fee 0.25%. c. The CTF is a climate change donor driven fund seeking the implementation of transformational low carbon options. CTF financial conditions are typically a 0.65% interest rate with a 20 to 40 year repayment period and 10 years of grace. Notes 1 According to the National Hydrocarbon Agency of Colombia (Agencia Nacional de Hidrocarburos, ANH) in 2009. 2 However, in 2008 there was an increase in registration of coal power projects (totaling 2,884 MW) and for the first time, fuel oil projects (totaling 305 MW of installed capacity). In contrast, 2,520 MW were natural gas, 7,770 MW were hydropower, and (as mentioned previously) 19.5 MW were wind. 3 Generally, the term reliability refers to the certainty that operators may have with regard to the future power output of their power plant. In the context of conventional and nonconventional power sources, although some may claim that conventional power sources are more reliable, xx Executive Summary others show that their reliability is hampered by the sudden shutdown of a power plant. Alternatively, nonconventional renewable power plants (such as wind farms) are claimed to be highly reliable because wind turbines do not all shut down simultaneously and instantaneously. As explained in this document, this is not a concept that has been integrated in the energy market in Colombia. It should be noted that in this document and for the case of Colombia, the term reliability is necessarily related to the reliability payment and the firm power output that power plants can produce during dry periods and in times of drought (this is further explained throughout the document). 4 However, current technical constraints do not allow a system to be fully based on wind power. 5 The level of emissions of the sector is well below the average in the United States, the European Union, Canada, and Mexico (0.35 ton CO2e/MWh). Some power plants that utilize renewable energies have already tapped into the international carbon trade (Jepírachi Wind Farm, Amoyá Run of River Power Plant) at an individual level, and new mechanisms are being developed globally to promote low carbon development paths. 6 Lawrence Berkeley National Laboratory (LBNL) estimates that this drop in costs could be due to the following: (a) O&M costs generally increase because as turbines age, component failures become more common; (b) as manufacturer warranties expire, projects installed more recently with larger turbines and more sophisticated designs may experience lower overall O&M costs on a per MWh basis; and (c) project size. To normalize for factors (a) and (b) above, LBNL produces other figures and analyses that can be found in the original publication but nonetheless reveal O&M cost declines. 7 The analysis is based on Jepírachi's operational record and wind data in meteorological stations in northern Colombia. 8 Vergara et al., 2008. 9 Note that the firm capacity of renewable energy is the capacity of conventional sources replaced, such that demands can be met with a specified reliability. The firm capacity of a renewable source depends on the correlated variations in demands and renewable supplies (Barrett 2007). 10 Firm energy is defined as the maximum monthly energy that can be produced without deficits during the analysis period which includes El Niño occurrences (this is further explained throughout the document). 11 It should be noted that simultaneously allowing for reliability charges and waiving CERE payments is not recommended. It would imply a logical contradiction because funds for the reliability charge come from CERE. CHAPTER 1 Introduction Context This report constitutes the second phase of an effort to identify and address barriers to the deployment of wind energy in Colombia's power sector. The first phase was reported in a document entitled Review of Policy Framework for Increased Reliance on Renewable Energy in Colombia, completed in February 2008 and discussed with high level energy authorities in Colombia. It concluded that (i) Colombia has a substantial nonconventional renewable energy resource endowment, in particular wind and solar but also significant prospects for geothermal, that complements the existing large hydropower potential; (ii) nonconventional energy options face important policy and regulatory barriers that prevent market entry; (iii) globally, several nonconventional renewable energy options are becoming financially more attractive as a result of a normal maturity process and commercialization of low carbon options; (iv) internalizing global and local externalities increases the competitiveness of selected nonconventional sources; and (v) options are available to decision makers to address barriers to the expansion of nonconventional power in the Colombian power mix. The report was designed to explore the impact of options identified for addressing these barriers. The wind regime in Colombia has been rated among the best in South America. Offshore regions of the northern part of Colombia have been classified with class seven winds (winds over nine meters per second [m/s] at heights of 50 meters). The only other region in South America with such high wind availability is the Patagonia region of Chile and Argentina. Colombia has an average estimated wind power potential of 18 GW in the La Guajira region, enough to meet the national power demand twice over (Pérez and Osorio 2002). However, the country only has an installed capacity of 19.5 MW of wind energy (Jepírachi Project, supported by the Bank) with a few additional projects under consideration, including a 200 MW project in Ipapure. Consequently, wind power today represents a small fraction of the installed capacity. In 2008 the installed capacity in Colombia (13.4 GW) was 67 percent hydro (including small hydro), 27 percent natural gas, 5 percent coal, and 0.3 percent wind and cogeneration. Figure 1.1 illustrates the installed capacity per technology type.1 The total annual electricity demand that same year was 54 TWh (UPME 2009). Colombia also has substantial reserves of natural gas and coal, which could be used to generate power. The natural gas reserves in 2007 were seven tera cubic feet, including proven and unproven reserves (Ministry of Mines and Energy 2008). The La Guajira region of Colombia supplies most of the demand, 62 percent in 2007, compared to the next highest supplier (Cusiana) with 26 percent. 1 2 World Bank Study Figure 1.1 Installed Capacity per Technology Type Other 26 MW Coal 700 MW Wind 18 MW Natural gas 3702 MW Hydropower 8994 MW Source: UPME 2009. Colombia's coal reserves are estimated at seven billion tons (or about 100 years of production at the present mining rate). These reserves are mostly located in the northern part of the country and are the largest coal reserves in South America. Most coal mined is anthracite, with very low ash and sulfur content, ideal for exports to the European market. Current production is 59 MMT (42 MTOE), with plans to increase production to 100 MMT by 2010.2 Most of Colombia's coal production is exported. Of the coal used internally (2.4 MMT in 2000), more than 75 percent goes to industrial uses and the rest goes to the power sector (equivalent to 378 KTOE or ~4,400 GWh). Colombia's power sector is maturing quickly, with relative stability in its regulations, an unbundled system, and a dispatch mechanism that closely resembles a well functioning competitive market. Competition is promoted and tools have been designed to attract cost effective capacity expansions that would promote reliability3 of service. (A fuller description of the system and its dispatch mechanism was included in the stage one report.) However, the interconnected system, if unguided, is not likely to promote nonconventional renewable energy resources such as wind, but rather maintain a high capacity share of hydropower or alternatively move toward a more carbon intensive energy resource mix (likely reliant on abundant coal reserves). In the latter case this would result in an increase in the carbon footprint of the economy from its current relatively low level of GHG emissions.4 The analysis focuses on wind power. Wind is currently the least cost nonconventional renewable energy alternative. There is also the possible complementarity of the wind regime with periods of low hydrology, which is further explored in this report. The World Bank was an early supporter of the wind option in Colombia through its participation in the Prototype Carbon Fund of the Jepírachi Wind Power Plant in the province of La Guajira. Wind Energy in Colombia 3 Structure of the Report After the introduction, Chapter 2 summarizes the main findings of the first phase. It describes Colombia's energy profile and presents the main barriers that limit the development of nonconventional renewable energy sources. Chapter 3 presents a comprehensive comparison of 37 energy technologies through levelized cost analyses. The analysis permits the identification of the technologies most likely to participate in the future expansion of the interconnected system. It also studies whether CO2 revenues change the least cost capacity ranking. Chapter 4 summarizes the cost evolution of wind energy units over time and provides an overview of the trends that define the future of this technology. Chapter 5 presents the complementarity of joint operation of wind and hydro in Colombia and explores the possible contribution of wind to firm energy. Chapter 6 introduces different policy options to facilitate the market entry of wind power, and Chapter 7 reviews the effectiveness of the selected policy options in creating the adequate incentives (that is, expected financial returns on equity) to attract potential investors. Key findings and conclusions are summarized in the Chapter 8. Notes 1 In 2008 there was an increase in the registration of prospective coal power projects (totaling 2,884 MW) and, for the first time, of fuel oil projects (totaling 305 MW of installed capacity). In contrast, 2,520 MW were natural gas, 7,770 MW were hydropower, and (as mentioned previously) 19.5 MW were wind. 2 Although there are plans to expand production, there is also a holdback based on fears that this would cause a drop in coal prices because Colombia is such an important player in the world's thermal coal market. 3 Generally, the term "reliability" refers to the certainty that operators may have with regard to the future power output of their power plants. In the context of conventional and nonconventional power sources, although some may claim that conventional power sources are more reliable, others show that their reliability is hampered by the sudden shutdown of a power plant. Alternatively, nonconventional renewable power plants (such as wind farms) are claimed to be highly reliable because wind turbines do not all shut down simultaneously and instantaneously. As explained in this document, this is not a concept that has been integrated in the energy market in Colombia. It should be noted that in this document and for the case of Colombia, the term "reliability" is necessarily related to the "reliability payment" and the "firm power" output that power plants can produce during dry periods and in times of drought (this is further explained throughout the document). 4 The sector's level of emissions is well below the average in the United States, the European Union, Canada, and Mexico (0.35 ton CO2e/MWh). Some power plants that utilize renewable energies have already tapped into the international carbon trade (Jepírachi Wind Farm, Amoyá Run of River Power Plant) at an individual level, and new mechanisms are being developed globally to promote low carbon development paths. CHAPTER 2 Summary of Findings from First Stage Report: Nonconventional Renewable Energy Barrier Analysis T his chapter summarizes the results of the first stage of the ESMAP funded Review of Policy Framework for Increased Reliance on Renewable Energy in Colombia. Its objective was to identify barriers to the development of nonconventional renewable energy resources in Colombia. Large hydro is not included as part of nonconventional energy resources because it is a well established option in Colombia. Large hydropower is also a relatively low cost renewable energy source and already constitutes the bulk of the base load in the power sector. This document emphasizes nonconventional renewable energy sources. Colombia is a net energy exporter. Colombia is not one of the world's leading energy producers, but it is a net energy exporter. Colombia's demand for energy has been increasing over the past decade and is expected to grow at an average of about 3.5 percent per year through 2020 (UPME 2009). The country's total energy production in 2006 was 3.3 QUADS (quadrillion1 BTU),2 while consumption was 1.2 QUADS, from which electricity consumption stood at 0.14 QUADS.3 This highlights the energy export nature of the Colombian economy. The difference between its energy production and consumption has been due mostly to oil and large coal exports. The country is a modest energy user and CO2 emitter. The power sector in Colombia already has a very low carbon footprint (0.35 tons/MWh generated4). Energy demand is characterized by growing requirements in the transport sector, followed by the industrial and domestic sectors. The average power use per capita is 923 kilowatt hours (kWh)/year. National carbon dioxide (CO2) emissions are 59.4 million metric tons (MMT), or 1.3 tons of CO2 (tCO2)/capita, less than half the world average. Colombia's energy intensiveness is 0.2 CO2/GDP (PPP) (kg CO2/2000 US$ PPP), according to the International Energy Agency (IEA) in 2006.5 This is much lower than that of countries in Europe and North America. Hydropower is the dominant source of energy and is likely to continue to characterize Colombia's power sector for the foreseeable future. Currently, about 64 percent of capacity and 81 percent of generation are hydro based. A generous hydrological regime and a favorable orography provide the basis for a large 4 Wind Energy in Colombia 5 hydropower potential. The most recent bid for power supply resulted in an overwhelming supply of new hydropower plants to meet the projected increase in demand in the immediate future. A largely hydro based power system may be susceptible to anticipated climate variability affecting rainfall patterns. A projected increase in the intensification of the water cycle and the possible intensification of extreme events (El Niño Southern Oscillation [ENSO] and La Niña) associated with temperature dipoles on the Pacific coast of Colombia may raise the vulnerability of the power sector by affecting the reservoir capacity of hydropower based plants. It is therefore prudent to examine how the sector's climate resilience could be strengthened. Colombia's oil reserves are more limited. The country has long relied on a generous endowment of fossil fuels, oil, coal, and gas to meet domestic energy needs and to contribute substantially to the balance of trade in international markets. However, recent discoveries have expanded reserves in number of years of supply, which until recently had been estimated at eight years (Ministry of Mines and Energy 2008). Natural gas supplies are sufficient for 27 years of supply at the current rate of consumption; however, bottlenecks in the gas distribution system limit its use in several areas of the country. The main transportation restrictions will be removed in the 2010­2012 period with new pipelines and transport loops that are under construction and that could facilitate natural gas transport from the main fields to the large natural gas markets. Prior to the use of nonconventional renewable resources in the power sector, there is a need to address a number of barriers that impede the wide deployment of these resources. These include: capital intensity, local financial market limitations, lack of regulations and regulatory uncertainty, lack of adequate data to assess resource availability, lack of clear rules for nonconventional energy sources, bias toward conventional technologies (for example, with the firm energy reliability payment), and limited strategic planning. The Government of Colombia (GOC) can play a significant role in facilitating the entry of nonconventional energy sources. Policy options include: (i) developing a strategic energy plan beyond 10 years that includes nonconventional energy resources; (ii) similarly, adopting least cost planning that includes environmental and social costs in decision making; (iii) modifying the regulatory framework to address obstacles that prevent a level playing field for nonconventional renewable power resources; (iv) facilitating information sharing on wind endowment; and (v) facilitating access to financial instruments available under climate change investment funds. This report focuses on alternatives to address (counter) the relatively higher capital intensity of the wind power option, which may result in a more attractive energy source in the country, provided that certain potential regulatory framework modifications are made. 6 World Bank Study Notes 1 1015; SI prefix peta (P). 2 3.3 QUADS or 85 MTOE (IEA 2006). 3 0.14 QUADS or 42 TWh (IEA 2006). 4 As estimated in the recently completed PDD for the Amoyá Environmental Services Project. 5 http://www.iea.org/Textbase/stats/indicators.asp?COUNTRY_CODE=CO&Submit=Submit. CHAPTER 3 Cost Comparison of Alternative Power Sources Based on the Expansion Plan for 2008­2025 B efore a detailed assessment is made of policy options to facilitate market entry for wind power, this chapter provides a cost comparison of available technologies for power generation, based on the generation expansion plan for 2008­2025 prepared by the Mines and Energy Planning Unit (UPME) of the Colombian Ministry of Mines and Energy. For this purpose, the analysis includes simple screening curves of 37 power generation technologies to compare with the results of the wind option. Hydropower is the dominant source in the National Interconnected System (NIS) and is expected to continue to be so for the foreseeable future. The large base load hydro capacity is complemented today by thermal power, mostly from domestic natural gas fired power plants and a much smaller amount from domestic coal fired power plants. Methodology for Technology Cost Comparison Due to data availability restrictions, the analysis is limited to a simple static analysis to provide indicative values. Projections of increase or change in capital cost of power plants are beyond the scope of this study, especially considering the rapid growth and volatility in capital costs experienced since the early part of the present decade. Therefore, the most recent capital costs available are used (2007/2008). Price assumptions, in line with national projections, are made as follows: coal at US$35 per ton, natural gas at US$4/MBTU, and residual fuel oil for power plants at US$51 per barrel. The calculation of levelized total plant costs (TPC) is based on the "Technical and Economic Assessment of Off grid, Mini grid and Grid Electrification Technologies" (World Bank 2007). The 37 electricity generation options are listed in table 3.1. Coal fired power plants are considered as equipped with flue gas desulfurization (FGD) and selective catalytic reduction (SCR). Although Colombia currently does not require FGD, equipping coal fired power plants with FGD and SCR represents best international practice even when low sulfur coal is used. In addition, equipping SCR and FGD is a prerequisite to make coal fired power plants ready for carbon capture and storage (CCS). Coal fired power plant options include those that are much less expensively made in China. Two metrics are used to assess the relative rating, as per 7 8 World Bank Study the procedure mentioned above: the cost of capacity of the plant per year (US$/kW per year) and the cost of generation (US$/kWh). Table 3.1. Power Generation Options Included in the Screening Curve Analyses Plant Type Subcritical (SC) coal-fired 300 MW/550MW Diesel 5 MW Supercritical (SPC) coal-fired 550 MW Hydro 400 MW/1200 MW Ultra supercritical (USPC) coal-fired 550 MW* Wind 10MW/300 MW Subcritical (SC) 300 MW/550 MW coal-fired carbon capture Subcritical (SC) Circulating Fluidized Bed (CFB) and storage (CCS) 300MW/500MW Supercritical (SPC) coal-fired 550 MW carbon capture and Subcritical (SC) Natural Gas Steam 300 MW storage (CCS) Ultra supercritical (USPC) coal-fired 550 MW carbon Subcritical (SC) Oil Steam to Coal 300 MW capture and storage (CCS) Integrated Gasification Combined Cycle (IGCC) 300 Subcritical (SC) Natural Gas Steam to Coal 300 MW MW/640 MW Integrated Gasification Combined Cycle (IGCC) carbon Subcritical (SC) 500 MW Rehabilitation capture and storage (CCS) 220 MW/555 MW Simple Cycle Gas Turbine (GT) 150 MW China subcritical (SC) 300 MW/550 MW Combined Cycle Gas Turbine (CCGT) 140 MW/560 MW China supercritical (SPC) 550 MW Combined Cycle Gas Turbine (CCGT) China ultrasupercritical (USPC) 550 MW Combined Cycle Gas Turbine (CCGT) carbon capture and China subcritical (SC) 300 MW/SC 550 MW carbon capture storage (CCS) 50 MW and storage (CCS) Combined Cycle Gas Turbine (CCGT) carbon capture and China supercritical (SPC) 550 MW carbon capture and storage (CCS) 482 MW storage (CCS) Fuel Oil Steam 300MW Source: Authors' data. Notes: CFB: Circulating Fluidized Bed. IGCC: Integrated Gasification Combined Cycle. CCS: Carbon Capture and Storage. CCGT: Combined Cycle Gas Turbine. SC: Subcritical. SPC: Supercritical. USPC: Ultra supercritical. *According to the World Coal Institute website in 2009, new pulverized coal combustion systems-- utilizing supercritical and ultrasupercritical technology--operate at increasingly higher temperatures and pressures and therefore achieve higher efficiencies and significant CO2 reductions than those of conventional pulverized coal fired units (www.worldcoal.org). As of 2006, nine coal fired power plants were installed in Colombia (totaling 700 MW); these were commissioned between 1963 and 1999. Although it is unclear whether these power plants have been rehabilitated to prolong their plant life, they are included in the analysis. Moreover, although a few hydropower plants operate at a high capacity factor of around 80 percent, it is assumed that, on average, the hydropower capacity factor is 60 percent. A 40 percent capacity factor is assumed for wind power.1 Within the screening curves, the electricity generation plants were ranked in order of least levelized cost per kW for different capacity factors. The levelized cost analysis is done with and without consideration of carbon revenues. The results are presented below. Wind Energy in Colombia 9 Least (Levelized) Cost Comparison Clearly, the low cost of hydropower in Colombia is evidenced by the high hydropower capacity reserve of its power system, in which many hydropower plants function as base load. The total hydropower net effective installed capacity is 13 GW with a peak power demand at 9 GW. With or without CO2e emission reduction revenues, large scale hydropower is the least cost power option. The rehabilitation of subcritical coal power plants and the fuel switch from oil or natural gas to coal fired power plants present the next lowest levelized costs at any capacity factor. However, these options do not add to installed capacity. The next low cost option is low cost manufactured coal fired power plants, without allowance for CCS. Likewise, Combined Cycle Gas Turbines (CCGT) are among the cheapest technology options. Wind power generation under current scenarios and conditions, and even with possible capacity factors of up to 40 percent, is not among the least cost choices. Similarly, Integrated Gasification Combined Cycle (IGCC) and CCS technologies are also not among the least cost options in Colombia. The most cost effective power generation options are presented in tables 3.2 and 3.3. The options presented are similar to the current generation picture of Colombia, but with more inclusion of coal power plants due to their lower cost. Abundant coal reserves would back up the development of this option. This assumes that the internalization of global environmental issues is not considered. Figures 3.12 and 3.2 provide a graphic representation of the results of the analysis. Figure 3.1 presents the results for the aggregate cost over a year; this figure increases as the capacity factor increases since it shows the amount of power generated over the year. Figure 3.2 presents the calculated generation costs, which decrease as the capacity factor increases. Table 3.2. Least-Cost Capacity Expansion Mix (without CO2e revenue) Electricity generation Base load Medium load Peak load Major additions Large and medium hydropower with modest backup Large and medium Large and of new capacity requirement of low-cost coal-fired SC, SPC and hydropower medium USPC power plants using most advanced clean coal hydropower technology Minor additions CCGT and old SC coal power plant rehabilitation CCGT (which could also Gas turbines of capacity using most advanced clean coal technology operate both base load and and diesel peaking, as backup) Additional 15% Large and medium hydropower Large and medium Large and for capacity hydropower medium reserve hydropower Source: Authors' data. 10 World Bank Study Table 3.3. Suggested Capacity Expansion Mix at US$18 per Ton CO2e Electricity generation Base load Medium load Peak load Major new Large and medium hydropower with modest backup Large and medium Large and capacity requirement of low-cost coal-fired SC, SPC and hydropower and wind medium USPC power plants using most advanced clean power hydropower and coal technology wind power Modest new CCGT and old SC coal power plant rehabilitation CCGT (which could also Gas turbines and capacity using most advanced clean coal technology operate both base load diesel and peaking, as backup) 15% or more Large and medium hydropower Large and medium Large and capacity reserve hydropower medium hydropower Source: Authors' data. Figure 3.1. Screening Curve for Levelized Total Costs Measured in Cost of Capacity of a Plant per Year (US$/kW-yr) at Different Capacity Factors 700 IGCC CCS 555MW USPC 550MW CCS 600 IGCC 640MW Simple Cycle GT 150 500 MW SC 550MW USPC 550MW US$/KW-yr 400 CCGT CCS 482MW China USPC 550MW 300 CCS CCGT 560MW 200 China USPC 550MW SC Oil Steam to Coal 300MW 100 Small Wind 10MW Wind 300MW 0 Small to Med Hydro 10 20 30 40 50 60 70 80 90 100 400MW Large Hydro 1200MW Capacity factor (%) Source: Authors' data. Note: Coal price US$35/ton; emission reductions US$18/ton CO2e. Wind Energy in Colombia 11 Figure 3.2. Screening Curve for Levelized Total Costs at Different Capacity Factors Measured in Terms of Generation Costs (US cents/kWh) 60 IGCC CCS 555MW USPC 550MW CCS IGCC 640MW 50 Simple Cycle GT 150 MW USPC 550MW 40 SC 550MW CCGT CCS Uscent/KWh 482MW 30 China USPC 550MW CCS CCGT 560MW 20 Small Wind 10MW Wind 300MW China USPC 10 550MW SC Oil Steam to Coal 300MW Small to Med Hydro 400MW 0 Large Hydro 10 20 30 40 50 60 70 80 90 100 1200MW Capacity factor (%) Source: Authors' data. Note: Coal price US$35/ton; Emission reductions US$18/ton CO2e. Coal Netback Calculations For coal prices ranging up to US$60 per ton, the rehabilitation of existing coal fired power plants (limited to a total of 700 MW) is among the least cost options for adding capacity. Rehabilitating existing coal fired power plants is a good option for the range of coal prices indicated.3 At a price of more than US$50 per ton of coal, and including US$18 per CO2e ton, new coal power plants are not a least cost option. Furthermore, if low cost coal fired power plant4 options are excluded, coal fired power plants become the least cost options only at very low coal prices from US$10 to US$20 per ton. 12 World Bank Study Allowing for CO2 revenues does not significantly change the least cost capacity expansion ranking. For analysis purposes it is assumed that CO2e is valued at US$18 per ton for the 37 options (the results are similar to those presented in table A1.2 of Appendix 1). For 2007 investment costs (base year used) even at a CO2e price of US$50, wind power is still not the least cost option. Within this range of revenues, carbon credits fail to effectively affect the ranking of options, proving that the CDM alone at the 2009 price level is not enough to promote alternative zero carbon energy under existing conditions in Colombia. Therefore, other policy options are required to facilitate market entry for wind power. From the results of the analysis, and under current and foreseeable conditions, large hydro remains the best option for power generation and guarantees a power sector that is relatively low in carbon footprint. Moreover, under the current scenario, coal seems an obvious backup option to the base load. Since this is a limited estimate, based on secondary data, a more comprehensive modeling exercise and impact analyses on low carbon growth should be conducted; this would include all other relevant costs (for example, transportation costs, transmission pipeline and distribution costs, transaction costs, environmental and social costs, institutional costs, logistical costs, and so forth). Tools available to perform this analysis include MARKAL.5 Moreover, although not directly assessed, the deployment of renewable sources, including hydro, reduces exposure to volatility in fossil fuel prices. Notes 1 A capacity factor of 40 percent is assumed: the winds on the northern coast of Colombia are class 7 and are constant. This number has been discussed with the utility that owns, maintains, and operates the only wind farm in Colombia. Values have been and can be obtained in the area (in a location near the site where a larger wind project can be located). 2 Figure 3.1 shows the cost per year of operation of a power plant operating at different plant factors. The higher the plant factor the higher the costs (although the cost per unit of energy generated decreases). On the other hand, figure 3.2 presents the average generation costs, which decrease as the capacity factor increases. 3 In Colombia, most coal power plants are old and have not been retrofitted (there has been a focus on building natural gas plants, rather than coal plants). These coal power plants could be modernized to achieve greater efficiencies. 4 New low cost coal fired power plants (imported from China, with operational reliability yet to be defined) result in least cost; this is especially true if a supercritical (SPC) coal fired power plant of 550 MW is installed. 5 MARKAL is a generic model tailored by the input data to represent the evolution over a period of usually 40 to 50 years of a specific energy system at the national, regional, state, provincial or community level. MARKAL was developed by the Energy Technology Systems Analysis Programme (ETSAP) of the International Energy Agency. Source: http://www.etsap.org/Tools/ MARKAL.htm. CHAPTER 4 Wind Power Costs Outlook T he results of the technology cost comparison show that under existing conditions (base year 2007) wind power is not a least cost option for power generation in Colombia, even at a CO2e price of US$50/ton and high capacity factors. However, wind power costs are expected to decrease with time as the technology matures. This chapter examines the trends in wind power costs and performance. Technical Viability of Wind Power In early 2009 wind power installed capacity worldwide reached 121 GW. Since the late 1990s, wind power installed capacity has increased by over 20 percent annually and is expected to continue increasing in 2009 and 2010 by similar magnitudes (figure 4.1). Figure 4.1. World Total Wind Power Installed Capacity (MW) Source: World Wind Energy Association 2009. 13 14 World Bank Study Efficiency Gains over Time Project capacity factors have increased in recent years due to technological advancements, higher hub height, and improved siting. The weighted average of capacity factors went from 22 percent for wind power projects installed before 1998 to 30­32 percent for projects installed from 1998 to 2003 and to 33­35 percent for projects installed from 2004 to 2006 (LBNL 2008). Even capacity factors above 40 percent can be found in excellent wind resource areas, such as those in northern Colombia. The following figure (4.2) presents the evolution of capacity factors by commercial operation date in the United States. Figure 4.2. Project Capacity Factors by Commercial Operation Date Source: Berkeley Lab database. A cost study conducted by the U.S. Department of Energy (DOE) Wind Program identified numerous opportunities for reductions in the life cycle cost of wind power (Cohen and Schweizer et al. 2008). Based on machine performance and cost, this study used advanced concepts to suggest pathways that integrate the individual contributions from component level improvements into system level estimates of the capital cost, annual energy production, reliability, operation, maintenance, and balance of station. The results indicate significant potential impacts on annual energy production increases (estimated with an average efficiency increase of 45 percent) and capital cost reductions of 10 percent. Changes in annual energy production are equivalent to changes in the capacity factor because the turbine rating was fixed. Capital Cost Evolution Figure 4.3 provides the trend in turbine costs in the U.S. market. Wind power project costs are a function of turbine prices. Turbine prices went from US$700/kW in 2000­ 2002 to US$1240/kW in 2007; these costs were even higher in 2008 (US$2,200/installed kW). Higher costs in 2006­2008 were likely due to the high demand for technology (shortages in certain turbine components and turbines, greater demand than supply), the high cost of materials/inputs (such as oil and steel), a general move by manufacturers to improve their profitability, the devaluation of the dollar in comparison to the euro, an upscaling of turbine size and hub height, and improved sophistication in turbine design such as improved grid interaction (LBNL 2008). Wind Energy in Colombia 15 Figure 4.3. Reported US Wind Turbine Transaction Prices over Time Source: LBNL 2008. After the peak values reached in 2008 (equivalent to unit investment costs around US$2,400/kW), new transactions indicate a return to a more competitive market. As of March 2009, the European Wind Energy Association reported that the average cost of recent projects is now back to around the level of 1,225/kW. This translates to approximately US$1,800/kW as the average 2009 transactions in the European market. This would continue the long term trend in capital cost reductions observed earlier. Operation and Maintenance Costs Are Decreasing Annual average O&M costs of wind power production have declined1 substantially since 1980. O&M cost declines can be observed in figure 4.4 for projects that were installed in 1980, until 2005. The figure specifically suggests that capacity weighted average 2000­2007 operation and maintenance costs for projects constructed in the 1980s equal US$30/MWh, dropping to US$20/MWh for projects installed in the 1990s, and to US$9/MWh for projects installed in the 2000s. Figure 4.4. Average Operation and Maintenance Costs for Available Data Years from 2000 to 2007, by Last Year of Equipment Installation Source: Berkeley Lab database; five data points suppressed to protect confidentiality. 16 World Bank Study Wind Power Grid Integration Integration of large capacities of wind energy into power systems is increasingly less of a concern (there is growing literature in this respect2). In fact, as an example, the European Wind Energy Association considers that integrating 300 GW of wind power by year 2030 into European power systems is not only a feasible option for the electricity supply, but it has the benefits of increasing the security of supply and could contribute to low and predictable electricity prices (European Wind Energy Association 2008). Furthermore, wind power has also been stated to help with system stability by providing Low Voltage Run Through (LVRT)3 and dynamic variable support to thus reduce voltage excursions and dampen swings (UWIG 2007). Moreover, by integrating wind power into the energy grid, the aggregation of wind turbines reduces variability in power generation;4 simultaneous loss of capacity does not occur in a broad geographic region (as shown by extensive modeling studies). Meso scale wind forecasting could provide some predictability of plant output within some margin of error; similarly, forecasts are improving (UWIG 2007). Turbine orders larger than 300 MW tend to result in lower costs than turbine orders of less than 100 MW (likely due to economies of scale and lower transaction costs/kW) (LBNL 2008). However, there seems to be a small difference in costs for projects between 30 and 200 MW; in general, variations in costs of wind projects are more likely due to regional differences such as development costs, site and permitting requirements, and construction expenses (URS 2008). Outlook Wind power has undergone a fast developmental phase. The unprecedented pace of growth during this decade has outpaced manufacturing capabilities, creating a seller's side market. Prices have also been affected by commodity price fluctuations, associated with the increasing levels of economic activity seen in the last five years and more recently by changes in the worldwide economy. Wind power capacity is expected to continue to rise significantly worldwide and to play an increasingly relevant role in meeting the growing energy demands of the future. Wind power installed capacity in Latin America is very low and is increasing slowly. However, the slow pace of growth is expected to change once the downward trend in prices induces more stable market conditions. The financial crisis might allow the industry to find opportunities for development and to deal with demand expectations. The threshold price for the wind power option (300 MW) to become competitive with large hydro power (1,200 MW), which is currently the least cost option, without reliance on incentives or other subsidies with the 30 or 40 percent capacity factor is when the levelized cost of wind energy is at US$940/KW and hydro power at US$1,200/KW. Both options then total for either US$136/KW/year at the capacity factor of 30 percent or US$139/KW/year at the capacity factor of 40 percent. Wind Energy in Colombia 17 Notes 1 LBNL estimates that this drop in costs could be due to the following: (a) O&M costs generally increase because as turbines age, component failures become more common; (b) as manufacturer warranties expire, projects installed more recently with larger turbines and more sophisticated designs may experience lower overall O&M costs on a per MWh basis; and (c) project size. To normalize for factors (a) and (b) above, LBNL produces other figures and analyses that can be found in the original publication but nonetheless reveal O&M cost declines. 2 See, for example, Boyle (2007). 3 Also called ride through faults, LVRTs are devices that may be required to be available when the voltage in the grid is temporarily reduced due to a fault or load change in the grid. Wind generators can serve as LVRTs. 4 Aggregation provides smoothing in the short term. However, there are significant benefits to geographical dispersion because dispersion provides smoothing in the long term. CHAPTER 5 Wind and Hydro in Colombia: Complementarity Analysis A lthough the levelized cost analysis indicates that under current conditions wind is not competitive with hydro, wind power under proper circumstances could complement the sector's large hydro based capacity. This chapter examines the extent to which the wind resource complements the hydro regime in the country. It also characterizes some of the climate vulnerabilities of a hydro based power sector to future climate change. Complementarity of the Wind and Hydro Regimes Does the wind energy potential in northern Colombia have a distribution that is complementary to the availability of hydropower? This question can be examined on the basis of Jepírachi's1 power generation records, available since it started operations in 2004,2 and on the analysis of wind data in meteorological stations in northern Colombia. Complementarity could also be measured by wind availability during extreme drought conditions associated with El Niño events, and through the analysis of independent and joint operation of the Jepírachi wind farm and hydropower plants on selected rivers in Colombia. This chapter presents the results of these analyses. Generation Data from Jepírachi Power generation data at hourly level were available for the Jepírachi plant during its operation period.3 These data make it possible to estimate the distribution of the average monthly generation under peak, medium, and base loads (table 5.1). For the dry period of December 1 to April 30 (as defined by the regulatory agency, CREG), Jepírachi produces 10 percent more energy than its yearly average. The historical generation in Jepírachi during the first four months of the year is 17 percent above the yearly monthly generation. 18 Wind Energy in Colombia 19 Table 5.1. Jepírachi Monthly Power Generation Jepírachi average generation (MWh) Ratio of average generation Demand Block Demand Block Total Peak Load Med Load Low Load Peak Load Med Load Low Load Jan 5098 232 4074 792 1.08 1.13 1.00 Feb 5338 258 4269 811 1.20 1.18 1.03 Mar 6414 313 5041 1060 1.46 1.40 1.34 Apr 4893 230 3737 926 1.07 1.03 1.17 May 4515 215 3439 861 1.00 0.95 1.09 Jun 4531 218 3558 755 1.01 0.99 0.96 Jul 6392 290 4768 1334 1.35 1.32 1.69 Aug 5123 248 3939 936 1.15 1.09 1.19 Sep 4046 194 3115 737 0.90 0.86 0.93 Oct 2492 107 1979 406 0.50 0.55 0.51 Nov 2830 130 2307 393 0.61 0.64 0.50 Dec 3722 143 3119 460 0.67 0.86 0.58 Total 55394 2578 43345 9471 1.00 1.00 1.00 "dry" period 25465 1176 20240 4049 1.09 1.12 1.03 "wet" period 29929 1402 23105 5422 0.93 0.91 0.98 Source: Consultants' study (see Appendix 6). Note: The calculations assume that peak load corresponds to the generation during the 20th hour of the day, medium load corresponds to generation during the 6th to 19th and 21th to 23rd hours, and base load corresponds to the remaining hours of the day. This distribution is very important since the medium and peak load hours (when energy is more costly) have a larger plant factor than the base load hours. Table 5.1 also shows the distribution of energy production during the NIS Peak Load, Medium Load, and Low Load periods. During the Peak Load period, defined as the hour of peak demand (8 p.m.), Jepírachi produces 17 percent more energy during the dry season in relation to production during the wet season. This could be interpreted as an indication of the ability of wind based power plants to contribute to peak demand when it is most needed. The contribution of wind farms is also higher during the dry season for all load conditions. While the hydro based system undergoes the dry season (low availability of water for generation), the wind farms in northern Colombia could produce well above their average output. Wind Data from Reference Stations Figures 5.1 to 5.3 present a graphic representation4of the temporal characteristics of the northern coast wind field in Colombia. Figure 5.1 illustrates the distribution of the reference stations used to describe the wind potential on the northern coast of Colombia. Wind data are summarized from Almirante Padilla airport in La Guajira (Station 6 in figure 5.1), the closest climate station to Jepírachi reported in the Wind Atlas, and three other climate stations along the northern Caribbean coast of Colombia (Galerazamba, Bolívar; E. Cortizzos Airport, Atlántico; and S. Bolívar Airport, Magdalena). 20 World Bank Study The Almirante Padilla Airport station provides data that are representative of the wind field found in Northern Guajira. Its graphic representation is shown in figure 5.2. The figure shows wind availability (speed above 4.0 m/s) from 8 or 9 a.m. until 5 to 7 p.m. on a consistent basis. Lower speeds are measured from August to December. Higher speeds are measured from December to April and then again during June and July. Figure 5.1. Stations Used to Characterize Wind Power in Colombia Source: UPME and IDEAM. Note: Station 6, Almirante Padilla Airport, Guajira; Station 12, Simón Bolívar Airport, Magdalena; Station 11, Soledad Airport, Atlántico; and Station 1, Galarezamba, Bolívar. Wind Energy in Colombia 21 Figure 5.2. Almirante Padilla Airport, Guajira Source: UPME and IDEAM. Data collected at other coastal sites along the Caribbean coast of Colombia were also analyzed (figure 5.3). The trade winds follow Colombia's northern coast from the northeast to the west during most of the year. This general circulation pattern remains year around, with changes in intensity (wind speeds). In all cases, wind intensity peaks between February and March. This is indicated in table 5.2. Table 5.2. Wind Speed as a Fraction of Mean Yearly Wind Speeds Month 1 2 3 4 5 6 7 8 9 10 11 12 Load Peak 1.27 1.38 1.34 1.15 1.00 0.88 0.96 0.88 0.61 0.69 0.81 1.04 Med 1.32 1.36 1.34 1.17 0.94 0.87 0.93 0.85 0.67 0.69 0.81 1.04 Low 1.36 1.39 1.26 1.13 0.99 0.81 0.90 0.81 0.69 0.75 0.88 1.04 W speed avg 7.78 8.05 7.77 6.80 5.60 5.01 5.42 4.94 3.95 4.15 4.86 6.13 Ratio to annual avg 1.33 1.37 1.32 1.16 0.95 0.85 0.92 0.84 0.67 0.71 0.83 1.04 Source: Authors' data. 22 World Bank Study On average, wind speed at 8 p.m. is above the annual average by 11 percent, and during the "dry" months of December to April the wind speeds are 37 percent above the annual average, a large increase given the fact that the power of wind energy is proportional to the cube of the wind speed. Wind power is available when its contribution to the national grid is most needed, that is, during the dry periods and to an extent during the afternoon when demand peaks. Figure 5.3 presents the wind conditions in three wind measuring stations. Figure 5.3. Graphic Representation of Wind Conditions in Northern Colombia Galerazamba, Bolívar E. Cortizzos Airport, Atlántico S. Bolivar Airport, Magdalena Wind intensities above five m/s are Wind speeds above 5 m/s are From 2 p.m. until 8 p.m., this observed year around in the observed in the afternoon, with station exhibits wind speeds above afternoon, with higher values-- average values close to 8 m/s. 5 m/s, with average values near 8 above 8 m/s--observed during the These strong winds are observed m/s for the months of January first three months of the year. Wind until early morning, especially through April. Winds are direction is predominantly from the during the first four months. Wind predominantly from the north. northeast. direction is mostly from the northeast, although winds from the north are significant. Source: UPME and IDEAM. Note: E. Cortizzos Airport, Atlántico, and S. Bolívar Airport, Magdalena stations are strongly affected by the Sierra Nevada de Santa Marta, which interrupts the wind flow to the stations (for which reason the winds blow predominantly from the north). Wind Energy in Colombia 23 Complementarity during El Niño Southern Oscillation (ENSO) Events Colombia's interconnected hydro based system is severely affected by large scale droughts. Historically, critical drought conditions are linked to El Niño events, such as those of 1991­1992 and 2002­2003. Table 5.3 shows the period of occurrence of El Niño events and their length. Thus, a key element for this analysis is whether there are complementarities between wind and hydropower during dry periods. Based on existing power generation data from Jepírachi (for the period from February 2004 to March 2009) and wind velocity records data from Puerto Bolívar, wind and generation data were extended to cover the period from 1985 to 2008. For the El Niño periods, the wind data were normalized so that positive values indicate above average conditions measured in standard deviations, and negative values indicate below average conditions. Table 5.3. El Niño Periods Start Jul-51 Mar-57 Jun-63 May-65 Oct-68 Aug-69 Apr-72 Aug-76 Aug-77 Finish Jan-52 Jul-58 Feb-64 May-66 Jun-69 Feb-70 Feb-73 Mar-77 Feb-78 Months 6 14 8 13 8 6 10 7 6 Start Apr-82 Jul-86 Apr-91 Feb-93 Mar-94 Apr-97 Apr-02 Jan-04 Aug-06 Finish Jul-83 Mar-88 Jul-92 Aug-93 Apr-95 May-98 Apr-03 Mar-05 Feb-07 Months 15 20 15 6 13 13 12 8 6 Source: IDEAM. A similar analysis was conducted for four rivers with hydropower development: Guavio, Nare, Cauca, and Magdalena. Results show negative values for the four rivers during most El Niño occurrences, while the Jepírachi generation resulted mostly in positive values. The most severe basin response corresponds to El Niño from April 1991 to July 1992 when energy rationing occurred, and from April 1997 to May 1998 when pool prices reached very high spot prices, forcing regulatory changes in the market. During these periods of extreme drought, the hydrology of the country was severely affected, resulting in a reduction of mean reservoir capacities, and the system had to rely on alternative generation capacity provided through the use of thermal capacity. During these periods the estimated generation from Jepírachi is well above the mean value. That is, during periods of extreme drought associated with El Niño phenomena, wind energy from northern Colombia is above average, emphasizing the possible role of wind power during these critical periods. This analysis is described in a separate report which can be found from Appendix 6. Table 5.4 shows that El Niño periods have historically lasted between 6 and 20 months; on average in the 1951­2006 period they have lasted 10.5 months. 24 World Bank Study Table 5.4. Wind and Hydro Complementary during El Niño ANALYSIS OF EL NIÑO OCCURRENCES Departure from mean value expressed as number of standard deviations El Niño occurrences Jul. 86 Abr. 91 Feb. 93 Mar. 94 Abr. 97 Abr. 02 Jun. 04 Ago. 06 Mar. 88 Jul. 92 Ago. 93 Abr. 95 May. 98 Abr. 03 Mar. 05 Feb. 07 Guavio River 1.03 0 .5 3 0.64 1.50 0.87 0.6 6 0.94 1.02 Nare River 0.73 1.39 0.71 0.64 1.86 0.90 0 .6 8 0.08 Cauca River 1.48 1.14 0.17 0.48 1.53 1.52 0.07 0.90 Magdalena River 0.51 1 .0 7 0.00 0.80 1.69 1.08 0.81 0.52 Jepirachi Powerplant 1.23 1.20 0.20 1.23 0 .56 1 .1 9 0.91 0.80 Source: Consultants' study (see Appendix 6). Wind and Hydro Generation Complementarity Complementarity was also explored through an analysis of the joint operation of a simple system consisting of a wind farm that operates with a hydropower plant of similar size for each of the rivers studied and a range of reservoir sizes. The results for each of the rivers are described in Appendix 6. Table 5.5 below presents the results from the joint analysis of Jepírachi and the Nare River. These results are similar to those found when Jepírachi is combined with the other rivers. The firm energy from the isolated operation of the hydropower plant and the wind farm is far below the firm energy resulting from their joint operation. This result holds for the wide range of possible reservoir sizes studied. It is therefore concluded that the joint operation of wind and hydropower plants exhibits a strong complementarity, which is not rewarded in the current regulatory system adopted by Colombia. Table 5.5. Complementarity of Joint Operation of Hydro Plant and Wind Farm; the Case of the Nare River FIRM ENERGY FOR NARE AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Nare 0 0.2 0.4 0.6 0.8 1 Nare River(isolated) 0.179 0.369 0.435 0.459 0.471 0.480 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Nare River + Jepirachi in isolated operation 0.268 0.458 0.524 0.548 0.560 0.569 Nare River + Jepirachi in joint operation 0.410 0.811 0.943 0.972 0.994 1.009 Source: Consultants' study (see Appendix 6). Wind Energy in Colombia 25 Firm Energy and Joint Operation of Wind and Hydroelectric Projects An analysis was conducted to understand the firm energy obtained from hydroelectric plants (with and without reservoir) in conjunction with the Jepírachi power plant under scenarios of joint and isolated operation (Colombian regulation estimates the reliability of individual power plants and does not consider joint operation). Firm energy is defined as the maximum monthly energy that can be produced without deficits during the analysis period which would include El Niño occurrences. The same results were obtained for the total energy obtained from the joint operation of the hydropower plants and the Jepírachi plant. The analysis was conducted using a simulation model that operates the plants and the reservoirs to provide a given energy target, adjusting this target until no deficits are generated. For this purpose, hypothetical hydroelectric plants with capacity similar to that of wind power plants were analyzed. Mean multiannual inflow to the hydroelectric power plants (expressed in energy) at the plant sites is equal to the same value for Jepírachi generation. This was done by multiplying river discharges by a factor to convert them to energy such that mean inflows are equal to mean Jepírachi generation. In order to avoid confusion with existing hydroelectric plants, the hypothetical plants analyzed will be named Guavio River, Nare River, Cauca River, and Magdalena River. Several reservoir sizes were analyzed; reservoir size (expressed as a fraction of mean annual inflow to the reservoir in energy) varies between 0 (run of river plant) to 1 (substantial regulation capacity). Results are shown below. An Example: The Guavio River Table 5.6 and figure 5.4 show results for the Guavio River. Firm energy has been normalized, with actual firm energy divided by the sum of mean energy for the Guavio River and Jepírachi. Table 5.6. Firm Energy Results for Guavio River Analyzed in Isolated and Joint Operation FIRM ENERGY FOR GUAVIO AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Guavio 0 0.2 0.4 0.6 0.8 1 Guavio River (isolated) 0.064 0.334 0.451 0.481 0.507 0.514 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Guavio River + Jepirachi in isolated operation 0.153 0.423 0.540 0.570 0.596 0.602 Guavio River + Jepirachi in joint operation 0.212 0.709 0.860 0.908 0.935 0.962 Source: Consultants' study (see Appendix 6). 26 World Bank Study Figure 5.4. Firm Energy for Guavio River as a Result of Isolated and Joint Operation 1.200 1.000 Firm energy/Mean energy Guavio R. + 0.800 Jepirachi (joint) 0.600 Guavio R. + Jepirachi (isolated) 0.400 Guavio R.(isolated) 0.200 Jepirachi (isolated) 0.000 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reservoir size/Mean annual energy Source: Consultants' study (see Appendix 6). In this case, the firm energy that results from the joint operation of the wind farm and the hypothetical hydropower plant is greater than the sum of the isolated operation of the two individual projects. Table 5.6 and figure 5.4 indicate an increase in firm energy when joint operation is considered. This is because critical periods for the Guavio River do not coincide with Jepírachi generation during the same period. Figures 5.5 and 5.6, showing reservoir operation both in isolated and joint operation, illustrate this in greater detail. Figure 5.5, corresponding to a reservoir size of 0.2, shows that in isolated operation the reservoir is emptied during the El Niño occurrence of April 1997­May 1998, while in joint operation the reservoir is emptied in April 2001. The El Niño occurrence of April 1997­April 1998 is balanced by large scale generation in the Jepírachi power plant, showing the complementarity of river discharges in the Guavio River and wind generation in the Jepírachi power plant. The analysis is also performed for the Nare and Magdalena Rivers and the results are similar to those presented here for the Guavio River (that is, in joint operation the firm energy is greater than in isolated operation). For purposes of simplification, only the Guavio River example, with a reservoir size of 0.2 and 0.5, is shown. Wind Energy in Colombia 27 Figure 5.5. Guavio River Reservoir Operation with a Reservoir Size of 0.2 in Isolated and Joint Operation 1.200 Reservoir volume/Resrvoir capacity 1.000 0.800 0.600 Joint 0.400 Isolated 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Consultants' study (see Appendix 6). Notes: 0=run of river plant to 1=substantial regulation capacity. The bars represent El Niño occurrences. Figure 5.6. Guavio River Reservoir Operation with a Reservoir Size of 0.5 in Isolated and Joint Operation 1.200 Reservoir volume/Resrvoir capacity 1.000 Joint 0.800 0.600 Isolated 0.400 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Consultants' study (see Appendix 6). Notes: 0=run of river plant to 1=substantial regulation capacity. The bars represent El Niño occurrences. Impact of Extreme Events on Hydropower Capacity Although there is still no consensus on how climate change may affect average precipitation in Colombia, there is a generally accepted notion that global warming 28 World Bank Study will result not only in changes in mean conditions but also in increases in the extent and frequency of extreme precipitation events. Changes in extremes would have an impact on the country's hydrological regime. Appendix 2 presents a summary of the results of an analysis conducted with the use of runoff data, derived from rainfall projections by the Earth Simulator to estimate the likelihood of extreme weather events around the end of the century (2090). This would result in an increase in stream flow during the high flow season and a decrease in the low flow season. The annual range of stream flow becomes larger, implying more floods in the wet season and droughts in the dry season. The anticipated changes in surface hydrology will affect hydropower potential by reducing the potential firm capacity of reservoirs. Notes 1 Jepírachi is a small wind farm, with 19.6 MW of nominal capacity, located in Northern Guajira, owned by EPM and in operation since 2004. 2 Note that the capacity factor of Jepírachi during the 2004­2008 period was lower than expected, nearly 32 percent. Communication with the wind farm's owners reveals that some wind turbines were turned off for maintenance and that there were periods (normally between midnight and 6 a.m.) in which the wind farm did not generate due to tension imbalances in the transmission lines to which the wind farm is connected. These issues have now been resolved but it is believed that without these issues the capacity factor for Jepírachi could have been higher than that experienced. 3 Data are from the Neon database with historical operation data created by Expertos en Mercados S.A. E.S.P. (XM), the Colombian hydrothermal system operator. 4 The information was compiled and published as a joint effort by Unidad de Planeamineto Minero Energética (UPME) and Institute of Hydrology, Meteorology and Environmental Studies of Colombia (IDEAM), part of the Ministry of Environment, Housing, and Territorial Development. CHAPTER 6 Options to Aid Market Entry of Wind Energy in the Country's Power Mix Introduction Under current circumstances wind based generation faces considerable obstacles to participate in the nation's power mix. Key obstacles, as described in the first phase report, include the current relatively high capital intensity and the structure of the regulatory system which does not acknowledge wind power's potential firm capacity.1 However, the wind resource along the northern coast appears to complement well the country's hydrological regime and could be part of a strategy to strengthen the climate resilience of the hydropower based sector. To promote wind power generation, the actions required would result in a positive impact on the financial performance of projects while minimizing distortions in the existing power market and the overall economy. This chapter reviews the typology of available options to address the higher capital cost of the wind option as well as an option to address the variable nature of wind energy. This chapter follows a microeconomics approach. The analysis is focused on potential investors as the key economic agents sought by the GOC. These investors base their investment decisions on important regulatory and financial aspects. This chapter describes the tools available to guide the market and the tools for government intervention in guiding the independent investors' decisions, while chapter 7 describes the financial analysis upon which the effectiveness of such tools is assessed. The interpretation of results provides guidance to decision makers on regulatory work. Options to Facilitate Market Entry of Wind Energy A number of options could be used to facilitate the market entry of wind power in Colombia. This chapter describes a typology of policy instruments, out of which a selection is made for further use in the analysis. The options are categorized in four groups: (i) price based policy instruments; (ii) policy options guiding renewable energy output (quantity based policy instruments); (iii) adjustments in the regulatory system; and (iv) instruments that provide incentives other than price. In addition, a proposal is detailed, providing a simple methodology to assess the contribution of wind powered plants to firm energy, an opening through which wind farms could be recipients of reliability payments. 29 30 World Bank Study Price-Based Policy Instruments Although many practitioners find these instruments very effective in promoting RET, their implementation may generate financial distortions. These instruments or policy tools have so far not been considered in Colombia, nor are they favored by market players and policy makers, because the country's generation requirements are currently being met by independent power producers without the need for government financing or intervention. Feed in tariff system or price based instrument. This approach forces utility companies to purchase all the electricity produced by renewable energy producers in their service area at a tariff determined by the authorities and guaranteed for a specific period of time (typically 10 to 20 years). Feed in tariffs offer a financial incentive for renewable project developers to exploit all available generating sites until the marginal cost of producing energy equals the proposed feed in tariff. Costs are recovered through a levy on all electricity consumers who purchase power from utilities. Fixed premium system (environmental kWh premium). This price based mechanism adds a fixed premium to the basic wholesale electricity price, making the total price received per kWh produced less predictable than in the feed in tariff described above. Valuing carbon emissions. Valuing carbon emissions could be achieved by taxing power plants' emissions of pollutants in accordance with standard principles of tax policy, or by imposing a discriminatory sales tax on electricity generated by polluting fossil fuels and using the revenue to pay a premium to generators that utilize nonpolluting renewable energy sources. Production tax credits. A production tax credit provides the investor or owner of a qualifying generating facility an annual tax credit based on the amount of electricity generated by that facility to encourage improved operating performance. Policy Options Guiding Renewable Energy Output (Quantity-Based Policy Instruments) Renewable energy mix targets. This instrument establishes a minimum percentage of renewable energy as part of the national energy portfolio. Electric utilities are required to procure a certain quantity of their electricity from renewable technologies as a percentage of the total or to install a certain capacity of renewable power. The renewable based generation increases with the overall increase in electricity demand. Producers could then decide either to implement the projects themselves or to put them out to tender from independent power producers. Suppliers may also choose competitive bidding from independent power producers and participate in green certification systems. However, inadequate administrative capacity for verification mechanisms, record keeping for transactions, and compliance may complicate their implementation. Several countries have adopted or are proposing national renewable energy targets. The European Union has collectively adopted a target of 22 percent of total electricity generation from renewables by 2010, with individual member states selecting their own targets. Japan has adopted a target of 3 percent of total primary energy by 2010. Recent legislative proposals in the United States would require 10 percent of electricity generation from renewables by 2020. Competitively awarded subsidies. Competitively awarded subsidies, that is, through auctions, could be offered to promote certain technologies and achieve Wind Energy in Colombia 31 predefined output targets. In Poland, the World Bank's Global Environment Facility (GEF) helped to develop markets and reduce costs for products through subsidies given to technically qualified domestic manufacturers. Adjustments in the Regulatory System Exemption from systems charges. Colombia has an unbundled electricity market. The concept of unbundling--separately pricing all of the services that comprise a utility service--could be a disadvantage for producers of nonconventional power when they have to pay transmission charges on a per capacity basis. Some countries, such as Brazil, have experimented with reducing prices of transmission wheeling for producers of renewable energy. To this end, exemption from systems charges could be implemented, exempting renewables from generation surcharges and considering these alternatives as load reduction technologies. For the Colombian system several policy instruments could be devised under this heading to encourage new renewable plants: waiving the charges paid for automatic generation control; elimination or reduction of environmental charges and/or contributions for the electrification of off grid regions; and excluding new renewable power plants from CERE payment obligations. Adjusting the "reliability payment" regulation. Colombia has developed a financial mechanism to produce an economic signal to investors as a price premium on reliable installed power capacity. This instrument aims at increasing the resilience ("firmness) of the national interconnected system to extreme weather events, especially during unusually dry periods. The reliability payment, or firm capacity charge, should promote an efficient mix of energy sources, without discriminating renewable sources. Unfortunately, the existing regulation does not have clear rules to assess the potential contribution of wind energy to the overall reliability of the interconnected system and thus favors conventional power plants. In practice this discriminatory treatment has been identified as a major barrier to further investments in the wind sector.2 Fortunately, however, it is straightforward enough to include all resources in a nondiscriminatory manner. All that is required is an objective method of estimating the firm energy capacity of the resource. The issue of reliability payment is analyzed in detail below. Policy Instruments that Provide Incentives other than Price These policy tools provide incentives for voluntary investments in renewable energy by waiving taxes and/or reducing the costs of investments through financial mechanisms. There are at least five broad categories of instruments that (i) reduce capital costs after purchase (through tax relief) or offset costs through a stream of payments based on power production (through production tax credits); (ii) reduce investment costs up front (through credits, subsidies, and rebates); (iii) provide public financing or public facilitation through concessionary loans, grants, and other financial assistance; and (iv) reduce capital and installation costs through economies of bulk procurement (Valencia 2008). The following policy instruments are applicable in the case of Colombia. Property tax incentives. These incentives are generally implemented in one of three ways: (i) renewable energy property is partially or fully excluded from property tax assessment, (ii) renewable energy property value is capped at the value of an 32 World Bank Study equivalent conventional energy system that provides the same service, or (iii) tax credits are awarded to offset property taxes. Experts have long argued in favor of imposing corporate and sales taxes on electricity on the grounds that it is a fairly price inelastic product. Reduction or elimination of import duties. Much of the equipment for renewable generation must be imported to host countries. High capital import duties and tariffs distort the market, artificially raising the price of renewable technologies and discouraging their adoption. Temporary or permanent waivers may contribute to reduce the impact of high initial investment costs and allow renewable technologies to compete in the market. Such waivers may be justified either on the basis that renewables are a pioneer (or start up) industry or on the basis that payment of such duties and tariffs by a generating company ultimately would have been passed on to the final consumer. Tax exemptions encourage investment. Financing of renewable energies. These may include: imposing a surcharge on electricity consumption, to be collected in a special purpose fund for renewable energy support (in which case larger consumers bear most of the burden); providing a tax credit to be assigned at the local and central levels on renewable energy produced; and taxing pollution, which raises the incremental cost of thermal generation and decreases the cost of competing renewable energy, as mentioned above. Other options could include a change in culture in which consumers would be willing to pay more for "green" electricity. Mexico has established a green fund to promote renewable energy. In this case a tax is collected from all power services and goes into a fund to support renewable energy projects. Grants and low cost loans. Many countries have offered grants for renewable energy purchases. In some developing countries, notably China, India, and Sri Lanka, multilateral loans by lenders such as the World Bank have provided financing for renewable energy, usually in conjunction with commercial lending (Valencia 2008). The newly established CTF falls into this option. Proposal to Address the Reliability Issue for Wind Energy As briefly explained earlier, the Colombian electricity market includes a reliability payment for each resource based on its ability to generate energy during unusually dry periods; this is called "firm energy." The product needed for reliability in Colombia's hydro dominated electricity market was introduced in Colombia to minimize the probability of brownouts and blackouts in the interconnected grid as a consequence of hydrological variability. This firm energy is expected to meet user demand under critical conditions (when the wholesale market price is larger than the scarcity price3). This is found in CREG Resolution 071 2006. In 2008, Colombia introduced an innovative and effective market in which auctions4 are held to commit enough firm energy to cover its needs (Cramton and Stoft 2007, 2008).5 The firm energy market coordinates investment in new resources to assure that sufficient firm energy is available in dry periods. The firm energy product includes both a financial call option and the physical capability to supply firm energy. The physical capability assures that there will be sufficient energy during dry periods. The call option protects load from high spot prices and improves the performance of the spot market during scarcity. Wind Energy in Colombia 33 To promote an efficient mix of resources and for the firm energy market to succeed in providing reliable electricity at least cost, all resources, including variable resources such as wind power, should be eligible to receive the same reliability payment based upon the resource's ability to provide firm energy. Including wind power and other variable resources in the firm energy market has three important benefits for Colombia. First, it leads to a more efficient mix of resources and thereby could eventually reduce electricity costs. Second, it reduces risk by establishing a more diversified portfolio of fuel types. Third, it reduces Colombia's reliance on coal and other fossil fuels to generate electricity during dry periods, thereby reducing Colombia's emissions from fossil fuels. At present, the economic signal favors conventional power plants, but fortunately, it is straightforward enough to include all resources in a nondiscriminatory manner. The key input required in the firm energy auction is an estimate of the resource's ability to supply firm energy. This is already done for all hydro and thermal resources. What is required is an analogous methodology to estimate firm energy for variable resources. For purposes of simplicity, the analysis focuses on wind power as a variable resource, but the same approach applies to all variable resources--all resources of any type. In many respects, wind power is actually simpler than hydro or thermal, since it is straightforward enough to estimate the energy output of the wind resource. This is a step already taken as part of the due diligence for any wind project. For hydro resources, the regulator estimates the firm energy of a hydro project using a time series of hydrological data, ideally five or more decades. For thermal resources, the firm energy rating is based on the unit's nameplate capacity, which is then reduced based on sustainable utilization rates. Estimating the firm energy of a wind resource is similar to that of a hydro resource, although it is suggested that a much shorter time series (perhaps initially based on Jepírachi's five year record of operation) should be sufficient to determine a good estimate of firm energy capability. Such a series would be produced as part of the standard due diligence of an investor in a wind power project. No investor would build a wind project without first having a fairly good idea of the project's average energy output. Even if this initial estimate is biased, there is little economic harm, since as described below the rating would be adjusted so that it reflects the project's long run performance, which is measured automatically by the system operator. As with other resources, the firm energy rating should be updated based on actual performance. This is difficult for hydro resources given the low frequency of unusually dry periods, roughly once every 10 years. Wind power does not face this problem. The operation of wind farms generates meaningful data on firm energy that integrates local site specific wind conditions with turbine efficiency. For this reason, it would make sense to have a periodic (yearly) automatic adjustment to the firm energy rating of wind resources based upon historical performance. For purposes of simplicity, it is recommended that the firm energy rating of a wind resource be adjusted annually based on the following exponential smoothing formula: firm energy rating in year t + 1 = ½ (firm energy rating in year t) + ½ (energy produced in year t). 34 World Bank Study The initial period for locating wind plants along the northern coast could use the five year period recorded by Jepírachi, to be updated annually thereafter. This simple approach assures that the firm energy rating of wind power closely tracks its actual performance. The key assumption in the formula is that wind power is not correlated with dry periods; that is, wind resources on average generate the same amount of energy in unusually dry periods as in normal periods. If the seasonality for wind power is correlated with dry seasons, then it would make sense to modify the formula above by replacing "energy produced in year t" with "energy produced in dry season of year t" and then scale up the level of output to an annual measure by multiplying by 12/(number of months in the dry season). Under this simple approach, the firm energy rating and therefore the reliability payment will quickly converge to the long run average firm energy capability, even if the firm energy rating in the initial year is poorly measured. An exercise was conducted to calculate the results of the firm capacity factor for the Jepírachi wind farm in Colombia, using the method proposed above. The analysis is based on observed wind data recorded at meteorological stations in northern Colombia. These data, together with generation data from Jepírachi, allowed the reconstruction of a 24 year data series on monthly wind data and generation. This database was then used to estimate the corresponding firm energy rating in Jepírachi. On average, the yearly firm energy rating was estimated at 0.38, with a range between 0.25 and 0.47.6 For the dry season, the average firm energy factor found was 0.4 (with an initial year rating of 0.37 and a maximum firm energy factor of 0.47). When this firm energy rating is acknowledged for the entire year, the project owners could receive an annual average of US$975,000 from the reliability payment, based on the auction defined value of US$13.9 per MWh. This of course translates into very attractive earnings, especially when the lifetime of the project is taken into consideration. For the 24 year time series considered here, this could mean total project earnings of US$23.4 million. The suggested approach to assess the reliability factor for wind farms is risk neutral. If the yearly estimate is used during the "dry period," the difference between the annual mean and the dry period mean could be interpreted as a risk reduction strategy. A more formal option, in tune with the general risk aversion characteristic of Colombia's regulatory framework, is to subtract standard deviation affected by some factor of the historical performance. Importantly, for wind power the call option portion of the firm energy product is the same as the call option for thermal resources. During scarcity periods in which the spot price exceeds the scarcity price, the wind resource has an obligation to generate energy over the day consistent with the resource's firm energy rating. Deviations from this daily obligation are resolved at the spot energy price. As a variable resource, the energy output of the unit will surely differ from the obligation on any particular day, but over the course of many days the unit should produce an amount roughly equal to its firm energy rating. Thus, the resource should meet its obligation on average, and if it does so, then its net payment for deviations would be approximately zero. Wind Energy in Colombia 35 Notes 1 Note that the firm capacity of renewable energy is the capacity of conventional sources replaced, so that demands can be met with a specified reliability. The firm capacity of a renewable source depends on the correlated variations in demands and renewable supplies (Barrett 2007). 2 For a thorough discussion of the effects, advantages and disadvantages of, and barriers to distributed generation, see COLCIENCIAS, ISAGEN, Universidad Nacional, and Universidad de los Andes 2006. 3 The scarcity price is determined by CREG and updated monthly, determining the wholesale market price from which firm energy obligations become mandatory and establishing the maximum price at which this energy is remunerated. 4 The firm energy auction under the reliability payment (cargo por confiabilidad) is a scheme that establishes long term commitments and is expected to be a component of the wholesale energy market indefinitely. The auctions are held during various years prior to firm energy obligations (time is provided between auctions and the start of firm energy obligations to allow new projects to be able to enter into operation). To this end, each year the Regulatory Commission (CREG) evaluates the balance of supply and demand of the firm energy projections and if necessary calls for an auction (XM 2009). Available online at: www.xm.com.co/Informes%20Empresariales/ InformeAnual_XM.pdf. The next firm energy auction has not been scheduled. 5 It is worth noting that although the reliability payment has been successful in getting projects registered and assigned to provide firm energy, many of the projects that participated in the firm energy auction lacked an environmental assessment of alternative projects (UPME 2009). This can lead to system, environmental, and investor risks (for example, if it is later found that the projects cannot be implemented due to more environmentally friendly alternatives). However, it is important to keep in perspective the lessons from similar cases in other countries where hydropower projects are waiting in the pipeline and are being replaced by coal power projects because it takes a long time to produce the environmental licenses of hydropower projects. This, of course, may lead to dire and unintended consequences. For this reason, to avoid the possible risk described above, it is recommended that there be high level coordination among ministries and expedited action by the Ministry of Environment to review environmental licenses (including a review of possible alternatives). 6 Lower values are associated with the start of the project and the technical learning curve of the operating agency. The range highlights the variability of the wind field. 7 As stated previously, even if the firm energy rating in the initial year is poorly measured, the initial firm energy rating (and therefore the reliability payment) of 30 percent will quickly converge to the long run average firm energy capability. CHAPTER 7 Assessing the Effectiveness of Policy Instruments and Policy Options: Impact on a 300 MW Wind Powered Power Plant Operating in the Wholesale Energy Market T his chapter aims at exploring the effectiveness of alternative policy instruments in facilitating market entry of the wind option. The consequences of the alternative instruments are measured in terms of the financial result expected by potential investors. A hypothetical 300 MW wind power project is used to estimate the impacts from the different alternatives. Wind resources were defined using historical records and data from Jepírachi. Performance and operational data are based on this pilot wind farm. (Details are available upon request.) Scenarios of the expected price energy production response of the Colombian wholesale energy market (MEM) from 2008 to 2025 are used. This step is both a necessary input for assessing the financial sustainability of the wind project and a useful methodology to help evaluate other projects. These estimates rely on UPME's July 2008 forecasts for the national energy market, and include the analyses of demand forecasts, natural gas prices, and the expected optimal (minimum cost) generation expansion adjusted to include the characteristics of the Colombian transmission grid. For the purpose of assessing the attractiveness of the wind farm investment through its financial return, the study kept the value of the reliability payment for plant energy remuneration constant at US$13.05/MWh up to November 2012, and then increased this to US$14.00/MWh through the planning horizon.1 The following chapters summarize the analyses made, relegating the more detailed technical studies to technical appendixes and supporting documentation. This section concludes with an examination of the options available to the government for the promotion of increased RET participation in the country's energy mix. 36 Wind Energy in Colombia 37 Baseline Information Domestic Demand Forecasts As stated above, demand forecasts for the National Interconnected System (NIS) were obtained from UPME's latest forecasts dated July 2008 (figure 7.1), before the global financial crisis ensued, and thus may be currently characterized as somewhat optimistic. Figure 7.1. Colombia NIS Demand Forecasts, 2007­2028 120,000 100,000 80,000 GWH 60,000 40,000 20,000 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 LOW 52,851 53,882 55,398 57,435 59,392 61,261 63,221 65,274 67,318 69,338 71,415 73,498 75,693 78,067 80 ,066 82,109 84,393 86,727 89,016 MEAN 52,851 54,086 56,060 58,567 60,907 63,313 65,754 68,279 70,897 73,611 76,372 79,297 82,386 85,218 88,136 91,377 94,203 97,006 100,128 HIGH 52,851 54,290 56,608 59,247 62,100 65,027 68,01 3 71,142 74,466 77,953 81,954 85,358 89,471 92,821 96,279 100,35 104,02 107,58 111,556 Low Mean High Source: UPME, July 2008. Wind Project Generation Based on the MEM projections, figure 7.2 shows the estimated monthly values for wind power generation, including average, low (P10), and high (P90) estimates.2 Wind conditions are average conditions, estimated based on the existing Jepírachi records. Figure 7.2. Wind Project Generation Estimates 2012­2025 RET PROJECT GENERATION (GWH) 250 200 (GWH) 150 Mean 100 P10 P90 50 0 3 6 9 2 5 2 4 5 7 8 0 1 3 4 12 15 18 21 24 l-1 l-1 l-1 l-2 l-2 t-1 r-1 t-1 r-1 t-1 r-2 t-2 r-2 t-2 n- n- n- n- n- Ju Ju Ju Ju Ju Oc Oc Oc Oc Oc Ap Ap Ap Ap Ja Ja Ja Ja Ja Source: Authors' data. 38 World Bank Study Pool Prices Pool prices in the wholesale market are formed by adding other variable costs (CERE, FAZNI, environmental, and Automatic Generation Control [AGC]) to the pure marginal cost. This is presented for the mean case scenario in figure 7.3 (the other scenarios analyzed are included in Appendix 3). Pool price comparisons of the mean, high and low scenarios are presented in figure 7.4. Figure 7.3. Pool Prices, Base Scenario POOL PRICES--Mean Scenario 120 100 80 US$/Mwh Mean 60 P10 40 P90 20 0 8 Ma 9 Ap 0 1 Fe 2 Ja 3 De 4 No 4 Oc 5 6 17 18 9 20 Ap 1 2 Fe 3 Ja 4 De 5 5 r-1 l-0 0 y-1 r-1 1 1 c-1 v-1 t-1 l-1 y-2 r-2 r-2 2 2 c-2 n- b- n- p- g- n- b- n- Ju Ju Ma Ma Ma Ju Se Au Ju Source: Authors' data. Figure 7.4. Comparison of Pool Prices for Base, High, and Low Scenarios AVERAGE POOL PRICES--Mean, High and Low Scenarios 140 120 100 US$/MWh 80 60 Mean 40 High Low 20 0 Fe 8 Se 09 Ap 09 No 10 Ju 0 Ja 11 Au 12 Ma 12 Oc 13 M a 13 De 4 Ju 4 Fe 5 Se 16 Ap 1 6 No 17 Ju 7 J a 18 Au 19 Ma 19 Oc 20 Ma 20 De 1 Ju 21 Fe 2 Se 23 Ap 23 No 24 Ju 4 25 l-0 v-1 y-1 c-1 l-1 v-1 y-2 l-2 v-2 n- b- p- r- n- g- r- t- b- p- r- n- n- g- r- t- c- b- p- r- n- Ju Source: Authors' data. Annual NIS Balances This analysis also projects annual energy balances for the NIS under the four scenarios considered. These projections show the magnitude of the effect of reduced hydrology generation versus official expected hydrology generation, with the corresponding increase in the gap to be met by alternative means, that is, thermal generation. (These balances can be found in Appendix 4, tables A4.1 to A4.4.) Wind Energy in Colombia 39 Baseline Results A threshold of 14 percent Internal Rate of Return (IRR) was used to indicate adequate return to potential investors based on experience with previous operations (Amoyá, Jepírachi) and on a comparison with international markets. Three scenarios were used to define the overall energy demand and its relation to fuel prices. The outcomes of these scenarios determine for an "investment project" the set of prices that the investor might expect. The overall indicative prices range from US$39.41/MWh for the base high hydro scenario (see table 7.1) to US$66.70/MWh for the high demand­high fuel prices scenario. The baseline scenario has an indicative price of US$50.60/MWh. Although all the cases were analyzed for all the basic scenarios, the presentation will focus on the baseline conditions. Table 7.1. Demand Scenarios for the Interconnected Grid and Resulting Indicative Prices Indicative price a Scenario Demand Fuel prices Hydro (US$/MWh) Low Low Low Revised 43.3 Baseline Base Base Revised b 50.6 High High High Revised 66.7 Source: Authors' data. a. Indicative average energy price over the 2007 to 2028 simulation period. b. Energy production factor for hydropower plants estimated from historical generation records from the NEON database. Table 7.2 presents the results obtained for the baseline analysis. The expected returns on equity are shown for each of the general scenarios considered for Colombia's interconnected system. In addition, given the importance of the investment costs in the policy analysis and in the financial returns, table 7.2 presents results for a wide range of unit investments (expressed in US$/kW). These results indicate that returns are sensitive to the general growth scenario and the general economic environment. Because investments in the power sector are long term, average conditions should be expected to dominate. The selected baseline scenario provides a conservative picture of potential returns, although with a medium risk. Table 7.2. Expected Returns on Equity before Taxes for a 300 MW Wind Farm in Colombia--Business-as-Usual Results (no government intervention) Capital cost per kW installed National Base Scenarios $2,400 $2,100 $1,800 Low 4.7% 6.2% 8.1% Medium/Baseline 5.8% 7.6% 9.9% High 9.2% 11.5% 14.8% Source: Authors' data. Note: The results assume access to Carbon Emission Reductions of US$18/tCO2. 40 World Bank Study As the unit investment costs decrease, the return increases should be expected. Nevertheless, it should be emphasized that under the business as usual scenario--that is, without policy intervention--wind energy investments are not attractive to potential investors. Thus, if the GOC aims to increase the proportion of its electricity from renewable sources, it is required to adopt policies to aid market entry of RET by creating the enabling environment for independent investors to develop nonconventional renewable source power projects. Impact of Selected Policy Options Not all of the available policy instruments are applicable to the case of Colombia. A selection was therefore made, considering those that would fit the regulatory framework and that focus on actions that would not distort the wholesale market. In order to assess the effectiveness of the options, the financial results of their deployment are quantified. The assessment of financial results from different options assists in the selection of policy instruments and the adoption of a coherent set of alternatives that individually or jointly accomplish the desired results for the potential investors. Selected Policy Options The options were grouped under common policy themes: Group I. Access international financial instruments to internalize global externalities in national and private decisions. The government can play a leading and active role in accessing bilateral and multilateral financial instruments aimed at reducing GHG emissions, such as the CDM (this instrument is already mainstreamed into Colombia's environmental policy). This would be complemented through: The government acting as a bridge to attract multilateral soft loans earmarked for alternative energies; and The government facilitating access to clean technology concessionary financing. Group II. Target subsidies and government fiscal mechanisms. Under this group of policy options the government uses fiscal measures for the benefit of potential investors. Specifically, the mechanisms identified include tax subsidies and waiving of dispatch control charges (like AGC). Group III. Reform the regulatory system. Under this policy package, the regulatory system is adjusted to be technologically neutral (creating a level playing field among technologies), and could be complemented to guide the country toward low carbon intensity development. The existing regulatory system has developed mechanisms to steer the market in order to provide a more resilient interconnected system (expressed by its capacity to deliver the demand even during the most difficult hydrological conditions). In doing so, RETs have not received adequate compensation for their contribution. This situation needs to be adjusted and new tools could be included to give greater flexibility to the government in fostering RET. This includes: Complementing the scope of the reliability charge to include RET, and wind in particular; Wind Energy in Colombia 41 Waiving payment of CERE to carbon free power options, as an extension of the existing option for small scale investments; and Creating an environmental sustainability charge (to internalize local environmental and social impacts) and support a low carbon development path. Within the Colombian energy regulatory system the CERE plays a pivotal function in fostering a more resilient interconnected system. The CERE payment (contribution by the generators) is the revenue source used to pay for the reliability payments. Each electricity generator contributes to a fund in proportion to the energy produced. At the same time each power plant receives payments from this fund, based on its contribution to the "firmness" of the system, to avert the possibility of brownouts and blackouts. If new policy options are developed, the approach followed could easily be replicated. This analysis would likely take place as the government further fine tunes its decision on how to proceed. Table 7.3 shows the institutional responsibilities associated with the selected options. For each the key implementation stakeholders are identified, their responsibilities are described, and the general source of funding, or who bears the costs, is also described. Not all options have similar implementation characteristics. Table 7.3. Policy Options, Allocation of Responsibilities and Associated Costs Policy instrument Stakeholders Responsibility Source of revenues Group I. Access to soft Min Energy Negotiations with Multilateral Pass through costs loans Min Finance Development Banks (MDB), Might impact national debt donors; national debt availability for other activities, thus competing with other allocation needs Group I. Access to Min Energy Negotiate with MDB/donors Pass through costs concessionary funds Min Finance Targeted commitments Might impact national debt (CTF) Min Planning, National availability for other activities, Allocate national debt Planning Department thus competing with other (DNP) allocation needs Group II. Waiving system Regulators (CREG) Promote and enact regulation Other wholesale market charges adjusting system charges participants Final consumer Group III. Adjust the Regulators (CREG) Promote and enact regulation Cost neutral "reliability charge" adjusting methodology to assess the "reliability factor" Group III. Waiving CERE Regulators (CREG) Promote and enact regulation Other wholesale market payment changes participants Final consumer Group III. Income tax Min Energy Might require approval by Impact on fiscal resources breaks Min Finance Congress Final consumer Min Planning (DNP) Group III. Green charge Min Environment Promote the internalization of Final consumer Min Finance environmental externalities Min Planning (DNP) Might require a new law Regulator (CREG) Source: Authors' data. 42 World Bank Study For example, access to concessionary funds might require the country to make targeted commitments as to GHG emission reductions to achieve by defined dates, as well as potential impacts on the national debt ceiling, with potential allocation conflicts with other sectors and national needs. Table 7.3 shows that in general the selected policy instruments are relative easy to implement, especially those related to adjusting the regulatory system. Results Tables 7.4a, b, and c present the calculated returns on investments resulting from the application of the policy instruments. The results are presented for a range of unit investments from US$1,800/kW to US$2,400/kW. Table 7.4a presents the results with 20 percent, table 7.4b with 30 percent, and table 7.4c with 36 percent reliability payment.3 The policy instruments used are classified in three types: financial instruments; government fees, including taxes; and regulatory instruments. The internal rate of return is calculated for the project and for equity, before and after taxes. The threshold to judge the policy effective is 14 percent. The tables are divided between two policy options: policy option A with using the reliability payment and policy option B without using the reliability payment. Both policy options use the same base and carbon revenues (US$18/tCO2) as a basis, and add one or more policy instruments when going toward the right in the columns. For purposes of simplicity, the results are summarized only for the medium case scenario. The analyses were also conducted for each market scenario, the results of which can be found in Appendix 4. The term "base" in the table indicates the status quo. Tables 7.4a, b, and c provide a summary of a possible set of policy options open to the GOC. The selection of the set of policy instruments needed depends on the expected level of investment costs associated with wind power projects in Colombia. The industry outlook seems to be that costs will decrease with time, but the reduction of costs alone does not make the wind power sector financially feasible. If wind receives reliability payments for its contribution to firm energy, the need for complementary inducements is a function of the methodology adopted to assess such contribution. If the suggested methodology is adopted, no further inducement is required. If a more risk averse estimate is used, other policy instruments are required, at least until the investment costs catch up the difference. The results indicate: All options considered improve the financial return on wind investments. Wind farms become attractive to the Colombian energy market when their unit investment costs (US$ per kW installed) are such that independent investors reach the target IRR of 14 percent. Under existing market and regulatory conditions (wind plants are not recipients of reliability payment), the investment cost threshold is estimated to be $1,250/kW. If wind farms benefit from reliability payments, the threshold unit investment cost increases as follows: for reliability factors of 20, 30, and 36 percent, the corresponding threshold unit investment costs are $1,660/kW; $1,820 /kW, and $1,880/kW, respectively. In the latter two cases, investment in wind projects becomes financially viable for existing wind technologies. Wind Energy in Colombia 43 Adjusting the reliability payment (leveling the regulatory playing field for nonconventional renewable energy technologies) is a very effective incentive. A reliability factor greater than 30 percent by itself allows wind farms to be financially feasible for low investment costs, such as those recently reported for Europe. Eliminating income taxes does not seem to be an effective instrument to attract investments to RET, given the criteria utilized to judge financial feasibility. It does not lead to a 14 percent IRR under the conditions considered. If reliability payment is not used, also eliminating fees (AGC, FAZNI, CERE) makes wind power attractive at a US$1,800/kW investment cost. Access to concessionary financing has a significant effect. This option requires clean technology concessionary funding4 for up to 40 percent of the total unit investment to reach a 14 percent IRR. As expected, the reduction in unit investment (US$2,400 versus US$1,800) improves return on investment. However, a reduction in investment costs alone falls short of reaching the 14 percent IRR target. In summary, under existing conditions wind farms are not financially attractive in Colombia even considering the drop in investment costs recorded during 2009. However, wind investments would become financially attractive if the benefits of reliability payments are extended to wind power, even under current investment costs. The government has other multiple policy instruments to steer independent investors toward RETs. Adopting several of these options, as detailed in the report, seems relatively simple and will not distort the market. Improving the conditions for market entry of the wind option will serve to prepare the sector for the anticipated improvement of conditions as investment costs for wind decrease over time. Finally, deployment of the wind option would help the sector to strengthen its climate resilience and be better prepared to face climate variability, without increasing its carbon footprint. To complement the incentive structure, the government has various instruments at its disposal. If it uses the capacity to partially waive CERE payments, the attractiveness to potential investors is increased and wind power projects could be implemented at a faster pace and for a wider set of international investment costs.5 The results for each set of policy instruments integrated into a policy option illustrate the advantages and limitations of such an approach. The GOC would do better by mixing policy options to obtain the desired results. This is the analysis introduced in the next section. 44 Table 7.4a. Financial Results for a 300 MW Wind Farm In Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 20 Percent Firm Energy Factor World Bank Study Source: Authors' data. Note: If no financing terms are mentioned, it is assumed that the investor must finance 70 percent of the project costs through commercial credits. *Income tax reduction of 15% after 2017. Table 7.4b. Financial Results for a 300 MW Wind Farm in Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 30 Percent Firm Energy Factor Wind Energy in Colombia Source: Authors' data. Note: If no financing terms are mentioned, it is assumed that the investor must finance 70 percent of the project costs through commercial credits. *Income tax reduction of 15% after 2017. 45 46 World Bank Study Table 7.4c. Financial Results for a 300 MW Wind Farm in Northern Colombia after Use of Financial Instruments; Reliability Payment Considered with a 36 Percent Firm Energy Factor Source: Authors' data. Note: If no financing terms are mentioned, it is assumed that the investor must finance 70 percent of the project costs through commercial credits. *Income tax reduction of 15% after 2017. Wind Energy in Colombia 47 Key Findings: Options to Foster Investment in Wind Power The analysis of the information generated in the previous section illustrates the alternatives available to the GOC for promotion of wind power. The higher the investment cost, the greater government intervention is needed to promote investment in RET. Moreover, for investors not paying for CERE it is the same as having a reliability factor of 0.4. This should be obvious: CERE is the fund used to remunerate the guaranteed firm energy. Recognizing the contribution of wind power to firm energy allows it to benefit from reliability payments, thus offsetting the expenditure incurred in paying CERE. At the conceptual level, policy makers have the option of either waiving CERE payment from wind power producers, or recognizing their project's firm capacity. In this case, it may be simpler to recognize the firm capacity of each project. Table 7.5 summarizes alternative enabling environments conducive to investments in the wind power sector under the three cases of reliability payments. Table 7.5. Key Findings: Combination of Policy Instruments to Reach a Financial Threshold Investment cost/kW If reliability payment (US$) considered at Required actions to reach a 14% IRR $2,400 20% Need 40% clean tech concessionary financing + 20% soft loans 6+ 10% commercial credits 30% Need 30% of clean tech concessionary financing + 30% soft loans + 10% commercial credits 36% Need 20% clean tech concessionary financing + 50% soft loans $2,100 20% Need 15% clean tech concessionary financing + 55% soft loans; or 20% of clean tech concessionary financing + 40% soft loans + 10% commercial credits 30% Need 5% clean tech concessionary financing + 65% soft loans; or 20% of clean tech concessionary financing + 10% soft loans + 40% commercial credits 36% Need 60% soft loans + 10% commercial credits $1,800 20% Need 40% soft loans + 30% commercial credits 30% No concessionary financing is needed 36% No concessionary financing is needed Source: Authors' data. If the GOC decides to promote wind power under a pessimistic investment cost outlook, high reliability factors, reduction in fees, and concessionary financing are required (individually or in conjunction). On the other hand, if investment costs are US$1,800/kW, then less concessionary financing and fewer policy instruments would be required. The results summarized in table 7.5 provide a guideline for the GOC in the selection of a long term policy option for various wind technology investment costs. A potential transition strategy would be to develop and apply long term policy options-- to capture all the complementarity benefits to the interconnected system--while creating conditions for some early entrants to give the energy market players and operators the opportunity to learn and gain experience in the operation and system maintenance of large scale wind projects. 48 World Bank Study Conclusions of the Estimated Impact of Alternative Policy Options for a 300 MW Wind Energy Power Plant in the MEM The analysis conducted and the results summarized in previous sections allow the following general conclusions and results: In conclusion, the analysis from the viewpoint of potential investors provides a good foundation for understanding the relative strength of different options. Under current policy, regulatory and market conditions, wind power projects are not attractive for private investment. The starting point to promote wind power should be to review the existing regulatory system in detail and remove any biases against renewable energy technologies. Of all the options available to the GOC to improve the financial performance of wind power plants, the reliability payment has the greatest influence on returns. If the reliability charge is applied at levels reflecting the historical contribution of Jepírachi's energy generation during the dry period, financial performance for wind power improves significantly. If investment costs for wind power continue decreasing from the high values observed in late 2008, as expected in the near future, the returns improve considerably. Therefore, some options could be seen as a bridge mechanism to be ready for future conditions under which wind power would be more competitive. Access to concessionary resources, such as those associated with clean technology multilateral funds and soft loans, could be very useful to promote early investments; and exempting some charges and payments used in the regulatory system is shown to be very effective in increasing the IRR. Internalizing costs of global externalities through clean technology concessionary loans would be enough to provide returns on equity over the selected threshold, for basically all investment costs (in the analysis the maximum US$2,400/kW is used). This holds true even if the generators have to pay all MEM charges. The results also indicate that the GOC has the possibility to target future expectations regarding the investment costs associated with wind energy technology. At one extreme the regulators might study the possibility of fostering RETs even at investment prices above US$2,200/kW, for example. Or they might consider a more conservative approach targeting wind projects only if investment costs fall below US$1,900/kW or a similar value. As previously indicated, the higher the investment costs, the greater the government intervention required. Waiving the payment of CERE by RET generators is equivalent to remunerating the contribution of wind power projects (for the conditions of the easterly wind fields in northern Colombia) at a reliability factor of around 0.4. That is, from the potential investor's viewpoint (expected financial returns on investment), waiving a project's obligation to make CERE contributions is financially equivalent to remunerating the project with a reliability factor of Wind Energy in Colombia 49 0.4. However, it should be noted that policy makers have the option of either waiving CERE payment from wind power producers, or recognizing their project's firm capacity. In this case, it may be simpler to recognize the firm capacity of each project. The GOC could also consider temporary incentives for RET initiatives. That is, the energy sector could benefit from the early implementation of wind projects as a mechanism to gain experience in operating the interconnected system for the possible case of when wind energy becomes a more significant contributor to the grid. Similarly, the energy sector would also benefit from having a well functioning regulatory system for this power technology. After a well defined "promotion and experimentation period," sufficient to give the technology time to further reduce its investment needs, the incentives could be eliminated or adjusted. Notes 1 As previously indicated, the reliability payment seeks to provide independent investors with an economic signal of the relative importance of reliable installed (firm) power capacity. The GOC conducted a public auction to allocate "reliability payments" for future power plants. A value of US$13,998/MWh has resulted from the firm energy auction held in May 2008. 2 P10 indicates the energy generated with a 10 percent probability of values being lower, and P90 indicates the value with a 90 percent probability of values being lower. These probabilities refer to monthly values and cannot be assumed for longer periods. 3 As explained in previous sections, estimates using the available information from Jepírachi, complemented by observational records from nearby wind measuring stations from 1985 to 2008, produced a reliability factor of 0.415 with a standard deviation of 0.055. For illustration, a reliability factor of 0.36 is used in this analysis, equivalent to the mean value reduced by one standard deviation. 4 The Clean Technology Fund (CTF) is a climate change donor driven fund seeking the implementation of transformational low carbon options. CTF financial conditions are typically a 0.65% interest rate with a 20 to 40 year repayment period and 10 years of grace. 5 It should be noted that simultaneously allowing for reliability charges and waiving CERE payments is not recommended. It would imply a logical contradiction because funds for the reliability charge come from CERE. 6 CTF conditions are those defined for the CTF (typically, a 0.65 percent interest rate with a 20 year repayment period and 10 years of grace. Soft loans here mean those with conditions typical of IBRD loans in Colombia: currently a 17 year repayment period, interest rate LIBOR + 1.05 percent, front end fee 0.25 percent. CHAPTER 8 Conclusions C olombia has a power sector that is quickly maturing, with relative stability in its regulations, an unbundled system, and a dispatch mechanism that closely resembles a well functioning competitive market. Competition is promoted and tools have been designed to attract cost effective capacity expansions that would promote reliability of service (a fuller description of the system and its dispatch mechanism was included in first stage of this project's report). The Colombian energy sector is characterized by low carbon intensity, below the world average. For the foreseeable future, hydropower will likely continue to provide the backbone of the power sector. However, a highly hydro dependent power system makes the system intrinsically vulnerable to severe droughts. This vulnerability could be addressed by diversification of the power mix. Wind Energy Resources Could Become an Important Energy Option in Colombia Colombia has considerable wind resources, estimated to exceed 14 GW, mostly on its northern coast. However, the potential development of this resource is limited by the high initial investment costs and provisions in the regulatory system that affect this energy source. Wind technology costs reached a historical low of US$1,600/kW in 2002 and since then costs soared to a high of US$2,400/kW by September 2008. This trend was reversed in 2009, with recent figures reporting average values around US$1,800/kW.1 This decreasing cost trend is expected to continue. The research in this study showed that costs of US$1,800 or below make wind a viable option even with less heavy intervention from the government. However, under current policy, regulatory, and market conditions, wind power projects are still not attractive for private investment. Some reforms and changes in the market conditions could therefore also be seen as a bridge mechanism to be ready when wind power becomes a more competitive option with decreasing investment costs in the future. The report highlights ways to assess the complementarity between wind and water resources and the potential contribution to firm energy production during "critical" dry periods. For the Colombian case, the results indicate that during the dry season (when water resources availability becomes a concern and electricity prices rise) the wind resources could produce above average, at least in the northern part of Colombia. More importantly for Colombia, during critical El Niño events wind contribution exceeds non El Niño years. This contribution should be recognized and remunerated as well as rewarded in the current regulatory system adopted by Colombia. 50 Wind Energy in Colombia 51 Policy Instruments There is a wide range of instruments through which governments could guide the functioning of selected markets. However, not all of the available instruments are applicable to the case of Colombia. Therefore, only a reduced subset was explored, namely those instruments that are compatible with and relatively easy to incorporate into the existing regulatory system in Colombia and have the effect of changing the financial results for a potential investor. The instruments have been classified as: (i) financial instruments; (ii) payments to government, fees, and charges; and (iii) adjustment to the existing regulatory system. Policy Options The existing regulatory system needs to be assessed and any biases against renewable energy technologies need to be removed in order to create a level playing field for all technologies. In addition, changes in financial and fiscal conditions could also make wind power competitive in Colombia. There is a wide range of options through which governments could guide the functioning of the sector. The instruments explored in this study have been classified as: (i) price and quantity based instruments; (ii) adjustment in the regulatory system; and (iii) financial incentives other than price. From assessing the effectiveness of the instruments, it was found that the single most effective policy instrument to promote wind power in Colombia is the granting of access to reliability payments, recognizing the firm energy and complementarity offered by wind. The implementation of this policy option is relatively easy to incorporate into the existing regulatory system. For new wind power plants with costs in the range of $1,800/kW installed, the adoption of the reliability payments is enough to attract independent investors, operating in wind fields with similar characteristics to those found in Northern Guajira. Higher capital costs require access to concessionary financial conditions, such as those provided under the CTF or fiscal incentives. Likewise, internalizing costs of global externalities through certified emission reductions, already used to some extent, would help to make the projects more viable. Exempting some charges and payments used in the regulatory system is also shown to be a very effective way of increasing the returns on investments. This is true in particular if CERE charges are exempted. However, it should be noted that CERE payments and reliability charges are two sides of the same coin, since the funds for reliability charges come from CERE. Temporary incentives for wind and other renewable energy could also be considered in order for the sector to benefit and gain experience from the early implementation of wind projects before wind energy becomes a more significant contributor to the grid. Lack of access to the benefit of "reliability (firm energy) payments" for wind powered plants is a serious limitation to their development. A simple method for calculating the firm energy rating of wind powered plants was introduced. It is recommended that the firm energy rating of a wind resource be adjusted annually based on the following exponential smoothing formula: firm energy rating in year t + 1 = ½ (firm energy rating in year t) + ½ (energy produced in year t). 52 World Bank Study Under this approach, the firm energy rating, and therefore the reliability payment, will quickly converge to the long run average firm energy capability, even if the firm energy rating in the initial year is poorly measured. The suggested approach to assess the reliability factor for wind farms is risk neutral. If the yearly estimate is used during the "dry period," the difference between the annual mean and the dry period mean could be interpreted as a risk reduction strategy. A more formal option, in tune with the general risk aversion characteristic of Colombia's regulatory framework, is to subtract standard deviation affected by some factor of the historical performance. Other Findings Reliable data are needed to assess the specific potential of wind throughout Colombia. Without these data, promoters and investors face high uncertainties, which translate into an additional barrier to future investments. For this reason, the governments of Colombia and of other countries in the region are encouraged to assign resources to the proper mapping of their wind resource endowment and to make this information available to the public. Other actions required to improve access to the market include open access to research and technology developments, as well as promotion of medium scale developments (at 100 MW or more installed capacity), allowing the grid operator to be prepared for necessary system adjustments and plan strategically for greater transmission requirements when investments in wind power are increased. Applicability of the Analysis Conducted Although the analysis has focused on Colombia, the approach is applicable to other countries, which could further explore their nonconventional renewable resources. Other countries could benefit from performing a similar analysis to understand possible complementarities and how renewable energy technologies can also play a larger role in energy provision. Notes 1 As of March 2009, the European Wind Energy Association reports that the average of recent projects fluctuates around 1,225/kW. This translates to approximately US$1,800 (see explanation of turbine cost reductions in chapter 4). References Barrett, Mark. 2007. "A Renewable Electricity System for the UK." In Renewable Electricity and the Grid: The Challenge of Variability, ed. Godfrey Boyle. Boyle, Godfrey, ed. 2007. Renewable Electricity and the Grid: The Challenge of Variability. London: Earthscan. Cohen, J., T. Schweizer, A. Laxson, S. Butterfield, S. Schreck, L. Fingersh, P. Veers, and T. Ashwill. 2008. "Technology Improvement Opportunities for Low Wind Speed Turbines and Implications for Cost of Energy Reduction." Report No. NREL/SR 500 41036. Golden, CO: NREL. Cramton, P., and S. Stoft. 2007. "Colombia Firm Energy Market." Proceedings of the Hawaii International Conference on System Sciences. January. ------. 2008. "Forward Reliability Markets: Less Risk, Less Market Power, More Efficiency." Utilities Policy 16: 194­201. Edigson, Pérez Bedoya, and Jaime A. Osorio. 2002. "Energía, Pobreza y Deterioro Ecológico en Colombia: Introducción a las Energías Alternativas" Todográficas. Medellín, Colombia. European Wind Energy Association. 2008. "Integrating 300 GW Wind Power in European Power Systems: Challenges and Recommendations." Frans Van Hulle, Technical Advisor. Presented at the World Bank's SDN Week. February 21­22. World Bank, Washington, DC. LBNL (Lawrence Berkley National Laboratory). 2008. Annual Report on U.S. Wind Power Installation, Cost, and Performance Trends: 2007. Washington, DC: U.S. Department of Energy. May. Ministry of Mines and Energy 2008. Memorias al Congreso de la República 2007­2008. Junio. ISSN 0120 0291. Nohara, D., A. Kitoh, M. Hosaka, and T. Oki. 2006. "Impact of Climate Change on River Discharge Projected by Multimodel Ensemble." Journal of Hydrometeorology 7: 1076 1089. Universidad Nacional de Colombia, Universidad de los Andes, COLCIENCIAS, ISAGEN. 2006. "Proyecto Regulación Para Incentivar las Energías Alternas y la Generación Distribuida en Colombia." Informe 3. September. Bogotá, Colombia. UPME. 2009. Plan de Expansión de Referencia­Generación­Transmisión 2009­2023. Mining Energy Planning Unit (Unidad de Planeación Minero Energética, UPME) of the Ministry of Mines and Energy. April. URS Corporation. 2008. "Study of Equipment Prices in the Energy Sector." Study prepared for the World Bank, Washington, DC. June. 54 World Bank Study UWIG (Utility Wind Integration Group.) 2007. "Integrating Wind into the Grid." Presented to NRC Panel on Electricity from Renewables. Washington, DC. December 6. Valencia, Adriana. 2008. "Missing Links: Demystifying Alternative Energy Use and Improving Decision Making for Increased Off grid Electrification in Colombia." PhD Dissertation. Energy and Resources Group (ERG), University of California, Berkeley. Vergara, et al. 2008. "Review of Policy Framework for Increased Reliance on Renewable Energy in Colombia." ESMAP­World Bank. Washington, DC. Vergara. W., ed. 2009. "Assessing the Consequences of Climate Destabilization in Latin America." SDWP 32. World Bank, Washington, DC. World Bank 2007. "Technical and Economic Assessment of Off grid, Mini grid and Grid Electrification Technologies." Energy Sector Management Assistance Program (ESMAP). Technical Paper 121/07. World Bank, Washington, DC. December. World Wind Energy Association. 2009. World Wind Energy Report 2008. Bonn, Germany. February. Appendixes 55 Appendix 1. Technology Cost Comparison In relation to chapter 3 of the report, the following tables provide a cost ranking of various technologies according to capacity factors. Table A1.1. Least Levelized Cost Ranking of Electricity Generation Plant by Capacity Factor (%) without the Cost of CO2 Emissions 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 SC SC SC SC Large Large Large Large Large Large 500 MW 500 MW 500 MW 500 MW Hydro Hydro Hydro Hydro Hydro Hydro Rehabili- Rehabili- Rehabili- Rehabili- 1200 MW 1200 MW 1200 MW 1200 MW 1200 MW 1200 MW tation tation tation tation 2 SC Oil SC Oil SC Oil SC Oil SC SC SC SC SC SC Steam to Steam to Steam to Steam to 500 MW 500 MW 500 MW 500 MW 500 MW 500 MW Coal 300 Coal Coal Coal Rehabili- Rehabili- Rehabili- Rehabili- Rehabili- Rehabili- MW 300 MW 300 MW 300 MW tation tation tation tation tation tation 3 Simple SC Nat Large Large SC Oil SC Oil SC Oil SC Oil SC Oil SC Oil Cycle GT Gas Hydro Hydro Steam to Steam to Steam to Steam to Steam to Steam to 150 MW Steam to 1200 MW 1200 MW Coal Coal Coal Coal Coal Coal Coal 300 MW 300 MW 300 MW 300 MW 300 MW 300 300 MW MW 4 SC Nat Large SC Nat SC Nat SC Nat SC Nat SC Nat SC Nat SC Nat SC Nat Gas Hydro Gas Gas Gas Gas Gas Gas Gas Gas Steam 1200 MW Steam to Steam to Steam to Steam to Steam to Steam to Steam to Steam to 300 MW Coal Coal Coal Coal Coal Coal Coal Coal 300 MW 300 MW 300 MW 300 MW 300 MW 300 MW 300 MW 300 MW 5 CCGT CCGT CCGT China SC China China China China China China 560 MW 560 MW 560 MW 550 MW SPC SPC SPC SPC SPC SPC 550 MW+ 550 MW 550 MW 550 MW 550 MW 550 MW China SC 550 MW 6 CCGT CCGT China SC China China SC China SC China SC China SC China SC Small to 140 MW 140 MW 550 MW+ SPC 550 550 MW+ 550 MW 550 MW 550 MW 550 MW Med CCGT MW China Hydro 400 140 MW USPC MW 550 MW 7 Large Simple China CCGT CCGT China China China China China SC Hydro Cycle GT SPC 5 560 MW 560 MW USPC USPC USPC USPC 550 MW+ 1200 MW 150 MW 50 MW 550 MW 550 MW 550 MW 550 MW China USPC 550 MW 8 Diesel China SC China SC China SC CCGT China SC China SC China SC Small to China SC 5 MW 550 MW 300 MW 300 MW 140 MW 300 MW 300 MW 300 MW Med 300 MW Hydro 400 MW 9 China SC China China China Small to CCGT Small to Small to China SC CCGT 550 MW SPC USPC USPC Med 560 MW Med Med 300 MW 560 MW 550 MW 550 MW 550 MW+ Hydro 400 Hydro 400 Hydro 400 CCGT MW MW MW 140 MW 10 China China SC Simple Simple China SC CCGT CCGT CCGT CCGT CCGT SPC 300 MW Cycle GT Cycle GT 300 MW 140 MW 560 MW 560 MW 560 MW 140 MW 550 MW 150 MW 150 MW CCS 11 China SC China Diesel Small to SC Small to CCGT CCGT CCGT SC 300 MW USPC 5 MW Med 550 MW Med 140 MW 140 MW 140 MW 550 MW+ 550 MW Hydro 400 Hydro 400 SPC MW MW 550 MW 12 China Diesel China SC Diesel SPC China SC China SC SC SC USPC USPC 5 MW 300 MW 5 MW 550 MW 300 MW 300 MW 550 MW 550 MW 550 MW+ 550 MW CCS CCS CCS China SC 300 MW CCS 13 China SC China SC SC China SC SC SC SC China SC SPC 5 SC CFB 300 MW 300 MW 550 MW + 300 MW 300 MW 550 MW 550 MW 300 MW 50 MW 500 MW CCS CCS China SC CCS CCS 550 MW CCS Source: Authors' data. 56 Wind Energy in Colombia 57 Table A1.2. Least Levelized Cost Ranking of Electricity Generation Plant by Capacity Factor (%) with US$18/Ton CO2 Emissions 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 SC Oil SC Large Large Large Large Large Large Large Large Steam to 500 MW Hydro Hydro Hydro Hydro Hydro Hydro Hydro Hydro Coal Rehabili- 1200 1200 MW+ 1200 MW+ 1200 MW+ 1200 MW+ 1200 MW 1200 MW 1200 MW 300 MW tation MW+ 2 SC Oil SC Oil SC SC SC SC Small to Small to Small to Small to Steam to Steam to 500 MW 500 MW 500 MW 500 MW Med Hydro Med Hydro Med Hydro Med Hydro Coal Coal Rehabili- Rehabili- Rehabili- Rehabili- 400 MW+ 400 MW 400 MW 400 MW 300 MW 300 MW tation tation tation tation SC 500 MW Rehabili- tation 3 Simple CCGT 560 SC Oil SC Oil SC Oil SC Nat SC Oil SC SC SC Cycle GT MW Steam to Steam to Steam to Gas Steam Steam to 500 MW 500 MW 500 MW 150 MW Coal Coal Coal to Coal Coal Rehabili- Rehabili- Rehabili- 300 MW 300 MW 300 MW 300 MW 300 MW tation tation tation 4 CCGT 560 SC Nat SC Nat SC Nat SC Nat Small to SC Nat SC Oil SC Oil China MW Gas Steam Gas Steam Gas Steam Gas Steam Med Hydro Gas Steam Steam to Steam to USPC to Coal to Coal to Coal to Coal 400 MW+ to Coal Coal Coal 550 MW 300 MW 300 MW 300 MW 300 MW SC Nat 300 MW 300 MW 300 MW Gas Steam to Coal 300 MW 5 SC Nat CCGT 140 CCGT 560 CCGT 560 CCGT 560 CCGT 560 CCGT 560 SC Nat SC Nat China SPC Gas Steam MW MW MW MW MW MW Gas Steam Gas Steam 550 MW to Coal to Coal to Coal 300 MW 300 MW 300 MW 6 CCGT 140 Simple CCGT 140 CCGT 140 CCGT 140 CCGT 140 CCGT 140 China China CCGT 560 MW Cycle GT MW MW MW MW MW USPC USPC MW 150 MW 550 MW 550 MW 7 Large China SC China SC China SPC China SPC China SPC China SPC China SPC China SPC China SC Hydro 550 MW 550 MW 550 MW 550 MW 550 MW 550 MW 550 MW 550 MW 300 MW 1200 CCS MW+ 8 Diesel China SPC China SPC China SC Small to China China CCGT 560 CCGT 560 China SC 5 MW 550 MW 550 MW 550 MW Med Hydro USPC USPC MW+ MW 300 MW 400 MW+ 550 MW 550 MW China SC Nat USPC Gas Steam 550 MW to Coal 300 MW 9 China SC China SC China SC China China SC China SC CCGT 140 China SPC China SC SC 550 MW 300 300 MW USPC 550 550 MW 550 MW MW 550 MW 550 MW 300MW MW MW CCS 10 China SPC China China China SC China China SC China SC China SC CCGT 140 CCGT 140 550 MW USPC 550 USPC 550 300 MW USPC 300 MW 550 MW 550 MW+ MW MW MW MW 550 MW CCGT 140 MW 11 China SC Diesel China SC Small to China SC China SC China SC China SC China SC China SC 300 MW 5 MW 300 MW Med Hydro 300 MW 300 MW 300 MW 300 MW 300 MW 550 MW 400 MW CCS CCS CCS CCS 12 China Small to Simple China SC China SC China SC China SC China SC China SC China USPC 550 Med Hydro Cycle GT 300 MW 300 MW 550 MW 300 MW 550 MW 300 MW USPC MW 400 MW+ 150 MW CCS CCS CCS CCS CCS 550 MW China SC CCS 550 MW CCS 13 China SPC CCGT Diesel Simple China SC SC China SC China SC China SPC 550 MW CCS 5 MW Cycle GT 550 MW 300 MW 550 MW USPC 300 MW 550 MW CCS 50 MW 150 MW CCS CCS CCS 550 MW CCS CCS CCS 14 China China SPC Small to Diesel China SPC China SPC China SPC China SPC USPC 550 550 MW Med Hydro 5 MW 550 MW 550 MW 550 MW USPC 550 MW + MW CCS CCS 400 MW CCS + CCS CCS 550 MW USPC 550 CCGT CCS MW CCS 482 MW Source: Authors' data. Appendix 2. Use of Earth Simulator to Estimate the Likelihood of Extreme Weather Events Earth Simulator AGCM (atmospheric general circulation model) runs, developed by the Meteorological Research Institute (MRI) of the Japan Meteorological Agency (JMA), were used to estimate the likelihood of extreme weather events to the end of the century. The Earth Simulator is a super high resolution atmospheric general circulation model with a horizontal grid size of about 20 km, offering an unequaled high resolution capability. The use of the Earth Simulator made this super high resolution model's long term simulation possible.1 Although the global 20 km model is unique in terms of its horizontal resolution for global change studies with an integration period up to 25 years, available computer power is still insufficient to enable ensemble simulation experiments; this limits its application to a single member experiment. To address this caveat, parallel experiments with lower resolution versions of the same model (60 km, 120 km, and 180 km mesh) were performed. In particular, ensemble simulations with the 60 km resolution have been performed and compared with the 20 km version for this study. Two extreme indices for precipitation are used to illustrate changes in precipitation extremes over Colombia, one for heavy precipitation and one for dryness. All over the country, RX5D is projected to increase in the future. Largest RX5D increases (rainfall intensification) are found over south eastern Colombia. At a higher resolution (20 km) the model projects even larger increases in RX5D. Figure A2.1. Changes in Maximum Five-Day Precipitation Total (mm) between the Present and the End of the 21st Century for (a) 60-km and (b) 20-km, Respectively For 60-km model, areas with the highest projected consistency in sign are hatched. Zero lines are contoured. Source: MRI of the JMA. 58 Wind Energy in Colombia 59 Likewise, Figure A2.2 shows the changes in maximum number of consecutive dry days. A "dry day" is defined as a day with precipitation less than 1 mm d 1. Consecutive dry day periods are projected to increase, in particular over the northern coast. Figure A2.2. The Same as In Figure A.2.1 Except for Consecutive Dry Days (day) Source: MRI of the JMA. Impact on River Steam Flow Using the runoff data, derived from rainfall projections under the Earth Simulator, stream flow in large rivers can be calculated. The analysis used a "GRiveT" river model.2 In the present day simulation, large rivers are well represented by this model. Although the analysis has yet to be made for basins in Colombia with large hydropower potential, a similar assessment made for rivers in the Amazon Basin indicates that the changes in extremes and in particular the concentration of rainfall and the lengthening of dry periods will increase the amplitude of stream flows, which in turn would affect the mean firm capacity of hydropower installations. Notes 1 This model is an operational short term numerical weather prediction model of JMA and part of the next generation climate models for long term climate simulation at MRI. 2 (GRiveT: Global Discharge model using TRIP, the 0.5 x 0.5 version with global data for discharge channels; Nohara et al. (2006). The river runoff assessed in the land surface model is horizontally interpolated as external input data into the TRIP grid so that the flow volume is saved. Appendix 3. Pool Prices under Various Scenarios Pool prices in the wholesale market are formed by adding other variable costs (CERE, FAZNI, environmental and Automatic Generati on Control AGC) to the pure marginal cost. The report presents this for the mean case scenario. Other scenarios are defined in the Table A3.1: Table A3.1 MEM Scenarios SCENARIO DEMAND FUEL PRICES HYDRO MEAN BASE BASE REVISED MEAN HIGH HYDRO BASE BASE XM FACTORS LOW LOW LOW REVISED HIGH HIGH HIGH REVISED Source: Authors' data. The following Figures (A3.1 and A3.2) present this for the mean high hydro and high scenario. Figure A3.3 compares the pool prices for base and base high hydro scenarios. Figure A3.1 Pool Prices, Base High Hydro Scenario POOL PRICES--Mean High Hydro Scenario 140 120 100 US$/Mwh Mean 80 P10 60 P90 40 20 0 8 Ma 9 Ap 0 1 Fe 2 Ja 3 De 4 No 4 Oc 5 Se 6 Au 7 18 9 20 Ap 1 2 Fe 3 Ja 4 De 5 5 r-1 l-0 0 y-1 r-1 1 1 c-1 v-1 t-1 1 l-1 y-2 r-2 r-2 2 2 c-2 n- b- n- p- g- n- b- n- Ju Ju Ma Ma Ma Ju Ju Source: Authors' data. Figure A3.2. Pool Prices, High Scenario POOL PRICES--High Scenario 140 120 100 US$/Mwh 80 Mean 60 P10 P90 40 20 0 8 Ma 9 Ap 0 1 Fe 2 Ja 3 De 4 No 4 Oc 5 Se 6 Au 7 18 9 20 Ap 1 2 3 Ja 4 De 5 5 r-1 l-0 0 y-1 r-1 1 1 c-1 v-1 t-1 1 l-1 y-2 r-2 r-2 2 2 c-2 n- b- n- p- g- n- b- n- Ju Ju Ma Ma Ma Ju Ju Fe Source: Authors' data. 60 Wind Energy in Colombia 61 Figure A3.3. Comparison of Pool Prices for Base and Base High Hydro Scenarios AVERAGE POOL PRICES--Mean and Mean High Hydro Scenarios 120 100 80 US$/MWh Mean 60 40 Mean High Hydro 20 0 Ma 8 Ma 9 Ja 0 No 1 Se 1 Ju 2 Ma 3 Ma 4 Ja 5 No 6 Se 6 Ju 7 Ma 8 Ma 9 Ja 0 No 1 Se 1 Ju 2 Ma 3 Ma 4 5 1 v-1 l-0 y-0 r-1 1 l-1 y-1 r-1 1 v-1 1 l-1 y-1 r-2 2 v-2 2 l-2 y-2 r-2 n- p- n- p- n- p- Ju Source: Authors' data. As can be observed in Figure A3.3, the average pool prices for the mean scenario are regularly higher than the mean high hydro scenario. Appendix 4. Results of the Expected Returns on Investments with the Individual Application of the Policy Instruments for Different Market Scenarios Tables A4.1­A4.5 depict the expected returns on investments with the individual application of the selected policy instruments discussed in chapter VII of the report. The analysis of the information contained in table A4.1 indicates: All policy instruments improve the financial outcome of the potential investment under consideration, as compared with the baseline condition. Individually, none attains the selected threshold of a return on equity of 14 percent before taxes. A generous access to concessionary financing (policy instrument C2) provides the greater inducement. This option requires clean technology concessionary funding for up to 50 percent of the total unit investment. Eliminating CERE payments (column F) is a very effective instrument. Adjusting the access to the reliability charge (or leveling the playing field for nonconventional renewable energy technologies) is also a very effective incentive, as indicated in column H, depending on the methodology used for selecting the reliability factor. Eliminating income taxes does not seem to be an effective instrument to attract investments to RET, given the criteria used to judge financial feasibility. As should be expected, the comparison of results presented in Table A4.1 indicates that a reduction in unit investment moves the expected returns closer to the defined threshold of 14 percent before taxes but falls short of reaching this target. The use of individual policy instruments is not sufficient incentive for potential investors. The following tables summarize the analysis conducted when assessing the likely impact on potential investors of the selected policy group options. This policy group option does not provide adequate incentives to potential investors if the investment costs are to remain high, at or above US$2,100/kW. However, this policy group offers interesting flexibility for low unit investment costs. In particular, if the reliability factor is estimated through the methodology indicated in section VII.3, this would be the only government intervention required to open the market to wind powered energy investments.1 62 Wind Energy in Colombia 63 Table A4.1. Effectiveness Analysis of Individual Policy Instruments Results expressed as financial returns on capital for a 300 MW wind farm in northern Colombia POLICY OPTIONS A B1 B2 C1 C2 D E F G H TYPE I Financial Instruments Carbon CERs 18 0 0 0 0 0 0 0 0 0 Access CTCF loans 0 0 0 0.3 0.5 0 0 0 0 0 Access to soft loans 0 0.4 0.7 0 0 0 0 0 0 0 TYPE III Government Fees Income taxes 0.33 0.33 0.33 0.33 0.33 0 0.33 0.33 0.33 0.33 Generator charges 1 1 1 1 1 1 0 1 1 1 TYPE V Regulatory Instruments Sustainability charge 0 0 0 0 0 0 0 0 5 0 CERE payments 1 1 1 1 1 1 1 0 1 1 Reliability charge 0 0 0 0 0 0 0 0 0 0.36 Investments Costs 1800 $/kW Project before taxes 7.5% 5.8% 5.8% 5.8% 5.8% 5.8% 6.2% 9.4% 7.1% 9.5% Project after taxes 6.1% 4.6% 4.6% 4.6% 4.6% 5.8% 5.0% 7.8% 5.8% 7.9% Equity before taxes 7.3% 5.4% 5.9% 7.4% 10.2% 4.9% 5.6% 10.0% 6.8% 10.1% Equity after taxes 5.6% 4.0% 4.4% 5.7% 8.3% 4.9% 4.1% 7.9% 5.1% 8.0% Investments Costs 2100 $/kW Project before taxes 6.1% 4.4% 4.4% 4.4% 4.4% 4.4% 4.9% 7.8% 5.7% 7.9% Project after taxes 4.9% 3.5% 3.5% 3.5% 3.5% 4.4% 3.9% 6.4% 4.6% 6.4% Equity before taxes 5.4% 3.6% 3.9% 5.2% 7.3% 3.3% 3.8% 7.7% 4.9% 7.8% Equity after taxes 3.9% 2.5% 2.8% 3.8% 5.7% 3.3% 2.6% 5.9% 3.5% 6.0% Investments Costs 2400 $/kW Project before taxes 4.9% 3.4% 3.4% 3.4% 3.4% 3.4% 3.8% 6.5% 4.6% 6.6% Project after taxes 3.9% 2.6% 2.6% 2.6% 2.6% 3.4% 3.0% 5.3% 3.6% 5.3% Equity before taxes 3.9% 2.2% 2.4% 3.4% 5.2% 1.9% 2.5% 6.0% 3.5% 6.1% Equity after taxes 2.7% 1.3% 1.5% 2.4% 3.9% 1.9% 1.5% 4.4% 2.3% 4.5% Source: Authors' data. Note: The policy instruments used are read in the upper half of the table, while the lower half indicates the expected financial returns. For example, policy instrument A corresponds to access to payments for the reduction of GHG at a price of US$18/ton CO2. Policy instrument D shows that income taxes are waived. Policy Groups Tables A4.2, A4.3, and A4.4 present the results obtained from the analysis of the three policy groups under consideration. In each case the analysis seeks to find a combination of instruments that jointly create the conditions for potential investors to move their capital toward RET initiatives. The tables retain the same general design used to describe the results of individual policy instruments. Reading the table from 64 World Bank Study left to right, the columns aggregate the instruments used to create the policy group of interest. For example, as shown in table A4.2 the Group Policy Options are built as follows: Baseline + Carbon CERs + Soft Loans (20, 40, and 70 percent) + access to clean technology concessionary financing (30 and 50 percent). The use of financial instruments to build a policy option provides great flexibility. In the particular case under study the threshold, or target financial rate of return (FRR), is not achieved if the investment costs approach US$2,400/kW. For the low investment cost scenario, potential investors require access to clean technology concessionary resources for nearly 30 percent of the expected cost. Table A4.2. Effectiveness Analysis of Policy Options: Use of Financial Instruments Financial results for a 300 MW wind farm in northern Colombia POLICY OPTIONS A B C D E F TYPE I Financial Instruments Carbon CERs 0 18 18 18 18 18 Access CTCF loans 0 0 0 0 0.3 0.7 Access to soft loans 0 0 0.3 0.7 0.4 0 TYPE III Government Fees Income taxes 0.33 0.33 0.33 0.33 0.33 0.33 Generator charges 1 1 1 1 1 1 TYPE V Regulatory Instruments Sustainability charge 0 0 0 0 0 0 CERE payments 1 1 1 1 1 1 Reliability charge 0 0 0 0 0 0 Investments Costs 1800 $/kW Project before taxes 5.8% 7.5% 7.5% 7.5% 7.5% 7.5% Project after taxes 4.6% 6.1% 6.1% 6.1% 6.1% 6.1% Equity before taxes 4.9% 7.3% 7.9% 8.7% 12.1% 20.3% Equity after taxes 3.5% 5.6% 6.0% 6.8% 9.9% 18.0% Investments Costs 2100 $/kW Project before taxes 4.4% 6.1% 6.1% 6.1% 6.1% 6.1% Project after taxes 3.5% 4.9% 4.9% 4.9% 4.9% 4.9% Equity before taxes 3.3% 5.4% 5.8% 6.4% 9.0% 15.9% Equity after taxes 2.2% 3.9% 4.2% 4.8% 7.1% 13.7% Investments Costs 2400 $/kW Project before taxes 3.4% 4.9% 4.9% 4.9% 4.9% 4.9% Project after taxes 2.6% 3.9% 3.9% 3.9% 3.9% 3.9% Equity before taxes 1.9% 3.9% 4.2% 4.6% 6.7% 12.5% Equity after taxes 1.1% 2.7% 2.9% 3.3% 5.2% 10.4% Source: Authors' data. Wind Energy in Colombia 65 The use of government fiscal mechanisms is explored in table A4.3 below. As indicated in the table, the group encompasses a wide range of fees and payments to the government. The following sequence was used, as indicated by reading the table from left to right: baseline + Carbon CERs + tax shelter + waiver of generator charges + elimination of the obligation to contribute to CERE. The results indicate that this policy group option alone cannot create the required incentives to attract potential investors to wind power projects. Table A4.3. Effectiveness Analysis of Policy Options: Use of Government Fees and Payments POLICY OPTIONS A B C D E F TYPE I Financial Instruments Carbon CERs 0 18 18 18 18 18 Access CTCF loans 0 0 0 0 0 0 Access to soft loans 0 0 0 0 0 0 TYPE III Government Fees Income taxes 33% 33% 0% 0% 0% 0% Generator charges 1 1 1 0 1 0 TYPE V Regulatory Instruments Sustainability charge 0 0 0 0 0 0 CERE payments 1 1 1 1 0 0 Reliability charge 0 0 0 0 0 0 Investments Costs 1800 $/kW Project before taxes 5.8% 7.5% 7.5% 8.0% 10.9% 11.3% Project after taxes 4.6% 6.1% 7.5% 8.0% 10.9% 11.3% Equity before taxes 4.9% 7.3% 7.3% 8.0% 12.3% 12.9% Equity after taxes 3.5% 5.6% 7.3% 8.0% 12.3% 12.9% Investments Costs 2100 $/kW Project before taxes 4.4% 6.1% 6.1% 6.5% 9.1% 9.5% Project after taxes 3.5% 4.9% 6.1% 6.5% 9.1% 9.5% Equity before taxes 3.3% 5.4% 5.4% 5.9% 9.6% 10.2% Equity after taxes 2.2% 3.9% 5.4% 5.9% 9.6% 10.2% Investments Costs 2400 $/kW Project before taxes 3.4% 4.9% 4.9% 5.3% 7.8% 8.1% Project after taxes 2.6% 3.9% 4.9% 5.3% 7.8% 8.1% Equity before taxes 1.9% 3.9% 3.9% 4.4% 7.7% 8.1% Equity after taxes 1.1% 2.7% 3.9% 4.4% 7.7% 8.1% Source: Authors' data. 66 World Bank Study The use of regulatory instruments comprises the last group of policy options. Under this group the following sequence of instruments is used, as depicted in table A4.3 below: Baseline and Carbon CERs + reliability charge (reliability factors of 0.20, 0.30 and 0.36) +CERE waiver (50 percent, 100 percent). The results summarized in table A4.3 indicate: Table A4.4. Effectiveness Analysis of Policy Options: Use of Regulatory Instruments POLICY OPTIONS A B1 B2 B3 C1 C2 C3 D1 D2 D3 TYPE I Financial Instruments Carbon CERs 18 18 18 18 18 18 18 18 18 18 Access CTCF loans 0 0 0 0 0 0 0 0 0 0 Access to soft loans 0 0 0 0 0 0 0 0 0 0 TYPE III Government Fees Income taxes 33% 33% 33% 33% 33% 33% 33% 33% 33% 33% Generator charges 1 1 1 1 1 1 1 1 1 1 TYPE V Regulatory Instruments Sustainability charge 0 0 0 0 0 0 0 0 0 0 CERE payments 1 1 0.5 0 1 0.5 0 1 0.5 0 Reliability charge 0 0.2 0.2 0.2 0.3 0.3 0.3 0.36 0.36 0.36 Investments Costs 1800 $/kW Project before taxes 7.5% 9.5% 11.1% 12.6% 10.4% 12.0% 13.5% 10.9% 12.5% 13.9% Project after taxes 6.1% 7.9% 9.3% 10.8% 8.7% 10.2% 11.5% 9.2% 10.6% 12.0% Equity before taxes 7.3% 10.2% 12.6% 15.2% 11.6% 14.1% 16.6% 12.4% 14.9% 17.5% Equity after taxes 5.6% 8.0% 10.2% 12.5% 9.2% 11.5% 13.8% 10.0% 12.3% 14.7% Investments Costs 2100 $/kW Project before taxes 6.1% 7.9% 9.3% 10.7% 8.7% 10.1% 11.5% 9.2% 10.6% 11.9% Project after taxes 4.9% 6.4% 7.7% 9.0% 7.2% 8.5% 9.7% 7.6% 8.9% 10.1% Equity before taxes 5.4% 7.8% 9.9% 12.0% 9.0% 11.1% 13.2% 9.7% 11.8% 13.9% Equity after taxes 3.9% 6.0% 7.8% 9.6% 7.0% 8.8% 10.7% 7.6% 9.5% 11.4% Investments Costs 2400 $/kW Project before taxes 4.9% 6.6% 7.9% 9.2% 7.4% 8.7% 9.9% 7.8% 9.1% 10.3% Project after taxes 3.9% 5.3% 6.5% 7.6% 6.0% 7.2% 8.3% 6.4% 7.5% 8.6% Equity before taxes 3.9% 6.1% 7.9% 9.7% 7.1% 9.0% 10.8% 7.8% 9.6% 11.4% Equity after taxes 2.7% 4.5% 6.1% 7.6% 5.4% 7.0% 8.5% 5.9% 7.5% 9.1% Source: Authors' data. Notes 1 As explained in the document, estimates using the available information from Jepírachi, complemented by observational records from nearby wind measuring stations from 1985 to 2008, produce a reliability factor of 0.415. A standard deviation of 0.055 results in the reliability factor of 0.36 used in this analysis. Appendix 5. Exempting CERE Payments by 50 or 100 Percent In addition, the analysis also considered the option of exempting 100 percent or 50 percent of the CERE payment. The results show that clean technology concessionary financing is still required if CERE is considered only at 50 percent and above a unit price of US$2,100. Alternatively, this type of financing is not necessary if the unit investment is US$1,800 and the CERE payment is exempted, even at 50 percent. In short, eliminating the CERE payment alone is also an effective instrument. If CERE payment is eliminated, a unit investment cost of US$1,800/kW allows the IRR to reach the 14 percent target. The results are summarized in table A5.1 below. Table A5.1. Financing Necessary if CERE Is Returned 50 Percent or 100 Percent, Depending on Investment Costs Investment cost/kW (US$) % Returned CERE In all cases it is assumed that there is 30% equity 100% Need 15% clean tech concessionary financing + 55% soft loans $2,400 Need 40% clean tech concessionary financing + 20% soft loans + 50% 10% commercial credits 100% Need 45% soft loans + 25% commercial credits $2,100 50% Need 15% clean tech concessionary financing + 55% soft loans 100% No additional financing required $1,800 50% Need 35% soft loans + 35% commercial credits Source: Authors' data. The results of analyzing the possibility of excluding the hypothetical 300 MW wind power project from paying CERE charges indicates that not paying for CERE charges results in a return of investment that is the same as if the reliability payment is recognized at 40 percent. Therefore, the policy maker has an option of either not charging the CERE payment to wind power producers, or recognizing their project's firm capacity. In this case, it might be simpler and in the country's interest to recognize the firm capacity of each project. 67 Appendix 6. Complementarity between Wind Power and Hydroelectric Resources Jose Manuel Mejia and Alberto Brugman Estudios Energeticos Limitada Chapter 1: Introduction This report presents the results of the studies conducted to analyze the complementarities existing between hydroelectric resources and wind power in Colombia, including synergies that can occur during El Niño occurrences. Colombia is a country with abundant natural resources for the production of renewable energy. Historically, power sector development has been based on hydroelectric energy (approximately 80 percent of energy consumption). The country also has abundant coal resources, which are largely exported and which represent a considerable energy reserve of strategic interest for the country. At the moment there is only one wind power farm in the country (Jepírachi), located on the Caribbean coast, in the province of La Guajira, with 19.5 MW installed capacity. Several wind power advantages in the Colombian power system have been mentioned. Among these, the complementarities with hydroelectric resources are investigated in this study. Specifically, preliminary analyses indicate that during the dry hydrological period (December to April), wind velocities in the Caribbean are above the annual. Likewise, it has also been argued that wind velocities are above the mean when El Niño occurs. This study aims to find an answer to the following questions: Does complementarity exist between water resources and wind power resources in Colombia (for example, in La Guajira)? What could be the contribution of wind resources to the reliability of the national electricity system? What is the natural variability of the wind resource (monthly and summer potential contribution)? What is the wind power contribution during the period of "extreme" summer, associated with the El Niño phenomenon? 68 Wind Energy in Colombia 69 Chapter 2: Methodology The main aspects of the methodology are to: 1. Use the Puerto Bolívar meteorological station as the basis of the analysis. Information beginning in 1986 is available. 2. Fill in abundant missing hourly data. 3. Conduct a statistical analysis of hourly wind velocity characteristics. 4. Convert hourly wind velocity data in hourly power generation using conversion factors corresponding to a particular wind turbine and a given capacity installation. 5. Estimate monthly generation information. 6. Select four discharge measurement stations of the National Interconnected System for analysis of synergies of joint hydroelectric power and wind turbine operation. 7. Analyze river discharges and Jepírachi generation during El Niño occurrences. 8. Estimate the firm energy obtained from the individual operation of hydroelectric plants (with and without reservoirs) and wind power plants, as well as their joint operation. Firm energy will be defined as the maximum energy that can be produced without deficits during the analysis period, which will include El Niño occurrences. The analysis will be conducted using a simulation model that will operate the plants to provide a given energy target, adjusting this target until no deficits are generated. The analysis will be conducted for each of the hydroelectric plants selected. 9. Measure synergetic gains due to the complementarity between hydroelectricity and wind power, as the difference between firm energy in a joint operation and the sum of firm energies in isolated operation. 70 World Bank Study Chapter 3: Data Base 3.1 Wind Velocity Information The World Bank obtained hourly data for two stations in the Colombian Caribbean from IDEAM. The first station is located in Puerto Bolívar, in the vicinity of the Jepírachi power plant. It covers the period between October 1986 and December 2008, with several missing records. (There are 162,124 hourly records out of a total of 195,072, representing 83 percent.) There is no clear behavior of the distribution of hourly velocities during the day for the different months of the year. The distribution of wind velocities in the different hours of the day is shown below. Figure A6.1. Hourly Wind Velocity: Puerto Bolívar HOURLY WIND VELOCITY PUERTO BOLIVAR 1.400 Hourly velicity/daily velocity 1.200 1.000 0.800 0.600 0.400 0.200 0.000 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 12 22 22 23 23 24 01 12 23 34 45 56 67 78 89 9 10 Hours of the day Source: IDEAM. Figure A6.1 shows the trend of larger wind velocities during peak electricity load hours, while smaller wind velocities tend to be concentrated during early morning hours which are the minimum load hours. Therefore, there is a complementarity of wind velocities with electricity load, which is a clear advantage for wind power. As seen in figure A6.2, large wind velocities occur from December to April, which are the months with lower river discharges. This represents a positive complementarity between wind power and hydroelectric power. Wind Energy in Colombia 71 Figure A6.2. Seasonal Behavior of Mean Wind Velocity MEAN WIND VELOCITY PUERTO BOLIVAR 9.0 8.0 7.0 Wind velocity m/seg 6.0 5.0 4.0 3.0 Series1 2.0 1.0 0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month Source: IDEAM. Figures A6.3 and A6.4 show similar results for the Barranquilla Airport where the second station is located. The results are similar for the Puerto Bolívar and Barranquilla Airports, although the difference between the minimum and maximum values is more accentuated for the Barranquilla Airport. Mean velocities at the Barranquila Airport are substantially lower than those at Puerto Bolívar and do not have a good correlation with the Puerto Bolívar station, due to the fact of the shading effect of the Sierra Nevada de Santa Marta. Therefore, this station was not used. Figure A6.3. Hourly Mean Velocity: Barranquilla Airport HOURLY MEAN VELOCITY BARRANQUILLA AIRPORT 1.600 Hourly velocity/daily velocity 1.400 1.200 1.000 0.800 0.600 0.400 0.200 0.000 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 12 22 22 23 23 24 01 12 23 34 45 56 67 78 89 Hours Source: IDEAM. 72 World Bank Study Figure A6.4. Mean Wind Velocity: Barranquilla Airport MEAN WIND VELOCITY BARRANQUILLA AIRPORT 6 Mean velocity m/seg 5 4 3 2 1 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month Source: IDEAM. 3.2 River Discharges Monthly data for four rivers associated with hydroelectric power plants were used in this study. The information was obtained from databases for simulation of the interconnected hydrothermal power system. Rivers considered were the Nare River at the Santa Rita Dam (1955­2009), the Guavio River at the Guavio Dam (1963­2009), the Cauca River at the Salvajina Dam (1946­2009), and the Magdalena River at the Betania Dam (1972­2009), representing a sample of geographical regions of the country. Table A6.1 shows mean monthly values for these rivers. Table A6.1. Mean Monthly Values for the Guavio, Nare, Cauca, and Magdalena Rivers MEAN MONTHLY VALUES m3/seg Guavio Nare Cauca Magdalena Jan 18.4 36.2 166.3 145.4 Feb 19.9 32.3 145.8 154.6 Mar 29.9 34.5 139.7 183.4 Apr 65.6 46.4 152.2 225.2 May 106.3 62.1 153.1 240.4 Jun 139.2 58.2 127.9 240.5 Jul 144.4 47.9 103.2 240.3 Aug 110.0 48.8 74.6 179.1 Sep 76.0 59.3 63.2 138.6 Oct 64.9 67.6 109.6 177.4 Nov 52.7 67.6 197.2 218.0 Dec 31.3 49.6 215.5 199.2 Source: Appendix authors' data. Wind Energy in Colombia 73 Figure A6.5 illustrates the diversity of meteorological condition shown by the rivers chosen. Figure A6.5. Normalized Monthly Discharges of the Four Rivers NORMALIZED MONTHLY DISCHARGES 2.5 Monthly Mean/Annual Mean 2 1.5 Guavio Nare 1 Cauca Magdalena 0.5 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: Appendix authors' data. 3.3 Technical Information for the Jepírachi Power Plant The Jepírachi wind farm power plant is located in the northern part of Colombia, on the Guajira peninsula on the Caribbean Sea. It has been equipped with 15 Nordex N60 aerogenetors (1,300 kW each), with a total installed capacity of 19.5 MW. The power curve (relating wind velocity with power delivered by the generator) for each unit is shown in figure A6.6. Figure A6.6. Power Curve for Each Unit POWER CURVE NORDEX N60 1600 1400 Factor kW/m/seg 1200 1000 800 600 400 200 0 0 5 10 15 20 25 Velocity m/seg Source: Appendix authors' data. 74 World Bank Study The power curve is valid for standard air density (1.225 kg/m3). For a different air density a correction has to be made. Air density at Jepírachi is 1.165 kg/m3). Velocity is at tower altitude (60 meters). Therefore, a correction must be made, taking into account a roughness factor (0.15), and the air density in the power curve of the manufacturer and Puerto Bolivar, since velocity measurements are made at a 10 meter altitude. 3.4 Jepírachi Generation Hourly Jepírachi generation was obtained from the Neon database operated by Xm, the system operator. The information was available between February 2004 and March 2009. The following tables summarize the information at monthly hour level. Table A6.2. Jepírachi Monthly Hour Generation kWh (1 to 12) JEPIRACHI MONTHLY HOUR GENERATION KWH (1 to 12) Hora 1 2 3 4 5 6 7 8 9 10 11 12 Feb 04 46409 44317 39107 35323 30761 31072 29592 30234 39381 45079 54153 39381 Mar 04 140936 125710 116189 113622 113044 109002 104041 115101 135418 137050 158091 135418 Apr 04 127838 121365 115335 108995 105044 105672 112597 131031 138595 154124 190674 138595 May 04 231996 233364 228772 225541 209674 192621 193550 209064 211206 206943 222462 211206 Jun 04 330876 312374 304223 298968 294805 300260 303424 313792 340632 377004 377463 340632 Jul 04 242488 227097 221345 203293 183794 170140 176278 190410 187934 202619 246582 187934 Aug 04 240662 220931 213009 213823 220730 197890 191253 229974 252703 263633 276345 252703 Sep 04 46563 37095 36999 36673 35451 21177 21253 24676 22056 27475 50192 22056 Oct 04 61729 55937 48722 49209 46999 38244 34866 53119 66744 77137 89306 66744 Nov 04 60176 50989 42348 57550 69987 64443 59364 64631 100781 108311 127689 100781 Dec 04 83709 82030 66083 66415 61500 60122 61066 73683 120077 141600 155051 120077 Jan 05 117439 112019 113308 107172 100759 93807 74290 90023 134068 173343 197178 134068 Feb 05 130617 121767 100697 88478 80324 79221 68862 91027 138653 171403 214050 138653 Mar 05 190359 162779 151840 131416 118121 108643 99305 123272 163227 182891 224965 163227 Apr 05 154972 150516 145551 132760 114786 114394 122887 146267 163255 173237 184477 163255 May 05 134648 127325 118908 107702 107549 107578 111381 134308 133262 135237 146744 133262 Jun 05 116432 93899 92139 90527 75461 77493 74215 94447 114563 130210 150332 114563 Jul 05 179538 181386 172213 143606 124863 122338 128007 137707 154542 162466 188642 154542 Aug 05 172650 150581 147546 135425 126542 119151 122975 133853 132222 146884 182514 132222 Sep 05 125927 128747 134390 127963 117870 103245 102499 122933 117671 119877 137928 117671 Oct 05 125927 128747 134390 127963 117870 103245 102499 122933 117671 119877 137928 117671 Nov 05 58513 46187 46865 45333 45406 50663 45348 58816 76657 83044 92404 76657 Dec 05 83773 69392 58543 56719 61611 59192 54709 61794 101314 147022 168341 101314 Jan 06 142530 143363 135094 137363 121598 106907 95523 100618 143210 193212 235846 143210 Feb 06 186891 176962 156182 146437 141071 134439 136036 174944 230149 268658 279571 230149 Mar 06 194406 192387 186007 167988 160020 162331 158345 196017 248324 279524 322582 248324 Apr 06 134752 125313 118589 117559 112926 107591 91856 127226 162933 178068 191372 162933 May 06 146913 152761 150091 134786 129561 127175 137489 160875 150092 168364 202209 150092 Jun 06 188191 181644 165323 153041 145705 142623 149388 158433 193370 216576 245133 193370 Jul 06 273952 264547 247057 231815 220277 225774 245223 277116 303914 328205 345305 303914 Aug 06 175006 161350 151769 162598 163303 152824 159408 190130 187454 212316 226024 187454 Sep 06 121836 112256 113384 115297 101732 93420 97781 122323 114837 110926 150603 114837 Oct 06 33764 33913 30844 28615 26543 36114 41069 72346 73642 72866 85308 73642 Nov 06 55494 46044 48843 44619 44781 42900 43367 59042 70262 78197 90236 70262 Dec 06 59470 51653 51913 44500 46813 39869 46792 66155 110377 157627 223358 110377 Jan 07 80766 76745 69222 62755 54673 59807 66482 103797 192493 264594 283181 192493 Feb 07 88263 83999 72931 65508 66247 66469 74406 108545 157399 193461 218661 157399 Mar 07 136785 117686 122976 113518 102743 94738 106547 177327 197942 207676 238591 197942 Apr 07 142636 137103 136786 130534 126182 125110 132306 144649 123012 119740 147654 123012 May 07 111543 107354 105961 106932 94222 90915 83770 91401 89206 75341 94698 89206 Jun 07 84322 64716 49950 44181 34957 36981 68150 105818 130863 169942 203073 130863 Jul 07 179391 156942 143594 155505 162125 168097 182542 217389 231059 226006 253728 231059 Aug 07 106610 101659 82770 78702 64441 58055 58989 62929 64588 73801 93390 64588 Sep 07 86064 76093 76628 77241 69730 58474 52812 66407 62516 56065 76632 62516 Oct 07 24349 17836 14402 12186 12462 13039 13080 21766 21746 20414 26371 21746 Nov 07 73889 66966 60180 55416 56802 62436 64525 92063 130866 135115 145750 130866 Dec 07 90271 78076 68034 69976 69645 71624 66075 77080 105941 136080 173809 105941 Jan 08 150988 139050 130543 137005 126762 108916 101454 118931 173812 220046 245562 173812 Feb 08 186657 185811 163518 148699 134753 134196 128160 160803 210344 232433 252266 210344 Mar 08 194511 183501 168074 160775 152081 139823 126657 158281 198509 207617 241720 198509 Apr 08 192195 174564 152676 149851 135665 127285 127594 147007 160517 152390 174664 160517 May 08 163070 154352 140156 139326 141993 135519 154812 179539 169203 163547 194966 169203 Jun 08 226274 200330 185411 178294 164696 166520 170894 196275 199596 207061 246916 199596 Jul 08 187039 172327 171895 160658 145228 135780 135324 165633 190247 219829 257446 190247 Aug 08 105591 89614 80009 76364 70617 65652 67979 82000 88168 94028 106812 88168 Sep 08 31755 27904 31262 29408 23547 23649 24996 30282 36672 43186 49006 36672 Oct 08 53756 53432 44587 50841 38415 38779 33396 38066 49877 60150 64042 49877 Nov 08 34799 34755 34418 32401 31708 38107 34486 38293 45255 60389 56459 45255 Dec 08 64346 62404 53736 54629 46539 53755 49296 56524 78246 116582 128353 78246 Source: Neon database operated by Xm Wind Energy in Colombia 75 Table A6.3. Jepírachi Monthly Hour Generation kWh (13 to 24) JEPIRACHI MONTHLY HOUR GENERATION KWH (13 to 24) Hora 13 14 15 16 17 18 19 20 21 22 23 24 Feb 04 98832 116252 121112 116121 115746 109831 102702 88486 75112 71280 66288 54858 Mar 04 209915 222360 226980 228988 230264 225482 204557 191778 184204 175480 157655 144963 Apr 04 287841 308331 319624 322955 312303 290319 257954 217061 199895 180377 168059 143401 May 04 278974 309557 320782 339221 342712 317194 294642 278238 265014 267321 259511 248826 Jun 04 409552 425714 447946 471293 475756 464108 438164 412355 400403 399448 385357 361131 Jul 04 355699 422982 457067 451330 447171 422930 380081 323514 299569 282693 269326 254858 Aug 04 354570 419574 456167 471056 460154 439583 382403 339621 280883 266809 263923 249949 Sep 04 97805 137998 189027 204910 213888 183621 148821 138255 123365 98224 89071 70124 Oct 04 147251 194101 223123 226578 231010 204510 172888 144304 117469 85971 69242 64860 Nov 04 205158 205387 209871 229119 232997 216665 170954 136850 98301 76084 64232 66332 Dec 04 231460 276812 299321 321964 310293 266583 179038 126340 97680 90461 87406 84458 Jan 05 258325 301063 321480 319443 318790 294302 238770 209083 186783 161498 153155 135287 Feb 05 315201 331622 341827 365001 342725 324492 281304 224905 211600 201198 170405 143229 Mar 05 376594 427460 455243 469330 450999 418348 379968 329361 291567 258868 229836 211167 Apr 05 271492 285599 319247 329261 321717 306514 261862 223342 193847 169473 164377 153722 May 05 215529 232751 259749 274240 277104 252820 220982 201209 181340 173999 162739 151609 Jun 05 176273 209676 220576 246222 252478 254471 228805 199115 170318 144279 127117 126592 Jul 05 265501 308665 346186 355825 358698 316074 281342 258556 241290 221569 203399 194914 Aug 05 304565 352928 385470 410458 393774 359458 310473 257302 229357 212566 199387 187263 Sep 05 239844 289204 302050 321373 299517 265394 213934 171449 152194 140641 137612 138227 Oct 05 239844 289204 302050 321373 299517 265394 213934 171449 152194 140641 137612 138227 Nov 05 121430 144074 176884 181349 176428 156629 126755 99778 86610 85577 75391 66819 Dec 05 235340 278407 298848 317624 308204 251596 184531 140527 118467 107466 100216 93312 Jan 06 297591 312213 335222 350897 362125 335277 265013 220422 195749 186607 173772 166502 Feb 06 353252 373926 391412 411281 405491 391225 339936 286458 257249 230442 205121 197894 Mar 06 460954 482631 488260 486242 477543 441109 381590 339041 298122 264119 239649 209951 Apr 06 279775 329916 377467 401883 393763 359292 301963 224752 192494 171885 149502 143365 May 06 319630 357492 384891 409886 406684 373414 302349 240642 193930 167322 151309 147432 Jun 06 310895 332181 400086 431605 430180 424080 360362 285688 237219 211937 201652 199751 Jul 06 401830 422321 457354 464218 469127 426919 386932 329370 275202 273212 271802 264170 Aug 06 333779 366160 409564 453522 457085 416253 363068 302381 254522 221891 211144 200374 Sep 06 293854 338058 379687 397428 378110 351788 293816 240502 191294 156274 144871 135047 Oct 06 175005 209381 207484 225583 237867 203258 135133 86640 48166 36688 35897 34898 Nov 06 191568 235337 270363 286155 295931 260626 194329 148383 113079 94403 83464 75348 Dec 06 327977 373835 403263 413182 394487 350348 202481 130575 103574 84822 75342 57249 Jan 07 420510 455144 467110 466372 451484 416899 353725 252184 173892 134679 94213 83736 Feb 07 341224 390256 421898 428223 425323 391610 329281 261810 194782 147320 120428 106426 Mar 07 356405 414973 453991 470945 447771 408009 356324 269582 203153 168954 162023 143045 Apr 07 282406 310738 334771 350266 349246 312744 269336 241341 201140 178680 166082 153390 May 07 230448 283151 310982 340678 327283 300680 239819 202271 168300 148287 144181 123500 Jun 07 233700 257474 290373 324978 346464 336987 274039 169109 138581 134795 116039 102221 Jul 07 348952 383832 407687 401533 391359 368189 328132 282425 247193 223914 210796 190559 Aug 07 230393 290796 334389 325578 326678 305269 249026 182901 154951 140690 127179 112397 Sep 07 174368 203981 240120 289150 282957 259729 219883 171180 140650 117127 106259 93724 Oct 07 47037 63818 86835 100704 101463 95600 78947 62255 44651 34891 27253 22520 Nov 07 208529 230499 241063 246768 240535 220457 177799 140615 118235 114625 95984 86350 Dec 07 230459 241534 264572 278771 263945 234513 196068 156999 130672 118176 103910 98726 Jan 08 340054 378326 406077 406508 391829 333922 276244 244967 232306 209806 188271 154815 Feb 08 326300 363704 380896 396105 393385 368342 314384 277194 241242 217727 206113 205375 Mar 08 389435 437416 464872 473405 454508 409841 347882 294388 270853 251649 227481 208901 Apr 08 340106 390412 421859 440491 430585 390649 337703 301237 264715 236788 218018 205065 May 08 319964 367183 399320 433539 419788 388254 320916 276542 243471 216244 203152 180406 Jun 08 313496 357607 431142 443789 430007 392750 343404 313940 287019 262303 250843 246133 Jul 08 370006 394993 417817 435181 430535 375194 314522 269597 250177 239403 229782 204162 Aug 08 174777 209503 233851 252589 252856 237003 181824 159600 145402 117937 107903 109202 Sep 08 94118 133119 149249 163896 158904 144601 118924 91800 71418 58785 52261 44988 Oct 08 100118 117826 151023 151582 155406 141521 126389 111171 96806 84618 69173 64148 Nov 08 116445 143008 150568 167772 167080 146407 115897 94145 76043 63186 52745 43469 Dec 08 188203 211681 234862 234010 218937 197061 157795 124273 103599 96777 76449 73344 Source: Neon database operated by Xm. The distribution of wind velocities at different hours of the day is shown in figure A6.7. A favorable complementarity with daily electricity load fluctuations is observed. Differences between this curve and the corresponding curve for wind velocity are due to the nonlinear nature of the relationship between wind velocity and power. 76 World Bank Study Figure A6.7. Jepírachi: Hourly Generation HOURLY JEPIRACHI GENERATION 2 Monthly Hour generation/Monthly generation 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 12 22 22 23 23 24 01 12 23 34 45 56 67 78 89 Hours Source: Neon database operated by Xm. Figure A6.8. Jepírachi: Monthly Mean Generation MONTHLY MEAN GENERATION JEPIRACHI 1.6 Monthly Generation/Annual Generation 1.4 1.2 1 0.8 0.6 0.4 0.2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month Source: Neon database operated by Xm. A favorable complementarity with river discharges is observed. Differences between this curve and the corresponding curve for wind velocity are due to the nonlinear nature of the relationship between wind velocity and power. Wind Energy in Colombia 77 Chapter 4: Extension of Jepírachi Information Limited generation information at the Jepírachi power plant due to its short operation period is an obstacle for an analysis of this plant's contribution to the firmness of the power system in a joint operation. Therefore, generation information was extended using the longer wind velocity records available at Puerto Bolívar. The procedure followed is described below: Power calculations using wind velocities data at Puerto Bolívar. x For each of the hours of existing data at Puerto Bolívar, power generation in a Nordex N60 turbine was calculated. x The calculation adjusted wind velocity to an altitude of 60 ms using an assumed roughness factor (ar). x Power corresponding to the adjusted velocity was calculated based on the power curve of the aerogenerator. It was adjusted to take into account differences between air densities at Jepírachi and the standard value. Regression between hourly estimated power at Puerto Bolívar and Jepírachi generation. x Common hourly data between Jepírachi generation reported by XM and Jepírachi generation computed using the methodology described in a. (above), were used to perform a regression analysis. This analysis was repeated using different values of the roughness coefficient, choosing the value giving the best fit. Missing hourly velocity information at Puerto Bolívar was filled in. x Initially, the correlation between wind velocities at the Puerto Bolívar and Barranquilla Airports was studied, but no significance was found. Therefore, missing data were filled in based on daily mean velocity, if available. Otherwise, monthly mean velocity was used and finally, multiannual monthly mean velocity was used. All these results were adjusted to consider the hourly seasonality observed in the data. d. Extension of Jepírachi generation. x Jepírachi generation information was extended (1985­2008) using the regression equation found and applied to the filled in Puerto Bolívar velocity records. Tables A6.4 and A6.5 show extended monthly generation values for Jepírachi. 78 World Bank Study Table A6.4. Extended Monthly Generation for Jepírachi (January to June) EXTENDED MONTHLY GENERATION FOR JEPIRACHI (KWH) Jan Feb Mar Abr May Jun 1985 5834634 5682658 6515124 5886177 5767205 6914523 1986 5834634 5682658 6515124 5886177 5767205 6914523 1987 5834634 7887850 6889132 6483911 5767205 7236743 1988 7268860 7500189 8912351 6991765 7381501 6608671 1989 5834634 5682658 6515124 5930856 6758656 7576098 1990 9451220 5305019 4739058 6958182 6607606 7778632 1991 6888466 6044854 6438107 6649830 7122249 7487965 1992 6151827 6929893 8137434 7015526 6335503 7483811 1993 6475069 5620726 7363741 6390443 4092576 7248022 1994 6418401 7009124 7217540 7328519 7059915 8586733 1995 6697520 5836699 6443392 6231644 6491717 6627188 1996 5370648 6182659 6476131 5931191 5767205 6914523 1997 4676298 7837674 7564978 6823904 5922530 7007094 1998 5774991 5591419 7138039 6586111 5526878 7276299 1999 5773035 5364206 6468408 7050548 6026792 6759032 2000 5834634 5885611 6515124 5886177 6363170 7978651 2001 6235307 1399111 1603269 1734096 2497808 8361813 2002 6742444 6307064 7893469 6571283 7252543 7122733 2003 6213564 7372513 6822730 6040200 8540665 7014330 2004 4417189 1630762 3947270 4669053 6122227 9037782 2005 4431175 4802542 6244397 4828971 4138032 3619306 2006 5030369 6179175 7233905 5023244 5386714 6027382 2007 5426555 5031539 5762332 4826728 4020576 3947717 2008 5502418 5912375 6480628 5917684 5767061 6500230 Source: Appendix authors' data. Wind Energy in Colombia 79 Table A6.5. Extended Monthly Generation for Jepírachi (July to December) EXTENDED MONTHLY GENERATION FOR JEPIRACHI (KWH) Jul Aug Sep Oct Nov Dec 1985 7734911 6152792 3710858 2721909 3403509 4825251 1986 7734911 6152792 3710858 4571029 5907804 7209327 1987 9115923 8104222 5159465 2721909 4541764 6151577 1988 9115498 4376842 5177179 3453668 3353160 4825251 1989 7940524 6146705 4396355 6007866 5039010 5520377 1990 7449251 7981166 5372166 1751368 3965017 5536141 1991 8293163 7556200 6611484 4786059 3860408 4825251 1992 8855269 7881288 5345987 5580324 4515810 4825251 1993 8031142 7856364 3710858 6098720 4384973 6404446 1994 9710702 8014871 6137318 3802563 3901304 5442340 1995 6927623 3109042 3523425 2325658 4327140 4696092 1996 7207639 5736949 4090081 2721909 3403509 5121455 1997 8740789 8016731 5505666 2770995 3656170 6710743 1998 7066678 6175066 3769276 4174693 3401869 4825251 1999 7604889 5483666 2420295 2147343 2330244 4750685 2000 7497365 7177243 3394830 4059808 3775646 5349653 2001 7717748 8186006 5364343 4789924 4050187 4968515 2002 8186261 7731705 4112645 4967635 5747633 6888561 2003 8378413 6973755 4221191 2511494 3088666 4628588 2004 6911378 7201819 2131765 2623595 2895604 3527309 2005 5263834 5401123 4179454 4179454 2251921 3560560 2006 7761522 6197955 4783490 2289704 2992357 4075575 2007 6150271 3770902 3173954 1005641 3245168 3529592 2008 6204898 3252952 1699921 1974458 1863222 2842356 Source: Appendix authors' data. 80 World Bank Study Chapter 5: Case Studies for Complementarity Analysis Complementarity between hydroelectric generation and wind generation at Jepírachi is due to two factors: noncoincidence in seasonal mean values of both variables, and synergy obtained on the lack of coincidence of extreme events for them. 5.1 Mean Monthly Discharges and Jepírachi Generation The following figures show normalized values (monthly mean divided by the annual mean) for wind velocities and river discharges for the four rivers in which complementarity with wind power was analyzed. Figure A6.9 shows the complementarity between these resources, since low water discharges during the drier months (January to April) correspond to high wind power. Figure A6.9. Mean Monthly Values at the Guavio River Dam Site 2.5 Guavio 2 Normalized monthly mean Jepirachi 1.5 1 0.5 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: Appendix authors' data. Figure A6.10. Mean Monthly Values at the Santa Rita Dam Site on the Nare River 1.6 1.4 Normalized monthly mean 1.2 1 0.8 0.6 Nare 0.4 Jepirachi 0.2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: Appendix authors' data. Wind Energy in Colombia 81 The graph shows very good complementarity between the Nare River and wind power at Jepírachi. Low discharges during the two dry seasons (December to March and July and August) correspond to high wind power; the opposite is also true. Figure A6.11. Mean Monthly Values at the Salvajina Dam Site on the Cauca River 1.8 1.6 1.4 Normalized nonthly mean 1.2 1 0.8 0.6 Cauca 0.4 Jepirachi 0.2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: Appendix authors' data. The Cauca River at the Salvajina Dam site presents a dry period from June to September which is complemented by high wind power at the Jepírachi site. Figure A6.12. Mean Monthly Values at the Salvajina Dam Site on the Magdalena River 1.6 1.4 Normalized nonthly mean 1.2 1 0.8 Magdalena 0.6 Jepirachi 0.4 0.2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source: Appendix authors' data. Discharges of the Magdalena River at Betania follow a similar pattern to wind power at Jepírachi, although some complementarity is observed during the first dry season occurring at the beginning of the year. 82 World Bank Study 5.2 El Niño occurrences Colombia's interconnected power system is severely affected by severe droughts due to its very large hydroelectric component. Historically, during these periods electricity prices rise due to the supply shortage and, in extreme cases, electricity rationing may occur. An example is the rationing in 1992, with severe economic and political consequences in the country. Droughts in Colombia occur due to a global climatological event known as El Niño that affects nearly the entire country. Next Table identifies the El Niño periods that have occurred since 1950, according to IDEAM. Table A6.6. El Niño Periods since 1950 "EL NIÑO" PERIODS Source: IDEAM Start Finish Months Jul 51 Jan 52 6 Mar 57 Jul 58 14 Jun 63 Feb 64 8 May 65 May 66 13 Oct 68 Jun 69 8 Aug 69 Feb 70 6 Apr 72 Feb 73 10 Aug 76 Mar 77 7 Aug 77 Feb 78 6 Apr 82 Jul 83 15 Jul 86 Mar 88 20 Apr 91 Jul 92 15 Feb 93 Aug 93 6 Mar 94 Apr 95 13 Apr 97 May 98 13 Apr 02 Apr 03 12 Jun 04 Mar 05 8 Aug 06 Feb 07 6 Source: Appendix authors' data. An analysis was conducted of the severity of El Niño occurrences in the four rivers selected (Nare, Guavio, Cauca and Magdalena) compared with energy delivered by the Jepírachi power plant. Initially, average historical values for river discharges and Jepírachi generation during El Niño periods were examined, as shown in following tables. An example will better illustrate better the analysis conducted. The first column of the first table analyzes the severity of the El Niño occurrence between July 1986 and March 1988. The series of mean discharge occurrences in all historical periods starting in July and finishing in March of the following year were analyzed (as shown in the table). The mean and standard deviation of these series were obtained (shown at the end of the table), and the departure from the mean value, expressed in terms of Wind Energy in Colombia 83 standard deviations, was obtained for the value corresponding to El Niño (July 1986 to March 1988) and is shown at the end of the table. The tables present the information for all El Niño occurrences from 1985 to December 2008 for the four rivers already mentioned, as well as historical and reconstructed generation at the Jepírachi power plant. El Niño occurrences are shown in gray in the tables. Table A6.7. Analysis of El Niño Occurrences in Guavio River Discharges (1986­1995) TABLE 1. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG GUAVIO RIVER Jul. 86 Abr. 91 Feb. 93 Mar. 94 Mar. 88 Jul. 92 Ago. 93 Abr. 95 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Jul 63 Mar 65 60.06 Apr 63 Jul 64 69.92 Feb 63 Aug 63 75.70 Mar 63 Apr 64 56.94 Jul 64 Mar 66 67.40 Apr 64 Jul 65 79.97 Feb 64 Aug 64 75.27 Mar 64 Apr 65 64.99 Jul 65 Mar 67 57.03 Apr 65 Jul 66 68.81 Feb 65 Aug 65 85.40 Mar 65 Apr 66 67.11 Jul 66 Mar 68 63.08 Apr 66 Jul 67 65.89 Feb 66 Aug 66 52.44 Mar 66 Apr 67 51.86 Jul 67 Mar 69 67.73 Apr 67 Jul 68 84.14 Feb 67 Aug 67 87.93 Mar 67 Apr 68 67.57 Jul 68 Mar 70 65.23 Apr 68 Jul 69 77.97 Feb 68 Aug 68 94.44 Mar 68 Apr 69 70.26 Jul 69 Mar 71 75.13 Apr 69 Jul 70 80.53 Feb 69 Aug 69 69.10 Mar 69 Apr 70 64.65 Jul 70 Mar 72 80.60 Apr 70 Jul 71 97.58 Feb 70 Aug 70 99.61 Mar 70 Apr 71 82.70 Jul 71 Mar 73 77.06 Apr 71 Jul 72 100.64 Feb 71 Aug 71 107.96 Mar 71 Apr 72 84.30 Jul 72 Mar 74 67.00 Apr 72 Jul 73 83.92 Feb 72 Aug 72 109.86 Mar 72 Apr 73 72.37 Jul 73 Mar 75 67.47 Apr 73 Jul 74 82.51 Feb 73 Aug 73 75.46 Mar 73 Apr 74 68.41 Jul 74 Mar 76 66.10 Apr 74 Jul 75 78.49 Feb 74 Aug 74 90.61 Mar 74 Apr 75 64.09 Jul 75 Mar 77 74.10 Apr 75 Jul 76 95.67 Feb 75 Aug 75 85.99 Mar 75 Apr 76 70.86 Jul 76 Mar 78 62.94 Apr 76 Jul 77 88.67 Feb 76 Aug 76 117.47 Mar 76 Apr 77 78.69 Jul 77 Mar 79 60.09 Apr 77 Jul 78 72.44 Feb 77 Aug 77 75.00 Mar 77 Apr 78 58.91 Jul 78 Mar 80 59.04 Apr 78 Jul 79 69.74 Feb 78 Aug 78 76.63 Mar 78 Apr 79 59.27 Jul 79 Mar 81 60.88 Apr 79 Jul 80 75.49 Feb 79 Aug 79 69.96 Mar 79 Apr 80 63.13 Jul 80 Mar 82 59.11 Apr 80 Jul 81 72.39 Feb 80 Aug 80 75.44 Mar 80 Apr 81 59.43 Jul 81 Mar 83 70.60 Apr 81 Jul 82 79.08 Feb 81 Aug 81 74.60 Mar 81 Apr 82 66.71 Jul 82 Mar 84 78.63 Apr 82 Jul 83 91.63 Feb 82 Aug 82 90.24 Mar 82 Apr 83 79.66 Jul 83 Mar 85 73.62 Apr 83 Jul 84 92.31 Feb 83 Aug 83 102.13 Mar 83 Apr 84 77.71 Jul 84 Mar 86 67.12 Apr 84 Jul 85 82.16 Feb 84 Aug 84 106.16 Mar 84 Apr 85 69.06 Jul 85 Mar 87 75.89 Apr 85 Jul 86 89.86 Feb 85 Aug 85 78.79 Mar 85 Apr 86 67.44 Jul 86 Mar 88 73.60 Apr 86 Jul 87 89.63 Feb 86 Aug 86 110.30 Mar 86 Apr 87 78.78 Jul 87 Mar 89 70.86 Apr 87 Jul 88 75.53 Feb 87 Aug 87 91.97 Mar 87 Apr 88 64.30 Jul 88 Mar 90 72.71 Apr 88 Jul 89 84.73 Feb 88 Aug 88 66.87 Mar 88 Apr 89 65.18 Jul 89 Mar 91 66.91 Apr 89 Jul 90 89.78 Feb 89 Aug 89 98.09 Mar 89 Apr 90 73.62 Jul 90 Mar 92 65.10 Apr 90 Jul 91 85.34 Feb 90 Aug 90 104.51 Mar 90 Apr 91 70.64 Jul 91 Mar 93 65.02 Apr 91 Jul 92 77.94 Feb 91 Aug 91 100.49 Mar 91 Apr 92 69.94 Jul 92 Mar 94 69.79 Apr 92 Jul 93 77.95 Feb 92 Aug 92 69.31 Mar 92 Apr 93 59.21 Jul 93 Mar 95 77.55 Apr 93 Jul 94 94.88 Feb 93 Aug 93 97.86 Mar 93 Apr 94 73.01 Jul 94 Mar 96 60.63 Apr 94 Jul 95 85.68 Feb 94 Aug 94 111.73 Mar 94 Apr 95 80.13 Jul 95 Mar 97 57.14 Apr 95 Jul 96 67.70 Feb 95 Aug 95 60.00 Mar 95 Apr 96 50.44 Jul 96 Mar 98 60.42 Apr 96 Jul 97 81.59 Feb 96 Aug 96 89.71 Mar 96 Apr 97 64.29 Jul 97 Mar 99 68.01 Apr 97 Jul 98 83.38 Feb 97 Aug 97 91.19 Mar 97 Apr 98 58.79 Jul 98 Mar 00 70.75 Apr 98 Jul 99 89.81 Feb 98 Aug 98 102.20 Mar 98 Apr 99 78.68 Jul 99 Mar 01 66.49 Apr 99 Jul 00 84.11 Feb 99 Aug 99 89.73 Mar 99 Apr 00 69.36 Jul 00 Mar 02 67.04 Apr 00 Jul 01 80.88 Feb 00 Aug 00 89.37 Mar 00 Apr 01 68.45 Jul 01 Mar 03 70.74 Apr 01 Jul 02 86.79 Feb 01 Aug 01 80.96 Mar 01 Apr 02 68.21 Jul 02 Mar 04 64.52 Apr 02 Jul 03 82.16 Feb 02 Aug 02 106.01 Mar 02 Apr 03 73.84 Jul 03 Mar 05 73.75 Apr 03 Jul 04 89.32 Feb 03 Aug 03 76.74 Mar 03 Apr 04 65.69 Jul 04 Mar 06 64.77 Apr 04 Jul 05 87.01 Feb 04 Aug 04 116.03 Mar 04 Apr 05 81.73 Jul 05 Mar 07 63.68 Apr 05 Jul 06 82.15 Feb 05 Aug 05 74.21 Mar 05 Apr 06 68.79 Jul 06 Mar 08 56.29 Apr 06 Jul 07 75.35 Feb 06 Aug 06 91.33 Mar 06 Apr 07 67.05 Jul 07 Mar 09 59.08 Apr 07 Jul 08 72.14 Feb 07 Aug 07 72.53 Mar 07 Apr 08 56.25 Jul 08 Mar 10 Apr 08 Jul 09 Feb 08 Aug 08 75.53 Mar 08 Apr 09 64.01 Average 67.13 Average 82.30 Average 87.89 Average 68.18 St, Dev. 6.29 St, Dev. 8.25 St, Dev. 15.63 St, Dev. 7.99 Deviation from mean 1.03 Deviation from mean 0.53 Deviation from mean 0.64 Deviation from mean 1.50 Source: Appendix authors' data. 84 World Bank Study Table A6.8. Analysis of El Niño Occurrences in Guavio River Discharges (1997­2007) TABLE 2. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG GUAVIO RIVER Abr. 97 Abr. 02 Jun. 04 Ago. 06 May. 98 Abr. 03 Mar. 05 Feb. 07 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Apr 63 May 64 63.31 Apr 63 Apr 64 60.12 Jun 63 Mar 64 55.07 Aug 63 Feb 64 44.83 Apr 64 May 65 72.89 Apr 64 Apr 65 68.37 Jun 64 Mar 65 65.95 Aug 64 Feb 65 26.30 Apr 65 May 66 68.76 Apr 65 Apr 66 71.16 Jun 65 Mar 66 68.78 Aug 65 Feb 66 17.20 Apr 66 May 67 55.79 Apr 66 Apr 67 52.96 Jun 66 Mar 67 54.88 Aug 66 Feb 67 30.40 Apr 67 May 68 71.22 Apr 67 Apr 68 71.11 Jun 67 Mar 68 66.63 Aug 67 Feb 68 13.75 Apr 68 May 69 74.78 Apr 68 Apr 69 73.92 Jun 68 Mar 69 72.64 Aug 68 Feb 69 17.00 Apr 69 May 70 70.75 Apr 69 Apr 70 68.19 Jun 69 Mar 70 63.99 Aug 69 Feb 70 27.25 Apr 70 May 71 88.02 Apr 70 Apr 71 85.55 Jun 70 Mar 71 82.06 Aug 70 Feb 71 27.10 Apr 71 May 72 91.50 Apr 71 Apr 72 86.42 Jun 71 Mar 72 82.82 Aug 71 Feb 72 41.45 Apr 72 May 73 76.71 Apr 72 Apr 73 75.02 Jun 72 Mar 73 69.98 Aug 72 Feb 73 12.25 Apr 73 May 74 76.01 Apr 73 Apr 74 72.72 Jun 73 Mar 74 73.57 Aug 73 Feb 74 18.95 Apr 74 May 75 68.72 Apr 74 Apr 75 66.57 Jun 74 Mar 75 61.98 Aug 74 Feb 75 10.55 Apr 75 May 76 79.56 Apr 75 Apr 76 73.17 Jun 75 Mar 76 73.55 Aug 75 Feb 76 18.90 Apr 76 May 77 80.13 Apr 76 Apr 77 81.25 Jun 76 Mar 77 78.28 Aug 76 Feb 77 13.60 Apr 77 May 78 63.75 Apr 77 Apr 78 61.97 Jun 77 Mar 78 63.85 Aug 77 Feb 78 12.15 Apr 78 May 79 63.44 Apr 78 Apr 79 62.76 Jun 78 Mar 79 59.79 Aug 78 Feb 79 9.75 Apr 79 May 80 66.56 Apr 79 Apr 80 66.22 Jun 79 Mar 80 65.09 Aug 79 Feb 80 15.85 Apr 80 May 81 65.73 Apr 80 Apr 81 62.51 Jun 80 Mar 81 61.18 Aug 80 Feb 81 13.45 Apr 81 May 82 72.69 Apr 81 Apr 82 70.11 Jun 81 Mar 82 63.00 Aug 81 Feb 82 17.40 Apr 82 May 83 85.02 Apr 82 Apr 83 83.65 Jun 82 Mar 83 76.36 Aug 82 Feb 83 38.75 Apr 83 May 84 79.00 Apr 83 Apr 84 78.18 Jun 83 Mar 84 76.96 Aug 83 Feb 84 46.05 Apr 84 May 85 74.46 Apr 84 Apr 85 71.91 Jun 84 Mar 85 76.24 Aug 84 Feb 85 11.05 Apr 85 May 86 72.19 Apr 85 Apr 86 71.53 Jun 85 Mar 86 72.91 Aug 85 Feb 86 21.30 Apr 86 May 87 81.07 Apr 86 Apr 87 81.78 Jun 86 Mar 87 89.00 Aug 86 Feb 87 22.45 Apr 87 May 88 67.74 Apr 87 Apr 88 67.54 Jun 87 Mar 88 72.91 Aug 87 Feb 88 13.05 Apr 88 May 89 74.24 Apr 88 Apr 89 69.38 Jun 88 Mar 89 73.11 Aug 88 Feb 89 26.65 Apr 89 May 90 81.60 Apr 89 Apr 90 74.98 Jun 89 Mar 90 69.63 Aug 89 Feb 90 22.10 Apr 90 May 91 72.36 Apr 90 Apr 91 70.81 Jun 90 Mar 91 62.57 Aug 90 Feb 91 16.25 Apr 91 May 92 71.50 Apr 91 Apr 92 73.25 Jun 91 Mar 92 76.71 Aug 91 Feb 92 12.90 Apr 92 May 93 65.30 Apr 92 Apr 93 62.74 Jun 92 Mar 93 63.53 Aug 92 Feb 93 14.95 Apr 93 May 94 81.11 Apr 93 Apr 94 75.67 Jun 93 Mar 94 72.74 Aug 93 Feb 94 14.15 Apr 94 May 95 84.00 Apr 94 Apr 95 83.85 Jun 94 Mar 95 83.13 Aug 94 Feb 95 13.55 Apr 95 May 96 57.55 Apr 95 Apr 96 52.17 Jun 95 Mar 96 49.24 Aug 95 Feb 96 25.75 Apr 96 May 97 71.45 Apr 96 Apr 97 66.82 Jun 96 Mar 97 62.65 Aug 96 Feb 97 23.90 Apr 97 May 98 66.70 Apr 97 Apr 98 61.93 Jun 97 Mar 98 56.73 Aug 97 Feb 98 11.35 Apr 98 May 99 85.03 Apr 98 Apr 99 83.49 Jun 98 Mar 99 77.69 Aug 98 Feb 99 28.25 Apr 99 May 00 77.70 Apr 99 Apr 00 72.42 Jun 99 Mar 00 66.67 Aug 99 Feb 00 23.75 Apr 00 May 01 74.14 Apr 00 Apr 01 71.73 Jun 00 Mar 01 70.02 Aug 00 Feb 01 15.95 Apr 01 May 02 76.74 Apr 01 Apr 02 71.95 Jun 01 Mar 02 69.77 Aug 01 Feb 02 16.20 Apr 02 May 03 80.08 Apr 02 Apr 03 76.43 Jun 02 Mar 03 70.08 Aug 02 Feb 03 10.20 Apr 03 May 04 75.61 Apr 03 Apr 04 68.75 Jun 03 Mar 04 62.41 Aug 03 Feb 04 16.10 Apr 04 May 05 86.50 Apr 04 Apr 05 84.02 Jun 04 Mar 05 76.66 Aug 04 Feb 05 20.45 Apr 05 May 06 76.09 Apr 05 Apr 06 72.65 Jun 05 Mar 06 62.54 Aug 05 Feb 06 16.35 Apr 06 May 07 69.01 Apr 06 Apr 07 67.68 Jun 06 Mar 07 57.59 Aug 06 Feb 07 10.80 Apr 07 May 08 60.02 Apr 07 Apr 08 58.74 Jun 07 Mar 08 58.73 Aug 07 Feb 08 17.30 Apr 08 May 09 Apr 08 Apr 09 67.76 Jun 08 Mar 09 71.00 Aug 08 Feb 09 19.95 Average 73.70 Average 71.13 Average 68.71 Average 19.95 St, Dev. 8.07 St, Dev. 8.06 St, Dev. 8.45 St, Dev. 9.00 Deviation from mean 0.87 Deviation from mean 0.66 Deviation from mean 0.94 Deviation from mean 1.02 Source: Appendix authors' data. Wind Energy in Colombia 85 Table A6.9. Analysis of El Niño Occurrences in Nare River Discharges (1986­1995) TABLE 1. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG NARE RIVER Jul. 86 Abr. 91 Feb. 93 Mar. 94 Mar. 88 Jul. 92 Ago. 93 Abr. 95 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Jul 55 Mar 57 Apr 55 Jul 56 Feb 55 Aug 55 Mar 55 Apr 56 Jul 56 Mar 58 46.34 Apr 56 Jul 57 56.21 Feb 56 Aug 56 61.69 Mar 56 Apr 57 57.79 Jul 57 Mar 59 32.95 Apr 57 Jul 58 37.60 Feb 57 Aug 57 40.71 Mar 57 Apr 58 37.63 Jul 58 Mar 60 33.64 Apr 58 Jul 59 31.68 Feb 58 Aug 58 31.56 Mar 58 Apr 59 30.28 Jul 59 Mar 61 40.42 Apr 59 Jul 60 38.61 Feb 59 Aug 59 30.00 Mar 59 Apr 60 35.31 Jul 60 Mar 62 42.88 Apr 60 Jul 61 41.01 Feb 60 Aug 60 37.51 Mar 60 Apr 61 40.84 Jul 61 Mar 63 52.44 Apr 61 Jul 62 48.93 Feb 61 Aug 61 33.60 Mar 61 Apr 62 41.99 Jul 62 Mar 64 49.32 Apr 62 Jul 63 56.69 Feb 62 Aug 62 55.69 Mar 62 Apr 63 56.65 Jul 63 Mar 65 44.83 Apr 63 Jul 64 46.53 Feb 63 Aug 63 49.24 Mar 63 Apr 64 45.92 Jul 64 Mar 66 45.08 Apr 64 Jul 65 43.83 Feb 64 Aug 64 40.91 Mar 64 Apr 65 43.26 Jul 65 Mar 67 46.64 Apr 65 Jul 66 43.56 Feb 65 Aug 65 35.59 Mar 65 Apr 66 41.83 Jul 66 Mar 68 47.91 Apr 66 Jul 67 50.76 Feb 66 Aug 66 39.31 Mar 66 Apr 67 46.55 Jul 67 Mar 69 46.77 Apr 67 Jul 68 48.89 Feb 67 Aug 67 50.07 Mar 67 Apr 68 45.64 Jul 68 Mar 70 48.86 Apr 68 Jul 69 49.41 Feb 68 Aug 68 46.21 Mar 68 Apr 69 49.92 Jul 69 Mar 71 54.32 Apr 69 Jul 70 48.59 Feb 69 Aug 69 38.64 Mar 69 Apr 70 44.57 Jul 70 Mar 72 66.16 Apr 70 Jul 71 65.99 Feb 70 Aug 70 43.51 Mar 70 Apr 71 58.11 Jul 71 Mar 73 55.01 Apr 71 Jul 72 67.64 Feb 71 Aug 71 76.86 Mar 71 Apr 72 66.13 Jul 72 Mar 74 48.54 Apr 72 Jul 73 45.06 Feb 72 Aug 72 52.69 Mar 72 Apr 73 45.02 Jul 73 Mar 75 60.69 Apr 73 Jul 74 56.09 Feb 73 Aug 73 36.21 Mar 73 Apr 74 52.97 Jul 74 Mar 76 61.30 Apr 74 Jul 75 59.13 Feb 74 Aug 74 55.84 Mar 74 Apr 75 57.80 Jul 75 Mar 77 53.27 Apr 75 Jul 76 61.31 Feb 75 Aug 75 50.69 Mar 75 Apr 76 60.09 Jul 76 Mar 78 39.04 Apr 76 Jul 77 41.66 Feb 76 Aug 76 49.01 Mar 76 Apr 77 41.51 Jul 77 Mar 79 49.93 Apr 77 Jul 78 50.83 Feb 77 Aug 77 35.10 Mar 77 Apr 78 44.46 Jul 78 Mar 80 48.50 Apr 78 Jul 79 53.37 Feb 78 Aug 78 60.59 Mar 78 Apr 79 52.84 Jul 79 Mar 81 42.76 Apr 79 Jul 80 46.91 Feb 79 Aug 79 44.39 Mar 79 Apr 80 48.06 Jul 80 Mar 82 51.01 Apr 80 Jul 81 46.15 Feb 80 Aug 80 33.06 Mar 80 Apr 81 37.21 Jul 81 Mar 83 49.59 Apr 81 Jul 82 60.96 Feb 81 Aug 81 53.56 Mar 81 Apr 82 58.96 Jul 82 Mar 84 39.28 Apr 82 Jul 83 44.58 Feb 82 Aug 82 52.61 Mar 82 Apr 83 44.44 Jul 83 Mar 85 49.38 Apr 83 Jul 84 46.66 Feb 83 Aug 83 38.23 Mar 83 Apr 84 40.25 Jul 84 Mar 86 53.20 Apr 84 Jul 85 54.26 Feb 84 Aug 84 51.06 Mar 84 Apr 85 52.94 Jul 85 Mar 87 46.37 Apr 85 Jul 86 50.79 Feb 85 Aug 85 48.39 Mar 85 Apr 86 50.66 Jul 86 Mar 88 43.10 Apr 86 Jul 87 42.14 Feb 86 Aug 86 44.23 Mar 86 Apr 87 42.37 Jul 87 Mar 89 57.32 Apr 87 Jul 88 46.19 Feb 87 Aug 87 36.36 Mar 87 Apr 88 42.99 Jul 88 Mar 90 61.80 Apr 88 Jul 89 62.64 Feb 88 Aug 88 48.97 Mar 88 Apr 89 61.71 Jul 89 Mar 91 48.79 Apr 89 Jul 90 52.78 Feb 89 Aug 89 50.71 Mar 89 Apr 90 52.76 Jul 90 Mar 92 41.68 Apr 90 Jul 91 46.19 Feb 90 Aug 90 41.71 Mar 90 Apr 91 44.55 Jul 91 Mar 93 35.89 Apr 91 Jul 92 37.53 Feb 91 Aug 91 40.36 Mar 91 Apr 92 37.89 Jul 92 Mar 94 42.95 Apr 92 Jul 93 37.80 Feb 92 Aug 92 30.19 Mar 92 Apr 93 35.01 Jul 93 Mar 95 46.40 Apr 93 Jul 94 48.09 Feb 93 Aug 93 38.97 Mar 93 Apr 94 46.84 Jul 94 Mar 96 49.49 Apr 94 Jul 95 46.95 Feb 94 Aug 94 41.31 Mar 94 Apr 95 42.69 Jul 95 Mar 97 57.65 Apr 95 Jul 96 60.15 Feb 95 Aug 95 52.27 Mar 95 Apr 96 53.49 Jul 96 Mar 98 41.42 Apr 96 Jul 97 55.28 Feb 96 Aug 96 64.27 Mar 96 Apr 97 57.90 Jul 97 Mar 99 43.61 Apr 97 Jul 98 34.03 Feb 97 Aug 97 40.31 Mar 97 Apr 98 31.94 Jul 98 Mar 00 68.16 Apr 98 Jul 99 60.18 Feb 98 Aug 98 35.40 Mar 98 Apr 99 55.04 Jul 99 Mar 01 70.02 Apr 99 Jul 00 75.00 Feb 99 Aug 99 67.37 Mar 99 Apr 00 71.34 Jul 00 Mar 02 49.22 Apr 00 Jul 01 62.89 Feb 00 Aug 00 74.29 Mar 00 Apr 01 66.10 Jul 01 Mar 03 38.28 Apr 01 Jul 02 41.34 Feb 01 Aug 01 37.43 Mar 01 Apr 02 38.95 Jul 02 Mar 04 40.88 Apr 02 Jul 03 43.82 Feb 02 Aug 02 42.76 Mar 02 Apr 03 40.02 Jul 03 Mar 05 51.20 Apr 03 Jul 04 48.73 Feb 03 Aug 03 43.21 Mar 03 Apr 04 45.25 Jul 04 Mar 06 57.63 Apr 04 Jul 05 60.48 Feb 04 Aug 04 47.91 Mar 04 Apr 05 55.87 Jul 05 Mar 07 54.55 Apr 05 Jul 06 60.24 Feb 05 Aug 05 53.67 Mar 05 Apr 06 56.35 Jul 06 Mar 08 61.07 Apr 06 Jul 07 59.86 Feb 06 Aug 06 57.03 Mar 06 Apr 07 55.89 Jul 07 Mar 09 73.96 Apr 07 Jul 08 73.12 Feb 07 Aug 07 56.59 Mar 07 Apr 08 67.00 Jul 08 Mar 10 Apr 08 Jul 09 Feb 08 Aug 08 76.61 Mar 08 Apr 09 75.74 Average 49.64 Average 50.94 Average 47.07 Average 49.19 St, Dev. 9.01 St, Dev. 9.68 St, Dev. 11.42 St, Dev. 10.14 Deviation from mean 0.73 Deviation from mean 1.39 Deviation from mean 0.71 Deviation from mean 0.64 Source: Appendix authors' data. 86 World Bank Study Table A6.10. Analysis of El Niño Occurrences in Nare River Discharges (1997­2007) TABLE 2. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG NARE RIVER Abr. 97 Abr. 02 Jun. 04 Ago. 06 May. 98 Abr. 03 Mar. 05 Feb. 07 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Apr 55 May 56 Apr 55 Apr 56 Jun 55 Mar 56 Aug 55 Feb 56 66.50 Apr 56 May 57 58.19 Apr 56 Apr 57 58.20 Jun 56 Mar 57 57.47 Aug 56 Feb 57 60.37 Apr 57 May 58 39.24 Apr 57 Apr 58 38.62 Jun 57 Mar 58 36.82 Aug 57 Feb 58 36.59 Apr 58 May 59 31.06 Apr 58 Apr 59 30.50 Jun 58 Mar 59 29.33 Aug 58 Feb 59 31.67 Apr 59 May 60 37.24 Apr 59 Apr 60 36.53 Jun 59 Mar 60 37.77 Aug 59 Feb 60 39.74 Apr 60 May 61 41.25 Apr 60 Apr 61 41.87 Jun 60 Mar 61 43.10 Aug 60 Feb 61 43.77 Apr 61 May 62 46.09 Apr 61 Apr 62 43.05 Jun 61 Mar 62 44.62 Aug 61 Feb 62 47.13 Apr 62 May 63 58.41 Apr 62 Apr 63 58.12 Jun 62 Mar 63 56.45 Aug 62 Feb 63 54.54 Apr 63 May 64 45.33 Apr 63 Apr 64 45.98 Jun 63 Mar 64 43.47 Aug 63 Feb 64 46.24 Apr 64 May 65 45.34 Apr 64 Apr 65 44.93 Jun 64 Mar 65 47.14 Aug 64 Feb 65 48.01 Apr 65 May 66 42.59 Apr 65 Apr 66 43.08 Jun 65 Mar 66 44.48 Aug 65 Feb 66 50.44 Apr 66 May 67 49.43 Apr 66 Apr 67 48.19 Jun 66 Mar 67 50.35 Aug 66 Feb 67 52.87 Apr 67 May 68 46.44 Apr 67 Apr 68 46.63 Jun 67 Mar 68 43.60 Aug 67 Feb 68 41.09 Apr 68 May 69 51.84 Apr 68 Apr 69 51.59 Jun 68 Mar 69 52.47 Aug 68 Feb 69 52.79 Apr 69 May 70 48.34 Apr 69 Apr 70 46.22 Jun 69 Mar 70 45.80 Aug 69 Feb 70 52.41 Apr 70 May 71 63.16 Apr 70 Apr 71 60.55 Jun 70 Mar 71 61.57 Aug 70 Feb 71 65.66 Apr 71 May 72 69.01 Apr 71 Apr 72 66.95 Jun 71 Mar 72 66.69 Aug 71 Feb 72 65.93 Apr 72 May 73 45.12 Apr 72 Apr 73 45.88 Jun 72 Mar 73 42.84 Aug 72 Feb 73 41.16 Apr 73 May 74 55.62 Apr 73 Apr 74 55.19 Jun 73 Mar 74 59.64 Aug 73 Feb 74 66.56 Apr 74 May 75 58.40 Apr 74 Apr 75 59.08 Jun 74 Mar 75 61.15 Aug 74 Feb 75 65.34 Apr 75 May 76 62.91 Apr 75 Apr 76 61.99 Jun 75 Mar 76 65.32 Aug 75 Feb 76 70.23 Apr 76 May 77 41.68 Apr 76 Apr 77 42.15 Jun 76 Mar 77 37.97 Aug 76 Feb 77 36.80 Apr 77 May 78 48.39 Apr 77 Apr 78 46.21 Jun 77 Mar 78 44.24 Aug 77 Feb 78 45.09 Apr 78 May 79 53.94 Apr 78 Apr 79 53.53 Jun 78 Mar 79 47.93 Aug 78 Feb 79 44.30 Apr 79 May 80 48.26 Apr 79 Apr 80 49.18 Jun 79 Mar 80 50.44 Aug 79 Feb 80 54.31 Apr 80 May 81 41.59 Apr 80 Apr 81 38.12 Jun 80 Mar 81 38.67 Aug 80 Feb 81 40.77 Apr 81 May 82 63.91 Apr 81 Apr 82 61.46 Jun 81 Mar 82 59.95 Aug 81 Feb 82 57.50 Apr 82 May 83 45.21 Apr 82 Apr 83 44.71 Jun 82 Mar 83 36.47 Aug 82 Feb 83 36.90 Apr 83 May 84 42.62 Apr 83 Apr 84 41.35 Jun 83 Mar 84 40.70 Aug 83 Feb 84 42.60 Apr 84 May 85 56.36 Apr 84 Apr 85 54.82 Jun 84 Mar 85 58.02 Aug 84 Feb 85 56.33 Apr 85 May 86 51.51 Apr 85 Apr 86 51.78 Jun 85 Mar 86 50.03 Aug 85 Feb 86 54.21 Apr 86 May 87 43.39 Apr 86 Apr 87 42.43 Jun 86 Mar 87 41.73 Aug 86 Feb 87 43.03 Apr 87 May 88 44.26 Apr 87 Apr 88 44.40 Jun 87 Mar 88 45.13 Aug 87 Feb 88 50.60 Apr 88 May 89 63.98 Apr 88 Apr 89 64.13 Jun 88 Mar 89 70.50 Aug 88 Feb 89 77.30 Apr 89 May 90 53.56 Apr 89 Apr 90 53.39 Jun 89 Mar 90 53.93 Aug 89 Feb 90 57.53 Apr 90 May 91 46.35 Apr 90 Apr 91 45.66 Jun 90 Mar 91 45.21 Aug 90 Feb 91 46.77 Apr 91 May 92 38.59 Apr 91 Apr 92 38.48 Jun 91 Mar 92 37.52 Aug 91 Feb 92 37.30 Apr 92 May 93 37.71 Apr 92 Apr 93 35.86 Jun 92 Mar 93 36.12 Aug 92 Feb 93 38.96 Apr 93 May 94 48.81 Apr 93 Apr 94 48.27 Jun 93 Mar 94 47.68 Aug 93 Feb 94 52.66 Apr 94 May 95 44.73 Apr 94 Apr 95 43.57 Jun 94 Mar 95 42.36 Aug 94 Feb 95 43.43 Apr 95 May 96 57.36 Apr 95 Apr 96 55.02 Jun 95 Mar 96 56.77 Aug 95 Feb 96 55.47 Apr 96 May 97 56.57 Apr 96 Apr 97 58.17 Jun 96 Mar 97 56.75 Aug 96 Feb 97 52.54 Apr 97 May 98 32.26 Apr 97 Apr 98 31.30 Jun 97 Mar 98 29.76 Aug 97 Feb 98 26.61 Apr 98 May 99 59.91 Apr 98 Apr 99 57.82 Jun 98 Mar 99 61.07 Aug 98 Feb 99 62.09 Apr 99 May 00 71.74 Apr 99 Apr 00 70.42 Jun 99 Mar 00 70.61 Aug 99 Feb 00 75.60 Apr 00 May 01 65.63 Apr 00 Apr 01 67.12 Jun 00 Mar 01 69.88 Aug 00 Feb 01 66.61 Apr 01 May 02 40.43 Apr 01 Apr 02 39.11 Jun 01 Mar 02 36.28 Aug 01 Feb 02 35.47 Apr 02 May 03 41.56 Apr 02 Apr 03 41.02 Jun 02 Mar 03 36.45 Aug 02 Feb 03 33.94 Apr 03 May 04 47.94 Apr 03 Apr 04 46.32 Jun 03 Mar 04 45.68 Aug 03 Feb 04 43.21 Apr 04 May 05 59.60 Apr 04 Apr 05 57.47 Jun 04 Mar 05 57.78 Aug 04 Feb 05 61.81 Apr 05 May 06 61.15 Apr 05 Apr 06 57.87 Jun 05 Mar 06 54.53 Aug 05 Feb 06 53.16 Apr 06 May 07 59.99 Apr 06 Apr 07 57.12 Jun 06 Mar 07 50.64 Aug 06 Feb 07 52.66 Apr 07 May 08 71.09 Apr 07 Apr 08 69.84 Jun 07 Mar 08 68.54 Aug 07 Feb 08 72.04 Apr 08 May 09 Apr 08 Apr 09 76.71 Jun 08 Mar 09 78.85 Aug 08 Feb 09 81.06 Average 50.68 Average 50.34 Average 49.97 Average 51.62 St, Dev. 9.93 St, Dev. 10.39 St, Dev. 11.42 St, Dev. 12.43 Deviation from mean 1.86 Deviation from mean 0.90 Deviation from mean 0.68 Deviation from mean 0.08 Source: Appendix authors' data. Wind Energy in Colombia 87 Table A6.11. Analysis of El Niño Occurrences in Cauca River Discharges (1986­1995) TABLE 1. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG CAUCA RIVER Jul. 86 Abr. 91 Feb. 93 Mar. 94 Mar. 88 Jul. 92 Ago. 93 Abr. 95 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Jul 46 Mar 48 Apr 46 Jul 47 Feb 46 Aug 46 Mar 46 Apr 47 Jul 47 Mar 49 122.05 Apr 47 Jul 48 120.06 Feb 47 Aug 47 83.00 Mar 47 Apr 48 119.43 Jul 48 Mar 50 152.67 Apr 48 Jul 49 118.81 Feb 48 Aug 48 115.14 Mar 48 Apr 49 119.00 Jul 49 Mar 51 219.10 Apr 49 Jul 50 215.75 Feb 49 Aug 49 122.29 Mar 49 Apr 50 192.93 Jul 50 Mar 52 160.95 Apr 50 Jul 51 205.88 Feb 50 Aug 50 295.57 Mar 50 Apr 51 232.36 Jul 51 Mar 53 125.95 Apr 51 Jul 52 132.94 Feb 51 Aug 51 144.86 Mar 51 Apr 52 137.86 Jul 52 Mar 54 133.10 Apr 52 Jul 53 118.63 Feb 52 Aug 52 123.14 Mar 52 Apr 53 121.86 Jul 53 Mar 55 167.52 Apr 53 Jul 54 148.56 Feb 53 Aug 53 104.00 Mar 53 Apr 54 148.14 Jul 54 Mar 56 190.62 Apr 54 Jul 55 177.75 Feb 54 Aug 54 144.14 Mar 54 Apr 55 176.00 Jul 55 Mar 57 170.10 Apr 55 Jul 56 184.56 Feb 55 Aug 55 163.57 Mar 55 Apr 56 192.64 Jul 56 Mar 58 124.86 Apr 56 Jul 57 147.88 Feb 56 Aug 56 152.00 Mar 56 Apr 57 149.07 Jul 57 Mar 59 92.10 Apr 57 Jul 58 105.13 Feb 57 Aug 57 125.43 Mar 57 Apr 58 109.86 Jul 58 Mar 60 110.38 Apr 58 Jul 59 97.00 Feb 58 Aug 58 84.71 Mar 58 Apr 59 90.29 Jul 59 Mar 61 117.76 Apr 59 Jul 60 119.94 Feb 59 Aug 59 91.43 Mar 59 Apr 60 120.50 Jul 60 Mar 62 111.62 Apr 60 Jul 61 109.75 Feb 60 Aug 60 114.86 Mar 60 Apr 61 112.00 Jul 61 Mar 63 127.29 Apr 61 Jul 62 117.63 Feb 61 Aug 61 98.86 Mar 61 Apr 62 111.64 Jul 62 Mar 64 130.52 Apr 62 Jul 63 144.38 Feb 62 Aug 62 122.29 Mar 62 Apr 63 144.36 Jul 63 Mar 65 123.86 Apr 63 Jul 64 131.25 Feb 63 Aug 63 159.29 Mar 63 Apr 64 129.00 Jul 64 Mar 66 120.67 Apr 64 Jul 65 129.81 Feb 64 Aug 64 121.14 Mar 64 Apr 65 132.43 Jul 65 Mar 67 149.76 Apr 65 Jul 66 111.06 Feb 65 Aug 65 94.00 Mar 65 Apr 66 108.14 Jul 66 Mar 68 161.86 Apr 66 Jul 67 168.13 Feb 66 Aug 66 94.86 Mar 66 Apr 67 168.07 Jul 67 Mar 69 130.33 Apr 67 Jul 68 134.88 Feb 67 Aug 67 147.00 Mar 67 Apr 68 139.36 Jul 68 Mar 70 137.00 Apr 68 Jul 69 136.63 Feb 68 Aug 68 133.86 Mar 68 Apr 69 137.29 Jul 69 Mar 71 161.86 Apr 69 Jul 70 137.31 Feb 69 Aug 69 128.14 Mar 69 Apr 70 137.50 Jul 70 Mar 72 179.00 Apr 70 Jul 71 180.69 Feb 70 Aug 70 131.57 Mar 70 Apr 71 186.93 Jul 71 Mar 73 123.00 Apr 71 Jul 72 156.75 Feb 71 Aug 71 179.57 Mar 71 Apr 72 167.00 Jul 72 Mar 74 143.10 Apr 72 Jul 73 100.56 Feb 72 Aug 72 133.43 Mar 72 Apr 73 102.21 Jul 73 Mar 75 178.90 Apr 73 Jul 74 171.50 Feb 73 Aug 73 89.71 Mar 73 Apr 74 173.36 Jul 74 Mar 76 174.43 Apr 74 Jul 75 151.69 Feb 74 Aug 74 159.86 Mar 74 Apr 75 157.21 Jul 75 Mar 77 145.24 Apr 75 Jul 76 178.94 Feb 75 Aug 75 155.29 Mar 75 Apr 76 187.86 Jul 76 Mar 78 97.76 Apr 76 Jul 77 102.69 Feb 76 Aug 76 152.14 Mar 76 Apr 77 110.29 Jul 77 Mar 79 111.86 Apr 77 Jul 78 108.19 Feb 77 Aug 77 77.14 Mar 77 Apr 78 104.71 Jul 78 Mar 80 123.76 Apr 78 Jul 79 119.25 Feb 78 Aug 78 92.00 Mar 78 Apr 79 112.71 Jul 79 Mar 81 118.10 Apr 79 Jul 80 124.44 Feb 79 Aug 79 118.00 Mar 79 Apr 80 132.64 Jul 80 Mar 82 142.24 Apr 80 Jul 81 128.19 Feb 80 Aug 80 105.14 Mar 80 Apr 81 119.00 Jul 81 Mar 83 144.95 Apr 81 Jul 82 167.63 Feb 81 Aug 81 151.71 Mar 81 Apr 82 168.86 Jul 82 Mar 84 132.19 Apr 82 Jul 83 141.88 Feb 82 Aug 82 171.29 Mar 82 Apr 83 153.00 Jul 83 Mar 85 155.00 Apr 83 Jul 84 148.25 Feb 83 Aug 83 128.57 Mar 83 Apr 84 141.64 Jul 84 Mar 86 157.14 Apr 84 Jul 85 163.19 Feb 84 Aug 84 158.29 Mar 84 Apr 85 169.21 Jul 85 Mar 87 130.90 Apr 85 Jul 86 147.00 Feb 85 Aug 85 117.14 Mar 85 Apr 86 144.50 Jul 86 Mar 88 96.57 Apr 86 Jul 87 110.50 Feb 86 Aug 86 148.14 Mar 86 Apr 87 120.00 Jul 87 Mar 89 139.76 Apr 87 Jul 88 98.63 Feb 87 Aug 87 80.29 Mar 87 Apr 88 89.14 Jul 88 Mar 90 158.19 Apr 88 Jul 89 167.38 Feb 88 Aug 88 101.86 Mar 88 Apr 89 167.71 Jul 89 Mar 91 117.14 Apr 89 Jul 90 127.88 Feb 89 Aug 89 147.57 Mar 89 Apr 90 135.07 Jul 90 Mar 92 107.52 Apr 90 Jul 91 116.38 Feb 90 Aug 90 129.00 Mar 90 Apr 91 118.36 Jul 91 Mar 93 101.48 Apr 91 Jul 92 103.06 Feb 91 Aug 91 111.86 Mar 91 Apr 92 110.07 Jul 92 Mar 94 122.95 Apr 92 Jul 93 105.63 Feb 92 Aug 92 78.71 Mar 92 Apr 93 99.86 Jul 93 Mar 95 125.52 Apr 93 Jul 94 136.75 Feb 93 Aug 93 121.86 Mar 93 Apr 94 140.43 Jul 94 Mar 96 129.10 Apr 94 Jul 95 120.63 Feb 94 Aug 94 134.57 Mar 94 Apr 95 123.86 Jul 95 Mar 97 149.90 Apr 95 Jul 96 144.25 Feb 95 Aug 95 106.29 Mar 95 Apr 96 141.50 Jul 96 Mar 98 115.71 Apr 96 Jul 97 146.75 Feb 96 Aug 96 152.00 Mar 96 Apr 97 154.50 Jul 97 Mar 99 124.76 Apr 97 Jul 98 94.31 Feb 97 Aug 97 131.86 Mar 97 Apr 98 92.43 Jul 98 Mar 00 192.90 Apr 98 Jul 99 161.88 Feb 98 Aug 98 91.57 Mar 98 Apr 99 162.07 Jul 99 Mar 01 166.79 Apr 99 Jul 00 196.44 Feb 99 Aug 99 177.14 Mar 99 Apr 00 207.86 Jul 00 Mar 02 103.21 Apr 00 Jul 01 119.82 Feb 00 Aug 00 186.86 Mar 00 Apr 01 136.77 Jul 01 Mar 03 92.33 Apr 01 Jul 02 102.17 Feb 01 Aug 01 88.36 Mar 01 Apr 02 103.56 Jul 02 Mar 04 92.67 Apr 02 Jul 03 94.24 Feb 02 Aug 02 97.04 Mar 02 Apr 03 90.42 Jul 03 Mar 05 118.19 Apr 03 Jul 04 105.12 Feb 03 Aug 03 91.73 Mar 03 Apr 04 106.50 Jul 04 Mar 06 136.31 Apr 04 Jul 05 122.75 Feb 04 Aug 04 88.13 Mar 04 Apr 05 125.39 Jul 05 Mar 07 131.57 Apr 05 Jul 06 141.19 Feb 05 Aug 05 119.00 Mar 05 Apr 06 144.29 Jul 06 Mar 08 144.85 Apr 06 Jul 07 132.30 Feb 06 Aug 06 148.51 Mar 06 Apr 07 133.14 Jul 07 Mar 09 176.39 Apr 07 Jul 08 176.43 Feb 07 Aug 07 130.51 Mar 07 Apr 08 172.94 Jul 08 Mar 10 Apr 08 Jul 09 Feb 08 Aug 08 179.73 Mar 08 Apr 09 182.54 Average 136.78 Average 136.54 Average 127.86 Average 139.02 St, Dev. 27.21 St, Dev. 29.27 St, Dev. 36.02 St, Dev. 31.37 Deviation from mean 1.48 Deviation from mean 1.14 Deviation from mean 0.17 Deviation from mean 0.48 Source: Appendix authors' data. 88 World Bank Study Table A6.12. Analysis of El Niño" Occurrences in Cauca River Flows (1997­2007) TABLE 2. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG CAUCA RIVER Abr. 97 Abr. 02 Jun. 04 Ago. 06 May. 98 Abr. 03 Mar. 05 Feb. 07 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Apr 46 May 47 Apr 46 Apr 47 Jun 46 Mar 47 Aug 46 Feb 47 Apr 47 May 48 123.71 Apr 47 Apr 48 122.85 Jun 47 Mar 48 126.30 Aug 47 Feb 48 132.86 Apr 48 May 49 120.14 Apr 48 Apr 49 117.08 Jun 48 Mar 49 105.90 Aug 48 Feb 49 107.14 Apr 49 May 50 211.14 Apr 49 Apr 50 198.54 Jun 49 Mar 50 197.70 Aug 49 Feb 50 197.86 Apr 50 May 51 217.86 Apr 50 Apr 51 221.54 Jun 50 Mar 51 205.30 Aug 50 Feb 51 196.14 Apr 51 May 52 137.07 Apr 51 Apr 52 134.31 Jun 51 Mar 52 131.60 Aug 51 Feb 52 135.29 Apr 52 May 53 122.07 Apr 52 Apr 53 121.62 Jun 52 Mar 53 115.60 Aug 52 Feb 53 121.14 Apr 53 May 54 151.29 Apr 53 Apr 54 151.85 Jun 53 Mar 54 150.00 Aug 53 Feb 54 169.43 Apr 54 May 55 180.86 Apr 54 Apr 55 179.92 Jun 54 Mar 55 176.20 Aug 54 Feb 55 184.71 Apr 55 May 56 188.21 Apr 55 Apr 56 191.31 Jun 55 Mar 56 195.80 Aug 55 Feb 56 210.57 Apr 56 May 57 150.79 Apr 56 Apr 57 147.31 Jun 56 Mar 57 150.90 Aug 56 Feb 57 151.43 Apr 57 May 58 108.29 Apr 57 Apr 58 108.23 Jun 57 Mar 58 98.10 Aug 57 Feb 58 94.00 Apr 58 May 59 94.00 Apr 58 Apr 59 92.00 Jun 58 Mar 59 91.50 Aug 58 Feb 59 98.00 Apr 59 May 60 124.57 Apr 59 Apr 60 124.92 Jun 59 Mar 60 129.60 Aug 59 Feb 60 134.14 Apr 60 May 61 109.71 Apr 60 Apr 61 111.31 Jun 60 Mar 61 105.50 Aug 60 Feb 61 114.57 Apr 61 May 62 116.57 Apr 61 Apr 62 114.23 Jun 61 Mar 62 114.50 Aug 61 Feb 62 111.29 Apr 62 May 63 148.86 Apr 62 Apr 63 144.23 Jun 62 Mar 63 139.10 Aug 62 Feb 63 141.29 Apr 63 May 64 128.21 Apr 63 Apr 64 127.23 Jun 63 Mar 64 107.50 Aug 63 Feb 64 111.29 Apr 64 May 65 137.57 Apr 64 Apr 65 137.23 Jun 64 Mar 65 137.50 Aug 64 Feb 65 141.71 Apr 65 May 66 111.86 Apr 65 Apr 66 110.46 Jun 65 Mar 66 107.90 Aug 65 Feb 66 121.00 Apr 66 May 67 172.07 Apr 66 Apr 67 174.77 Jun 66 Mar 67 192.10 Aug 66 Feb 67 217.43 Apr 67 May 68 134.86 Apr 67 Apr 68 135.62 Jun 67 Mar 68 131.70 Aug 67 Feb 68 128.71 Apr 68 May 69 140.50 Apr 68 Apr 69 137.46 Jun 68 Mar 69 127.80 Aug 68 Feb 69 131.43 Apr 69 May 70 141.64 Apr 69 Apr 70 141.31 Jun 69 Mar 70 134.50 Aug 69 Feb 70 132.57 Apr 70 May 71 187.93 Apr 70 Apr 71 186.08 Jun 70 Mar 71 193.00 Aug 70 Feb 71 212.43 Apr 71 May 72 161.07 Apr 71 Apr 72 162.23 Jun 71 Mar 72 151.40 Aug 71 Feb 72 158.57 Apr 72 May 73 99.36 Apr 72 Apr 73 99.00 Jun 72 Mar 73 91.50 Aug 72 Feb 73 87.00 Apr 73 May 74 180.07 Apr 73 Apr 74 182.62 Jun 73 Mar 74 204.30 Aug 73 Feb 74 225.71 Apr 74 May 75 152.93 Apr 74 Apr 75 150.46 Jun 74 Mar 75 153.20 Aug 74 Feb 75 160.14 Apr 75 May 76 186.21 Apr 75 Apr 76 187.85 Jun 75 Mar 76 192.40 Aug 75 Feb 76 206.71 Apr 76 May 77 104.93 Apr 76 Apr 77 104.08 Jun 76 Mar 77 90.40 Aug 76 Feb 77 85.14 Apr 77 May 78 110.93 Apr 77 Apr 78 108.77 Jun 77 Mar 78 107.50 Aug 77 Feb 78 119.86 Apr 78 May 79 118.86 Apr 78 Apr 79 116.62 Jun 78 Mar 79 109.40 Aug 78 Feb 79 110.14 Apr 79 May 80 129.79 Apr 79 Apr 80 131.69 Jun 79 Mar 80 131.90 Aug 79 Feb 80 135.29 Apr 80 May 81 127.14 Apr 80 Apr 81 118.31 Jun 80 Mar 81 111.40 Aug 80 Feb 81 112.29 Apr 81 May 82 174.21 Apr 81 Apr 82 170.00 Jun 81 Mar 82 151.80 Aug 81 Feb 82 144.57 Apr 82 May 83 148.79 Apr 82 Apr 83 146.69 Jun 82 Mar 83 121.20 Aug 82 Feb 83 121.71 Apr 83 May 84 148.50 Apr 83 Apr 84 143.54 Jun 83 Mar 84 128.50 Aug 83 Feb 84 135.14 Apr 84 May 85 170.00 Apr 84 Apr 85 170.54 Jun 84 Mar 85 170.00 Aug 84 Feb 85 188.43 Apr 85 May 86 148.50 Apr 85 Apr 86 148.85 Jun 85 Mar 86 148.00 Aug 85 Feb 86 150.14 Apr 86 May 87 115.14 Apr 86 Apr 87 114.00 Jun 86 Mar 87 109.30 Aug 86 Feb 87 108.57 Apr 87 May 88 92.07 Apr 87 Apr 88 91.38 Jun 87 Mar 88 86.80 Aug 87 Feb 88 91.57 Apr 88 May 89 174.93 Apr 88 Apr 89 175.15 Jun 88 Mar 89 194.20 Aug 88 Feb 89 204.29 Apr 89 May 90 132.36 Apr 89 Apr 90 128.31 Jun 89 Mar 90 120.10 Aug 89 Feb 90 120.00 Apr 90 May 91 117.93 Apr 90 Apr 91 117.31 Jun 90 Mar 91 102.90 Aug 90 Feb 91 101.57 Apr 91 May 92 106.43 Apr 91 Apr 92 108.92 Jun 91 Mar 92 106.40 Aug 91 Feb 92 112.00 Apr 92 May 93 106.79 Apr 92 Apr 93 102.23 Jun 92 Mar 93 100.80 Aug 92 Feb 93 101.43 Apr 93 May 94 141.14 Apr 93 Apr 94 140.54 Jun 93 Mar 94 131.00 Aug 93 Feb 94 137.43 Apr 94 May 95 123.50 Apr 94 Apr 95 121.62 Jun 94 Mar 95 110.00 Aug 94 Feb 95 114.14 Apr 95 May 96 145.43 Apr 95 Apr 96 145.54 Jun 95 Mar 96 143.40 Aug 95 Feb 96 146.71 Apr 96 May 97 148.71 Apr 96 Apr 97 150.54 Jun 96 Mar 97 151.30 Aug 96 Feb 97 157.71 Apr 97 May 98 91.86 Apr 97 Apr 98 89.00 Jun 97 Mar 98 78.30 Aug 97 Feb 98 66.00 Apr 98 May 99 169.57 Apr 98 Apr 99 170.31 Jun 98 Mar 99 172.00 Aug 98 Feb 99 180.43 Apr 99 May 00 205.29 Apr 99 Apr 00 205.85 Jun 99 Mar 00 205.30 Aug 99 Feb 00 225.14 Apr 00 May 01 124.74 Apr 00 Apr 01 127.22 Jun 00 Mar 01 115.95 Aug 00 Feb 01 109.57 Apr 01 May 02 103.11 Apr 01 Apr 02 102.03 Jun 01 Mar 02 98.72 Aug 01 Feb 02 105.20 Apr 02 May 03 92.83 Apr 02 Apr 03 91.22 Jun 02 Mar 03 77.57 Aug 02 Feb 03 71.90 Apr 03 May 04 108.64 Apr 03 Apr 04 108.44 Jun 03 Mar 04 104.03 Aug 03 Feb 04 108.37 Apr 04 May 05 129.41 Apr 04 Apr 05 129.38 Jun 04 Mar 05 132.23 Aug 04 Feb 05 139.90 Apr 05 May 06 144.29 Apr 05 Apr 06 141.38 Jun 05 Mar 06 137.94 Aug 05 Feb 06 147.89 Apr 06 May 07 133.75 Apr 06 Apr 07 128.62 Jun 06 Mar 07 108.26 Aug 06 Feb 07 103.66 Apr 07 May 08 179.88 Apr 07 Apr 08 177.17 Jun 07 Mar 08 170.51 Aug 07 Feb 08 178.63 Apr 08 May 09 Apr 08 Apr 09 180.40 Jun 08 Mar 09 175.81 Aug 08 Feb 09 181.46 Average 139.49 Average 139.02 Average 134.86 Average 139.53 St, Dev. 31.07 St, Dev. 31.54 St, Dev. 35.09 St, Dev. 39.82 Deviation from mean 1.53 Deviation from mean 1.52 Deviation from mean 0.07 Deviation from mean 0.90 Source: Appendix authors' data. Wind Energy in Colombia 89 Table A6.13. Analysis of El Niño occurrences in Magdalena River discharges (1986­ 1995) TABLE 1. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG MAGDALENA RIVER Jul. 86 Abr. 91 Feb. 93 Mar. 94 Mar. 88 Jul. 92 Ago. 93 Abr. 95 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Jul 72 Mar 74 198.51 Apr 72 Jul 73 161.18 Feb 72 Aug 72 215.87 Mar 72 Apr 73 158.22 Jul 73 Mar 75 256.77 Apr 73 Jul 74 261.46 Feb 73 Aug 73 152.23 Mar 73 Apr 74 248.98 Jul 74 Mar 76 247.04 Apr 74 Jul 75 236.34 Feb 74 Aug 74 297.01 Mar 74 Apr 75 236.84 Jul 75 Mar 77 253.34 Apr 75 Jul 76 285.58 Feb 75 Aug 75 234.67 Mar 75 Apr 76 273.19 Jul 76 Mar 78 193.41 Apr 76 Jul 77 232.01 Feb 76 Aug 76 304.99 Mar 76 Apr 77 238.92 Jul 77 Mar 79 168.98 Apr 77 Jul 78 190.93 Feb 77 Aug 77 176.37 Mar 77 Apr 78 183.85 Jul 78 Mar 80 183.66 Apr 78 Jul 79 193.09 Feb 78 Aug 78 178.44 Mar 78 Apr 79 174.26 Jul 79 Mar 81 181.18 Apr 79 Jul 80 215.63 Feb 79 Aug 79 218.23 Mar 79 Apr 80 214.96 Jul 80 Mar 82 208.41 Apr 80 Jul 81 202.89 Feb 80 Aug 80 204.24 Mar 80 Apr 81 179.19 Jul 81 Mar 83 224.60 Apr 81 Jul 82 261.77 Feb 81 Aug 81 223.89 Mar 81 Apr 82 247.41 Jul 82 Mar 84 191.81 Apr 82 Jul 83 215.84 Feb 82 Aug 82 283.74 Mar 82 Apr 83 229.51 Jul 83 Mar 85 209.93 Apr 83 Jul 84 207.09 Feb 83 Aug 83 182.39 Mar 83 Apr 84 193.59 Jul 84 Mar 86 226.24 Apr 84 Jul 85 235.73 Feb 84 Aug 84 226.76 Mar 84 Apr 85 220.95 Jul 85 Mar 87 220.01 Apr 85 Jul 86 245.43 Feb 85 Aug 85 208.51 Mar 85 Apr 86 219.66 Jul 86 Mar 88 171.06 Apr 86 Jul 87 208.83 Feb 86 Aug 86 285.93 Mar 86 Apr 87 225.78 Jul 87 Mar 89 177.14 Apr 87 Jul 88 167.64 Feb 87 Aug 87 156.34 Mar 87 Apr 88 141.64 Jul 88 Mar 90 187.12 Apr 88 Jul 89 208.44 Feb 88 Aug 88 173.71 Mar 88 Apr 89 194.78 Jul 89 Mar 91 165.73 Apr 89 Jul 90 186.99 Feb 89 Aug 89 194.80 Mar 89 Apr 90 172.93 Jul 90 Mar 92 159.59 Apr 90 Jul 91 182.13 Feb 90 Aug 90 211.97 Mar 90 Apr 91 172.99 Jul 91 Mar 93 153.53 Apr 91 Jul 92 165.45 Feb 91 Aug 91 184.37 Mar 91 Apr 92 162.81 Jul 92 Mar 94 186.16 Apr 92 Jul 93 169.35 Feb 92 Aug 92 140.97 Mar 92 Apr 93 147.02 Jul 93 Mar 95 211.82 Apr 93 Jul 94 244.79 Feb 93 Aug 93 207.71 Mar 93 Apr 94 220.15 Jul 94 Mar 96 179.88 Apr 94 Jul 95 218.29 Feb 94 Aug 94 296.24 Mar 94 Apr 95 226.53 Jul 95 Mar 97 186.36 Apr 95 Jul 96 199.99 Feb 95 Aug 95 158.97 Mar 95 Apr 96 180.45 Jul 96 Mar 98 151.98 Apr 96 Jul 97 202.57 Feb 96 Aug 96 248.69 Mar 96 Apr 97 200.88 Jul 97 Mar 99 161.50 Apr 97 Jul 98 150.57 Feb 97 Aug 97 186.06 Mar 97 Apr 98 128.23 Jul 98 Mar 00 209.02 Apr 98 Jul 99 210.25 Feb 98 Aug 98 157.86 Mar 98 Apr 99 199.95 Jul 99 Mar 01 177.97 Apr 99 Jul 00 220.76 Feb 99 Aug 99 237.87 Mar 99 Apr 00 220.04 Jul 00 Mar 02 130.18 Apr 00 Jul 01 159.87 Feb 00 Aug 00 233.33 Mar 00 Apr 01 164.61 Jul 01 Mar 03 138.89 Apr 01 Jul 02 160.18 Feb 01 Aug 01 157.27 Mar 01 Apr 02 139.78 Jul 02 Mar 04 133.77 Apr 02 Jul 03 160.73 Feb 02 Aug 02 192.90 Mar 02 Apr 03 153.91 Jul 03 Mar 05 143.48 Apr 03 Jul 04 148.24 Feb 03 Aug 03 145.21 Mar 03 Apr 04 141.28 Jul 04 Mar 06 157.62 Apr 04 Jul 05 155.63 Feb 04 Aug 04 145.46 Mar 04 Apr 05 151.70 Jul 05 Mar 07 176.31 Apr 05 Jul 06 190.83 Feb 05 Aug 05 158.74 Mar 05 Apr 06 175.89 Jul 06 Mar 08 205.89 Apr 06 Jul 07 209.21 Feb 06 Aug 06 233.24 Mar 06 Apr 07 196.84 Jul 07 Mar 09 253.50 Apr 07 Jul 08 255.37 Feb 07 Aug 07 212.16 Mar 07 Apr 08 229.54 Jul 08 Mar 10 Apr 08 Jul 09 Feb 08 Aug 08 261.97 Mar 08 Apr 09 274.84 Average 188.40 Average 203.36 Average 207.81 Average 195.73 St, Dev. 33.89 St, Dev. 35.52 St, Dev. 46.66 St, Dev. 38.70 Deviation from mean 0.51 Deviation from mean 1.07 Deviation from mean 0.00 Deviation from mean 0.80 Source: Appendix authors' data. 90 World Bank Study Table A6.14. Analysis of El Niño Occurrences in Magdalena River Discharges (1997­2007) TABLE 2. ANALYSIS OF "EL NIÑO" OCCURRENCES RIVER DISCHARGES IN M3/SEG MAGDALENA RIVER Abr. 97 Abr. 02 Jun. 04 Ago. 06 May. 98 Abr. 03 Mar. 05 Feb. 07 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Apr 72 May 73 155.96 Apr 72 Apr 73 154.37 Jun 72 Mar 73 131.63 Aug 72 Feb 73 110.30 Apr 73 May 74 263.41 Apr 73 Apr 74 262.03 Jun 73 Mar 74 279.30 Aug 73 Feb 74 288.46 Apr 74 May 75 231.73 Apr 74 Apr 75 225.95 Jun 74 Mar 75 219.19 Aug 74 Feb 75 209.10 Apr 75 May 76 280.36 Apr 75 Apr 76 276.30 Jun 75 Mar 76 275.74 Aug 75 Feb 76 277.21 Apr 76 May 77 234.94 Apr 76 Apr 77 235.78 Jun 76 Mar 77 215.10 Aug 76 Feb 77 199.49 Apr 77 May 78 190.36 Apr 77 Apr 78 189.50 Jun 77 Mar 78 175.23 Aug 77 Feb 78 174.39 Apr 78 May 79 183.64 Apr 78 Apr 79 179.32 Jun 78 Mar 79 155.50 Aug 78 Feb 79 131.64 Apr 79 May 80 211.74 Apr 79 Apr 80 212.75 Jun 79 Mar 80 198.14 Aug 79 Feb 80 187.31 Apr 80 May 81 198.64 Apr 80 Apr 81 181.30 Jun 80 Mar 81 169.86 Aug 80 Feb 81 150.37 Apr 81 May 82 261.22 Apr 81 Apr 82 254.11 Jun 81 Mar 82 232.97 Aug 81 Feb 82 227.03 Apr 82 May 83 227.59 Apr 82 Apr 83 225.99 Jun 82 Mar 83 193.28 Aug 82 Feb 83 178.54 Apr 83 May 84 201.66 Apr 83 Apr 84 196.80 Jun 83 Mar 84 176.30 Aug 83 Feb 84 192.64 Apr 84 May 85 225.25 Apr 84 Apr 85 226.62 Jun 84 Mar 85 227.59 Aug 84 Feb 85 240.34 Apr 85 May 86 225.78 Apr 85 Apr 86 228.61 Jun 85 Mar 86 232.66 Aug 85 Feb 86 193.00 Apr 86 May 87 216.29 Apr 86 Apr 87 215.65 Jun 86 Mar 87 215.15 Aug 86 Feb 87 185.46 Apr 87 May 88 146.08 Apr 87 Apr 88 145.81 Jun 87 Mar 88 134.43 Aug 87 Feb 88 138.06 Apr 88 May 89 209.62 Apr 88 Apr 89 204.78 Jun 88 Mar 89 223.75 Aug 88 Feb 89 189.40 Apr 89 May 90 176.95 Apr 89 Apr 90 165.11 Jun 89 Mar 90 154.06 Aug 89 Feb 90 145.20 Apr 90 May 91 178.29 Apr 90 Apr 91 176.77 Jun 90 Mar 91 160.47 Aug 90 Feb 91 133.59 Apr 91 May 92 158.67 Apr 91 Apr 92 163.41 Jun 91 Mar 92 162.55 Aug 91 Feb 92 164.14 Apr 92 May 93 160.81 Apr 92 Apr 93 153.82 Jun 92 Mar 93 152.84 Aug 92 Feb 93 125.54 Apr 93 May 94 229.96 Apr 93 Apr 94 219.87 Jun 93 Mar 94 203.01 Aug 93 Feb 94 187.73 Apr 94 May 95 221.67 Apr 94 Apr 95 224.13 Jun 94 Mar 95 197.61 Aug 94 Feb 95 164.83 Apr 95 May 96 187.24 Apr 95 Apr 96 184.72 Jun 95 Mar 96 175.37 Aug 95 Feb 96 148.77 Apr 96 May 97 194.89 Apr 96 Apr 97 191.48 Jun 96 Mar 97 190.73 Aug 96 Feb 97 173.97 Apr 97 May 98 134.79 Apr 97 Apr 98 129.55 Jun 97 Mar 98 112.84 Aug 97 Feb 98 79.47 Apr 98 May 99 214.17 Apr 98 Apr 99 210.75 Jun 98 Mar 99 203.99 Aug 98 Feb 99 186.63 Apr 99 May 00 232.73 Apr 99 Apr 00 220.71 Jun 99 Mar 00 205.14 Aug 99 Feb 00 200.50 Apr 00 May 01 157.36 Apr 00 Apr 01 155.56 Jun 00 Mar 01 127.80 Aug 00 Feb 01 121.43 Apr 01 May 02 141.89 Apr 01 Apr 02 138.68 Jun 01 Mar 02 131.14 Aug 01 Feb 02 116.09 Apr 02 May 03 158.58 Apr 02 Apr 03 154.68 Jun 02 Mar 03 148.39 Aug 02 Feb 03 114.97 Apr 03 May 04 145.56 Apr 03 Apr 04 144.25 Jun 03 Mar 04 132.23 Aug 03 Feb 04 129.37 Apr 04 May 05 158.50 Apr 04 Apr 05 158.36 Jun 04 Mar 05 151.66 Aug 04 Feb 05 144.40 Apr 05 May 06 177.11 Apr 05 Apr 06 176.20 Jun 05 Mar 06 163.69 Aug 05 Feb 06 155.50 Apr 06 May 07 194.76 Apr 06 Apr 07 190.65 Jun 06 Mar 07 171.75 Aug 06 Feb 07 148.36 Apr 07 May 08 241.32 Apr 07 Apr 08 239.09 Jun 07 Mar 08 238.13 Aug 07 Feb 08 214.46 Apr 08 May 09 Apr 08 Apr 09 276.02 Jun 08 Mar 09 289.07 Aug 08 Feb 09 263.03 Average 198.04 Average 197.01 Average 187.25 Average 172.72 St, Dev. 37.52 St, Dev. 39.22 St, Dev. 44.13 St, Dev. 47.26 Deviation from mean 1.69 Deviation from mean 1.08 Deviation from mean 0.81 Deviation from mean 0.52 Source: Appendix authors' data. Wind Energy in Colombia 91 Table A6.15. Analysis of El Niño Occurrences at Jepírachi Power Plant (1986­1995) TABLE 1. ANALYSIS OF "EL NIÑO" OCCURRENCES ENERGY IN KWH JEPIRACHI POWERPLANT Jul. 86 Abr. 91 Feb. 93 Mar. 94 Mar. 88 Jul. 92 Ago. 93 Abr. 95 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Jul 85 Mar 87 5764185 Apr 85 Jul 86 5715773 Feb 85 Aug 85 6379056 Mar 85 Apr 86 5539347 Jul 86 Mar 88 6422022 Apr 86 Jul 87 6441876 Feb 86 Aug 86 6379056 Mar 86 Apr 87 6247520 Jul 87 Mar 89 6132963 Apr 87 Jul 88 6816347 Feb 87 Aug 87 7354998 Mar 87 Apr 88 6631787 Jul 88 Mar 90 5864084 Apr 88 Jul 89 6095130 Feb 88 Aug 88 7269545 Mar 88 Apr 89 6011369 Jul 89 Mar 91 6062719 Apr 89 Jul 90 6475338 Feb 89 Aug 89 6650089 Mar 89 Apr 90 6306075 Jul 90 Mar 92 6182776 Apr 90 Jul 91 6395260 Feb 90 Aug 90 6688416 Mar 90 Apr 91 6011417 Jul 91 Mar 93 6402382 Apr 91 Jul 92 6756367 Feb 91 Aug 91 7084624 Mar 91 Apr 92 6561814 Jul 92 Mar 94 6253623 Apr 92 Jul 93 6441281 Feb 92 Aug 92 7519818 Mar 92 Apr 93 6559013 Jul 93 Mar 95 6480640 Apr 93 Jul 94 6721780 Feb 93 Aug 93 6657573 Mar 93 Apr 94 6396776 Jul 94 Mar 96 5632175 Apr 94 Jul 95 6577503 Feb 94 Aug 94 7846772 Mar 94 Apr 95 6600790 Jul 95 Mar 97 5233897 Apr 95 Jul 96 5506845 Feb 95 Aug 95 5952472 Mar 95 Apr 96 5333111 Jul 96 Mar 98 5810455 Apr 96 Jul 97 5966733 Feb 96 Aug 96 6316614 Mar 96 Apr 97 5733818 Jul 97 Mar 99 5729205 Apr 97 Jul 98 6257190 Feb 97 Aug 97 7416243 Mar 97 Apr 98 6272154 Jul 98 Mar 00 5229874 Apr 98 Jul 99 5865564 Feb 98 Aug 98 6480070 Mar 98 Apr 99 5756883 Jul 99 Mar 01 4937749 Apr 99 Jul 00 5658389 Feb 99 Aug 99 6393935 Mar 99 Apr 00 5368818 Jul 00 Mar 02 5195507 Apr 00 Jul 01 5064481 Feb 00 Aug 00 6757620 Mar 00 Apr 01 4926389 Jul 01 Mar 03 6429024 Apr 01 Jul 02 6109140 Feb 01 Aug 01 4499979 Mar 01 Apr 02 5484855 Jul 02 Mar 04 5687418 Apr 02 Jul 03 6810213 Feb 02 Aug 02 7295008 Mar 02 Apr 03 6637391 Jul 03 Mar 05 4780761 Apr 03 Jul 04 5508310 Feb 03 Aug 03 7306087 Mar 03 Apr 04 5206022 Jul 04 Mar 06 4601699 Apr 04 Jul 05 4903049 Feb 04 Aug 04 5645756 Mar 04 Apr 05 4955349 Jul 05 Mar 07 4954198 Apr 05 Jul 06 5004060 Feb 05 Aug 05 4899743 Mar 05 Apr 06 4795268 Jul 06 Mar 08 4566048 Apr 06 Jul 07 4981479 Feb 06 Aug 06 6258557 Mar 06 Apr 07 5201357 Jul 07 Mar 09 Apr 07 Jul 08 4747240 Feb 07 Aug 07 4787152 Mar 07 Apr 08 4517570 Jul 08 Mar 10 Apr 08 Jul 09 Feb 08 Aug 08 5719404 Mar 08 Apr 09 Average 5652427.35 Average 5948667 Average 6481608 Average 5784995.29 St, Dev. 624131.16 St, Dev. 671731 St, Dev. 885823 St, Dev. 665029.00 Deviation from mean 1.23 Deviation from mean 1.20 Deviation from mean 0.20 Deviation from mean 1.23 Source: Appendix authors' data. Table A6.16. Analysis of El Niño Occurrences at Jepírachi Power Plant (1997­2007) TABLE 2. ANALYSIS OF "EL NIÑO" OCCURRENCES ENERGY IN KWH JEPIRACHI POWERPLANT Abr. 97 Abr. 02 Jun. 04 Ago. 06 May. 98 Abr. 03 Mar. 05 Feb. 07 Start Finish Average Start Finish Average Start Finish Average Start Finish Average Apr 85 May 86 5485924 Apr 85 Apr 86 5464287 Jun 85 Mar 86 5349617 Aug 85 Feb 86 4618802 Apr 86 May 87 6194097 Apr 86 Apr 87 6226935 Jun 86 Mar 87 6281286 Aug 86 Feb 87 5896328 Apr 87 May 88 6666956 Apr 87 Apr 88 6611991 Jun 87 Mar 88 6671300 Aug 87 Feb 88 5921141 Apr 88 May 89 5857533 Apr 88 Apr 89 5788216 Jun 88 Mar 89 5494269 Aug 88 Feb 89 4671913 Apr 89 May 90 6312681 Apr 89 Apr 90 6289994 Jun 89 Mar 90 6212223 Aug 89 Feb 90 5980936 Apr 90 May 91 6181645 Apr 90 Apr 91 6109291 Jun 90 Mar 91 5920517 Aug 90 Feb 91 5362740 Apr 91 May 92 6554485 Apr 91 Apr 92 6571330 Jun 91 Mar 92 6463968 Aug 91 Feb 92 5817303 Apr 92 May 93 6270095 Apr 92 Apr 93 6437596 Jun 92 Mar 93 6394728 Aug 92 Feb 93 5749208 Apr 93 May 94 6375075 Apr 93 Apr 94 6322395 Jun 93 Mar 94 6437959 Aug 93 Feb 94 5983270 Apr 94 May 95 6548945 Apr 94 Apr 95 6553348 Jun 94 Mar 95 6457344 Aug 94 Feb 95 5690374 Apr 95 May 96 5284812 Apr 95 Apr 96 5247704 Jun 95 Mar 96 4956561 Aug 95 Feb 96 4219238 Apr 96 May 97 5694275 Apr 96 Apr 97 5676717 Jun 96 Mar 97 5527502 Aug 96 Feb 97 4798268 Apr 97 May 98 6126576 Apr 97 Apr 98 6172706 Jun 97 Mar 98 6091264 Aug 97 Feb 98 5432388 Apr 98 May 99 5677508 Apr 98 Apr 99 5650640 Jun 98 Mar 99 5429478 Aug 98 Feb 99 4783342 Apr 99 May 00 5361301 Apr 99 Apr 00 5284234 Jun 99 Mar 00 4973152 Aug 99 Feb 00 4121783 Apr 00 May 01 4639438 Apr 00 Apr 01 4804179 Jun 00 Mar 01 4847088 Aug 00 Feb 01 4484514 Apr 01 May 02 5888374 Apr 01 Apr 02 5783438 Jun 01 Mar 02 6438151 Aug 01 Feb 02 5772640 Apr 02 May 03 6683619 Apr 02 Apr 03 6540770 Jun 02 Mar 03 6516598 Aug 02 Feb 03 6147751 Apr 03 May 04 5155986 Apr 03 Apr 04 5081660 Jun 03 Mar 04 4681166 Aug 03 Feb 04 3924521 Apr 04 May 05 4968975 Apr 04 Apr 05 5032894 Jun 04 Mar 05 4980736 Aug 04 Feb 05 3944830 Apr 05 May 06 4734004 Apr 05 Apr 06 4683796 Jun 05 Mar 06 4689910 Aug 05 Feb 06 4397437 Apr 06 May 07 4971834 Apr 06 Apr 07 5045007 Jun 06 Mar 07 5034841 Aug 06 Feb 07 4399596 Apr 07 May 08 4517908 Apr 07 Apr 08 4421820 Jun 07 Mar 08 4271867 Aug 07 Feb 08 3734293 Apr 08 May 09 Apr 08 Apr 09 Jun 08 Mar 09 Aug 08 Feb 09 Average 5745741 Average 5730476 Average 5657458 Average 5037070 St, Dev. 685203 St, Dev. 678529 St, Dev. 745717 St, Dev. 799741 Deviation from mean 0.56 Deviation from mean 1.19 Deviation from mean 0.91 Deviation from mean 0.80 Source: Appendix authors' data. 92 World Bank Study The following table summarizes the results. One can see that the four rivers show negative values for most El Niño occurrences, while Jepírachi generation is positive in most of them. The most severe occurrences for the rivers analyzed are April 1991­July 1992 (when a severe rationing occurred in the country) and April 1997­May 1998 (when pool prices rose significantly, forcing regulatory changes in the market). During these periods Jepírachi generation is well above the mean value, complementing hydroelectric generation. Table A6.17. Summary of El Niño occurrences, 1986­2007 ANALYSIS OF "EL NIÑO" OCCURRENCES Depa rture from mean va lue e xpresse d as number of standard deviations "EL NIÑO" OCCURRENCES Jul. 86 Abr. 91 Fe b. 93 Mar. 94 Abr. 97 Abr. 02 Jun. 04 Ago. 06 Mar. 88 Jul. 92 Ago. 93 Abr. 95 May. 98 Abr. 03 Mar. 05 Feb. 07 Guavio River 1.03 0.53 0.64 1.50 0.87 0.66 0.94 1.02 Nare River 0.73 1.39 0.71 0.64 1.86 0.90 0.68 0.08 Cauca River 1.48 1.14 0.17 0.48 1.53 1.52 0.07 0.90 Magdalena River 0.51 1.07 0.00 0.80 1.69 1.08 0.81 0.52 Jepirachi Powerplant 1.23 1.20 0.20 1.23 0.56 1.19 0.91 0.80 Source: Appendix authors' data. 5.3 Firm Energy An analysis of firm energy was obtained from hydroelectric plants (with and without reservoirs) and the Jepírachi power plant in isolated operation, and compared with joint operation of hydro and wind power plants. Firm energy is defined as the maximum monthly energy that can be produced without deficits during the analysis period, which will include El Niño occurrences. The same results were obtained for the total energy obtained from the joint operation of the hydroelectric power plants and the Jepírachi plant. The analysis was conducted using a simulation model that operates the plants and the reservoirs to provide a given energy target, adjusting this target until no deficits are generated. The analysis was conducted for each of the selected hydroelectric plants. Hypothetical hydroelectric plants of similar capacity to that of wind power plants were analyzed. Mean multiannual inflow to the hydroelectric power plants (expressed in energy) at the plant sites is equal to the same value for the Jepírachi generation. This was done by multiplying river discharges by a factor to convert them to energy so that mean inflows are equal to Jepírachi's mean generation. In order to avoid confusion with existing hydroelectric plants, the hypothetical plants analyzed will be named Guavio River, Nare River, Cauca River, and Magdalena River. Several reservoir sizes were analyzed; reservoir size (expressed as a fraction of mean annual inflow to the reservoir in energy) varies between 0 (run of river plant) to 1 (substantial regulation capacity). Results are shown in the following chapters. 5.3.1 Guavio River Table A6.18 and figure A6.13 show results for the Guavio River. Firm energy has been normalized dividing actual firm energy by the sum of mean energy for the Guavio River and Jepírachi. Wind Energy in Colombia 93 Table A6.18. Firm Energy for Guavio and Jepírachi in Isolated and Joint Operation FIRM ENERGY FOR GUAVIO AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Guavio 0 0.2 0.4 0.6 0.8 1 Guavio River (isolated) 0.064 0.334 0.451 0.481 0.507 0.514 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Guavio River + Jepirachi in isolated operation 0.153 0.423 0.540 0.570 0.596 0.602 Guavio River + Jepirachi in joint operation 0.212 0.709 0.860 0.908 0.935 0.962 Source: Appendix authors' data. Figure A6.13. Firm Energy for Guavio and Jepírachi in Isolated and Joint Operation 1.200 Guavio R. + 1.000 Jepirachi (joint) Firm energy/Mean energy 0.800 Guavio R. + Jepirachi (isolated) 0.600 Guavio R.(isolated) 0.400 Jepirachi (isolated) 0.200 0.000 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reservoir size/Mean annual energy Source: Appendix authors' data. The substantial increase in firm energy when joint operation is considered can be seen both in the table and the figure. This is because critical periods for the Guavio River do not coincide with Jepírachi generation during the same period. The following figures, showing reservoir operation both in isolated and joint operation, illustrate this fact. Figure A6.14, corresponding to a reservoir size of 0.2, shows that in isolated operation the reservoir is emptied during the El Niño occurrence of April 1997­May 1998, while in joint operation the reservoir is emptied in April 2001. The El Niño occurrence of April 1997­April 1998 is balanced by large scale generation in the Jepírachi power plant, showing the complementarity of river discharges in the Guavio River and wind generation in the Jepírachi power plant. Figure A6.15, corresponding to a reservoir size of 0.5, illustrates the same effect. 94 World Bank Study Figure A6.14. Guavio River Reservoir Operation with Reservoir Size 0.2 1.200 1.000 Reservoir volume/Resrvoir capacity 0.800 0.600 Joint 0.400 Isolated 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. Figure A6.15. Guavio River Reservoir Operation with Reservoir Size 0.5 1.200 1.000 Reservoir volume/Resrvoir capacity 0.800 Joint 0.600 Isolated 0.400 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. 5.3.2 Nare River The following tables and graphs show the same results for the Nare River as those shown for the Guavio River. One can see the similarity of results with those for the Guavio River. Wind Energy in Colombia 95 Table A6.19. Firm Energy for Nare and Jepírachi in Isolated and Joint Operation FIRM ENERGY FOR NARE AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Nare 0 0.2 0.4 0.6 0.8 1 Nare River(isolated) 0.179 0.369 0.435 0.459 0.471 0.480 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Nare River + Jepirachi in isolated operation 0.268 0.458 0.524 0.548 0.560 0.569 Nare River + Jepirachi in joint operation 0.410 0.811 0.943 0.972 0.994 1.009 Source: Appendix authors' data. Figure A6.16. Firm Energy for Nare and Jepírachi in Isolated and Joint Operation 1.200 1.000 Firm energy/Mean energy 0.800 Nare R. + Jepirachi (joint) 0.600 Nare R. + Jepirachi (isolated) 0.400 Nare R.(isolated) 0.200 0.000 Jepirachi (isolated) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reservoir size/Mean annual energy Source: Appendix authors' data. Figure A6.17. Nare River Reservoir Operation with Reservoir Size 0.2 1.200 1.000 Reservoir volume/Resrvoir capacity 0.800 0.600 Isolated 0.400 Joint 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. 96 World Bank Study Figure A6.18. Nare River Reservoir Operation with Reservoir Size 0.5 1.200 Reservoir volume/Resrvoir capacity 1.000 0.800 0.600 Isolated 0.400 Joint 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. 5.3.3 Cauca River The following tables and figures show the same results for the Cauca River as those shown for the Guavio and Nare Rivers. Once again, one can easily see the similarity of results with those for the Guavio and Nare Rivers. Table A6.20. Firm Energy for Cauca and Jepírachi in Isolated and Joint Operation FIRM ENERGY FOR CAUCA AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Cauca 0 0.2 0.4 0.6 0.8 1 Cauca River (isolated) 0.146 0.381 0.417 0.443 0.466 0.489 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Cauca River + Jepirachi in isolated operation 0.234 0.470 0.506 0.532 0.555 0.578 Cauca River + Jepirachi in joint operation 0.346 0.824 0.903 0.922 0.941 0.957 Source: Appendix authors' data. Wind Energy in Colombia 97 Figure A6.19. Firm Energy for Cauca and Jepírachi in Isolated and Joint Operation 1.200 1.000 Firm energy/Mean energy 0.800 Cauca R. + 0.600 Jepirachi (joint) Cauca R. + 0.400 Jepirachi (isolated) Cauca R.(isolated) 0.200 Jepirachi (isolated) 0.000 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reservoir size/Mean annual energy Source: Appendix authors' data. Figure A6.20. Cauca River Reservoir Operation with Reservoir Size 0.2 1.200 1.000 Reservoir volume/Resrvoir capacity 0.800 0.600 Isolated 0.400 Joint 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. 98 World Bank Study Figure A6.21. Cauca River Reservoir Operation with the Reservoir Size 0.5 1.200 1.000 Reservoir volume/Resrvoir capacity 0.800 0.600 Isolated 0.400 Joint 0.200 0.000 Dec-84 Nov-85 Nov-86 Nov-87 Nov-88 Nov-89 Nov-90 Nov-91 Nov-92 Nov-93 Nov-94 Nov-95 Nov-96 Nov-97 Nov-98 Nov-99 Nov-00 Nov-01 Nov-02 Nov-03 Nov-04 Nov-05 Nov-06 Nov-07 Nov-08 Source: Appendix authors' data. 5.3.4 Magdalena River The following tables and figures show the same results for the Magdalena River as those shown for the Guavio River. One can see the similarity of results with those for the Guavio River. Table A6.21. Firm Energy for Magdalena and Jepírachi in Isolated and Joint Operation FIRM ENERGY FOR MAGDALENA AND JEPIRACHI IN ISOLATED AND JOINT OPERATION Firm Energy/Mean Energy Reservoir volume expressed as a fraction of mean energy inflow to Magdalena 0 0.2 0.4 0.6 0.8 1 Magdalena River (isolated) 0.082 0.354 0.429 0.447 0.465 0.484 Jepirachi (isolated) 0.089 0.089 0.089 0.089 0.089 0.089 Magdalena River + Jepirachi in isolated operation 0.170 0.442 0.518 0.536 0.554 0.572 Magdalena River + Jepirachi in joint operation 0.350 0.770 0.869 0.910 0.929 0.948 Source: Appendix authors' data. Figure A6.22. Firm Energy for Magdalena and Jepírachi in Isolated and Joint Operation 1.000 0.900 Firm energy/Mean energy 0.800 Magdalena R. + 0.700 Jepirachi (joint) 0.600 Magdalena R. + 0.500 Jepirachi (isolated) 0.400 Magdalena 0.300 R.(isolated) 0.200 Jepirachi (isolated) 0.100 0.000 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reservoir size/Mean annual energy Source: Appendix authors' data. Reservoir volume/Resrvoir capacity Reservoir volume/Resrvoir capacity 0.000 0.200 0.400 0.600 0.800 1.000 1.200 0.000 0.200 0.400 0.600 0.800 1.000 1.200 Dec-84 Dec-84 Nov-85 Nov-85 Nov-86 Nov-86 Nov-87 Nov-87 Nov-88 Nov-88 Source: Appendix authors' data. Source: Appendix authors' data. Nov-89 Nov-89 Nov-90 Nov-90 Nov-91 Nov-91 Nov-92 Nov-92 Nov-93 Nov-93 Nov-94 Nov-94 Nov-95 Nov-95 Nov-96 Nov-96 Nov-97 Nov-97 Nov-98 Nov-99 Nov-98 Nov-00 Nov-99 Nov-01 Nov-00 Nov-02 Nov-01 Nov-03 Nov-02 Nov-04 Nov-03 Nov-05 Nov-04 Nov-06 Nov-05 Nov-07 Nov-06 Nov-08 Nov-07 Nov-08 Figure A6.24. Magdalena River Reservoir Operation with Reservoir Size 0.5 Figure A6.23. Magdalena River Reservoir Operation with Reservoir Size 0.2 Joint Joint Wind Energy in Colombia Isolated Isolated 99 Eco-Audit Environmental Benefits Statement The World Bank is committed to preserving Endangered Forests and natural resources. We print World Bank Working Papers and Country Studies on postconsumer recycled paper, processed chlorine free. The World Bank has formally agreed to follow the recommended standards for paper usage set by Green Press Initiative--a nonprofit program supporting publishers in using fiber that is not sourced from Endangered Forests. For more information, visit www.greenpressinitiative.org. In 2008, the printing of these books on recycled paper saved the following: Net Greenhouse Trees* Solid Waste Water Total Energy Gases 289 8,011 131,944 27,396 92 mil. *40 feet in height and Pounds CO2 6­8 inches in Pounds Gallons BTUs Equivalent diameter W ind Energy in Colombia is part of the World Bank Study series. These papers are published to communicate the results of the Bank's ongoing research and to stimulate public discussion. The purpose of this study is to provide decision makers--analysts, planners, operators, genera- tors, and policy makers--in Colombia as well as in other countries with a set of policy options to promote the use of wind power. Of the renewable energy alternatives available in Colombia, wind power was chosen for its technical maturity, its relatively low cost compared to other options, and Colombia's previous experiences and wind power potential. The possible policy options include financial instruments, government fiscal mechanisms, and adjustments to the regulatory system. The single most effective policy instrument to promote wind power in Colombia consists of recog- nizing its potential to complement hydroelectric power. This study is part of an ongoing project to identify and address barriers to the development of nonconventional renewable energy resources in Colombia's power sector. World Bank Studies are available individually or on standing order. This World Bank Study series is also available online through the World Bank e-library (www.worldbank.org/newelibrary). ISBN 978-0-8213-8504-3 SKU 18504