MIT Study: The Future of Natural Gas
Upcoming SlideShare
Loading in...5

MIT Study: The Future of Natural Gas



A 174-page study issued by the Massachusetts Institute of Technology in June 2011 evaluating natural gas and how it fares as an energy source in a political environment where carbon is taxed. That is, ...

A 174-page study issued by the Massachusetts Institute of Technology in June 2011 evaluating natural gas and how it fares as an energy source in a political environment where carbon is taxed. That is, how competitive is natural gas when considering lower carbon emissions standards? The study finds, among other things, that hydraulic fracturing technology as used in shale gas drilling is safe.



Total Views
Views on SlideShare
Embed Views



4 Embeds 1,589 1586 1 1 1


Upload Details

Uploaded via as Adobe PDF

Usage Rights

CC Attribution-ShareAlike LicenseCC Attribution-ShareAlike License

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

MIT Study: The Future of Natural Gas MIT Study: The Future of Natural Gas Document Transcript

  • Foreword and AcknowledgementsThe Future of Natural Gas is the fourth in a invested in advising us. However, the study is theseries of MIT multidisciplinary reports examin- responsibility of the MIT study group and theing the role of various energy sources that may advisory committee members do not necessarilybe important for meeting future demand under endorse all of its findings and recommendations,carbon dioxide (CO2) emissions constraints. In either individually or collectively.each case, we explore the steps needed to enablecompetitiveness in a future marketplace condi- Finally, we are very appreciative of the supporttioned by a CO2 emissions price or by a set of from several sources. First and foremost, weregulatory initiatives. This report follows an thank the American Clean Skies Foundation.interim report issued in June 2010. Discussions with the Foundation led to the conclusion that an integrative study on theThe first three reports dealt with nuclear power future of natural gas in a carbon-constrained(2003), coal (2007) and the nuclear fuel cycle world could contribute to the energy debate in(2010 and 2011). A study of natural gas is more an important way, and the Foundation steppedcomplex than these previous reports because forward as the major sponsor. MIT Energygas is a major fuel for multiple end uses — Initiative (MITEI) members Hess Corporationelectricity, industry, heating — and is increasingly and Agencia Naçional de Hidrocarburosdiscussed as a potential pathway to reduced oil (Colombia), the Gas Technology Institute (GTI),dependence for transportation. In addition, Exelon, and an anonymous donor providedthe realization over the last few years that the additional support. The Energy Futures Coali-producible unconventional gas resource in the tion supported dissemination of the studyU.S. is very large has intensified the discussion results, and MITEI employed internal funds andabout natural gas as a “bridge” to a low-carbon fellowship sponsorship to support the study asfuture. Recent indications of a similarly large well. As with the advisory committee, theglobal gas shale resource may also transform the sponsors are not responsible for and do notgeopolitical landscape for gas. We have carried necessarily endorse the findings and recommen-out the integrated analysis reported here as dations. That responsibility lies solely with thea contribution to the energy, security and MIT study group.climate debate. We thank Victoria Preston and RebeccaOur primary audience is U.S. government, Marshall-Howarth for editorial support andindustry and academic leaders, and decision Samantha Farrell for administrative support.makers. However, the study is carried out withan international perspective.This study is better as a result of comments andsuggestions from our distinguished externalAdvisory Committee, each of whom broughtimportant perspective and experience to ourdiscussions. We are grateful for the time they MIT Study on the Future of Natural Gas iii
  • Study ParticipantsSTUDY CO-CHAIRSERNEST J. MONIZ — CHAIR MELANIE A. KENDERDINECecil and Ida Green Professor of Physics Executive Director, MITEI and Engineering Systems, MITDirector, MIT Energy Initiative (MITEI) FRANCIS O’SULLIVAN Research Engineer, MITEIHENRY D. JACOBY — CO-CHAIRProfessor of Management, MIT SERGEY PALTSEV Principal Research Scientist, Joint Program onANTHONY J. M. MEGGS — CO-CHAIR the Science and Policy of Global Change, MITVisiting Engineer, MITEI JOHN E. PARSONSSTUDY GROUP Senior Lecturer, Sloan School of Management, MIT Executive Director, Joint Program on theROBERT C. ARMSTRONG Science and Policy of Global Change andChevron Professor, Department of Chemical Center for Energy and Environmental Engineering, MIT Policy Research, MITDeputy Director, MITEI IGNACIO PEREZ-ARRIAGADANIEL R. COHN Professor of Electrical Engineering,Senior Research Scientist, Plasma Science Comillas University, Spain and Fusion Center, MIT Visiting Professor, Engineering Systems Division, MITExecutive Director, Natural Gas Study JOHN M. REILLYSTEPHEN R. CONNORS Senior Lecturer, Sloan School of Management, MITResearch Engineer, MITEI Co-Director, Joint Program on the Science and Policy of Global Change, MITJOHN M. DEUTCHInstitute Professor, CAROLYN SETO Department of Chemistry, MIT Clare Boothe Luce Postdoctoral Fellow, Department of Chemical Engineering, MITQUDSIA J. EJAZPostdoctoral Associate, MITEI MORT D. WEBSTER Assistant Professor, Engineering SystemsJOSEPH S. HEZIR Division, MITVisiting Engineer, MITEI YINGXIA YANGGORDON M. KAUFMAN MITEIMorris A. Adelman Professor of Management (Emeritus), MIT MIT Study on the Future of Natural Gas v
  • Advisory Committee MembersTHOMAS F. (MACK) MCLARTY, III — SENATOR (ret.) J. BENNETT JOHNSTONCHAIRMAN Chairman, Johnston AssociatesPresident & CEO, McLarty Associates VELLO A. KUUSKRAADENISE BODE President, Advance Resources International, Inc.CEO, American Wind Energy Association MIKE MINGRALPH CAVANAGH Oklahoma Secretary of EnergySenior Attorney and Co-Director of Energy Program, Natural Resource Defense Council THEODORE ROOSEVELT IV Managing Director & Chairman, Barclays CapitalSUNIL DESHMUKH Clean Tech InitiativeFounding Member, Sierra Club India Advisory Council OCTAVIO SIMOES Vice President of Commercial Development,JOSEPH DOMINGUEZ Sempra EnergySenior Vice President, Exelon Corporation GREG STAPLERON EDELSTEIN CEO, American Clean Skies FoundationDirector, Regulatory and Government Relations, GTI PETER TERTZAKIAN Chief Energy Economist and Managing Director,NEAL ELLIOTT ARC FinancialAssociate Director for Research, American Council for an Energy-Efficient Economy DAVID VICTOR Director, Laboratory on International LawJOHN HESS and Regulation, University of California, San DiegoChairman and CEO, Hess Corporation ARMANDO ZAMORAJAMES T. JENSEN Director, ANH-Agencia Nacional De HidrocarburosPresident, Jensen AssociatesWhile the members of the advisory committee provided invaluable perspective and advice to the study group,individual members may have different views on one or more matters addressed in the report. They are notasked to individually or collectively endorse the report findings and recommendations. MIT Study on the Future of Natural Gas vii
  • AbstractNatural gas is finding its place at the heart of the The U.S. natural gas supply situation hasenergy discussion. The recent emergence of enhanced the substitution possibilities forsubstantial new supplies of natural gas in the natural gas in the electricity, industry, buildings,U.S., primarily as a result of the remarkable and transportation sectors.speed and scale of shale gas development, hasheightened awareness of natural gas as a key In the U.S. electricity supply sector, the costcomponent of indigenous energy supply and has benchmark for reducing carbon dioxide emissionslowered prices well below recent expectations. lies with substitution of natural gas for coal,This study seeks to inform discussion about the especially older, less efficient units. Substitutionfuture of natural gas, particularly in a carbon- through increased utilization of existing combinedconstrained economy. cycle natural gas power plants provides a rela- tively low-cost, short-term opportunity to reduceThere are abundant supplies of natural gas in U.S. power sector CO2 emissions by up to 20%,the world, and many of these supplies can be while also reducing emissions of criteria pollut-developed and produced at relatively low cost. ants and mercury.In North America, shale gas development overthe past decade has substantially increased Furthermore, additional gas-fired capacity willassessments of resources producible at modest be needed as backup if variable and intermittentcost. Consequently, the role of natural gas is renewables, especially wind, are introduced on alikely to continue to expand, and its relative large scale. Policy and regulatory steps are neededimportance is likely to increase even further to facilitate adequate capacity investment forwhen greenhouse gas emissions are constrained. system reliability and efficiency. These increas-In a carbon-constrained world, a level playing ingly important roles for natural gas in thefield — a carbon dioxide (CO2) emissions price electricity sector call for a detailed analysis of thefor all fuels without subsidies or other preferen- interdependencies of the natural gas and powertial policy treatment —maximizes the value to generation infrastructures.society of the large U.S. natural gas resource. The primary use of natural gas in the U.S.There are also a number of key uncertainties: manufacturing sector is as fuel for boilers andthe extent and nature of greenhouse gas emission process heating, and replacement with newmitigation measures that will be adopted; the higher efficiency models would cost-effectivelymix of energy sources as the relative costs of fuels reduce natural gas use. Natural gas could alsoand technologies shift over time; the evolution substitute for coal in boilers and process heatersof international natural gas markets. We explore and provide a cost-effective alternative forhow these uncertainties lead to different out- compliance with Environmental Protectioncomes and also quantify uncertainty for natural Agency (EPA) Maximum Achievable Controlgas supply and for the U.S. electricity fuel mix. Technology standards.The environmental impacts of shale development In the residential and commercial buildingsare challenging but manageable. Research and sector, transformation of the current approach toregulation, both state and Federal, are needed to efficiency standards to one based on full fuelminimize the environmental consequences. cycle analysis will enable better comparison of different energy supply options (especially MIT Study on the Future of Natural Gas xiii
  • natural gas and electricity). Efficiency metrics The evolution of global gas markets is unclear. should be tailored to regional variations in A global “liquid” natural gas market is benefi- climate and electricity supply mix. cial to U.S. and global economic interests and, at the same time, advances security interests Within the U.S. market, the price of oil (which through diversity of supply and resilience to is set globally) compared to the price of natural disruption. The U.S. should pursue policies that gas (which is set regionally) is very important encourage the development of such a market, in determining market share when there is the integrate energy issues fully into the conduct of opportunity for substitution. Over the last U.S. foreign policy, and promote sharing of decade or so, when oil prices have been high, know-how for strategic global expansion of the ratio of the oil price to the natural gas price unconventional gas production. has been consistently higher than any of the standard rules of thumb. If this trend is robust, Past research, development, demonstration, and use of natural gas in transportation, either deployment (RDD&D) programs supported with through direct use or following conversion to a public funding have led to significant advances liquid fuel, could in time increase appreciably. for natural gas supply and use. Public-private partnerships supporting a broad natural gas research, development, and demonstration (RD&D) portfolio should be pursued.xiv MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Chapter 1: Overview and ConclusionsPURPOSE AND OUTLINE OF THE STUDYDespite its vital importance to the national Having explored supply volumes and costs, weeconomy, natural gas has often been overlooked, use integrated models to examine the role thator at best taken for granted, in the debate about natural gas could play in the energy system underthe future of energy in the U.S. Over the past different carbon-constraining mechanisms ortwo or three years this has started to change, and policies. It is important to recognize that the studynatural gas is finding its place at the heart of the does not set out to make predictions or forecastsenergy discussion. of the likelihood or direction of CO2 policy in the U.S. Rather, we examine a number of differentThere are a number of reasons for this shift. scenarios and explore their possible impacts onThe recent emergence of substantial new sup- the future of natural gas supply and demand.plies of natural gas in the U.S., primarily as aresult of the remarkable speed and scale of shale Natural gas is important in many sectors of thegas development, has heightened awareness of economy — for electricity generation, as annatural gas as a key component of indigenous industrial heat source and chemical feedstock,energy supply and lowered prices well below and for water and space heating in residentialrecent expectations. Instead of the anticipated and commercial buildings. Natural gas competesgrowth of natural gas imports, the scale of domes- directly with other energy inputs in these sectors.tic production has led producers to seek new But it is in the electric power sector — wheremarkets for natural gas, such as an expanded role natural gas competes with coal, nuclear, hydro,in transportation. Most importantly for this wind and solar — that inter-fuel competition isstudy, there has been a growing recognition that most intense. We have, therefore, explored inthe low carbon content of natural gas relative to depth how natural gas performs in the electricother fossil fuels could allow it to play a signifi- power sector under different scenarios. We havecant role in reducing carbon dioxide (CO2) emis- also taken a close look at the critical interactionsions, acting as a “bridge” to a low-carbon future. between intermittent forms of renewable energy, such as wind and solar, and gas-fired power as aWithin this context, the MIT study of The Future reliable source of backup capacity.of Natural Gas seeks to inform the discussionaround natural gas by addressing a fundamental We look at the drivers of natural gas use in thequestion: what is the role of natural gas in a industrial, commercial and residential sectors,carbon-constrained economy? and examine the important question of whether natural gas, in one form or another, could be aIn exploring this question, we seek to improve viable and efficient substitute for gasoline or dieselgeneral understanding of natural gas, and in the transportation sector. We also examine theexamine a number of specific issues. How much possible futures of global natural gas markets, andnatural gas is there in the world, how expensive the geopolitical significance of the ever-expandingis it to develop, and at what rate can it be pro- role of natural gas in the global economy. Finally,duced? We start from a global perspective, and we make recommendations for research andthen look in detail at U.S. natural gas resources, development priorities and for the means bypaying particular attention to the extent and cost which public support should be provided.of shale gas resources, and whether these sup-plies can be developed and produced in anenvironmentally sound manner. Chapter 1: Overview and Conclusions 1
  • HIGH-LEVEL FINDINGS 5. Increased utilization of existing natural gas combined cycle (NGCC) power plants The findings and recommendations of the provides a relatively, low-cost short-term study are discussed later in this chapter, and opportunity to reduce U.S. CO2 emissions covered in detail in the body of this report. by up to 20% in the electric power sector, Nevertheless, it is worth summarizing here or 8% overall, with minimal additional the highest level conclusions of our study: capital investment in generation and no new technology requirements. 1. There are abundant supplies of natural gas in the world, and many of these supplies 6. Natural gas-fired power capacity will play can be developed and produced at relatively an increasingly important role in providing low cost. In the U.S., despite their relative backup to growing supplies of intermittent maturity, natural gas resources continue to renewable energy, in the absence of a grow, and the development of low-cost and breakthrough that provides affordable abundant unconventional natural gas utility-scale storage. But in most cases, resources, particularly shale gas, has a increases in renewable power generation material impact on future availability will be at the expense of natural gas-fired and price. power generation in the U.S. 2. Unlike other fossil fuels, natural gas plays 7. The current supply outlook for natural gas a major role in most sectors of the modern will contribute to greater competitiveness economy — power generation, industrial, of U.S. manufacturing, while the use of commercial and residential. It is clean and more efficient technologies could offset flexible. The role of natural gas in the world increases in demand and provide cost- is likely to continue to expand under almost effective compliance with emerging envi- all circumstances, as a result of its availability, ronmental requirements. its utility and its comparatively low cost. 8. Transformation of the current approach 3. In a carbon-constrained economy, the to appliance standards to one based on full relative importance of natural gas is likely fuel cycle analysis will enable better com- to increase even further, as it is one of the parison of different energy supply options most cost-effective means by which to in commercial and residential applications. maintain energy supplies while reducing CO2 emissions. This is particularly true in 9. Natural gas use in the transportation sector the electric power sector, where, in the U.S., is likely to increase, with the primary natural gas sets the cost benchmark against benefit being reduced oil dependence. which other clean power sources must Compressed natural gas (CNG) will play compete to remove the marginal ton of CO2. a role, particularly for high-mileage fleets, but the advantages of liquid fuel in trans- 4. In the U.S., a combination of demand portation suggest that the chemical conver- reduction and displacement of coal-fired sion of gas into some form of liquid fuel may power by gas-fired generation is the lowest- be the best pathway to significant market cost way to reduce CO2 emissions by up to penetration. 50%. For more stringent CO2 emissions reductions, further de-carbonization of the energy sector will be required; but natural gas provides a cost-effective bridge to such a low-carbon future.2 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 10. International gas trade continues to grow natural gas allows high recoveries from conven- in scope and scale, but its economic, tional reservoirs at relatively low cost, and also security and political significance is not yet enables natural gas to be economically recov- adequately recognized as an important ered from even the most unfavorable subsurface focus for U.S. energy concerns. environments, as recent developments in shale formations have demonstrated.11. Past research, development, demonstration and deployment (RDD&D) programs These physical characteristics underpin the supported with public funding have led to current expansion of the unconventional significant advances for natural gas supply resource base in North America, and the and use. potential for natural gas to displace more carbon-intensive fossil fuels in a carbon-BACKGROUND constrained world.The Fundamental Characteristics On the other hand, because of its gaseous formof Natural Gas and low energy density, natural gas is uniquely disadvantaged in terms of transmission andFossil fuels occur in each of the three funda- storage. As a liquid, oil can be readily trans-mental states of matter: in solid form as coal; ported over any distance by a variety of means,in liquid form as oil and in gaseous form as and oil transportation costs are generally a smallnatural gas. These differing physical character- fraction of the overall cost of developing oilistics for each fuel type play a crucial part in fields and delivering oil products to market. Thisshaping each link in their respective supply has facilitated the development of a truly globalchains: from initial resource development and market in oil over the past 40 years or more.production through transportation, conversionto final products and sale to customers. Their By contrast, the vast majority of natural gasphysical form fundamentally shapes the supplies are delivered to market by pipeline, andmarkets for each type of fossil fuel. delivery costs typically represent a relatively large fraction of the total cost in the supply chain.Natural gas possesses remarkable qualities. These characteristics have contributed to theAmong the fossil fuels, it has the lowest carbon evolution of regional markets rather than aintensity, emitting less CO2 per unit of energy truly global market in natural gas. Outsidegenerated than other fossil fuels. It burns cleanly North America, this somewhat inflexibleand efficiently, with very few non-carbon pipeline infrastructure gives strong politicalemissions. Unlike oil, natural gas generally and economic power to those countries thatrequires limited processing to prepare it for control the pipelines. To some degree, theend use. These favorable characteristics have evolution of the spot market in Liquefiedenabled natural gas to penetrate many markets, Natural Gas (LNG) is beginning to introduceincluding domestic and commercial heating, more flexibility into global gas markets andmultiple industrial processes and electrical stimulate real global trade. The way this tradepower. may evolve over time is a critical uncertainty that is explored in this report.Natural gas also has favorable characteristicswith respect to its development and production.The high compressibility and low viscosity of Chapter 1: Overview and Conclusions 3
  • The Importance of Natural Gas in the coming from indigenous resources. Perhaps of Energy System more significance, is the very important role that natural gas plays in all sectors of the economy, Natural gas represents a very important, and with the exception of transport. Very approxi- growing, part of the global energy system. mately, the use of natural gas is divided evenly Over the past half century, natural gas has between three major sectors: industrial, residen- gained market share on an almost continuous tial and commercial, and electric power. The 3% basis, growing from some 15.6% of global share that goes to transport is almost all associ- energy consumption in 1965 to around 24% ated with natural gas use for powering oil and today. In absolute terms, global natural gas gas pipeline systems, with only a tiny fraction consumption over this period has grown from going into vehicle transport. around 23 trillion cubic feet (Tcf) in 1965 to 104 Tcf in 2009, a more than fourfold increase. In the Residential and Commercial sectors, natural gas provides more than three-quarters Within the U.S. economy, natural gas plays a of the total primary energy, largely as a result vital role. Figure 1.1 displays the sources and uses of its efficiency, cleanliness and convenience of natural gas in the U.S. in 2009, and it reveals a for uses such as space and hot water heating. number of interesting features that It is also a major primary energy input into the are explored in more detail in the body of this Industrial sector, and thus the price of natural report. At 23.4 quadrillion British thermal units gas has a very significant impact on the com- (Btu)1, or approximately 23 Tcf, gas represents a petitiveness of some U.S. manufacturing little under a quarter of the total energy supply industries. While natural gas provided 18% of in the U.S., with almost all of this supply now the primary fuel for power generation in 2009, Figure 1.1 Sources and Use of Primary Energy Sources in the U.S. with Natural Gas Highlighted (quadrillion Btu), 2009 Supply Sources Demand Sources 94.6 Quads Percent of Source Percent of Sector 72% 94% Petroleum 35.3 22% 27.0 Transportation 1% 3% 3% 5% 41% 3% Natural Gas 23.4 32% 40% 18.8 Industrial 11% 30% 35% 7% 17% 76% 10.6 Residential & 1% 7% 1% Commercial Coal 19.7 <1% 93% 12% 26% 48% 18% 1% 9% 11% Electric Renewables 7.7 38.6 53% 22% Power 100% Nuclear 8.3 Source: EIA, Annual Energy Outlook, 20094 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • it provided 23% of the produced electricity, By the mid 1990s, wholesale electricity marketsreflecting the higher efficiency of natural gas and wellhead natural gas prices had beenplants. As will be seen later in this report, deregulated; new, highly efficient and relativelynatural gas-fired capacity represents far more inexpensive combined cycle gas turbines hadthan 23% of total power generating capacity, been deployed and new upstream technologiesproviding a real opportunity for early action in had enabled the development of offshorecontrolling CO2 emissions. natural gas resources. This contributed to the perception that domestic natural gas suppliesA Brief History of Natural Gas in the U.S. were sufficient to increase the size of the U.S. natural gas market from around 20 Tcf/year toThe somewhat erratic history of natural gas much higher levels. New gas-fired powerin the U.S. over the last three decades or so capacity was added at a rapid pace.provides eloquent testimony to the difficultiesof forecasting energy futures, particularly for Between 1989 after the repeal of the FUA andnatural gas. It also serves as a reminder of the 2009, the U.S. added 306 GW of generationneed for caution in the current period of capacity, 88% of which was gas fired and 4%supply exuberance. was coal fired.2 Today, the nameplate capacity of this gas-fired generation is significantly under-The development of the U.S. natural gas market utilized, and the anticipated large increase inwas facilitated by the emergence of an interstate natural gas use has not materialized.natural gas pipeline system, supplying localdistribution systems. This market structure was By the turn of the 21st century, a new set ofinitially viewed as a natural monopoly and was concerns arose about the adequacy of domesticsubjected to cost-of-service regulation by both natural gas supplies. Conventional suppliesthe Federal government and the states. Natural were in decline, unconventional natural gasgas production and use grew considerably resources remained expensive and difficult tounder this framework in the 1950s, 1960s and develop and overall confidence in gas plum-into the 1970s. meted. Natural gas prices started to rise, becom- ing more closely linked to the oil price, whichThen came a perception of supply scarcity. itself was rising. Periods of significant naturalAfter the first oil embargo, energy consumers gas price volatility were experienced.sought to switch to natural gas. However, thecombination of price controls and tightly This rapid buildup in natural gas price, andregulated natural gas markets dampened perception of long-term shortage, createdincentives for domestic gas development, economic incentives for the accelerated devel-contributing to a perception that U.S. natural opment of an LNG import infrastructure. Sincegas resources were limited. In 1978, convinced 2000, North America’s rated LNG capacity hasthat the U.S. was running out of natural gas, expanded from approximately 2.3 billion cubicCongress passed the Power Plant and Industrial feet (Bcf)/day to 22.7 Bcf/day, around 35% ofFuel Use Act (FUA) that essentially outlawed the nation’s average daily requirement.the building of new gas-fired power plants.Between 1978 and 1987 (the year the FUA was This expansion of LNG capacity coincided withrepealed), the U.S. added 172 Gigawatts (GW) an overall rise in the natural gas price and theof net power generation capacity. Of this, market diffusion of technologies to developalmost 81 GW was new coal capacity, around affordable unconventional gas. The game-26% of today’s entire coal fleet. About half of changing potential of these technologies,the remainder was nuclear power. combined with the large unconventional Chapter 1: Overview and Conclusions 5
  • resource base, has become more obvious over The likely technology mix in a carbon- the last few years, radically altering the U.S. constrained world, particularly in the power supply picture. We have once again returned to sector: the relative costs of different tech- a period where supply is seen to be abundant. nologies may shift significantly in response New LNG import capacity goes largely unused to RD&D, and a CO2 emissions price will at present, although it provides a valuable affect the relative costs. Moreover, the tech- supply option for the future. nology mix will be affected by regulatory and subsidy measures that will skew economic These cycles of perceived “feast and famine” choices. demonstrate the genuine difficulty of forecast- ing the future and providing appropriate policy The ultimate size and production cost of the support for natural gas production and use. natural gas resource base, and the environ- They underpin the efforts of this study to mental acceptability of production methods: account for this uncertainty in an analytical much remains to be learned about the perfor- manner. mance of shale gas plays, both in the U.S. and in other parts of the world. Indeed, even higher Major Uncertainties risk and less well-defined unconventional natural gas resources, such as methane Looking forward, we anticipate policy and hydrates, could make a contribution to geopolitics, along with resource economics and supply in the later decades of the study’s technology developments, will continue to play time horizon. a major role in determining global supply and market structures. Thus, any analysis of the The evolution of international natural gas future of natural gas must deal explicitly with markets: very large natural gas resources are multiple uncertainties: to be found in several areas outside the U.S., and the role of U.S. natural gas will be The extent and nature of the greenhouse gas influenced by the evolution of this market — (GHG) mitigation measures that will be particularly the growth and efficiency of adopted: the U.S. legislative response to the trade in LNG. Only a few years back, U.S. climate threat has proved quite challenging. industry was investing in facilities for sub- However, the Environmental Protection stantial LNG imports. The emergence of the Agency (EPA) is developing regulations domestic shale gas resource has depressed under the Clean Air Act, and a variety of this business in the U.S., but in the future, local, state and regional GHG limitation the nation may again look to international programs have been put in place. At the markets. international level, reliance upon a system of voluntary national pledges of emission Of these uncertainties, the last three can be reductions by 2020, as set out initially in the explored by applying technically grounded Copenhagen Accord, leaves uncertainty analysis: lower cost for carbon capture and concerning the likely structure of any future sequestration (CCS), renewables and nuclear agreements that may emerge to replace the power; producible resources of different levels Kyoto Protocol. The absence of a clear and regional versus global integrated markets. international regime for mitigating GHG In contrast, the shape and size of GHG mitiga- emissions in turn raises questions about the tion measures is likely to be resolved only likely stringency of national policies in both through complex ongoing political discussions industrialized countries and major emerging at the national level in the major emitting economies over coming decades. countries and through multilateral negotiations.6 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • The analysis in this study is based on three Unconventional natural gas, and particularlypolicy scenarios: shale gas, will make an important contribution to future U.S. energy supply and CO2 emission-1. A business-as-usual case, with no significant reduction efforts. Assessments of the recover- carbon constraints; able volumes of shale gas in the U.S. have increased dramatically over the last five years,2. GHG emissions pricing, through a cap- and continue to grow. The mean projection of and-trade system or emissions tax, leading the recoverable shale gas resource in this report to a 50% reduction in U.S. emissions below is approximately 650 Tcf, with low and high the 2005 level, by 2050. projections of 420 Tcf and 870 Tcf, respectively. Of the mean projection, approximately 400 Tcf3. GHG reduction via U.S. regulatory could be economically developed with a natural measures without emissions pricing: a gas price at or below $6/MMBtu at the wellhead. renewable portfolio standard and measures While the pace of shale technology development forcing the retirement of some coal plants. has been very rapid over the past few years, there are still many scientific and technologicalOur analysis is long term in nature, with challenges to overcome before we can bea 2050 time horizon. We do not attempt to confident that this very large resource base ismake detailed short-term projections of being developed in an optimum manner.volumes, prices or price volatility, but ratherfocus on the long-term consequences of the Although there are large supplies, global conven-carbon mitigation scenarios outlined above, tional natural gas resources are concentratedtaking into account the manifold uncertainties geographically, with 70% in three countries:in a highly complex and interdependent Qatar, Iran and Russia. There is considerableenergy system. potential for unconventional natural gas supply outside North America, but these resources areMAJOR FINDINGS AND largely unproven with very high resourceRECOMMENDATIONS uncertainty. Nevertheless, unconventional supplies could provide a major opportunity forIn the following section we summarize the diversification and improved security of supplymajor findings and recommendations for each in some parts of the world.chapter of the report. The environmental impacts of shale develop-Supply ment are challenging but manageable. Shale development requires large-scale fracturingGlobally, there are abundant supplies of of the shale formation to induce economicnatural gas, much of which can be developed production rates. There has been concern thatat relatively low cost. The mean projection of these fractures can also penetrate shallowremaining recoverable resource in this report freshwater zones and contaminate them withis 16,200 Tcf, 150 times current annual global fracturing fluid, but there is no evidence thatnatural gas consumption, with low and high this is occurring. There is, however, evidenceprojections of 12,400 Tcf and 20,800 Tcf, of natural gas migration into freshwater zonesrespectively. Of the mean projection, approxi- in some areas, most likely as a result of sub-mately 9,000 Tcf could be developed economi- standard well completion practices by a fewcally with a natural gas price at or below $4/ operators. There are additional environmentalMillion British thermal units (MMBtu) at theexport point. Chapter 1: Overview and Conclusions 7
  • challenges in the area of water management, development through both research and particularly the effective disposal of fracture regulation. Transparency is key, both fluids. Concerns with this issue are particularly for fracturing operations and for water acute in regions that have not previously management. Better communication of experienced large-scale oil and natural gas oil- and gas-field best practices should development, especially those overlying the massive Marcellus shale, and do not have a be facilitated. Integrated regional water well-developed subsurface water disposal usage and disposal plans and disclosure of infrastructure. It is essential that both large and hydraulic fracture fluid components should small companies follow industry best practices; be required. that water supply and disposal are coordinated on a regional basis and that improved methods The U.S. should support unconventional are developed for recycling of returned fracture natural gas development outside U.S., fluids. particularly in Europe and China, as a means of diversifying the natural gas supply base. Natural gas trapped in the ice-like form known as methane hydrate represents a vast potential The U.S. government should continue to resource for the long term. Recent research is sponsor methane hydrate research, with a beginning to provide better definition of the particular emphasis on the demonstration overall resource potential, but many issues of production feasibility and economics. remain to be resolved. In particular, while there have been limited production tests, the long- term producibility of methane hydrates U.S. Natural Gas Production, Use and remains unproven, and sustained research Trade: Potential Futures will be required. In a carbon-constrained world, a level playing M A J O R R E CO M M E N D AT I O N S field — a CO2 emissions price for all fuels Government-supported research on the without subsidies or other preferential policy fundamental challenges of unconventional treatment — maximizes the value to society of the large U.S. natural gas resource. natural gas development, particularly shale gas, should be greatly increased Under a scenario with 50% CO2 reductions to in scope and scale. In particular, support 2050, using an established model of the global should be put in place for a comprehensive economy and natural gas cost curves that and integrated research program to include uncertainty, the principal effects of the build a system-wide understanding of associated CO2 emissions price are to lower all subsurface aspects of the U.S. shale energy demand and displace coal with natural gas in the electricity sector. In effect, gas-fired resource. In addition, research should be power sets a competitive benchmark against pursued to reduce water usage in fracturing which other technologies must compete in a lower and to develop cost-effective water carbon environment. A major uncertainty that recycling technology. could impact this picture in the longer term is technology development that lowers the costs A concerted coordinated effort by industry of alternatives, in particular, renewables, and government, both state and Federal, nuclear and CCS. should be organized so as to minimize the environmental impacts of shale gas8 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • A more stringent CO2 reduction of, for exam- Natural gas can make an importantple, 80% would probably require the complete contribution to GHG reduction in comingde-carbonization of the power sector. This decades, but investment in low-emissionmakes it imperative that the development of technologies, such as nuclear, CCS andcompeting low-carbon technology continues renewables, should be actively pursuedapace, including CCS for both coal and naturalgas. It would be a significant error of policy to to ensure that a mitigation regime can becrowd out the development of other, currently sustained in the longer term.more costly, technologies because of the newassessment of the natural gas supply. Con- Natural Gas for Electric Powerversely, it would also be a mistake to encourage,via policy and long-term subsidy, more costly In the U.S., around 30% of natural gas istechnologies to crowd out natural gas in the consumed in the electric power sector. Withinshort to medium term, as this could signifi- the power sector, gas-fired power plants playcantly increase the cost of CO2 reduction. a critical role in the provision of peaking capacity, due to their inherent ability to respondThe evolution of global natural gas markets is rapidly to changes in demand. In 2009, 23% ofunclear; but under some scenarios, the U.S. the total power generated was from natural gas,could import 50% or more of its natural gas while natural gas plants represented over 40%by 2050, despite the significant new resources of the total generating capacity.created in the last few years. Imports canprevent natural gas-price inflation in future In a carbon-constrained world, the poweryears. sector represents the best opportunity for a significant increase in natural gas demand, inM A J O R R E CO M M E N D AT I O N S direct competition with other primary energyTo maximize the value to society of the sources. Displacement of coal-fired power by gas-fired power over the next 25 to 30 years issubstantial U.S. natural gas resource base, the most cost-effective way of reducing CO2U.S. CO2 reduction policy should be designed emissions in the power create a “level playing field,” where allenergy technologies can compete against As a result of the boom in the constructioneach other in an open marketplace of gas-fired power plants in the 1990s, thereconditioned by legislated CO2 emissions is a substantial amount of underutilized NGCCgoals. A CO2 price for all fuels without capacity in the U.S. today. In the short term, displacement of coal-fired power by gas-firedlong-term subsidies or other preferential power provides an opportunity to reduce CO2policy treatment is the most effective way emissions from the power sector by about 20%,to achieve this result. at a cost of less than $20/ton of CO2 avoided.In the absence of such policy, interim energy This displacement would use existing generating capacity, and would, therefore, require little inpolicies should attempt to replicate as the way of incremental capital expenditure forclosely as possible the major consequences new generation capacity. It would also signifi-of a “level playing field” approach to carbon- cantly reduce pollutants such as sulfur dioxideemissions reduction. At least for the near (SO2), nitrous oxide (NOX), particulates andterm, that would entail facilitating energy mercury (Hg).demand reduction and displacement ofsome coal generation with natural gas. Chapter 1: Overview and Conclusions 9
  • Natural gas-fired power generation provides the END USE GAS DEMAND major source of backup to intermittent renew- able supplies in most U.S. markets. If policy In the U.S., around 32% of all natural gas support continues to increase the supply of consumption is in the Industrial sector, where intermittent power, then, in the absence of its primary uses are for boiler fuel and process affordable utility-scale storage options, addi- heat; and 35% of use is in the Residential and tional natural gas capacity will be needed to Commercial sectors, where its primary applica- provide system reliability. In some markets, tion is space heating. Only 0.15% of natural gas existing regulation does not provide the is used as a vehicle transportation fuel. appropriate incentives to build incremental capacity with low load factors, and regulatory Industrial, Commercial and Residential changes may be required. Within the Industrial sector, there are opportu- In the short term, where a rapid increase in nities for improved efficiency of the Industrial renewable generation occurs without any boiler fleet, replacing less-efficient natural gas adjustment to the rest of the system, increased boilers with high-efficiency, or super-high renewable power displaces gas-fired power efficiency boilers with conversion efficiencies up generation and thus reduces demand for to 94%. There are also opportunities to improve natural gas in the power sector. In the longer the efficiency of natural gas use in process term, where the overall system can adjust heating and to reduce process heating require- through plant retirements and new construc- ments through changes in process technologies tion, increased renewables displace baseload and material substitutions. generation. This could mean displacement of coal, nuclear or NGCC generation, depending Our analysis suggests that conversion of on the region and policy scenario under coal-fired boilers in the Industrial sector to consideration. For example, in the 50% CO2 high-efficiency gas boilers could provide a reduction scenario described earlier, where cost-effective option for compliance with new gas-fired generation drives out coal generation, hazardous air pollutant reductions and create increased renewable penetration as a result of significant CO2 reduction opportunities at cost reduction or government policy will modest cost, with a potential to increase natural reduce natural gas generation on a nearly gas demand by up to 0.9 Tcf/year. one-for-one basis. Natural gas and natural gas liquids (NGLs) are M A J O R R E CO M M E N D AT I O N S a principal feedstock in the chemicals industry The displacement of coal generation with and a growing source of hydrogen production NGCC generation should be pursued as for petroleum refining. Our analysis of selected the most practical near-term option for cases indicates that a robust domestic market for natural gas and NGLs will improve the significantly reducing CO2 emissions from competitiveness of manufacturing industries power generation. dependent on these inputs. In the event of a significant penetration of intermittent renewable production in Natural gas has significant advantages in the the generation technology mix, policy and Residential and Commercial sectors due in part regulatory measures should be developed to its cleanliness and life cycle energy efficiency. to facilitate adequate levels of investment However, understanding the comparative in natural gas generation capacity to ensure cost-effectiveness and CO2 impacts of different energy options is complex. Comparison of system reliability and efficiency.10 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • end use or “site” energy efficiencies can be Transportationmisleading, since it does not take into accountfull system energy use and emissions (such as The ample domestic supply of natural gas hasthe efficiency and emissions of electricity stimulated interest in its use in transportation.generation). However, quantitatively account- There are multiple drivers: the oil-natural gasing for the full system impacts from the “source” price spread on an energy basis generally favorscan be challenging, requiring a complex natural gas, and today that spread is at histori-end-to-end, full fuel cycle (FFC) analysis that cally high levels; an opportunity to lessen oilis not generally available to the consumer dependence in favor of a domestically suppliedor to the policy maker. fuel, including natural gas-derived liquid fuels with modest changes in vehicle and/or infra-Consumer and policy maker choices are further structure requirements and reduced CO2complicated by the influence of local climatic emissions in direct use of natural gas.conditions and regional energy markets. Theprimary energy mix of the regional generation Compressed natural gas (CNG) offers a signifi-mix fundamentally affects “site versus source” cant opportunity in U.S. heavy-duty vehiclesenergy and emissions comparisons. And the used for short-range operation (buses, garbagelocal climate has a major influence on the best trucks, delivery trucks), where payback timeschoice of heating and cooling systems, particu- are around three years or less and infrastructurelarly the appropriate use of modern space issues do not impede development. However,conditioning technologies such as heat pumps. for light passenger vehicles, even at 2010Consumer information currently available to oil-natural gas price differentials, high incre-consumers does not facilitate well-informed mental costs of CNG vehicles lead to longdecision making. payback times for the average driver, so signifi- cant penetration of CNG into the passengerExpanded use of combined heat and power fleet is unlikely in the short term. Payback(CHP) has considerable potential in the Indus- periods could be reduced significantly if thetrial and large Commercial sectors. However, cost of conversion from gasoline to CNG couldcost, complexity and the inherent difficulty of be reduced to the levels experienced in otherbalancing heat and power loads at a very small parts of the world such as Europe.scale make residential CHP a much moredifficult proposition. LNG has been considered as a transport fuel, particularly in the long-haul trucking sector.M A J O R R E CO M M E N D AT I O N S However, as a result of operational and infra-Improved energy efficiency metrics, which structure considerations as well as high incre- mental costs and an adverse impact on resaleallow consumers to accurately compare value, LNG does not appear to be an attractivedirect fuel and electricity end uses on a full option for general use. There may be anfuel cycle basis, should be developed. opportunity for LNG in the rapidly expandingOver time, these metrics should be tailored segment of hub-to-hub trucking operations, where infrastructure and operational challengesto account for geographical variations in the can be overcome.sources of electric power supply and localclimate conditions. Chapter 1: Overview and Conclusions 11
  • Energy density, ease of use and infrastructure U.S. production profiles, with supplies generally considerations make liquid fuels that are stable shifting from offshore Gulf of Mexico back to at room temperature a compelling choice in the onshore; shifts in U.S. population, generally Transportation sector. The chemical conversion from the Northeast and Midwest to the South of natural gas to liquid fuels could provide an and West and growth in global LNG markets, attractive alternative to CNG. Several pathways driven by price differences between regional are possible, with different options yielding markets. different outcomes in terms of total system CO2 emissions and cost. Conversion of natural gas The system generally responds well to market to methanol, as widely practiced in the chemi- signals. Changing patterns of supply and cals industry, could provide a cost-effective demand have led to a significant increase in route to manufacturing an alternative, or infrastructure development over the past few supplement, to gasoline, while keeping CO2 years with West to East expansions dominating emissions at roughly the same level. Gasoline pipeline capacity additions. Infrastructure engines can be modified to run on methanol limitations can temporarily constrain produc- at modest cost. tion in emerging production areas such as the Marcellus shale — but infrastructure capacity M A J O R R E CO M M E N D AT I O N S expansions are planned or underway. Demand The U.S. government should consider increases and shifts in consumption and production are expected to require around revision to its policies related to CNG $210 billion in infrastructure investment over vehicles, including how aftermarket CNG the next 20 years. conversions are certified, with a view to reducing up-front costs and facilitating Much of the U.S. pipeline infrastructure is CNG-gasoline capacity. old — around 25% of U.S. natural gas pipelines are 50 years old or older — and recent incidents The U.S. government should implement an demonstrate that pipeline safety issues are a open fuel standard that requires automobile cause for concern. The Department of Trans- manufacturers to provide tri-flex fuel portation (DOT) regulates natural gas pipeline (gasoline, ethanol and methanol) operation safety and has required integrity management in light-duty vehicles. Support for methanol programs for transmission and distribution fueling infrastructure should also be pipelines. The DOT also supports a small pipeline safety research program, which seems considered. inadequate given the size and age of the pipe- line infrastructure. Infrastructure Increased use of natural gas for power genera- tion has important implications for both The continental U.S. has a vast, mature and natural gas and electric infrastructures, includ- robust natural gas infrastructure, which ing natural gas storage. Historically, injections includes: over 300,000 miles of transmission and withdrawals from natural gas storage have lines; numerous natural gas-gathering systems; been seasonal. Increased use of natural gas for storage sites; processing plants; distribution power generation may require new high- pipelines and LNG import terminals. deliverability natural gas storage to meet more variable needs associated with power generation. Several trends are having an impact on natural gas infrastructure. These include changes in12 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • M A J O R R E CO M M E N D AT I O N S M A J O R R E CO M M E N D AT I O N SAnalysis of the infrastructure demands The EPA and the U.S. Department of Energyassociated with potential shift from coal to (DOE) should co-lead a new effort to review,gas-fired power should be undertaken. and update as appropriate, the methane emission factors associated with naturalPipeline safety technologies should be gas production, transmission, storageincluded in natural gas RD&D programs. and distribution. The review should have broad-based stakeholder involvement andEND USE EMISSIONS VERSUS should seek to reach a consensus on theSYSTEM-WIDE EMISSIONS appropriate methodology for estimating methane emissions rates. The analysisWhen comparing GHG emissions for different should, to the extent possible: (a) reflectenergy sources, attention should be paid to the actual emissions measurements; (b) addressentire system. In particular, the potential forleakage of small amounts of methane in the fugitive emissions for coal and oil as well asproduction, treatment and distribution of coal, natural gas; and (c) reflect the potential foroil and natural gas has an effect on the total cost-effective actions to prevent fugitiveGHG impact of each fuel type. The modeling emissions and venting of methane.analysis in Chapter 3 addresses the system-wideimpact, incorporating methane leakage fromcoal, oil and natural gas production, processing MARKETS AND GEOPOLITICSand transmission. In Chapter 5 we do notattempt to present detailed full-system account- The physical characteristics of natural gas,ing of CO2 (equivalent) emissions for various which create a strong dependence on pipelineend uses, although we do refer to its potential transportation systems, have led to localimpact in specific instances. markets for natural gas – in contrast to the global markets for oil.The CO2 equivalence of methane is conven-tionally based on a Global Warming Potential There are three distinct regional gas markets:(GWP)3 intended to capture the fact that each North America, Europe and Asia, with moreGHG has different radiative effects on climate localized markets elsewhere. The U.S. gasand different lifetimes in the atmosphere. market is mature and sophisticated, andIn our considerations, we follow the standard functions well, with a robust spot market.Intergovernmental Panel on Climate Change Within the U.S. market, the price of oil, (which(IPCC) and EPA definition that has been widely is set globally) compared to the price of naturalemployed for 20 years. Several recently published gas (which is set regionally) is very importantlife cycle emissions analyses do not appear to be in determining market share when there is thecomprehensive, use common assumptions or opportunity for substitution. Over the lastrecognize the progress made by producers to decade or so, when oil prices have been high,reduce methane emissions, often to economic the ratio of the benchmark West Texas Inter-benefit. We believe that a lot more work is mediate oil price to the Henry Hub natural gasrequired in this area before a common under- price has been consistently higher than any ofstanding can be reached. Further discussion the standard rules of thumb.can be found in Appendix 1A. Chapter 1: Overview and Conclusions 13
  • International natural gas markets are in the early stages of integration, with many impedi- pipelines and pipeline routes is intense in ments to further development. While increased key regions. LNG trade has started to connect these mar- kets, they remain largely distinct with respect to supply patterns, pricing and contract struc- ability of the natural gas infrastructure. tures, and market regulation. If a more inte- grated market evolves, with nations pursuing M A J O R R E CO M M E N D AT I O N S gas production and trade on an economic basis, The U.S. should pursue policies that there will be rising trade among the current encourage the development of an efficient regional markets and the U.S. could become and integrated global gas market with a substantial net importer of LNG in future decades. transparency and diversity of supply. Natural gas issues should be fully integrated Greater international market liquidity would into the U.S. energy and security agenda, be beneficial to U.S. interests. U.S. prices for and a number of domestic and foreign natural gas would be lower than under current regional markets, leading to more gas use in the policy measure should be taken, including: U.S. Greater market liquidity would also contribute to security by enhancing diversity conduct of U.S. foreign policy, which will of global supply and resilience to supply require multiagency coordination with disruptions for the U.S. and its allies. These factors ameliorate security concerns about leadership from the Executive Office of import dependence. the President; As a result of the significant concentration of Energy Agency (IEA) to place more atten- conventional gas resources globally, policy and geopolitics play a major role in the develop- tion on natural gas and to incorporate the ment of global supply and market structures. large emerging markets (such as China, Consequently, since natural gas is likely to play India and Brazil) into the IEA process as a greater role around the world, natural gas integral participants; issues will appear more frequently on the U.S. energy and security agenda. Some of the specific security concerns are: expansion of unconventional resources; and of allies, could constrain U.S. foreign policy options, especially in light of the unique cyber-security as the global gas delivery American international security system becomes more extended and responsibilities. interconnected. impediments to the development of transparent markets.14 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • RD&D While natural gas can provide a cost-effective bridge to a low carbon future, it is vital thatThere are numerous RD&D opportunities to efforts continue to improve the cost andaddress key objectives for natural gas supply, efficiency of low or zero carbon technologiesdelivery and end use: for the longer term. This will require sustained RD&D and subsidies of limited duration to encourage early deployment. development as an important contributor to the public good; M A J O R R E CO M M E N D AT I O N S The Administration and Congress should support RD&D focused on environmentally natural gas production, delivery and use; responsible domestic natural gas supply. This should entail both a renewed applications for public policy purposes, such DOE program, weighted towards basic as emissions reductions and diminished oil research, and a complementary industry- dependence; led program, weighted towards applied research, development and demonstration, that is funded through an assured funding infrastructure; stream tied to energy production, delivery - and use. The scope of the program should sion and end-use so as to use the resource be broad, from supply to end-use. most effectively. Support should be provided through RD&D, and targeted subsidies of limited duration,Historically, RD&D funding in the natural gasindustry has come from a variety of sources, for low-emission technologies that have theincluding private industry, the DOE, and prospect of competing in the long run. Thisprivate/public partnerships. In tandem with would include renewables, carbon capturelimited tax credits, this combination of support and sequestration for both coal and gasplayed a major role in development of uncon- generation, and nuclear power.ventional gas. It has also contributed to moreefficient end-use, for example in the develop-ment of high-efficiency gas turbines.Today government funded RD&D for naturalgas is at very low levels. The elimination ofrate-payer funded RD&D has not been com-pensated by increased DOE appropriationsor by a commensurate new revenue streamoutside the appropriations process. The totalpublic and public-private funding for naturalgas research is down substantially from its peakand is more limited in scope, even as naturalgas takes a more prominent role in a carbon-constrained world. Chapter 1: Overview and Conclusions 15
  • CONCLUSION Over the past few years, the U.S. has developed an important new natural gas resource that fundamentally enhances the nation’s long-term gas supply outlook. Given an appropriate regulatory environment, which seeks to place all lower carbon energy sources on a level competitive playing field, domestic supplies of natural gas can play a very significant role in reducing U.S. CO2 emissions, particularly in the electric power sector. This lowest cost strategy of CO2 reduction allows time for the continued development of more cost-effective low or zero carbon energy technology for the longer term, when gas itself is no longer sufficiently low carbon to meet more stringent CO2 reduction targets. The newly realized abundance of low cost gas provides an enor- mous potential benefit to the nation, providing a cost effective bridge to a secure and low carbon future. It is critical that the additional time created by this new resource is spent wisely, in creating lower cost technology options for the longer term, and thereby ensuring that the natural gas bridge has a safe landing place in a low carbon future. NOTES EIA 2009 Annual Energy Review, Figure 45. 2 Global-warming potential (GWP) is a relative 3 1 One quadrillion Btu (or “quad”) is 1015 or measure of how much heat a given greenhouse gas 1,000,000,000,000,000 British thermal units. Since traps in the atmosphere. one standard cubic foot of gas is approximately 1,000 Btu, then 1 quad is approximately 1 Tcf of gas.16 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Chapter 2: SupplyINTRODUCTION AND CONTEXT Thermogenic1 natural gas, which is formed by the application, over geological time, of enormousIn this chapter, we discuss various aspects of heat and pressure to buried organic matter,natural gas supply: how much natural gas exists exists under pressure in porous rock formationsin the world; at what rate can it be produced thousands of feet below the surface of the earth.and what it will cost to develop. Following the It exists in two primary forms: “associated gas” isintroduction and definitions, we look at produc- formed in conjunction with oil, and is generallytion history, resource volumes and supply costs released from the oil as it is recovered from thefor natural gas — first from a global perspective, reservoir to the surface — as a general rule theand then focusing in more detail on the U.S., gas is treated as a by-product of the oil produc-paying particular attention to the prospects for tion process; in contrast, “non-associated gas” isshale gas. We then discuss the science and found in reservoirs that do not contain oil, and istechnology of unconventional gas, the environ- developed as the primary product. While associ-mental impacts of shale gas development and ated gas is an important source, the majority offinally the prospects for methane hydrates. gas production is non-associated; 89% of the gas produced in the U.S. is non-associated.NATURAL GAS AND THERECOVERY PROCESS Non-associated gas is recovered from the forma- tion by an expansion process. Wells drilled intoThe primary chemical component of natural gas the gas reservoir allow the highly compressedis methane, the simplest and lightest hydrocar- gas to expand through the wells in a controlledbon molecule, comprised of four hydrogen (H) manner, to be captured, treated and transportedatoms bound to a single carbon (C) atom. In at the surface. This expansion process generallychemical notation, this is expressed as CH4 leads to high recovery factors from conventional,(the symbol for methane). Natural gas may good-quality gas reservoirs. If, for example, thealso contain small proportions of heavier average pressure in a gas reservoir is reducedhydrocarbons: ethane (C2H6); propane (C3H8) from an initial 5,000 pounds per square inchand butane (C4H10); these heavier components (psi) to 1,000 psi over the lifetime of the field,are often extracted from the producing stream then approximately 80% of the Gas Initially Inand marketed separately as natural gas liquids Place (GIIP) will be recovered. This is in contrast(NGL). In the gas industry, the term “wet gas” to oil, where recovery factors of 30% to 40% areis used to refer to natural gas in its raw unpro- more typical.cessed state, while “dry gas” refers to natural gasfrom which the heavier components have been Gas is found in a variety of subsurface locations,extracted. with a gradation of quality as illustrated in the resource triangle in Figure 2.1. Chapter 2: Supply 17
  • Figure 2.1 GIIP as a Pyramid in Volume and Quality. Conventional reservoirs are at the top of the pyramid. They are of higher quality because they have high permeability and require less technology for development and production. The unconventional reservoirs lie below the conventional reservoirs in this pyramid. They are more abundant in terms of GIIP but are currently assessed as recoverable resources — and commercially developed — primarily in North America. They have lower permeability, require advanced technology for production and typically yield lower recovery factors than conventional reservoirs. Increasing Technology/Decreasing Recovery Factor Conventional Resources High-Quality Reservoirs Low-Quality Reservoirs Unconventional Resources Tight Gas Sands Coal Bed Methane Shale Gas Methane Hydrates Volume Adapted from Holditch 2006 Conventional resources exist in discrete, sandstone formations, coal beds (coal bed well-defined subsurface accumulations (reser- methane or CBM) and shale formations. voirs), with permeability2 values greater than Unconventional resource accumulations tend a specified lower limit. Such conventional to be distributed over a larger area than con- gas resources can usually be developed using ventional accumulations and usually require vertical wells, and generally yield the high advanced technology such as horizontal wells recovery factors described above. or artificial stimulation in order to be economi- cally productive; recovery factors are much By contrast, unconventional resources are lower — typically of the order of 15% to 30% found in accumulations where permeability is of GIIP. The various resource types are shown low. Such accumulations include “tight” schematically in Figure 2.2.18 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.2 Illustration of Various Types of Gas Resource Schematic geology of natural gas resource Land surface Conventional non-associated Coal bed methane gas Conventional associated gas Seal Oil Sandstone Tight sand gas Gas-rich shale Source: U.S. Energy Information AdministrationRESOURCE DEFINITIONS Gas resources are an economic concept — a function of many variables, in particular the cost of exploration,The complex cross-dependencies betweengeology, technology and economics mean that production and transportation relative to the pricethe use of unambiguous terminology is critical of sale to users.when discussing natural gas supply. In thisstudy, the term “resource” will refer to the sum Figure 2.3 illustrates how proved reserves,of all gas volumes expected to be recoverable in reserve growth and undiscovered resourcesthe future, given specific technological and combine to form the “technically recoverableeconomic conditions. The resource can be resource,” that is, the total volume of naturaldisaggregated into a number of sub-categories; gas that could be recovered in the future,specifically, “proved reserves,” “reserve growth” using today’s technology, ignoring economic(via further development of known fields) and constraints.“undiscovered resources,” which represent gasvolumes that are expected to be discovered inthe future via the exploration process. Chapter 2: Supply 19
  • Figure 2.3 Modified McKelvey Diagram, Showing the Interdependencies between Geology, Technology and Economics and Their Impacts on Resource Classes; Remaining Technically Recoverable Resources Are Outlined in Red Discovered/Identified Undiscovered Confirmed Unconfirmed Cumulative Production Economic Increasing Economic Viability Recoverable Inferred Technically Undiscovered Reserves/ Technically Reserve Recoverable Reserves Growth Resources Sub-economic Unrecoverable Technically Increasing Geologic Knowledge The methodology used in analyzing natural gas associated with the underlying geological, supply for this study places particular emphasis technological, economic and political in two areas: conditions. The analysis of natural gas supply in this study has been carried out 1. Treating gas resources as an economic in a manner that frames any single point concept — recoverable resources are a resource estimate within an associated function of many variables, particularly the uncertainty envelope, in order to illustrate ultimate price that the market will pay. A set the potentially large impact this ever- of supply curves has been developed using present uncertainty can have. the ICF3 Hydrocarbon Supply Model with volumetric and fiscal input data supplied by The volumetric data used as the basis of the ICF and MIT. These curves describe the analysis for both the supply curve development volume of gas that is economically recover- and the volumetric uncertainty analysis was able for a given gas price. These curves form compiled from a range of sources. In particular, a primary input to the integrated economic use has been made of data from work at the modelling in Chapter 3 of this report. United States Geological Survey (USGS), the Potential Gas Committee (PGC), the Energy 2. Recognizing and embracing uncertainty — Information Agency (EIA), the National uncertainty exists around all resource Petroleum Council (NPC) and ICF International. estimates due to the inherent uncertainty20 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • GLOBAL SUPPLY to demonstrating the overwhelming scale of the United States and Russia compared to otherProduction Trends producing countries, this figure illustrates the very significant growth rates in other countries.Over the past two decades, global production The substantial growth of new gas producingof natural gas has grown significantly, rising countries over the period reflects the relativeby almost 42% overall from approximately immaturity of the gas industry on a global74 trillion cubic feet (Tcf )4 in 1990 to 105 Tcf basis outside Russia and North America, thein 2009. This is almost twice the growth rate expansion of gas markets and the rise in globalof global oil production, which increased by cross-border gas trade.around 22% over the same period. Much of thegas production growth has been driven by the Between 1993 and 2008, global cross-border gasrapid expansion of production in areas that trade almost doubled, growing from aroundwere not major gas producers prior to 1990. 18 Tcf (25% of global supply) to around 35 TcfThis trend is illustrated in Figure 2.4, which (32% of global supply). Most of the world’s gasshows how growth in production from regions supply is transported from producing fields tosuch as the Middle East, Africa and Asia & market by pipeline. However, the increase inOceania has significantly outpaced growth in global gas trade has been accelerated by thethe traditional large producing regions, includ- growing use of Liquefied Natural Gas (LNG),ing North America and Eurasia (primarily which is made by cooling natural gas to aroundRussia). -162°C. Under these conditions, natural gas becomes liquid, with an energy density 600Figure 2.5 compares the 1990 and 2009 annual times that of gas at standard temperature andproduction levels for the 10 largest gas-producing pressure — and it can be readily transportednations (as defined by 2009 output). In addition over long distances in specialized ocean-goingFigure 2.4 Trends in Annual Global Dry Gas Production by Region between 1990 and 2009 Tcf of Gas 120 100 80 S. & C. America 60 Africa Europe 40 Middle East Asia & Oceania 20 North America Eurasia 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: MIT; U.S. Energy Information Administration Chapter 2: Supply 21
  • Figure 2.5 Comparison of 1990 and 2009 Natural Gas Production Levels for the Top 10 Natural Gas Producing Nations (as defined by 2009 output) Tcf of Gas 25 1990 2009 20 15 10 5 0 United States Russia Canada Iran Norway Quatar Algeria Netherlands Saudi Arabia China Source: MIT; U.S. Energy Information Administration Figure 2.6 Global Cross-Border Gas Trade Tcf of Gas 35 LNG Pipeline 30 8.0 25 20 3.0 15 10 14.8 26.4 5 0 1993 2008 Source: MIT; U.S. Energy Information Administration22 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • LNG tankers. Over this 15-year period, global Figure 2.7 depicts the estimated remaininggas trade doubled, while LNG trade increased recoverable gas resources, together with esti-even more rapidly, as shown in Figure 2.6. mated uncertainty,6 broken down by regions as defined by the Emissions Prediction and PolicyRESOURCES5 Analysis (EPPA) model employed in Chapter 3 of this report. Figure 2.8 depicts the geographi-Global natural gas resources are abundant. The cal distribution of EPPA regions, together withmean remaining resource base is estimated to the mean resource estimate for each 16,200 Tcf, with a range between 12,400 Tcf The resources are comprised of three major(with a 90% probability of being exceeded) and components defined above: reserves, reserve20,800 Tcf (with a 10% probability of being growth and yet-to-find resources. For the U.S.exceeded). The mean projection is 150 times and Canada, we have also included a fourththe annual consumption in 2009. With the category, unconventional resources. As discussedexception of Canada and the U.S., this estimate later, due to the very high levels of uncertaintydoes not include any unconventional supplies. at this stage, we have not included unconven-The global gas supply base is relatively imma- tional resource estimates for other regions.ture; outside North America only 11% of theestimated ultimately recoverable conventionalresources have been produced to date.Figure 2.7 Global Remaining Recoverable Gas Resource by EPPA Region, with Uncertainty Middle East Russia United States Africa Central Asia and Rest of Europe Canada Rest of Americas EU27 and Norway Dynamic Asia Brazil Rest of East Asia Total Reserves Australia and Oceania Reserve Growth (Mean) Conventional Undiscovered Gas (Mean) China Unconventional Gas (Mean) Mexico Low High India RRR RRR 0 1,000 2,000 3,000 4,000 5,000 6,000 Tcf of gas Source: MIT analysis based on data and information from: Ahlbrandt et al. 2005; United States Geological Survey 2010; National Petroleum Council 2003; United States Geological Survey n.d.; Potential Gas Committee 1990; Attanasi & Coburn 2004; Energy Information Administration 2009 1.0 0.8 0.6 0.4 0.2 0.0 Chapter 2: Supply 23
  • Although resources are large, the supply base is even more geographically concentrated than oil concentrated geographically, with an estimated supplies. Political considerations and individual 70% in only three regions: Russia, the Middle country depletion policies play at least as big a East (primarily Qatar and Iran) and North role in global gas resource development as America (where North American resources geology and economics, and dominate the also include unconventional gas). By some evolution of the global gas market. measures, global supplies of natural gas areFigure 2.8 Map of EPPA Regions, and Mean Resource Estimates Middle East [4670 TCF] Canada [820 TCF] Rest of East Asia [240 TCF] Russia [3410 TCF] Rest of Americas [800 TCF] Austria and Oceania [225 TCF] United States [2150 TCF] EU27 and Norway [720 TCF] China [210 TCF] Africa [1050 TCF] Dynamic Asia [480 TCF] India/Mexico [95/50 TCF] Eastern Europe and Central Asia [940 TCF] Brazil [350 TCF] Japan [~0 TCF] Source: EPPA, MIT24 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • SUPPLY COSTS7 of illustration, resources that can be economically developed at a gas price of $1 or $2/million Figure 2.9 depicts a set of global supply curves, British thermal units (MMBtu) may well require which describe the resources of gas that can be an additional $3 to $5/MMBtu of transport costs developed economically at given prices at the to get to their ultimate destination. These high point of export. The higher the price, the more transportation costs are also a significant factor gas will ultimately be developed. Much of the in the evolution of the global gas market. global supply can be developed economically with relatively low prices at the wellhead or the Figure 2.10 depicts the mean gas supply curves point of export.8 However, the cost of delivering for those EPPA regions that contain significant this gas to market is generally considerably gas resources. Again, this illustrates the significant higher. concentration of gas resources in the world. In contrast to oil, the total cost of delivering gas In contrast to oil, the total cost of getting gas to international markets is strongly influenced by transportation costs, either via long-distance to international markets is strongly influenced pipeline or as LNG. Transportation costs will by the cost of transportation — a significant factor obviously be a function of distance, but by way in the evolution of the global gas market. Figure 2.9 Global Gas Supply Cost Curve, with Uncertainty; 2007 Cost Base Breakeven gas price at point of export: $/MMBtu 20.00 Example LNG value chain 18.00 costs incurred during gas delivery 16.00 $/MMBtu 14.00 Liquefaction $2.15 12.00 Shipping $1.25 Regasification $0.70 10.00 Total $4.10 8.00 Volumetric uncertainty around Low mean of 16,200 Tcf 6.00 Mean 4.00 High Low High 12,400 20,800 2.00 0 0 3,000 6,000 9,000 12,000 15,000 18,000 Source: MIT; ICF Global Hydrocarbon Supply Model Tcf of gas 1.0 0.8 0.620000 0.4 0.2 0.015000 Chapter 2: Supply 25
  • Figure 2.10 Global Gas Supply Cost Curve by EPPA Region; 2007 Cost Base Breakeven Gas Price Africa $MMBtu Australia and Oceania $20.00 Brazil China $15.00 Dynamic Asia Europe India $10.00 Mexico Middle East $5.00 North America Rest of Americas Rest of East Asia $0 Rest of Europe 0 500 1,000 1,500 2,000 2,500 3,000 3.500 4,000 4,500 5,000 and Central Asia Russia Source: MIT; ICF Global Hydrocarbon Supply Model UNCONVENTIONAL RESOURCES9 Given the concentrated nature of conventional Africa supplies and the high costs of long-distance Outside of Canada and the U.S., there has been Australia transportation, there may be considerable and Oceania very little development of the unconventional strategic and economic value in the development gas supply base — indeed there has been little of unconventional resources in those Brazil regions need when conventional resources are so China that are currently gas importers, such as Europe abundant. But due to this lack of development, Dynamic Asia and China. It would be in the strategic interest unconventional resource estimates are sparse Europe of the U.S. to see these indigenous supplies and unreliable. developed. As a market leader in this India technol- ogy, the U.S. could play a significant role in Mexico Based on an original estimate by Rogner10, facilitating this development. Middle East there may be of the order of 24,000 Tcf of Rest of Americas unconventional GIIP outside North America. R E CO M M E N D AT I O N Rest of East Asia Applying a nominal 25% recovery factor, this U.S. policy should encourage the strategic Rest of Europe would imply around 6,000 Tcf of unconven- and Central Asia development of unconventional gas tional recoverable resources. However, these Russia supplies in regions which currently depend global estimates are highly speculative, almost completely untested and subject to very wide on imported gas, in particular, Europe bands of uncertainty. There is a long-term need and China. for basin-by-basin resource evaluation to provide credibility to the GIIP estimates and, most importantly, to establish realistic estimates of recoverable resource volumes and costs11.26 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • UNITED STATES SUPPLY production figures reported by the EIA. Small volumes of gas are exported from Alaska toProduction Trends Japan as LNG.There is significant geographical variation in Figure 2.12 illustrates the regional breakdownU.S. natural gas production levels. For the of dry natural gas production in the U.S. sincepurposes of this discussion of U.S. production, 2000. Some level of production occurs inwe will use the U.S. EIA pipeline regions all eight regions, but the dominance of the(Figure 2.11). Southwest, Gulf of Mexico and Central regions is clearly shown. The dynamics of the produc-Natural gas production in the U.S. has tradi- tion levels across these major regions havetionally been associated with the Southwest differed appreciably over the past decade. In theregion and the Gulf of Mexico. However, Southwest, the largest gas producing region,significant production also takes place in Alaska annual production levels remained relativelyand in the Central region. In the case of Alaska, flat at about 9.3 Tcf from 2000 to 2005. Sincethe vast majority of the gas is associated with 2005, output from the region has increased,oil production on the North Slope, and due to growing by 21% to 11.4 Tcf in 2008. Much ofthe lack of an export mechanism, this gas is this growth in the latter half of the decade is there-injected to enhance recovery from Alaskan result of rapid expansion in the production ofoil fields. These gas production volumes are gas from shale plays.therefore not included in the national gasFigure 2.11 EIA Natural Gas Pipeline Regions for the L48 States; the State of Alaska andthe U.S. Offshore Territory in the Gulf of Mexico Form Two Additional Regions Central WA Midwest Northeast ME MT ND MN VT NH OR WI MA NY SD MI ID RI WY CT PA NJ IA IN OH NE IL DE NV UT WV VA MD CO CA KS MO KY NC AZ TN OK NM AR SC Western MS AL GA Southeast TX LA FL SouthwestSource: U.S. Energy Information Administration Chapter 2: Supply 27
  • Figure 2.12 Regional Breakdown of Annual Dry Gas Production in the U.S. between 2000 and 2009 Tcf of Gas 22 20 18 Western 16 Southwest 14 Southeast 12 Northeast 10 Midwest 8 GOM 6 Central 4 Alaska 2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: MIT; U.S. Energy Information Administration Since 2000, the Central region has seen the PRODUCTION TRENDS BY RESOURCE greatest percentage growth in production TYPE IN THE UNITED STATES among U.S. regions. Annual dry gas output has risen from 2.6 Tcf to 4.5 Tcf, an overall increase In a global context, U.S. gas production by type of 75%. Unlike the Southwest region, produc- is extremely diverse. Both conventional and tion from the Central region has grown con- unconventional gas output is significant, with tinuously since 2000, with output increasing the contribution of unconventional gas growing from all resource types. In marked contrast, gas steadily year-on-year. output from offshore fields in the Gulf of Mexico has fallen dramatically from approxi- Figure 2.13a plots contributions to production mately 5 Tcf in 2000 to 2.4 Tcf in 2008, the from conventional, unconventional and result of fewer new wells being brought online associated gas. This breakdown illustrates the in the Gulf to replace those older wells that are marked shift towards unconventional resources now in decline or have been taken off produc- that has been a feature of gas production in the tion. This decline is an indication of the U.S. over the past decade and more. In 2000, maturity of the conventional resource base the combined gross production of conventional in the Gulf of Mexico. and associated gas in the L48 states was 14.6 Tcf (71% of total output). By 2009, the combined conventional and associated output had fallen to 11.4 Tcf (52% of the total). In concert with this fall in conventional and associated gas production, there has been continuous expan- sion in the production of unconventional gas, with approximately 4.5 Tcf more unconven- tional gas being produced in 2009 than in 2000.28 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.13a Breakdown by Type of Figure 2.13b Percentage Breakdown byAnnual Gross Gas Production in the L48 Type of Gross Gas Production in the L48U.S. between 2000 and 2009 U.S. in 2000 and 2009 Tcf of Gas 25 1% 7% 14% 20 21% 9% 15 Shale 16% 25% CBM 10 Tight 11% Associated Conventional 5 55% 41% 0 2001 2003 2005 2007 2009 2000 2009 Source: MIT; HPDI production databaseHistorically, tight gas has been the most signifi- emergence of shale gas. Although shalecant source of unconventional gas production resources have been produced in the U.S. sincein the U.S., and is likely to remain so for some 1821, the volumes have not been significant.time. Tracking tight gas production can be This situation changed fundamentally duringdifficult because it can exist in a continuum the past decade as technological advanceswith conventional gas. However, a review of enabled production from shales previouslyoutput from known tight plays shows a growth considered uneconomical. Expansion in shalein annual output from 4.5 Tcf to 5.6 Tcf gas output is illustrated in Figures 2.13a andbetween 2000 and 2009, an increase from 21% 2.13b. From 2000 to 2009, the contribution ofto 25% of total gross production as shown in shale gas to overall production grew from 0.1Figure 2.13b. Commercial production of CBM Tcf, or less than 1%, to 3.0 Tcf, or nearly 14%.began at the end of the 1980s, and grew sub- This growth is all the more remarkable instantially during the 1990s from an output of that 80% of it was driven by one play, the0.2 Tcf in 1990 to 1.3 Tcf in 1999. This growth Barnett shale, located in Texas’ Fort Worthmoderated during the last decade, with 2009 Basin. Activity in other shale plays has alsoCBM output standing at 1.92 Tcf or 9% of been increasing, with appreciable volumesthe total. now being produced from the Fayetteville and Woodford shales in the Arkoma Basin, theAside from the fall in conventional production, Haynesville shale in the East Texas Basin andthe most striking feature of the gas production as of the end of 2009, the Marcellus shale in thein the U.S. this past decade has been the Appalachian Basin. Chapter 2: Supply 29
  • U.S. RESOURCES12 Around 15% of the U.S. resource is in Alaska, and full development of this resource will Table 2.1 illustrates mean U.S. resource esti- require major pipeline construction to bring mates from a variety of resource assessment the gas to market in the L48 states. Given the authorities. These numbers have tended to abundance of L48 supplies, development of the grow over time, particularly as the true poten- pipeline is likely to be deferred yet again, but tial of the unconventional resource base has this gas represents an important resource for started to emerge over the past few years. the future. For this study, we have assumed a mean In the L48, some 55% to 60% of the resource remaining resource base of around 2,100 Tcf. base is conventional gas, both onshore and This corresponds to approximately 92 times the offshore. Although mature, the conventional annual U.S. consumption of 22.8 Tcf in 2009. resource base still has considerable potential. We estimate the low case (with a 90% probabil- Around 60% of this resource is comprised of ity of being met or exceeded) at 1,500 Tcf, and proved reserves and reserve growth, with the the high case (with a 10% probability of being remainder — of the order of 450 to 500 Tcf — met or exceeded) at 2,850 Tcf. from expected future discoveries. Table 2.1 Tabulation of US Resource Estimates by Type, from Different Sources NPC USGS/MMS PGC ICF (2003) (Various Years) (2006) (2008) (2009) L48 Conventional 691 928 693 869 Tight 175 190 966 174 Shale 35 85 616 631 CBM 58 71 108 99 65 Total L48 959 1,274 1,074 1,584 1,563 Alaska Conventional 237 357 194 237 Tight – – 194 Shale – – – – CBM 57 18 57 57 57 Total Alaska 294 375 251 251 294 U.S. Conventional 929 1,284 930 1,063 Tight 175 190 1,160 174 Shale 35 85 616 631 CBM 115 89 165 156 122 Total U.S. 1,254 1,648 1,325 1,835 1,857 Proved Reserves 184 245 204 245 245 Total (Tcf) 1,438 1,893 1,529 2,080 2,102 Source: National Petroleum Council 2003; United States Geological Survey 2010; Minerals Management Service 2006; Potential Gas Committee 2007; Potential Gas Committee 2009; Energy Information Administration 200930 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.14a Volumetric Uncertainty of U.S. Gas Figure 2.14b Breakdown of Mean U.S. Gas SupplySupply Curves; 2007 Cost Base Curve by Type; 2007 Cost BaseBreakeven Gas Price Breakeven Gas Price$/MMBtu $/MMBtu40.00 40.0036.00 36.0032.00 32.0028.00 28.0024.00 24.0020.00 20.0016.00 16.0012.00 12.00 Conventional Low 8.00 Tight 8.00 Mean Shale 4.00 4.00 CBM High 0 0 0 500 1,000 1,500 2,000 2,500 3,000 0 100 200 300 400 500 600 700 800 900 1,000 Tcf of gas Tcf of gas Source: MIT; ICF North American Hydrocarbon Supply Model Source: MIT; ICF North American Hydrocarbon Supply ModelFigure 2.14a represents the supply curves for Despite the relative maturity of the U.S. gas 1.0 1.0the aggregate of all U.S. resources, depicting the 0.8 0.6 supply, estimates of remaining resources have 0.8 0.6mean estimate and the considerable range of continued to grow over time — with an acceler- 0.4 0.4 0.2 0.2 0.0uncertainty. Figure 2.14b illustrates the mean ating trend in recent years, mainly attributable to 0.0supply curves, broken down by resource type. It unconventional gas, especially in the shales.clearly shows the large remaining conventionalresource base, although it is mature and some of The PGC, which evaluates the U.S. gas resourceit will require high gas prices to become eco- on a biannual cycle, provides perhaps the bestnomical to develop. These curves assume current historical basis for looking at resource growthtechnology. In practice, future technology over time. According to this data, remainingdevelopment 3000 will enable these costs to be driven resources have grown by 77% since 1990, 1200down over time, allowing a larger portion of the despite a cumulative production volume duringresource base2500 to be economically developed. that time of 355 Tcf. 1000 Don’t Use This One, has all data, real one is on “chart to cut”Figure 2.14b also demonstrates the consider- As a subset of this growth process, Useappli- Don’t the This Oneable potential of shale supplies. Using a 2007 2000 cation of horizontal drilling and hydraulic 800cost base, a substantial portion of the estimated fracturing technology to the shales has causedshale resource base is economic at prices 1500 resource estimates to grow over a five-year 600between $4/MMBtu and $8/MMBtu. As we see period from a relatively minor 35 Tcf (NPC,in the current U.S. gas markets, some of the 2003), to a current estimate of 615 Tcf (PGC,shale resources will displace higher-cost con- 1000 2008), with a range of 420 to 870 Tcf. This 400ventional gas in the short to medium term,exerting downward pressure on gas prices. 500 200 0 0 Chapter 2: Supply 31
  • resource growth is a testament to the power This variability in performance is incorporated of technology application in the development in the supply curves on the previous page, as of resources, and also provides an illustration well as in Figure 2.15. Figure 2.15a shows initial of the large uncertainty inherent in all resource production and decline data from three major estimates. U.S. shale plays, illustrating the substantial differences in average well performanceAccording to Potential Gas Committee data, U.S. between the plays. Figure 2.15b shows a prob- ability distribution of initial flow rates from thenatural gas remaining resources have grown by 77% Barnett formation. While many refer to shalesince 1990, a testament to the power of technology, development as more of a “manufacturingand an illustration of the large uncertainty inherent process,” where wells are drilled on a statisticalin all resource estimates. basis — in contrast to a conventional explora- tion, development and production process, The new shale plays represent a major where each prospective well is evaluated on an contribution to the resource base of the U.S. individual basis — this “manufacturing” still However, it is important to note that there is occurs within the context of a highly variable considerable variability in the quality of the subsurface environment. resources, both within and between shale plays.Figure 2.15a Illustration of Variation in Mean Figure 2.15b Illustration of Variation in InitialProduction Rates between Three Shale Plays Production Rates of 2009 Vintage Barnett WellsProduction Rate IP Rate ProbabilityMcf/day (30-day average) (Barnett 2009 Well Vintage)9,000 0.128,000 0.107,0006,000 0.085,000 0.064,0003,000 Haynesville 0.04 Marcellus2,000 Barnett 0.021,000 0 0 0 1 2 3 4 5 0 1,000 2,000 3,000 4,000 5,000 Year IP Rate Mcf/day (30-day avg) Source: MIT analysis; HPDI production database and various Source: MIT analysis; HPDI production database and various industry sources industry sources 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 IP Rate Probability32 MIT STUDY ON THE FUTURE OF NATURAL GAS (Barnett ’09 Well Vintage)
  • This high level of variability in individual well amounts of liquid, which can have a consider-productivity clearly has consequences with able effect on the breakeven economics — par-respect to the variability of individual well ticularly if the price of oil is high compared toeconomic performance.13 This is illustrated in the price of gas.Table 2.2, which shows the variation in break-even gas price as a function of initial productiv- The liquid content of a gas is often measuredity for the five major U.S. shale plays. The P20 in terms of the “condensate ratio,” expressed in30-day initial production rate represents the terms of barrels of liquid per million cubic feetrate that is equaled or exceeded by only 20% of of gas (bbls/MMcf). Figure 2.16 shows thethe wells completed in 2009; the P80 represents change in breakeven gas price for varyingthe initial rate equaled or exceeded by 80% of condensate ratios in a typical Marcellus well,14completed wells. assuming a liquids price of $80/bbl. It can be seen that for a condensate ratio in excess ofAnother major driver of shale economics is the approximately 50 bbls/MMcf in this particularamount of hydrocarbon liquid produced along case, the liquid production alone can providewith gas. The results in Table 2.2 assume dry an adequate return on the investment, even ifgas with no liquid co-production; however, the gas were to realize no market value.some areas contain wet gas with appreciableTable 2.2 Full-Cycle 2009 Well Vintage P20, P50 and P80 30-Day Average InitialProduction (IP) Rates and Breakeven Prices (BEP) for Each of the Major U.S. Shale PlaysAssuming Mid Case Costs Barnett Fayetteville Haynesville Marcellus Woodford IP BEP IP BEP IP BEP IP BEP IP BEP Mcf/d $/Mcf Mcf/d $/Mcf Mcf/d $/Mcf Mcf/d $/Mcf Mcf/d $/Mcf P20 2700 $4.27 3090 $3.85 12630 $3.49 5500 $2.88 3920 $4.12 P50 1610 $6.53 1960 $5.53 7730 $5.12 3500 $4.02 2340 $6.34 P80 860 $11.46 1140 $8.87 2600 $13.42 2000 $6.31 790 $17.04Source: MIT analysis Chapter 2: Supply 33
  • Figure 2.16 Estimated Breakeven Gas Price ($/MMBtu) for a Mean Performing 2009 Vintage Marcellus Shale Well, with Varying Condensate Ratio (bbl/MMcf), Assuming a Liquids Price of $80/bbl Breakeven gas price $/MMBtu 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 5 10 15 20 25 30 35 40 45 50 Condensate Ratio Source: MIT analysis bbl/MMcf The effects described above create an interest- Figure 2.17 indicates how production from the ing dynamic in U.S. gas supply. Gas prices have top five shale plays might grow, if drilling were been driven to low levels in 2009 and 2010, at to continue at 2010 levels for the next 20 years. least in part as a result of the abundance of This illustrates the very significant production relatively low-cost shale gas. Meanwhile oil potential of the shale resource.15 The current prices, determined by global market forces, rapid growth in shale production can continue have remained high. This has led producers to for some time — but in the longer run produc- seek liquid rich gas plays, such as certain areas tion growth tapers off as high initial produc- of the Marcellus or the Eagle Ford play in Texas, tion rates are offset by high initial decline rates, where condensate ratios can be well in excess of and the quality of drilling prospects declines as 100 bbl/MMcf. These plays then enable more the plays mature. gas production, even at low gas prices, thus putting further downward pressure on gas The large inventory of undrilled shale acreage, prices. together with the relatively high initial produc- tivity of many shale wells, allows a rapid In addition to understanding the resource production response to any particular drilling volumes, it is important to understand the effort, provided that all wells can be completed contribution that the new shale resources could and tied in. However, this responsiveness will make to the overall production capacity within change over time as the plays mature, and the U.S. significant drilling effort is required just to maintain stable production against relatively high inherent production decline rates.34 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.17 Potential Production Rate that Could Be Delivered by the Major U.S. ShalePlays up to 2030 — Given 2010 Drilling Rates and Mean Resource Estimates 30 Marcellus 25 Haynesville Woodford Fayetteville 20 Barnett Bcf/day 15 10 5 0 2000 2005 2010 2015 2020 2025 2030 Source: MIT analysisUNCONVENTIONAL GAS SCIENCE reservoirs, there is a long history of productionAND TECHNOLOGY 16 from a wide variety of depositional, mineral- 1.0 ogical and geomechanical environments, suchEach unconventional gas resource type — 0.8 0.6 that analogues can be developed and statisticaltight gas, CBM and shale — presents it own 0.4 0.2 predictions about future performance can beproduction challenges, although they also share 0.0 developed. This is not yet the case in the shalesome common characteristics. In particular, all plays.three types have low intrinsic permeabilitywithin the rock matrix itself — and thus Gas shales refer to any very fine-grained rockrequire enhancement of the connectivity capable of storing significant amounts of gas.between the reservoir and the wellbore to Gas may be present as free gas stored in theenable gas flow at commercial rates. A second natural fracture and macroporosity, adsorbedcommon characteristic is that the resources onto the kerogen17 and internal surfaces of thetend to be distributed over large geographical pores or dissolved in the kerogen and bitumen.areas, saturating pore space often hundreds of The highly variable definition of gas shales hassquare kilometres in areal extent, rather than led to uncertainty in defining controllingwithin the tightly defined boundaries of factors that constitute an economic develop-conventional gas reservoirs. This means that ment. Values of the key parameters used inexploration risk is very low; the challenges lie identifying potential shale resources varyin achieving commercial production rates. widely between shale plays, making it difficult to apply analogues and expand shale gasShale resources represent a particular challenge, exploration and development outside estab-because of their complexity, variety and lack of lished basins.long-term performance data. In conventional Chapter 2: Supply 35
  • Production in shales is a multi-scale and In order to ensure the optimal development of multi-mechanism process. Fractures provide these important national assets, it is necessary the permeability for gas to flow, but contribute to build a comprehensive understanding little to the overall gas storage capacity. The of geochemistry, geological history, multiphase porosity of the matrix provides most of the flow characteristics, fracture properties and storage capacity, but the matrix has very low production behavior across a variety of shale permeability. Gas flow in the fractures occurs plays. It is also important to develop tools in a different flow regime than gas flow in the that can enable the scaling up of pore-level matrix. Because of these differing flow regimes, physics to reservoir-scale performance predic- the modeling of production performance in tion, and make efforts to improve core analysis fractured shale formations is far more complex techniques to allow accurate determination than for conventional reservoirs, and scaling of the recoverable resource. modeling results up to the field level is very challenging. This in turn makes it difficult to R E CO M M E N D AT I O N confidently predict production performance The U.S. Department of Energy (DOE) and devise optimal depletion strategies for should sponsor additional Research and shale resources. Development (R&D), in collaboration with Production behavior in shale wells is marked industry and academia, to address some by a rapid decline from initial production rates, of the fundamental challenges of shale as seen in Figure 2.15a. Early gas production gas science and technology, with the goal is dominated by free gas depleted from the of ensuring that this national resource is fractures and the macroporosity. This rapid exploited in the optimal manner. initial decline is followed by a long term, much slower decline. As the pressure is lowered, gas desorbs from the organic matter in the matrix Resource assessment and diffuses into the fracture system. During this stage, desorption and diffusion through It is in the national interest to have the best the matrix drive production. The long-term possible understanding of the size of the U.S. production behavior of a shale gas well is natural gas resource. For conventional reser- dependent on the time scale of flow from the voirs, statistically based resource assessment matrix relative to flow in the fracture network. methodologies have been developed and tested over many years. In contrast, the assessment In addition to the complexities of modelling methodology for the “continuous” unconven- performance, core analysis techniques devel- tional resources is less well developed. There oped for conventional gas, CBM and tight would be real benefit in improving the method- gas do not work well in shale reservoirs, ology for unconventional resource assessments. because they implicitly assume that the same production mechanisms are applicable. The R E CO M M E N D AT I O N determination of initial parameters such as The USGS should continue, and even permeability, porosity and initial gas-in-place accelerate, its efforts to develop improved can be misleading, contributing to uncertainty assessment methodologies for in resource size and production performance. unconventional resources.36 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Technology injected into the reservoir preferentially displaces methane molecules, allowing forThe development of unconventional resources enhanced gas production while storing CO2in general, and shale resources in particular, has permanently in the subsurface. While pilotbeen enabled by the application of existing projects have successfully demonstratedtechnology — horizontal drilling and hydraulic enhanced recovery from this technique, therefracturing — in a new setting. The objective is are significant challenges associated withto create very large surface areas in the forma- making this a commercial-scale process.tion that are in communication with thewellbore. Horizontal wells place 4,000 feet or -more of well directly into the formation, while niques are now commonly used to estimatemultistage fracturing along the horizontal the length and orientation of inducedsection then creates additional surface area in fractures in the reservoir during fracturingcommunication with the wellbore. operations; this technique is useful for improving fracturing effectiveness. At a moreImprovements in drilling and fracturing macroscopic level, there is a need to developperformance are currently rapid, coming from seismic techniques that allow the characteri-improved know-how rather than specific sation of large areas, to identify formationtechnology breakthroughs. The repetitive nature “sweet spots,” natural fracture orientationof the shale drilling and completion process and other properties that would be invaluableprovides an ideal environment for continuous in improving overall resource development.improvement of drilling and completion times,and fracturing performance. These improve- SHALE GAS ENVIRONMENTAL CONCERNSments can serve to enhance well economics andincrease the ultimate resource base. BackgroundThere are a number of areas of technology The rapid development of shale gas resourcesdevelopment that could enhance unconven- in the U.S. over the past few years has arousedtional gas recovery in the longer term: concern, and a perception in some quarters that this development is causing significant environ- mental problems. A good deal of attention has a high well density for full development. been focused on the high-volume hydraulic Technology that can reduce well costs and fracturing that is an essential component of increase wellbore contact with the reservoir shale gas development, with a major concern can make a significant impact on costs, being that the fracturing process risks injecting production rates and ultimate recovery. toxic fracture fluids into shallow groundwater Multi-lateral drilling, whereby a number of aquifers, which are in many cases the source of horizontal sections can be created from a potable water for public use. More broadly, single vertical wellbore, and coiled tubing there are concerns about water management drilling to decrease costs represent potential and in particular the proper disposal of poten- options for future unconventional gas tially toxic wastewater from the fracturing development. procedure. 2 enhanced recovery — simultaneous These concerns have led to restrictions on recovery of natural gas while sequestering drilling in some areas and proposed regulatory CO2 provides an interesting, although as yet action. Activity is currently restricted in poten- unproven, possibility for enhancing gas tially productive areas of the Marcellus shale in recovery while reducing environmental the Delaware River Basin, New York State and footprint. In enhanced CBM production, CO2 Pennsylvania State Forest land. The U.S. Chapter 2: Supply 37
  • Figure 2.18 Typical Shale Well Construction (Not to Scale) Feet Below Surface Conductor Casing 100 — Aquifer 1000 — Cement Surface Casing Drilling Mud 2000 — Salt Water Zone Intermediate Casing Cement Production Casing Production Tubing 7100 — Kickoff Point Production Zone Source: Based on Modern Shale Gas Development in the United States – a Primer Environmental Protection Agency (EPA) 3. Drilling and casing — as shown in is conducting an extensive review of hydraulic Figure 2.18, casing is cemented into the well fracturing, and legislation in the form of the at various stages in order to maintain the Fracturing Responsibility and Awareness of integrity of the wellbore, and to ensure that Chemicals (FRAC) Act was introduced in the fluids within the various strata are contained 2009–2010 Congress.18 within those strata. The drilling and casing process usually entails several stages: The Shale Drilling and Completion Process High Quality (i) Drill and set conductor casing — large Reservoirs Marcellus In order to appreciate the risks associated with diameter casing set at shallow depths. Haynesville shale development, and to understand appro- Low Quality Woodford priate risk mitigation techniques, it is helpful to Reservoirs (ii) Drill through shallow freshwater zones, Fayetteville understand the major steps involved in well set and cement surface casing — the Barnett construction: most critical phase with respect to the Tight Gas protection of groundwater resources. Sands Shale Gas 1. Well permitting — states require an operator to obtain a permit to drill a well. (iii) Drill and cement intermediate casing. 2. Well site construction — typically involves (iv) Drill and cement production casing. cleaning and grading an area of around four acres in the case of a single well site, or five to six acres in the case of a multi-well site.38 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 4. Perforate and fracture the well, usually in 3. Contamination as a result of inappropriate multiple stages. off-site wastewater disposal.5. Flowback fracture fluid. 4. Excessive water withdrawals for use in high-volume fracturing.6. Place well into production. 5. Excessive road traffic and impact on airPotential Risks quality.With over 20,000 shale wells drilled in the last Before examining these risks in more detail,10 years, the environmental record of shale gas it is instructive to look at data that attempt todevelopment has for the most part been a good summarize available information on recordedone. Nevertheless, it is important to recognize incidents relating to gas well drilling in the U.S.the inherent risks of the oil and gas business L48 onshore. It is beyond the scope of thisand the damage that can be caused by just one study to examine multiple state archives topoor operation; the industry must continuouslystrive to mitigate risk and address public With over 20,000 shale wells drilled in the last 10 years,concerns. Particular attention should be paid the environmental record of shale gas developmentto those areas of the country which are notaccustomed to oil and gas development, and has for the most part been a good one — but it iswhere all relevant infrastructure, both physical important to recognize the inherent risks and theand regulatory, may not yet be in place. In this damage that can be caused by just one poor operation.context, the Marcellus shale, which represents35% to 40% of the U.S. shale resource, is the review individual well incident reports. Instead,primary concern. to provide a high-level view we have extracted and combined the results from a number ofWithin the stages of well construction outlined reports that have reviewed drilling-relatedabove, the primary risks are as follows: incidents in the U.S. over the past few years. Table 2.3 indicates the results of this analysis,1. Contamination of groundwater aquifers while Appendix 2E provides a fuller description with drilling fluids or natural gas while of the data set. The data set does not purport drilling and setting casing through the to be comprehensive, but is intended to give a shallow zones. sense of the relative frequency of various types of incidents.2. On-site surface spills of drilling fluids, fracture fluids and wastewater from fracture flowbacks.Table 2.3 Widely Reported Incidents Involving Gas Well Drilling; 2005 – 2009 Type of Incident Number Reported Fraction of Total Groundwater contamination by natural gas or drilling fluid 20 47% On-site surface spills 14 33% Off-site disposal issues 4 9% Water withdrawal issues 2 4% Air quality 1 2% Blowouts 2 4% Chapter 2: Supply 39
  • Of the 43 widely reported incidents, almost half and casing protect the freshwater zones as the appear to be related to the contamination of fracture fluid is pumped from the surface down shallow water zones primarily with natural gas. into the shale formation. This protection is Another third of reported incidents pertain to tested at high pressures before the fracturing on-site surface spills. In the studies surveyed, fluids are pumped downhole. Once the fractur- no incidents are reported which conclusively ing process is underway, the large vertical demonstrate contamination of shallow water separation between the shale sections being zones with fracture fluids. fractured and the shallow zones prevents the growth of fractures from the shale formation The Fracturing Process into shallow groundwater zones. Table 2.4 describes the typical separations in the major The fracturing process entails the pumping of shale plays; in all but one case there are several fracture fluids, primarily water with sand thousand feet of rock — typically sandstones proppant and chemical additives, at sufficiently and shales, many of which have very low high pressure to overcome the compressive permeability — separating the fractures shale stresses within the shale formation for the formation and the groundwater zones. It duration of the fracturing procedure. Each should be noted here that only shallow zones stage is typically of the order of a few hours. contain potable water; as depths increase, the The process increases formation pressure above salinity of the groundwater increases to the the critical fracture pressure, creating narrow point that it has no practical utility. fractures in the shale formation. The sand proppant is then pumped into these fractures A recently published report summarizes the to maintain a permeable pathway for fluid flow results of a large number of fracturing opera- after the fracture fluid is withdrawn and the tions in the Barnett and the Marcellus shales operation is completed. (Fisher, 2010). Figure 2.19 illustrates these results for the Marcellus shale, showing that in The fracturing process itself poses minimal risk all cases the highest growth of the fractures to the shallow groundwater zones that may remains separated from the groundwater exist in the upper portion of the wellbore. As aquifers by thousands of feet of formation. described previously, multiple layers of cement Table 2.4 Separation Distance between Gas Shales and Shallow Freshwater Aquifers in Major Plays Basin Depth to Shale (ft) Depth to Aquifer (ft) Barnett 6,500–8,500 1,200 Fayetteville 1,000–7,000 500 Marcellus 4,000–8,500 850 Woodford 6,000–11,000 400 Haynesville 10,500–13,500 40040 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.19 Fracture Growth in the Marcellus; Marcellus Shale Mapped FractureTreatments (TVD) 0 Deepest Aquifier Depth fracTOP 1,000 perfTOP Perf Midpoint perfBTM 2,000 fracBTM 3,000 4,000 Depth (ft) 5,000 6,000 7,000 OH 8,000 PA WV 9,000 1 51 101 151 201 251 301 351 Frac Stages (Sorted on Perf Midpoint) Source: Pinnacle Halliburton Service, Kevin Fisher, Data Confirm Safety of Well Fracturing from July 2010The physical realities of the fracturing process, practice, governed by existing regulations,combined with the lack of reports from the should provide an adequate level of protec-many wells to date of fracture fluid contamina- tion from these problems.tion of groundwater, supports the assertion thatfracturing itself does not create environmental Nevertheless, regulations vary by state, as aconcerns. However, this simple statement does function of local conditions and historicalnot address the full range of environmental precedent — best practice involves settingconcerns listed earlier: cement all the way to surface, and conduct- ing pressure tests and cement-bond logs1. Leakage of natural gas or drilling fluids to ensure the integrity of the surface casing. into shallow zones: this appears to be the A detailed comparison of state-by-state most common cause of reported incidents, regulation would facilitate the widespread and it is generally associated with drilling adoption of best practice. and setting the surface casing. There are three potential risks during this phase of R E CO M M E N D AT I O N operation: (1) overweight drilling mud Conduct an inter-state regulatory causing some drilling fluid leakage into review and, within constraints of local groundwater zones; (2) unexpected considerations, adopt best practice for encounters with shallow gas zones with the possibility of gas migration into groundwa- drilling and high-volume hydraulic ter zones and (3) poor quality cementing of fracturing. the surface casing, allowing a potential fluid pathway into the groundwater zones during subsequent operations. The protection of groundwater aquifers is one of the primary objectives of state regulatory programs, and it should be emphasized that good oil field Chapter 2: Supply 41
  • 2. On-site surface spills: the drilling and operation — these additives will vary as a completion process involves the handling function of the well type and the prefer- of many thousands of barrels of fluids ences of the operators. While there has on-site, in particular drilling mud and been concern about the transparency of fracture fluids. Spills can occur as a result information as regards the make-up of of failure of equipment such as pumps and these additives, there has been considerable hoses; in addition, there is potential for progress on this issue. Although precise overflow of tanks and surface pits. Issues formulations remain proprietary, informa- will arise if the volume of spilled material is tion is now becoming available for all the such that local waterways could be con- chemical compounds contained within the taminated. These issues are not specifically fluids. associated with the fracturing process, and avoiding spills is a normal part of good In addition to greater transparency about the oil field management practice. The high compounds, there is also progress towards volumes of fluid associated with shale elimination of the toxic components from fracturing may increase spill potential. the additives. Again, state regulations stipulate the R E CO M M E N D AT I O N requirements for protecting surface waters Require the complete disclosure of all against leaks and spills, with regulation fracture fluid components. varying from state to state. Shale fracture fluid or “slickwater,” is R E CO M M E N D AT I O N largely composed of water, which generally constitutes over 99% of the liquid com- Continue efforts to eliminate toxic ponent. As described in Table 2.5, a number components of fracture fluids. of additives are mixed in with the water to increase the effectiveness of the fracturing Table 2.5 Typical Fracture Fluid Additives Purpose Chemical Common Use clean up damage from initial HCl swimming pool cleaner drilling, initiate cracks in rock gel agents to adjust viscosity guar gum thickener in cosmetics, toothpaste, sauces viscosity breakers ammonium persulfate, potassium, bleach agent in detergent and sodium peroxydisulfate hair cosmetics biocides gluteraldehyde, 2,2-dibromo3- medical disinfectant nitrilophopionamide surfactant isopropanol glass cleaner, antiperspirant corrosion inhibitor n, n-dimethylformamide pharmaceuticals clay stabilizer potassium chloride low sodium table salt substitute Source: Kaufman et al. 200842 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 3. Off-site wastewater disposal — another Much effort is now focused on addressing potential issue is the disposal of waste from this issue where disposal problems exist. fracturing operations, in particular the One approach is to recycle the flow-back fracture fluid and formation water that is fluid: using the flow-back fluid from one returned from the well when it is back- well as a component in the fracture fluid flowed upon completion of the fracturing of the next well. This has the additional operation, prior to start of production. advantage of reducing the total amount Typically, less than 100% of the injected of water that must be imported to site. fluid will be recovered, and it will generally In addition, techniques are also being be mixed with some volume of displaced developed to clean up wastewater prior formation brine. This fluid must be dis- to disposal. posed of appropriately. 4. Water withdrawal — large quantities of Every year the onshore U.S. industry safely water, typically of the order of 100,000 disposes of approximately 18 billion barrels barrels, are required for high-volume of produced water. By comparison, a hydraulic fracturing, and this has raised high-volume shale fracturing operation concerns about the impact on local water may return around 50 thousand barrels of resources. fracture fluid and formation water to the surface. The challenge is that these volumes While there may be temporary impacts on are concentrated in time and space. local resources, the overall impact is small, as can be seen when the volumes are placed The optimum method for disposal of oil in the context of total water usage. Table 2.6 field wastewater is injection into a deep looks at water usage for shale gas opera- saline aquifer through an EPA regulated tions as a fraction of total water usage in a Underground Injection Control (UIC) number of major shale plays — in all cases water disposal well. Problems can occasion- shale development water usage represents ally arise if there are insufficient wastewater less than 1% of total water usage in the disposal wells, as appears to be the case in affected areas. Pennsylvania. Waste can be disposed of at wastewater treatment plants, but problems can arise if the fluid for disposal is of high salinity or contains other contaminants19; this may cause the effluent from the treatment plant to exceed desired limits. Chapter 2: Supply 43
  • Table 2.6 Comparative Water Usage in Major Shale Plays Total Public Industrial/ Water Use Play Supply Mining Irrigation Livestock Shale Gas (Bbbls/yr) Barnett TX 82.7% 3.7% 6.3% 2.3% 0.4% 11.1 Fayetteville AR 2.3% 33.3% 62.9% 0.3% 0.1% 31.9 Haynesville LA/TX 45.9% 13.5% 8.5% 4.0% 0.8% 2.1 Marcellus NY/PA/WV 12.0% 71.7% 0.1% <0.1% <0.1% 85.0 Source: ALL Consulting Indeed, the “water intensity” of shale gas 5. Road traffic and environmental dis- development, at around 1 gallon of water turbance — oil and gas operations have consumed for every MMBtu of energy pro- the potential to be disruptive to local duced, is low compared to many other energy communities in the field development sources. By way of contrast, several thousand phase of well drilling and completion, gallons of water per MMBtu of energy pro- particularly in those areas not accustomed duced can be used in the irrigation of corn to routine oil field operations. As indicated grown for ethanol. in Table 2.7, the large volumes of water involved in fracturing operations can create Nevertheless, careful planning and coordina- high volumes of road traffic. tion is necessary to ensure that episodic water withdrawals do not disrupt local supply It should be emphasized that the large sources. number of traffic movements shown on this table are really worst-case numbers. In R E CO M M E N D AT I O N particular, re-use of flowback wastewater Prepare integrated regional water usage can and does significantly reduce the road traffic associated with hauling water, which and disposal plans for the major shale areas. represents much of the traffic movement. Furthermore, large-scale operators are also using pipelines to transport water to site, R E CO M M E N D AT I O N further reducing the amount of road traffic Undertake collaborative R&D to reduce very substantially. water usage and develop cost-effective water recycling.44 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 2.7 Truck Journeys for a Typical Shale Well Drilling and Completion Activity 1 Rig, 1 Well 2 Rigs, 8 Wells Pad and Road Construction 10 – 45 10 – 45 Drilling Rig 30 60 Drilling Fluid and Materials 25 – 50 200 – 400 Drilling Equipment (casing, drill pipe, etc.) 25 –50 200 – 400 Completion Rig 15 30 Completion Fluid and Materials 10 – 20 80 – 160 Completion Equipment (pipe, wellhead, etc.) 5 10 Fracturing Equipment (pump trucks, tanks, etc.) 150 – 200 300 – 400 Fracture Water 400 – 600 3,200 – 4,800 Fracture Sand 20 – 25 160 – 200 Flowback Water Disposal 200 – 300 1,600 – 2,400 Total 890 – 1,340 5,850 – 8,905Source: NTC ConsultingIn conclusion, it is clear that oil and gas Methane hydrates are an ice-like form ofdevelopment is not without risk to the natural methane and water stable at the pressure-environment. State and Federal regulations are temperature conditions common in the shallowdesigned to mitigate those risks. However, sediments of permafrost areas and continentalthough not the result of risks inherent to the margins. Globally, the total amount of methanefracturing of shale gas wells, operational errors sequestered in these deposits probably exceedsand poor drilling practice do result in a signifi- 100,000 Tcf, of which ~99% occurs in oceancant number of incidents. Implementation of sediments. Most of this methane is trapped inthe recommendations described above, highly disseminated and/or low saturationtogether with rigorous enforcement of all methane hydrates that are unlikely to ever beapplicable regulations, should reduce the a commercially viable gas source. An estimatednumber of incidents and ensure that shale 10,000 Tcf may be technically recoverable fromdevelopment can proceed with minimum high-saturation gas hydrate deposits (Boswellimpact on the environment. and Collett, 2010), primarily concentrated in permeable (likely sand-rich) sediments.METHANE HYDRATES20Methane hydrates are not considered in theresource estimates and supply curves describedabove, as they are still at a very early stage interms of resource definition and understand-ing. Nevertheless, gas hydrates could representa significant long-term resource option, possi-bly in North America but particularly in someother parts of the world. Chapter 2: Supply 45
  • Figure 2.20 USGS Database of Locations at which Gas Hydrate Has Been Recovered (circles) or StronglyInferred Based on Drilling-Based Evidence (squares) from Permafrost Areas (black labels) or from DepthsGreater than 50 m below the Seafloor (white labels). The color-coding refers to the primary (outer symbol)and, where relevant, the secondary (inner symbol) type of gas hydrate reservoir, using terminology fromthe gas hydrate resource pyramid (Figure 2.21 in MITEI report). Academic drill sites where deep gas hydratewas recovered but for which reservoir type has not been determined are designated by ODP/DSDP. Source: Ruppel, C., Collett, T. S., Boswell, R., Lorenson, T., Buczkowski, B., and Waite, W., 2011, A new global gas hydrate drilling map based on reservoir type, Fire in the Ice, DOE-NETL Newsletter, May edition, vol. 11(1), 13–17. To date, there have been few formal quantita- 85.4 Tcf (median) for permafrost-associated tive assessments of methane sequestered in gas gas hydrates on the Alaskan North Slope hydrates at regional scales. A recent assessment (Collett et al., 2008). Outside the U.S., the only of in-place resources in northern Gulf of formal assessment covers ~10% of the area Mexico yielded 6,717 Tcf (median) for sands associated with a certain gas hydrates seismic (Frye, 2008), and other assessments based on marker in the Nankai Trough and yielded 20 similar methodology are expected soon for the Tcf methane in-place in the high saturation U.S. Atlantic Margin and other U.S. margins. section (Fujii et al., 2008). The only assessment of technically recoverable methane hydrates ever completed calculated46 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 2.21 The Methane Hydrate Resource Pyramid, After Boswell and Collett (2006) e.g., 85 Tcf technically recoverable on Alaskan North Slope (Collett et al., 2008) Arctic (permafrost-associated) sand reservoirs e.g., 6,717 Tcf in-place Northern Marine sand reservoirs Gulf of Mexico sands (Frye, 2008) Non-sand marine reservoirs (including fracture filling) Massive seafloor/shallow hydrates at seeps Increasing in-place resources Decreasing reservoir quality Decreasing resource estimate accuracy Marine shales Increasing production challenges (low permeability) Likely decreasing recovery factor Source: After Boswell and Collett (2006)Several research challenges remain before gas associated wells in the U.S. and Canadianhydrate assessments become routine. The Arctic. Before 2015, the first research-scale,greatest need is geophysical methods that can long-term (several months or longer) produc-detect gas hydrates and constrain their in situ tion tests could beMarcellus out by the U.S. DOE carried Haynesvillesaturations more reliably than seismic surveys on the Alaskan North Woodfordalone and less expensively than direct drilling Slope and by the Methane hydrates are unlikely Fayettevilleand borehole logging. Electromagnetic (EM) Japanese MH21 project to reach commercial viability Barnettmethods have shown some promise in deep for Nankai Trough for global markets for at leastmarine settings, but refinements in seismic deepwater gas hydrates.techniques (e.g., full waveform inversions, The goals of these tests 15 to 20 years.seismic attribute analysis) may yet prove even are to investigate themore useful than routinely combining EM and optimal mix of production techniques toseismic surveys. sustain high rates of gas flow over the lifetime of a well and to assess the environmentalMethane hydrates are unlikely to reach com- impact of production of methane from gasmercial viability for global markets for at least hydrates.15 to 20 years. Through consortia of govern-ment, industry and academic experts, the U.S., Producing gas from methane hydrates requiresJapan, Canada, Korea, India, China and other perturbing the thermodynamic stabilitycountries have made significant progress on conditions to drive dissociation (breakdown)locating and sampling methane hydrates. of the deposits into their constituent gas andNo short-term production test has ever been water. The gas can then be extracted usingattempted in a marine gas hydrate setting, but well-established production methods. Depres-several short-term tests (few hours to a few surization of the formation is the preferreddays) have been completed in permafrost- technique for driving gas hydrate dissociation Chapter 2: Supply 47
  • since it yields a relatively sustainable and are shallower than those associated with well-controlled flow of gas. Thermal stimula- (deepwater) conventional gas, rendering gas tion through direct heating or injection of hydrate well control less of a challenge. Gas heated fluids can be used to drive episodic hydrate dissociation is also a self-regulating dissociation during longer-term depressuriza- process in most cases, so there is little danger tion, but requires significant energy expendi- of runaway dissociation. Changes in bulk ture. Injection of inhibitors (e.g., seawater sediment volume and sediment strength are or some chemicals) can also dissociate gas expected if high-saturation gas hydrates are hydrates in the formation, although this dissociated, but the impact of these changes technique has numerous disadvantages and is will depend on many factors, including the unlikely to be practical at large scales. A final geologic setting, the depth of the deposits and production method will be tested on the the fate of produced water. In short, the risks Alaskan North Slope in 2012 by ConocoPhillips associated with gas production from methane and could in theory produce methane as well as hydrates located beneath permafrost or deep sequester CO2: CO2 injected into methane within marine sediments are either largely hydrate deposits should liberate methane while known from existing gas operations or consid- simultaneously trapping the CO2 within stable ered manageable. gas hydrates (Yezdimer et al., 2002; Farrell et al., 2010). R E CO M M E N D AT I O N Continue methane hydrates research At present, most conventional oil and gas program to develop methods for remote producers avoid intersecting gas hydrate detection of highly concentrated deposits; deposits to prevent long-term damage to the borehole due to unintended dissociation. conduct formal resource assessments; Producing gas from methane hydrates will and prove the resource potential through instead require targeted drilling into high- long-term production testing. saturation deposits and careful management of potentially large amounts of co-produced water. The depths at which gas hydrate occurs48 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • APPENDICES SUPPLEMENTARY PAPERS ON MITEI WEBSITE:2A: Additional resource data tables and maps2B: Methodology for creating resource ranges SP 2.1 Natural Gas Resource Assessment2C: Additional supply curves and background Methodologies – Dr. Qudsia Ejaz information SP 2.2 Background Material on Natural2D: Shale gas economics Gas Resource Assessments with2E: Analysis of reported gas drilling incidents Major Resource Country Reviews – Dr. Qudsia Ejaz SP 2.3 Role of Technology in Unconventional Gas Resources – Dr. Carolyn Seto SP 2.4 Methane Hydrates – Dr. Carolyn Ruppel Chapter 2: Supply 49
  • REFERENCES Ahlbrandt, Thomas S., Ronald R. Charpentier, Farrell, H., Boswell, R., Howard, J., and Baker, T. R. Klett, James W. Schmoker, Christopher J. R., 2010, CO2-CH4 exchange in natural gas Schenk and Gregory F. Ulmishek. Global hydrate reservoirs: potential and challenges, in Resource Estimates from Total Petroleum DOE-NETL Fire in the Ice, Methane Hydrate Systems. AAPG, 2005. Newsletter, March 2010, p. 19–21. http://www. ALL Consulting: Water and Shale Gas Develop- ment; Presentation to National Association of Hydrates/Newsletter/MHNews_2010_03. Royalty Owners Annual Conference; October pdf#page=19 2010. Fisher, Kevin; Data Confirm Safety of Well Attanasi, E. D., and T. C. Coburn. “A bootstrap Fracturing; The American Oil and Gas approach to computing uncertainty in inferred Reporter; July 2010. oil and gas reserve estimates.” Natural Resources Frye, M. Preliminary evaluation of in-place Research 13, no. 1 (2004): p. 45–52. Gas Hydrate Resources: Gulf of Mexico Outer Boswell, R., and T. Collett, 2006. The gas Continental Shelf. Minerals and Management hydrates resource pyramid: Fire in the Ice, Services, 2008. Methane Hydrate Newsletter, U.S. Department Fujii, T., T. Saeiki, T. Kobayashi, T. Inamori, M. of Energy, Office of Fossil Energy, National Hayashi, O. Takano, T. Takayama, T. Kawasaki, Energy Technology Laboratory, Fall Issue, S. Nagakubo, M. Nakamizu and K. Yokoi, 2008. p. 5–7. Resource assessment of methane hydrate in the oil-gas/publications/Hydrates/Newsletter/ Nankai Trough, Japan, Offshore Technology HMNewsFall06.pdf#page=5 Conference, Houston, TX, Paper 19310. Boswell, R., and T. Collett. “Current Perspectives Ground Water Protection Council; ALL on Gas Hydrate “Resources”. Energy Environ. Consulting; Modern Shale Gas Development Sci., 2011. doi: 10.1039/C0EE00203H. in the United States: A Primer; U.S. Depart- Cadmus Group; Preliminary Analysis of ment of Energy; National Energy Technology Recently Reported Contamination; Prepared Laboratory. for: Drinking Water Protection Division Kaufman, P., Penny, G. S., Paktinat, J.: Critical (DWPD) Office of Ground Water and Drinking evaluations of additives used in shale slickwater Water (OGWDW) U.S. Environmental Protec- fracs. Paper SPE 119900 presented at the tion Agency (EPA) Hydraulic Fracturing; 2008 SPE Shale Gas Production Conference, September 2009. Ft. Worth TX, 16–18 November 2008. Collett, T., W. Agena, M. Lee, M. Zyrianova, Minerals Management Service. Assessment of T. Charpentier, D. Houseknecht, T. Klett, Undiscovered Technically Recoverable Oil and R. Pollastro and C. Schenk. Assessment of Gas Resources of the Nation’s Outer Continental Gas Hydrate Resources on the North Slope. U.S. Shelf, 2006 (Summary Brochure). Minerals Geological Survey Factsheet. United States Management Service, February 2006. Geological Survey, 2008. National Petroleum Council. Balancing Natural Energy Information Administration. U.S. Crude Gas Policy – Fueling the Demands of a Growing Oil, Natural Gas and Natural Gas Liquids Economy. National Petroleum Council, Reserves Report. Energy Information Adminis- September 2003. tration, February 2009. oil_gas/natural_gas/data_publications/crude_ oil_natural_gas_reserves/cr.html.50 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • NTC Consultants: Impact on Community United States Geological Survey. “National OilCharacter of Horizontal Drilling and High and Gas Assessment, USGS-ERP,” 2010. http://Volume Hydraulic Fracturing in Marcellus and Other Low-Permeability Gas ———. “World Petroleum Assessment-Reservoirs; September 2009. Information, Data and Products, USGS-ERP,”New York State Department of Environmental n.d. Division of Mineral Resources; we/index.jsp.Supplemental Generic Environmental Impact Vidas, E.H., Hugman, R.H. & Haverkamp, D.S.,Statement on the Oil, Gas and Solution Mining 1993. Guide to the Hydrocarbon Supply Model:Regulatory Program; Well Permit Issuance 1993 Update. Gas Research Institute, Reportfor Horizontal Drilling and High-Volume GRI-93/0454.Hydraulic Fracturing to Develop the Marcellus Watershed; Craig Michaels, Program Director;Shale and Other Low-Permeability Gas Reser- James L. Simpson, Senior Attorney; Williamvoirs; September 2009. Wegner, Staff Scientist; Fractured Communities –Potential Gas Committee. Potential Supply of Case Studies of the Environmental Impacts ofNatural Gas in the United States – Report of the Industrial Gas Drilling; September 2010.Potential Gas Committee (December 31, 2006). Yezdimer, E., Cummings, P., Chalvo, A., 2002.Potential Supply of Natural Gas in the United “Extraction of methane from its gas clathrateStates. Potential Gas Agency, Colorado School by carbon dioxide sequestration – determinationof Mines; November 2007. of the Gibbs Free Energy of gas replacement and———. Potential Supply of Natural Gas in molecular simulation.” J. Phys. Chem A., 106the United States – Report of the Potential Gas p. 7982–7987.Committee (December 31, 2008). PotentialSupply of Natural Gas in the United States.Potential Gas Agency, Colorado School ofMines; December 2009.———. Potential Supply of Natural Gas inthe United States – Report of the Potential GasCommittee (December 31, 2008) – AdvanceSummary. Potential Supply of Natural Gasin the United States. Potential Gas Agency,Colorado School of Mines; June 2009.Reservoir Research Partners and TudorPickering & Holt; Frac Attack: Risk, Hypeand Financial Reality of Hydraulic Fracturingin the Shale Plays; July 2010.Rogner, H. H. “An Assessment of WorldHydrocarbon Resources.” Annual Review ofEnergy and the Environment, 22, no. 1 (1997):p. 217–262. Chapter 2: Supply 51
  • NOTES At the time of writing, new more detailed estimates 11 of global unconventional resources are starting to 1 Thermogenic gas is formed by the application of be published. See, for example, World Shale Gas heat and pressure on organic matter; natural gas Resources: An Initial Assessment of 14 Regions can also be formed through a biogenic process, Outside the United States. Produced by Advanced in which microbial action in an anaerobic (oxygen Resources International (ARI) for the U.S. EIA free) environment creates methane from organic April 2011. matter — for example, in swamps, land-fills and Appendix 2A provides additional maps and 12 shallow formations. This chapter of the report detailed data tables concerning gas resource is focused on thermogenic gas. estimates. 2 Permeability is a measure of the ability of a porous Appendix 2D contains a detailed discussion of the 13 medium, such as that found in a hydrocarbon economic performance of the major U.S. shale reservoir, to transmit fluids, such as gas, oil or plays. water, in response to a pressure differential across the medium. In petroleum engineering, These are illustrative calculations only, not based 14 permeability is usually measured in units of on actual “wet” well performance. The calculations millidarcies (mD). Unconventional formations, assume that well performance, costs, etc., are by definition, have permeability less than 0.1mD. unchanged by increasing levels of liquids produc- tion. In practice, gas production may be affected 3 ICF International is a consulting firm whose by liquid co-production. services were used in preparation of supply curves for this study. This is not a forecast of production — but rather 15 an illustration of the production potential at an 4 In the US, natural gas volumes are typically assumed drilling rate and assuming a median measured in Standard Cubic Feet (Scf), where the estimate of resources. volume is measured at a temperature of 60°F and a pressure of one atmosphere (14.7 pounds per A detailed discussion of the science and technology 16 square inch). 1 trillion cubic feet (Tcf) = of unconventional gas resources can be found in 1,000,000,000,000 (or 1012) Scf. Outside North the Supplementary Paper SP 2.3 “Role of America, natural gas volumes are typically Technology in Unconventional Gas Resources,” by measured in cubic meters. 1 cubic meter 35.3 Dr. Carolyn Seto, published on the MITEI website. cubic feet. Kerogen and bitumen are comprised of organic 17 5 Appendix 2A provides additional maps and matter that occurs in hydrocarbon source rocks, detailed data tables concerning gas resource formed from the application of heat and pressure estimates. Supplementary Paper SP 2.2 to buried organic material over geological time. “Background Material on Natural Gas Resource Kerogen is insoluble in normal organic solvents, Assessments with Major Resource Country while bitumen is soluble. Reviews,” by Dr. Qudsia Ejaz, published on the The Fracture Responsibility and Awareness of 18 MITEI website, provides additional material. Chemicals (FRAC) Act of 2009 proposed to 6 Appendix 2B provides details on the methodology regulate fracturing under the Underground used to create the uncertainty estimates shown in Injection Control provisions of the Safe Water this chapter. Drinking Act, and to mandate full disclosure of the chemical constituents of all fracture fluid additives. 7 Appendix 2C provides further details of cost curves The Bill did not make it out of Committee during prepared for this study. the 2009–2010 session of Congress. 8 Supply curves shown here are based on oil field Flowback fluid can contain: dissolved solids 19 costs in 2007. There has been considerable oil field (chlorides, sulfates, and calcium); metals (calcium, cost inflation, and some recent deflation, in the last magnesium, barium, strontium) suspended solids; 10 years. We have estimated cost curves on a 2004 mineral scales (calcium carbonate and barium base (the end of a long period of stable costs) and sulfate); acid producing bacteria and sulfate a 2007 base (reasonably comparable to today’s reducing bacteria; friction reducers; iron solids costs, 70% higher than the 2004 level, and (iron oxide and iron sulfide); dispersed clay fines, continuing to decline). colloids and silts; acid gases (carbon dioxide, 9 Appendix 2A contains further details on global hydrogen sulfide); radionuclides (New York unconventional resources. Generic Environmental Impact Statement). 10 Rogner, “An Assessment of World Hydrocarbon A detailed discussion of methane hydrates can be 20 Resources”, 1997. found in the Supplementary Paper SP 2.4 “Methane Hydrates,” by Dr. Carolyn Ruppel, published on the MITEI website.52 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Chapter 3: U.S. Gas Production, Use andTrade: Potential FuturesINTRODUCTION BOX 3.1 GLOBAL AND U.S. ECONOMICAs discussed in other sections of this report, MODELSmany factors will influence the future role of Projections in this section were made using thenatural gas in the U.S. energy system. Here we MIT Emissions Prediction and Policy Analysisconsider the most important of these: Greenhouse (EPPA) model and the U.S. Regional EnergyGas (GHG) mitigation policy; technology Policy (USREP) model.1 Both are multi-region,development; size of gas resources; and global multi-sector representations of the economymarket developments. And we examine how that solve for the prices and quantities ofthey will interact to shape future U.S. gas use, energy and non-energy goods and projectproduction and trade over the next few decades. trade among regions. The core results for this study are simulatedWe investigate the importance of these factors using the EPPA model — a global model withand their uncertainties by applying established the U.S. as one of its regions. The USREP modelmodels of the U.S. and global economy (see is nearly identical in structure to EPPA, butBox 3.1). Alternative assumptions about the represents the U.S. only — segmenting it intofuture allow us to create a set of scenarios that 12 single and multi-state regions. In the USREPprovides bounds on the future prospects for gas model, foreign trade is represented through import supply and export demand functions,and illustrate the relative importance of different broadly benchmarked to the trade response infactors in driving the results. the EPPA model. Both models account for all Kyoto gases.The conditions explored include the High,Mean and Low ranges of gas resource estimates The advantage of models of this type is theirdescribed in Chapter 2. We show the impacts of ability to explore the interaction of those factors underlying energy supply and demandvarious policy alternatives, including: no new that influence markets. The models canclimate policy; a GHG emission reduction target illustrate the directions and relative magni-of 50% by 2050, using a price-based policy (such tudes of influences on the role of gas, provid-as a cap-and-trade system or emissions tax) and ing a basis for judgments about likely futurean emissions policy that uses a set of non-price developments and the effects of governmentregulatory measures. policy. However, results should be viewed in light of model limitations. Projections, espe-Several assumptions have a particularly impor- cially over the longer term, are naturallytant effect on the analysis. Long-term natural subject to uncertainty. Also, the cost ofgas supply curves, distinguishing the four gas technology alternatives, details of markettypes for the U.S. and Canada, are drawn from organization and the behavior of individualChapter 2. U.S. economic growth is assumed industries (e.g., various forms of gas contracts,to be 0.9% per year in 2005 to 2010, 3.1% in political constraints on trade and technology choice) are beneath the level of model aggre-2010 to 2020 (to account for recovery) and 2.4% gation. The five-year time step of the modelsfor 2020 to 2050. means that the effects of short-term price volatility are not represented. Chapter 3: System Studies 53
  • Table 3.1 Levelized Cost of Electricity (2005 cents/kWh) Reference Sensitivity Coal 5.4 Advanced Natural Gas (NGCC) 5.6 Advanced Nuclear 2 8.8 7.3 Coal/Gas with CCS 3 9.2/8.5 6.9/6.6 Renewables Wind 6.0 Biomass 8.5 Solar 19.3 Substitution elasticity 1.0 3.0 (Wind, Biomass, Solar) Wind+Gas Backup 10.0 Source: EPPA, MIT Influential cost assumptions are shown in increasing cost of production as their share Table 3.1. The first column contains technology increases, to be limited by the cost with backup. costs imposed in the main body of the analysis, These energy sector technologies, like others in as documented in Appendix 3A. The right- the model, are subject to cost reductions over most column shows values to be employed in time through improvements in labor, energy sensitivity tests to be explored later, where we and (where applicable) land productivity. vary the costs of competing generation tech- nologies (nuclear, coal and gas with carbon The potential role of compressed natural gas capture and storage and renewables). The in vehicles is considered separately, drawing on intermittent renewables (wind and solar) are estimates of the cost of these vehicles from distinguished by scale. At low penetration Chapter 5 of this report. levels, they enter as imperfect substitutes for conventional electricity generation, and the We also consider two possible futures for estimates of the levelized cost of electricity international gas markets: one where they (LCOE4) apply to early installations when continue in their current pattern of regional renewables are at sites with access to the best trading blocs and an alternative where there quality resources and to the grid, and storage or develops a tightly integrated global gas market backup is not required. Through the elasticity similar to that which now exists for crude oil. of substitution, the model imposes a gradually54 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • THE ROLE OF U.S. CLIMATE POLICY — THREE ALTERNATIVE SCENARIOSTo explore the future of U.S. gas use in a that a price-based policy is imposed on all U.S.carbon-constrained world, we analyze three GHG emissions with a target of a 50% reduc-scenarios of greenhouse control, with very tion by 2050, as can be seen in Figure 3.1.different implications for the energy sector as a Scenario 3 imposes no economy-wide target,whole. Scenario 1 establishes a baseline, with no but considers two measures proposed for theGHG policy measures beyond those in place electric power sector: a renewable energytoday. Emissions grow by some 50% over the standard and measures to force retirement ofperiod, as shown in Figure 3.1. Scenarios 2 and coal-fired power plants. As seen in Figure 3.1,3 are constructed to span a wide range of this scenario of a regulatory approach essen-possible approaches to climate policy, and tially stabilizes U.S. GHG emissions, yieldingpotential effects on gas use. Scenario 2 assumes only about 10% increase by 2050.Figure 3.1 U.S. Greenhouse Gas Emissions under Alternative Scenarios 10,000 8,000 Mt CO2e 6,000 4,000 2,000 0 2010 2020 2030 2040 2050 1.0 0.8 0.6 0.4 0.2 0.0 Year No Policy Regulatory Price-Based Source: EPPA, MIT Haynesville Woodford Fayetteville Barnett 1.0 Policy (net of carbo 0.8 0.6 0.4 Policy (incl. carbon 0.2 0.0 10000 Chapter 3: System Studies 55 8000
  • Scenario 1 — No Additional GHG The availability of shale gas resources has a Mitigation Policy substantial effect on these results. If the Mean estimate for other gas resources is assumed, and Unless gas resources are at the Low end of the this same projection is made omitting the shale resource estimates in Chapter 2, domestic gas gas component of supply, U.S. production use and production are projected to grow peaks around 2030 and declines to its 2005 level substantially between now and 2050. This result by 2050. is shown in Figure 3.2, from EPPA model simulations, on the assumption that global gas Given the continued existence of regional markets remain fragmented in regional trading trading blocs for gas, there is little change in blocs. Under the Mean resource estimate, U.S. the role played by imports and exports of gas. gas production rises by around 40% between Imports (mainly from Canada) are roughly 2005 and 2050, and by a slightly higher 45% constant over time, though they increase when under the High estimate. It is only under the U.S. resources are Low. Exports (principally to Low resource outcome that resource availability Mexico) are also maintained over the period substantially limits growth in domestic produc- and grow somewhat if U.S. gas resources are tion and use. In that case, gas production and at the High estimate. use plateau around 2030 and are in decline by 2050. Figure 3.2 U.S. Gas Use, Production and Imports & Exports (Tcf), and U.S. Gas Prices above Bars ($/1000 cf) for Low (L), Mean (M) and High (H) U.S. Resources. No Climate Policy and Regional International Gas Markets 45 40 9.5 10.4 35 9.2 8.7 30 7.9 8.0 10.9 8.6 15.5 7.0 6.9 6.8 25 Tcf 20 15 10 5 0 L M H L M H L M H L M H 2020 2030 2040 2050 Year Imports Production–Exports Exports Source: EPPA, MIT Marcellus Barnett 1.0 0.8 0.6 0.4 0.256 0.0 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Gas prices (2005 U.S. dollars), shown at the top While measures taken abroad are not of directof the bars in Figure 3.2, rise gradually over interest for this study, such policies or the lacktime as the lower-cost resources are depleted; of them will affect the U.S. energy systemthe lower the resource estimate, the higher the through international trade. If the U.S. were toprices. The difference in prices across the range pursue this aggressive GHG mitigation policy,of resource scenarios is not great for most we assume that it would need to see similarperiods. In 2030, for example, the High measures being taken abroad. Thus, a similarresource estimate yields a price 2% below that pattern of reductions is assumed for otherfor the Mean estimate, while the Low resource developed countries, with lagged reductionscondition increased the price by 7%. The in China, India, Russia, Mexico and Brazil thatdifference increases somewhat over time, start in 2020 on a linear path to 50% belowespecially for the Low resource case. By 2050, their 2020 levels by 2070. The rest of thefor example, the price is 8% lower if the High developing countries are assumed to delayresource conditions hold, but 50% higher if action to beyond 2050. We assume no emis-domestic resources are at the Low estimate. sions trading among countries.Underlying these estimates are developments The broad features of Even under the pressure of anon the demand side. Under Mean resources, U.S. gas markets under assumed CO2 emissions policy,electricity generation from natural gas would the assumed emissionsrise by about 70% over the period 2010 to 2050 restriction are not total U.S. gas use is projectedthough coal would continue to dominate, with substantially different to increase up to 2050.only a slightly growing contribution projected from the no-policyfrom nuclear power and renewable sources scenario, at least through 2040 (Figure 3.3).(wind and solar). National GHG emissions rise Gas production and use grows somewhat moreby about 40% from 2005 to 2050. More detailed slowly, reducing use and production by a fewresults for the scenarios with Mean resources Trillion cubic feet (Tcf) in 2040 compared withare provided in Appendix 3B. the case without climate policy. After 2040, however, domestic production and use beginScenario 2 — Price-Based Climate Policy to fall. This decline is driven by higher gas prices, Carbon Dioxide (CO2) charge inclusive,An incentive (or price) based GHG emissions that gas users would see. The price reachespolicy that establishes a national price on GHG about $22 per thousand cubic feet (cf) withemissions serves to level the emissions reduc- well over half of that price reflecting the CO2tion playing field by applying the same penalty charge. While gas is less CO2 intensive than coalto emissions from all sources and all uses. or oil, at the reduction level required by 2050, its CO2 emissions are beginning to representThe policy explored here gradually reduces an emissions U.S. GHG emissions, measured in CO2equivalents (CO2-e)5, to 50% below the 2005 However, even under the pressure of thelevel by 2050. The scenario is not designed to assumed emissions policy, total gas use isrepresent a particular policy proposal and no projected to increase from 2005 to 2050 evenprovision is included for offsets. for the Low estimate of domestic gas resources. Chapter 3: System Studies 57
  • Figure 3.3 U.S. Gas Use, Production and Imports & Exports (Tcf), and U.S. Gas Prices ($/1000 cf) for Low (L), Mean (M) and High (H) U.S. Resources, Price-Based Climate Policy and Regional International Gas Markets. Prices are shown without (top) and with (bottom) the emissions charge. 45 40 35 8.6 8.2 16.9 30 10.0 17.3 7.5 7.4 8.8 8.3 7.9 18.5 13.3 13.2 21.9 21.8 6.5 6.4 13.7 25 6.6 11.6 9.6 9.5 9.4 Tcf 23.6 20 15 10 5 0 L M H L M H L M H L M H 2020 2030 2040 2050 Year Imports Production–Exports Exports Source: EPPA, MIT A major effect of the economy-wide, price- substitution effect, mainly gas generation for based GHG policy is to reduce energy use coal generation, outweighs the demand reduc- (Figure 3.4). The effect in the electric sector is tion effect. For total energy (Figure 3.4b) the to effectively flatten demand, holding it near its demand reduction effect is even stronger, current 4 Trillion killowatt hour (TkWh) level leading to a decline in U.S. energy use of nearly (Figure 3.4a). Based on the cost assumptions 20 quadrillion (1015) British thermal units (Btu). underlying the simulation (see Appendix 3A) The reduction in coal use is evident, and oil and nuclear, Carbon Capture and Storage (CCS) and current-generation biofuels (included in oil) renewables are relatively expensive compared begin to be replaced by advanced biofuels. with generation from gas. Conventional coal is Because national energy use is substantially driven from the generation mix by the CO2 reduced, the share represented by gas is pro- prices needed to meet the economy-wide jected to rise from about 20% of the current emissions reduction targets. Natural gas is the national total to around 40% in 2040. substantial winner in the electric sector: the58 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 3.4 Energy Mix under a Price-Based Climate Policy, Mean Natural Gas Resources 3.4a Electric Sector (TkWh) 7 Reduced Use 6 Gas CCS 5 Coal CCS Renew 4 TkWh Hydro 3 Nuclear 2 Gas 1 Oil Coal 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 Year 3.4b Total Energy Use (qBtu) 140 120 Reduced Use Renew 100 Hydro 80 Nuclear qBtu 60 Gas Biofuel 40 Oil 20 Coal 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 Year Source: EPPA, MITUnder this policy scenario, the U.S. emissions increased (Figure 3.5a). The CO2 charge isprice is projected to rise to $106 per ton CO2-e nearly half of the user price of gas.6in 2030 and to $240 by 2050. The macroeco-nomic effect is to lower U.S. Gross Domestic Even in the No-Policy case, electricity pricesProduct (GDP) by 1.7% in 2030 and 3.5% in are projected to rise by 30% in 2030 and about2050. (Other measures of cost are provided in 45% over the period to 2050 (Figure 3.5b).Appendix 3A.) A selection of resulting U.S. The assumed emissions mitigation policy isdomestic prices is shown in Figure 3.5. Natural projected to cause electricity prices to rise bygas prices, exclusive of the CO2 price, are almost 100% in 2030 and by two and one-halfreduced slightly by the mitigation policy, but times by 2050 compared with current prices.the price inclusive of the CO2 charge is greatly Chapter 3: System Studies 59
  • Figure 3.5 U.S. Natural Gas and Electricity Prices under Alternative Policy Scenarios, Mean Gas Resources3.5a Natural Gas Prices ($/1000 cf) 3.5b Electricity Prices ($/kWh)$/1000 cf $/kWh 25 0.30 0.25 20 0.20 15 0.15 10 0.10 5 0.05 0 0.00 2000 2010 2020 2030 2040 2050 2060 2000 2010 2020 2030 2040 2050 2060 Year Year No Policy Price-Based Policy (net of carbon) No Policy Priced-Based Policy Price-Based Policy (incl. carbon) Source: EPPA, MIT Haynesville Woodford Fayetteville Barnett Haynes As noted earlier, a set of alternative cost Tcf respectively, as availability of cheaper renew- 1.0 1.0 0.8 assumptions was explored for low-carbon ables displaces nuclear power which by that time 0.8 0.6 0.6 Policy (net of carbon) Policy (incl. carbon) Policy (net of 0.4 0.2 technologies in the electricity sector, including starts to replace gas in the electric sector. With 0.4 0.2 0.0 less costly CCS, nuclear and renewables less-costly CCS, gas use increases in the electric 0.0 (Table 3.1). sector by nearly 3 Tcf. This is because both gas generation with CCS and coal generationThe biggest projected Of these, the biggest impact on gas with CCS become economic and share the low-impact on gas use in use in electricity results from low-cost carbon generation market (with about 25% of nuclear generation. Focusing on 2050, electricity produced by gas with CCS by 2050electricity results when the effects of alternative assump- and another 25% by coal with CCS). Gas usefrom an assumption tions are the largest, a low-cost nuclear in the economy as a whole increases even more,of low-cost nuclear assumption reduces annual gas use in by 4.2 Tcf.7 0.30generation. the electric sector by nearly 7 Tcf. 25 Economy-wide gas use falls by only Many other combinations of technological 0.25 about 5 Tcf, however, because the resulting uncertainties could be explored. For example, 0.20 20lower demand for gas in electricity leads to a breakthrough in large-scale electric storage a lower price and more use in other sectors would improve the competitiveness of inter- 0.15 15 of the economy. mittent sources. A major insight to be drawn from these few model experiments, however, is 0.10 Lower-cost renewables yield a reduction in gas that, under a policy based on emissions pricing 10 use in the electric sector by 1.8 Tcf in 2030, to mitigate greenhouse gas emissions, natural 0.05 but total gas use falls by only 1.2 Tcf. In 2050, gas is in a strong competitive position unless 5 a difference in gas use is smaller, 0.5 Tcf and 0.1 competing technologies are much less expen- 0.00 sive than we now anticipate. 060 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • The simulations shown in Figures 3.3–3.5 do resources are developed is material to thenot include the Compressed Natural Gas (CNG) patterns of production and distribution of gasvehicle. When this policy case is repeated with in the U.S. To identify regional patterns ofthis technology included, applying optimistic production and use within the U.S., we applycost estimates drawn from Chapter 4 of this the USREP model andreport, the result depends on the assumption report results for seven Some U.S. regions that haveabout the way competing biofuels, and their regions of the country not traditionally been gaspotential indirect land-use effects, are accounted. for 2006 and 2030 under producers do have significantEven with advanced biofuels credited as a the 50% climate policyzero-emissions option, however, CNG vehicles target and the Mean gas shale gas resources, and therise to about 15% of the private vehicle fleet by resources (Figure 3.6). extent to which these resources2040 to 2050. They consume about 1.5 Tcf of gas Gas production increases are developed is material to theat that time which, because of the effect of the most in those regions patterns of production andresulting price increase on other sectors, adds with the new shale distribution of gas in the U.S.approximately 1.0 Tcf to total national use.8 resources — by more than 78% in the Northeast region (New EnglandSome U.S. regions that have not traditionally through the Great Lakes States) and by aboutbeen gas producers have significant shale gas 50% in the South Central area that includesresources, and the extent to which these Texas. In regions without new shale resources,Figure 3.6 Natural Gas Production and Consumption by Region in the U.S.,2006 and 2030, Price-Based Policy Scenario, Mean Gas Resources WASHINGTON HAWAII MAINE NORTH MINNESOTA 0.3 – 2.7 = -2.4 MONTANA DAKOTA 0.4 – 1.2 = -0.8 MICHIGAN VERMONT OREGON NEW HAMPSHIRE 0.3 – 2.9 = -2.6 SOUTH DAKOTA WISCONSIN 0.9 – 6.2 = -5.3 MASSACHUSETTS IDAHO 0.2 – 2.1 = -1.9 MICHIGAN NEW YORK WEST WYOMING IOWA 1.6 – 7.2 = -5.6 RHODE ISLAND CONNECTICUT NEVADA MOUNTAIN NORTH CENTRAL INDIANA OHIO PENNSYLVANIA NEW JERSEY UTAH NEBRASKA CALIFORNIA 5.1 – 2.0 = 3.1 NORTH EAST DELAWARE ILLINOIS WEST MARYLAND COLORADO VIRGINIA KANSAS MISSOURI 4.3 – 1.8 = 2.5 VIRGINIA KENTUCKY 0.7 – 2.5 = -1.8 NEW MEXICO OKLAHOMA NORTH CAROLINA TENNESSEE ARIZONA SOUTH CENTRAL 0.9 – 4.2 = -3.3 ARKANSAS SOUTH CAROLINA TEXAS SOUTH EAST 11.6 – 6.6 = 5.0 MISSISSIPPI ALABAMA GEORGIA LOUISIANA 12.2 – 4.5 = 7.7 ALASKA FLORIDA 0.5 – 0.4 = 0.1 ALASKA 0.3 – 0.2 = 0.1 Production – Consumption < 0 (Net Imports) Year 2006 Production – Consumption > 0 (Net Imports) Year 2030 Figures refer to annual production and consumption in Tcf.Source: USREP, MIT Chapter 3: System Studies 61
  • production changes little, showing slight To explore this prospect, we formulate a increases or decreases. In the Northeast, the scenario with a renewable energy standard production increase comes close to matching (RES) mandating a 25% share of electric the projected growth in gas use. generation by 2030, and holding at that level through 2050, and measures to force retirement The most substantial potential need for addi- of coal-fired power plants starting in 2020, so tional interregional gas flows, on the regional that coal plants accounting for 55% of current definition of Figure 3.6, is from the Texas/South production are retired by 2050. Mean gas Central region which increases net exports by a resources are assumed, as are the reference combined 2.7 Tcf, with shipment to other levels of all technology costs. This case results regions except the Northeast.9 Compared to the in approximately a 50% reduction in carbon 2030 interregional flows absent climate policy, emissions in the electricity sector by 2050, but the assumed emissions target lowers the need it does not provide incentives to reduction in for new capacity largely because of the expan- non-electric sectors so these measures only sion of supply in the Northeast. hold total national GHG emissions to near the 2005 level, as shown in Figure 3.1.Among the most obvious measures that could have One evident result of these mitigation measuresa direct impact on CO2 emissions would be those is that the reduction in energy demand is lessrequiring renewable energy and one encouraging than under the assumed price-based policy,a phase-out of existing coal-fired power plants. either in the electric sector (Figure 3.7a) or in total energy (Figure 3.7b). Also, the measures represented here achieve less emissions reduc- Scenario 3 — Regulatory Emissions tion in the electricity sector than does the Reductions price-based policy. In the price-based policy, reductions in the electricity sector are about If emissions reductions are sought by regula- 70% by 2050, even though the national target tory and/or subsidy measures, with no price on is only a 50% reduction, because it is less costly emissions, many alternatives are available. to abate there than in the rest of the economy. Among the most obvious measures that could The difference in total national energy use is have a direct impact on CO2 emissions would more dramatic (Figure 3.7b compared with be those requiring renewable energy and one Figure 3.4b) because the all-sector effect of the encouraging a phase-out of existing coal-fired universal GHG price is missing. power plants.62 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 3.7 Energy Mix under a Regulatory Policy, Mean Gas Resources 3.7a Electric Sector (TkWh) 7 Reduced Use 6 Renew 5 Hydro Nuclear 4 TkWh Gas 3 Oil 2 Coal 1 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 Year 3.7b Total Energy Use (qBtu) 140 Reduced Use 120 Renew 100 Hydro Nuclear 80 qBtu Gas 60 Oil 40 Coal 20 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 Year Source: EPPA, MITThese regulatory measures yield a projection regulatory measures as in the case withoutof total U.S. gas use very similar to that under additional GHG-policy shown in Figure 3.5aa no-policy assumption, shown in Figure 3.2. (again see Appendix 3B for a comparison).Under the Mean resource estimate the 2050 Electricity prices do differ from the no-policylevel is almost identical between the two scenario, however, as higher generation costsscenarios (see Appendix 3B), and the figure are passed along to consumers. The result iswould look essentially the same for the High presented in Figure 3.8, where by 2050 the coaland Low cases as well. Also, U.S. natural gas and renewable regulations raise the electricityprices are essentially the same with these price by 50% over its level without GHG policy. Chapter 3: System Studies 63
  • In this case, the effects on natural gas, compared decades of the period, while the reduction in to a no-policy assumption, are concentrated in coal use opens up opportunities for gas. The the electric sector as the non-electric sectors net result is a pattern of gas use over time not face roughly the same gas price in both cases. different from the no-policy case, as noted earlier. Naturally, the net impact on gas use in the electric sector depends on the stringency ofNatural gas remains resilient under a wide range of the two regulatory measures and their relativepotential approaches to U.S. climate policy. pace of implementation, and compared to the assumed price-based approach, they have the In the electric sector, the forced expansion of potential to reduce the use of gas in the sector. renewables tends to squeeze out gas-based Nonetheless, for this regulatory scenario, like electric generation, particularly in the early the more ambitious policy-based case, U.S. natural gas demand remains resilient, continu-Figure 3.8 Electricity Prices ($/kWh) under No-Policy and ing to make a major contribution to nationalRegulatory Scenarios, Mean Gas Resources energy use. 0.25 0.20 0.15$/kWh 0.10 0.05 0.00 2000 2010 2020 2030 2040 2050 2060 1.0 0.8 0.6 0.4 0.2 0.0 Year No Policy Regulatory Policy Source: EPPA, MIT Regulatory Policy Haynesville Woodford Fayetteville Barnett 1.0 0.8 0.6 0.4 Policy (net of carbon) Policy (incl. carbon) 0.2 0.0 0.3064 MIT STUDY ON THE FUTURE OF NATURAL GAS 0.25
  • THE ROLE OF INTERNATIONAL GAS MARKETSCurrently world gas trade is concentrated in three In this scenario, gas suppliers and consumersregional markets: North America; Europe — are assumed to operate on an economic basis.served by Russia and Africa; and Asia — with That is, no effective gas cartel is formed, anda link to the Middle East. There are significant suppliers exploit their gas resources for maxi-movements of gas within each of these markets, mum national economic gain.but limited trade among them. Projected effects on U.S. production and tradeDifferent pricing structures hold within these are shown in Figure 3.9 for the 50% reductionregional markets. For some transactions, prices and High, Mean and Low gas resources cases.are set in liquid competitive markets; in others This result may be compared with the Regionalthey are dominated by contracts linking gas Markets case shown in Figure 3.3.prices to prices of crude oil and oil products.As a result, gas prices can differ substantially In 2020, U.S. net imports are lowered to 1.6 Tcfamong the regions. (versus 4.1 Tcf in the Regional Markets case). Because in the Integrated Global MarketThese relatively isolated, regionalized markets scenario the EPPA model resolves for the netcould be sustained for many more decades. trade only, a decrease in net imports might beOn the other hand, it is possible that LNG or interpreted as a potential for small gas exportspipeline transport could grow, linking these from the U.S. while keeping imports constant.three regions, with the effect of increasing Beginning in the period 2020 to 2030, the costinterregional gas competition, loosening price of U.S. gas begins to rise above that of suppliescontracts tied to oil products and moderating from abroad and the U.S. becomes morethe price deviations among the regions. dependent on imports of gas. In the Mean resource case, the U.S. depends on imports forSuch a process could go in many directions about 50% of its gas by 2050 and U.S. gas usedepending on the development of supply rises to near the level in the no-policy case,capacity by those nations with very large because prices are lower. As the emergence ofresources (mainly Russia and countries in the an integrated global market would lead ulti-Middle East) or perhaps the expansion of mately to greater reliance on imports, U.S. gasnon-conventional sources elsewhere. To the use — and prices — are much less affected byextent the structure evolves in this direction, the level of domestic resources. Thus, thehowever, there are major implications for U.S. development of a highly integrated interna-natural gas production and use. tional market, with decisions about supply and imports made on an economic basis, wouldTo investigate the end-effect of possible evolu- have complex effects: it would benefit the U.S.tion of an integrated global market akin to economically, limiting the development ofcrude oil, we simulate a scenario where market domestic resources but would lead to growingintegration and competition lead to equaliza- import dependence.tion of gas prices among markets except forfixed differentials that reflect transport costs. Chapter 3: System Studies 65
  • Figure 3.9 U.S. Gas Use, Production and Imports & Exports (Tcf) and U.S. Gas Prices ($/1000 cf) for Low (L), Mean (M) and High (H) U.S. Resources, Price-Based Climate Policy and Global Gas Markets. Prices are shown without (top) and with (bottom) the emissions charge. 45 40 6.6 6.4 6.2 7.3 7.0 6.8 14.5 14.3 14.2 17.2 17.0 16.9 35 5.8 5.7 5.6 30 11.5 11.4 11.3 5.1 5.1 5.1 8.1 8.1 8.1 25 Tcf 20 15 10 5 0 L M H L M H L M H L M H 2020 2030 2040 2050 Year Imports Production Source: EPPA, MIT In the Regional Markets case, global demand Union it increases by about 35%. Assumption for gas increases from the current demand of of an Integrated Global Market changes the about 100 Tcf, to about 150 Tcf by 2050. In the growth in Asia to 135%, while U.S. and Integrated Global Markets scenario, gas avail- European use grows by about 70%. A growth ability increases globally, reducing gas prices, in the Rest of the World (ROW) is mostly and as a result, gas demand rises to about driven by an increase in the gas usage in the 190 Tcf in 2050. Figure 3.10 shows the Middle East and the rest of Americas, where projected increase in gas use. In the Regional assumptions about the different market Markets case, gas use in U.S. and Asia grows structures affect the results to a lesser degree. by around 50% from 2010 to 2050, while in Europe and countries of the former Soviet66 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 3.10 Gas Use (Tcf) in Regional Markets (reg) and Integrated Global Markets (gl)scenarios for USA, Asia, Europe and former Soviet Union (EUR+FSU) and the Rest of theWorld (ROW) 200 180 160 140 120Tcf 100 80 60 40 20 0 reg gl reg gl reg gl reg gl 2020 2030 2040 2050 Year USA EUR & FSU Asia ROW Source: EPPA, MITPossible international gas trade flows that are of bilateral trade in gas are highly uncertain.consistent with U.S. and global demand under The Regional Markets case tends to increasethe Regional and Integrated Global Markets trade among partners where trade alreadycases are shown in Figure 3.11. Under Regional exists, locking in patterns determined in partMarket conditions (Figure 3.11a), we can see by historical political considerations.that trade flows are large within gas marketregions but small among them. To avoid a If a highly integrated Global Market is assumedcluttered map, small trade flows (less than to develop (Figure 3.11b), a very different1 Tcf) are not shown. Except for the “Middle pattern of trade emerges. The U.S. is projectedEast to Europe” flow of 1.8 Tcf, interregional to import from the Middle East as well as frommovements among the three regions specified Canada and Russia, and movements from theabove are less than 0.6 Tcf in any direction Middle East to Asia and Europe would increasein 2030. implying a substantial expansion of LNG — facilities. Russian gas would begin to move intoTrade flows can be particularly sensitive to the Asian markets, via some combination ofdevelopment of transportation infrastructure pipeline transport and LNG.and political considerations, and so projections Chapter 3: System Studies 67
  • Figure 3.11 Major Trade Flows of Natural Gas among the EPPA Regions in 2030, No New Policy (Tcf)3.11a Regional Markets Source: EPPA, MIT3.11b Global Market Source: EPPA, MIT68 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • The precise patterns of trade that might LONGER-TERM PROSPECTS FOR GASdevelop to 2030 and beyond will be influenced UNDER DEEPER EMISSIONS CUTSby the economics of the energy industry, ascaptured by the EPPA model, and also by While current investment and policy decisionsnational decisions regarding gas production, appropriately focus on a shorter horizon, policyimports and transport infrastructure. There- decisions related to atmospheric stabilization offore, the numbers shown are subject to a GHG concentrations inevitably involve a verynumber of uncertainties, prominent among long-term perspective. Though gas frequentlywhich is the willingness of Middle East and is touted as a “bridge” to the future, continuingRussian suppliers to produce and export on the effort is needed to prepare for that future, lestmodeled economic basis.10 If potential supplies the gift of greater domestic gas resources turnare not forthcoming, then global prices would out to be a bridge with no landing point on thebe higher and the U.S. would import less than far bank.projected and perhaps increase exports. Thebroad insight to be drawn is nonetheless To explore this issue, we conducted modelevident: to the degree that economics are simulations extending the horizon to 2100allowed to determine the global gas market, assuming GHG emissions cuts that deepen totrade in this fuel is set to increase over coming 80% below 2005 levels. The result is that, untildecades, with implications for investment and gas with CCS begins to penetrate after 2060, thepotential concerns about import dependence. cost of CO2 emissions from gas generation becomes too high to support its use in genera-The assumptions about the gas markets also tion (Figure 3.12). Nuclear is cheaper than coalaffect the carbon price and GDP impacts in the or gas with CCS for much of the period and soGHG mitigation scenario. While the difference it expands to meet the continuing electricityis small initially (in 2030, a U.S. carbon price is demand. Different cost assumptions well withindecreased from $106 to $103 per ton CO2-e and the range of uncertainty would lead to aU.S. GDP loss is decreased from 1.7% to 1.6%), different mix of low-CO2 generation sources,it grows over time (in 2050, a U.S. carbon price but the picture for gas without CCS wouldis decreased from $240 to $180 and U.S. GDP remain the same.loss is decreased from 3.5% to 2.6%).Figure 3.12 Energy Mix in Electric Generation under a Price-Based Climate Policy,Mean Natural Gas Resources and Regional Natural Gas Markets (TkWh) 8 Reduced Use 7 Renew 6 Hydro 5 Nuclear TkWh 4 Gas CCS 3 Gas 2 Oil Coal CCS 1 Coal 0 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 Source: EPPA, MIT Year Chapter 3: System Studies 69
  • One implication of this longer-term experi- CONCLUSIONS ment is that while we might rely on plentiful supplies of domestic gas in the near term, this The outlook for gas over the next several must not detract us from preparing for a future decades is in general very favorable. In the with even One implication of this longer-term electric generation sector, given the unproven experiment is that while we might rely on and relatively high cost of other low-carbon plentiful supplies of domestic gas in the near generation alternatives, gas could well be the term, this must not detract us from preparing preferred alternative to coal. for a future with even greater GHG emissions A multi-sector GHG pricing policy wouldTo the degree that economics are allowed to determine increase gas use in generation but reduce its usethe global gas market, trade in this fuel is set to increase in other sectors, on balance increasing gas useover coming decades, with implications for investment substantially from the present level. A regula- tory approach, applied to renewable and coaland import dependence. use in the electric sector, could lead to even greater growth in gas use while having a more constraints. Barriers to the expansion of limited effect on national GHG emissions. nuclear power or coal and/or gas generation Most important, in all cases studied — no new with CCS must be resolved over the next few climate policy and a wide range of approaches decades so that over time these energy sources to GHG mitigation — natural gas is positioned will be able to replace natural gas in power to play a growing role in the U.S. energy economy. generation. Without such capability, it would not be possible to sustain an emissions mitiga- International gas resources are likely less costly tion regime. than those in the U.S. except for the lowest-cost domestic shale resources, and the emergence of an integrated global gas market could result in significant U.S. gas imports. The shale gas resource is a major contributor to domestic resources but far from a panacea over the longer term. Under deeper cuts in CO2 emissions, cleaner technologies are needed. Gas can be an effective bridge to a lower CO2 emissions future but investment in the develop- ment of still lower CO2 technologies remains an important priority. 70 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • NOTES 7 For more details about sensitivity tests see Paltsev, S., H. Jacoby, J. Reilly, Q. Ejaz, F. O’Sullivan,1 Citations to documentation of the EPPA model and J. Morris, S. Rausch, N. Winchester and O. Kragha. features related to this study are provided in Paltsev, 2010: The Future of U.S. Natural Gas Production, S., H. Jacoby, J. Reilly, O. Kragha, N. Winchester, J. Use and Trade, MIT Joint Program on the Science Morris and S. Rausch, 2010: The Future of U.S. and Policy of Global Change, Report 186, Natural Gas Production, Use, and Trade. MIT Joint Cambridge, MA. Program on the Science and Policy of Global 8 Substitution for motor fuel is the likely target of Change, Report 186, Cambridge, MA. The USREP possible expansion of gas-to-liquids technology model is described by Rausch, S., G. Metcalf, J. Reilly (see Chapter 4). Its market penetration would and S. Paltsev, 2010: Distributional Impacts of depend on competition not only with oil products Alternative U.S. Greenhouse Gas Control Measures. but also with direct gas use, biofuels and electricity MIT Joint Program on the Science and Policy of which reduce CO2 emissions while liquids from gas Global Change, Report 185, Cambridge, MA. would not.2 Reference costs from the U.S. EIA (see Appendix 9 Gas production and use with the USREP model is 3A). The lower sensitivity estimate is based on somewhat lower than the EPPA projection. Update of the 2003 Future of Nuclear Power: An Compared to EPPA, the USREP model has the Interdisciplinary MIT study, Massachusetts advantage of capturing inter-regional differences in Institute of Technology, Cambridge, MA. coal and gas prices, and better reflecting differences3 Reference costs from the U.S. EIA (see Appendix in renewable costs among regions, but it does not 3A). The lower sensitivity estimate for coal with represent foreign trading partners. This variation CCS draws on The Future of Coal: An Inter- introduced by the different model structures is well disciplinary MIT study, Massachusetts Institute of within the range of other uncertainties. Technology, Cambridge, MA; that for gas with CCS 10 For additional scenarios about the long-term comes from McFarland, J., S. Paltsev and H. Jacoby, prospects for Russian natural gas, see Paltsev S., 2009: Analysis of the Coal Sector under Carbon (2011). Russia’s Natural Gas Export Potential up Constraints, Journal of Policy Modeling, 31(1), to 2050. MIT Joint Program on the Science and 404–424. Policy of Global Change Report (forthcoming).4 LCOE is the cost per kWh that over the life of the plant fully recovers operating, fuel, capital and financial costs.5 CO2 equivalent emissions for all greenhouse gases are calculated using 100-year global warming potentials (GWPs). See Appendix 1A for discussion. The simulations in this chapter account for fugitive methane emissions from the gas supply system.6 Because of the limited opportunities for gas-oil substitution the current price premium in the U.S. of oil products over gas (on an energy basis) is maintained and even grows over time. One substitution option not modeled here is the possibility of conversion of gas to liquids, which might become economic and perhaps be further stimulated by security concerns, even though making no contribution to CO2 reduction. Such a development would raise U.S. gas use and prices, and lower oil demand with some moderating effect on the world oil price. Chapter 3: System Studies 71
  • Chapter 4: Electric Power GenerationINTRODUCTIONThe low-carbon emissions and low capital cost Natural gas provides flexibility to the power systemof natural gas generation compared to other largely through the three types of generationfossil fuel generation, combined with abundant technologies: highly efficient natural gas com-gas supplies and current relatively low prices, bined cycle (NGCC) units, steam turbines, andmake natural gas an attractive option in a gas turbines. Gas turbines are generally used tocarbon-constrained environment, such as that meet peak demand levels and to handle weather,contemplated in the analysis in Chapter 3. In time of day, seasonal and unexpected changes inaddition to its increasingly important role as a demand. NGCCs and steam turbines can act asprimary fuel for electricity generation, natural baseload or intermediate-load units, althoughgas will continue to perform a unique function the majority of gas capacity in the U.S. nowin the power sector by providing both baseload operates in load-following (intermediate) orpower and the system flexibility that is required peaking meet variation in power demand and supplyfrom intermittent sources. Currently, natural gas is second only to coal in total generation, fueling 23% of U.S. electricityThe focus of this chapter is on the role of natural production. Natural gas, however, has the highestgas in helping to reduce CO2 emissions from the percentage of nameplate1 generation capacity ofpower sector and the interaction of gas use with any fuel, at 41% compared to 31% for coal, whichprojected growth in wind and solar generation. is the next highest (Figure 4.1). This difference between nameplate capacity and generation isFigure 4.1 % Nameplate Capacity Compared to % Net Generation, U.S., 2009* 50 44 41 40 Percentage (%) 30 30 23 20 20 9 9 10 7 6 5 4 1 <1 <1 0 Natural Gas Petroleum Coal Hydroelectric Nuclear Other Other Renewables % Nameplate Capacity % Net Generation *Numbers are rounded Source: MIT from EIA data Chapter 4: Electricity 73
  • explained in part by the overbuilding of BOX 4.1 MODELS EMPLOYED TO EXAMINE NGCC units in the mid-1990s. It also shows THE U.S. ELECTRICITY SYSTEM that NGCC units are operating well below The MARKAL (MARKet ALlocation) model of their optimum operating value. Finally, it the U.S. electricity sector enables a granular highlights the unique role of gas and steam understanding of generation technologies, turbines, which in 2009 had an average time-of-day and seasonal variations in electric- capacity factor of 10% (see Table 4.1). This ity demand and the underlying uncertainties low-capacity factor illustrates the peaking of demand. It was originally developed at function of these units, particularly the gas Brookhaven National Laboratory (L.D. Hamilton, turbines, that are routinely used only to meet G. Goldstein, J.C. Lee, A. Manne, W. Marcuse, peak demand levels and which, absent break- S.C. Morris, and C-O Wene, “MARKAL-MACRO: throughs in storage, are essential for following An Overview,” Brookhaven National Laboratory, time-varying electricity demand and accom- #48377, November 1992). The database for the modating the intermittency associated with U.S. electric sector was developed by the National Risk Management Laboratory of the wind and solar power. U.S. Environmental Protection Agency (EPA). Historically, because of its higher fuel price The Renewable Energy Deployment System compared with nuclear, coal and renewables, (ReEDS) model is used to project capacity natural gas has typically had the highest expansions of generation, incorporating trans- marginal cost and has been dispatched after mission network impacts, associated reliability considerations and dispatch of plants as operat- other generation sources. Consequently, natural ing reserves. It also captures the stochastic nature gas has set the clearing price for electricity in of intermittent generation as well as temporal much of the country. Lower natural gas prices, and spatial correlations in the generation mix the opportunities created by abundant and demand. It has been developed by the relatively low-cost supplies of unconventional National Renewable Energy Laboratory (NREL) shale gas, increased coal costs and impending (J. Logan, P. Sullivan, W. Short, L. Bird, T.L. James, environmental regulations that will add to the M. R. Shah, “Evaluating a Proposed 20% National cost of coal generation are, however, changing Renewable Portfolio Standard,” 35 pp. NREL the role of gas in power generation. Report No. TP-6A2-45161, 2009). The Memphis model realistically simulates the The Emissions Prediction and Policy Analysis hourly operation of existing generation plants (EPPA) model employed in Chapter 3 is in the presence of significant volumes of wind designed to study multi-sector, multi-region and solar generation. It was developed by the effects of alternative policy and technology Institute for Research in Technology of Comillas assumptions, and as a result it only approxi- University (Madrid, Spain) for the Spanish mates the complexities of electric system Electricity Transmission System Operator (Red dispatch. In this chapter, we analyze in greater Eléctrica de España) to integrate renewable depth two of the cases studied there, employing energies. (A. Ramos, K. Dietrich, J.M. Latorre, L. a more detailed model of the electric sector — Olmos, I.J. Pérez-Arriaga, “Sequential Stochastic Unit Commitment for Large-Scale Integration of MARKAL (see Box 4.1). This model is also RES and Emerging Technologies,” 20th Interna- used to further explore the implications of tional Symposium of Mathematical Program- uncertainty in fuel and technology choices as ming (ISMP) Chicago, IL, USA, August 2009. they influence natural gas demand in this sector, extending the uncertainty analysis in Chapter 3 which considers only the uncer- tainty in gas resources.74 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 4.1 2009 Average Capacity Factors by Select Energy Source, U.S. (numbers rounded) Natural Natural Hydroelectric Other All Energy Coal Petroleum Nuclear Gas CC Gas Other Conventional Renewables Sources 64 8 42 10 90 40 34 45Source: EIA, Table 5.2, Average Capacity Factors by Energy SourceThis chapter then considers two questions Study of these two questions is approachedabout gas use in U.S. power generation: with the use of two additional electric sector(1) What is the potential for reducing CO2 by models, each designed to simulate the powerchanging the current generation dispatch order2 system and its operations in detail over a rangeto favor NGCC over coal generation? (2) What of conditions and timescales (see Box 4.1),will be the effect of increased penetration of enabling the following analyses:wind and solar generation on natural gaspower generation? - sion constraints, which helps to isolate andTo answer the first question it is important to understand the total generation required atunderstand NGCC utilization patterns. NGGC points in time to meet demand for electricityunits are designed to be operated at capacity and maintain operating reserve capacity andfactors of up to around 85% rather than the adequate installed capacity margins. Wecurrent national average of 42%. This suggests employ ReEDS for this analysis, which usespossible opportunities for displacing some coal multiple time periods for any given year andwith gas generation, thereby lowering CO2 reports results by geographic regions.emissions from the sector. We examine howmuch of this capacity could actually be appliedto this purpose without diminishing system hourly level, which takes into considerationreliability. An important by-product of such details of real-time problems, such as uncer-a change, also analyzed, would be associated tainty and variability in demand and inreductions in criteria pollutant emissions. generation patterns for intermittent tech- nologies, and start-up and shut-downTo explore the second question, the interaction characteristics for plant cycling. Here we usebetween intermittent renewables and natural the Memphis model.gas use is analyzed from two viewpoints: onein the short term when additional intermittentcapacity is introduced into a system with othersources fixed; and the other in the longer termwhen the overall supply structure has time toadjust to growth in intermittent capacity. Inthis regard, we note that, at a more granularlevel than is presented in Table 4.1, windturbines have an average capacity factor of27%, solar thermal, 19%, and solar PV, 14%,and gas combustion turbines and steamturbines (used to balance load) have averagecapacity factors of 5% and 14%, respectively.3 Chapter 4: Electricity 75
  • ELECTRICITY SYSTEM OVERVIEW The expansion planning and operation of electric power systems involve several decisions The electricity system is complex; this overview at different timescales, generally based on of how the system works, including the regimes economic efficiency and system reliability under which power plants operate and the criteria. This process has a hierarchical structure, hierarchy of decision-making that influences where the solutions adopted at higher levels are the capacity and generation mixes, is intended passed on to the lower levels incorporating to enhance the understanding of the implica- technical or operational restrictions at that level:5 tions of the modeling and analysis discussed later in the chapter. Long-term decisions are part of a multi-year process (3 years up to 10 or more years) that Electricity is produced from diverse energy involves investments in generation and sources, varied technologies and at all scales. network required to expand the system. Sources for electric generation include a mix of renewables (sun, wind, hydro resources, among Medium-term decisions are taken once the others), fossil fuels (oil, natural gas, coal) and expansion decisions have been made. They uranium. As such, the generation of electricity are part of an annual process (up to 3 years) comprises a variety of technologies with the that determines the generation unit and grid type of fuel being used, and characterized by a maintenance schedule, fuel procurement wide range of investment and operating costs. and long-term hydro resource scheduling. Conventional power plants are operated under different regimes, mainly depending on their Short-term decisions are a taken on a weekly variable operating costs and operating time frame. They determine the hourly flexibility.4 production of thermal and hydroelectric plants for each day of the week (or month), Baseload plants are characterized by expensive subject to availability of the plants and to capital costs and low variable costs, and they hydro production quotas determined at the are operated most of the time during the year. upper decision level, and considering not They tend to be inflexible plants as they only variable operating costs, but also the cannot easily change their operational level technology’s own technical characteristics over a wide usage. such as start-up and shut-down cost and conditions, a plant’s technical minima and Peaking plants are characterized by low ramping times. In addition, these short-term capital costs and higher variable costs, and decisions are subject to generating reserve they are operated a few hours per year when capacity needed to immediately respond to the electric load is the highest. They can be unexpected events. characterized as flexible plants because of their quick operating response. Real-time decisions involve the actual opera- tion of the system (seconds to minutes). They Intermediate plants have variable costs that involve the economic dispatch of generation fall in between those of peaking and baseload units, the control of frequency so that technologies, and they are operated accord- production and demand are kept in balance ingly. They can be characterized as cycling at all times, while maintaining the system plants, i.e., plants that operate at varying levels components within prescribed safe tolerances during the course of the day and perhaps of voltages and power flows, accounting also shut down during nights and weekends. for possible contingencies.76 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Finally, meeting reliably the consumption of For consistency with the analysis in Chapter 3,electric power at all times requires having both certain MARKAL inputs are taken from theadequate installed capacity and secure opera- EPPA model results, including electricitytion procedures. A reliable operation involves demand, supply curves for natural gas and coalusing ancillary services at different levels, and the reference costs of generation technolo-maintaining sufficient capacity in reserve gies. Also, two of the same policy cases are(quick-start units, spinning reserves) and with considered: Scenario A, which assumes no newenough flexibility to respond to deviations in GHG policy; and Scenario B, which imposesthe forecast of demand or intermittent genera- a Price-Based mitigation measure. For thetion, and to unexpected events, such as the Price-Based case, a cap on CO2 emissions forsudden loss of lines or generation plants. the electric sector in MARKAL is set based on the results for that scenario in Chapter 3.THE ROLE OF GAS GENERATIONUNDER A CO2 LIMIT The underlying technology mix computed by the more-detailed electric sector model canThe EPPA model simulations in Chapter 3 be illustrated by annual load duration curves,provide insights into both the economy-wide which show the mix of generation dispatcheduse of natural gas and its market share in at different times to meet changes in the levelelectric power under various assumptions of electricity demand in the contiguous U.S.about greenhouse gas (GHG) mitigation. electric system over the course of a year. TheseApplication of the MARKAL model, with its curves for the year 2030, with and withoutgreater electric sector detail, provides a check a policy of carbon constraints, are shown inon the adequacy of the EPPA approximations Figure 4.2. In the absence of a carbon policyfor the power sector. MARKAL considers a (Panel a), generation from hydro, coal andmore complete listing of the generation alter- nuclear occur at all times of the year whilenatives, and it addresses the variation in the generation from wind and hydro are suppliedlevel of electricity demand, as a result of the whenever they are available.6diurnal, weekly and seasonal cycles (whichEPPA only roughly approximates). This varia- Without a carbon policy (Panel a), natural gastion is important because different technologies generation from combined cycle and steamare needed to run different numbers of hours turbines occurs for less than half of the timeper year — a pattern that changes over years over the course of the year during periods ofwith demand growth and new investment. Also, higher demand; and natural gas combustionthe MARKAL model allows for a more com- turbines are used for only a few hours per yearplete exploration of uncertainty in gas use in at the peak demand hours.the power sector. Under the carbon price policy (Panel b), NGCC technology largely substitutes for coal to provide baseload generation along with nuclear generation. Chapter 4: Electricity 77
  • Figure 4.2 Time blocks approximation to the Load Duration Curve for the (a) No Policy and (b) 50% Carbon Reduction Policy Scenarios in 2030. Three seasons have been con- sidered: summer, winter and spring/autumn. Within each season, there are four blocks: peak time, daytime PM, daytime AM, and nighttime, as shown in the graphs. The peak time block is very narrow. 4.2a 1200 1000 Solar 950 Oil 900 1000 Landfill 850 Geothermal 800 Diesel 800 750 Biomass 700 Municipal Waste GW 600 Gas Steam Gas Combustion Turbine Gas Combined Cycle 400 Wind Coal Steam 200 Hydro Nuclear 0 0 2000 4000 Hours 6000 8000 Summer Winter Spring & Autumn 4.2b 1000 800 750 800 700 650 600 600 GW 400 200 0 0 2000 4000 Hours 6000 8000 Summer Winter Spring & Autumn Source: MIT analysis78 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • The change over time in the energy mix in the because MARKAL is a forward-looking modelelectric sector is shown in Figure 4.3 for both and sees higher prices in the future whereas thethe No Policy and the Price-Based cases. In the recursive dynamic (myopic) EPPA model doesNo Policy case, under reference assumptions for not. Farther out in time coal is no longer in thefuel prices, electricity demand and technology mix, and under a continuously tightening CO2costs — and mean gas resources — these results constraint conventional gas generation beginsshow the same pattern of increasing gas use as to be replaced by non-carbon generation sourcesthe simulation studies in Chapter 3. The gas use such as nuclear, renewables and/or coal or gasin this sector in 2025 is essentially the same in with carbon capture and sequestration (CCS).the two studies. Toward the end of the simula-tion period, MARKAL projects one-quarter to The EPPA model expands nuclear generationone-third more gas-based generation than whereas MARKAL introduces natural gas withEPPA, though gas generation is still small CCS, yielding about a one-quarter greater levelrelative to coal. of gas use. The outlook for gas in this sector is consistently positive across the two studies, andUnder the Price-Based policy the overall the difference in details of load dispatch is to bepattern of change remains the same as in EPPA: expected for models of such different math-coal is forced out and replaced by gas. In the ematical structure, and well below the level ofperiod to 2025 MARKAL projects a more rapid uncertainty in either (see Figure 4.3).phase-out of coal than does EPPA, in partFigure 4.3 Future Energy Mix in Electricity Sector Figure 4.3a With No Climate Policy Figure 4.3b With Price-Based Climate Policy 7 7 6 6 Electricity generation (TkWh) Electricity generation (TkWh) Others 5 5 Renewable 4 Hydro 4 Nuclear 3 Natural Gas 3 Coal 2 2 1 1 0 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 2010 2015 2020 2025 2030 2035 2040 2045 2050 Source: MIT Analysis Chapter 4: Electricity 79
  • The systems studies in Chapter 3 consider only NEAR-TERM OPPORTUNITIES FOR uncertainty in the estimates of gas resources REDUCING CO2 EMISSIONS BY (Figures 3.2, 3.3 and 3.0). Applying the ENVIRONMENTAL DISPATCH MARKAL model and the reference assumptions discussed above, a study was carried out of the Near-term opportunities for CO2 emission effect on gas use of uncertainties not only in reductions in the power sector are limited by resources but in other prices, electricity the current generation mix and transmission demand and technology costs. The same two infrastructure, the cost of renewables and other cases were considered: No Policy; and the low-emission sources and technologies, as well as Price-Based policy. Here we describe results for the lag times associated with siting and building a 50% confidence interval: i.e., a 25% chance any new generation capacity. The re-ordering of of gas use above the high level as shown, and a generation between coal and gas units (modeled 25% chance of use below the low level. Details here as a form of environmental dispatch forced of the analysis are provided in Appendix 4B. by a CO2 constraint7) may be the only option for large-scale CO2 emissions reduction from the By 2030, with no additional mitigation policy, power sector which is both currently available the gas demand by the electric sector runs and relatively inexpensive. 17% above and 19% below the mean value of 6.3 trillion square feet (Tcf) (50% confidence As noted, the current fleet of NGCC units has interval). The main factors leading to this range an average capacity factor of 41%, relative to are the demand for electricity, the prices of a design performance of approximately 85%. natural gas and coal and the costs of new An electric system requires capacity to meet technologies, in particular the cost of new coal peak demands occurring only a few hours per steam and IGCC technologies. year, plus an operating reserve, so the system always includes some generation units that run Under the Price-Based policy the uncertainty is at capacity factors below their design value. substantially greater, ranging from 47% above However, the U.S. has enough spare capacity to 42% below the mean value of 12.8 Tcf (50% in other technologies to allow dispatching more confidence interval). The main influence NGCC generation, displacing coal and reduc- behind this greater uncertainty is in the costs of ing CO2 emissions, without major capital technologies that might substitute at large scale investment. An additional benefit of this for fossil-based generation, such as wind, solar approach would be to substantially reduce and advanced nuclear generation technologies. emissions of air pollutants such as sulfur The share of natural gas in the generation mix dioxide (SO2), nitrogen oxide (NOx), mercury is a result of the interplay between technologies (Hg) and particulates. that both compete with and complement each other at the same time as they supply different NGCC Potential if Fully Dispatched segments of demand over the year. Figure 4.4 suggests the scale and location of the The uncertainty ranges given here are intended potential for shifting among generation units. to caution the reader against giving too much Plotted there is the geographic distribution of weight to the actual numbers in future projec- fully-dispatched NGCC potential (FDNP), tions in this chapter and elsewhere in the defined as the difference between the electricity report. Rather, the critical insights are about the that would be produced by NGCC plants at trends and relationships, which are more robust an 85% capacity factor and their actual 2008 across a wide range of possible futures. generation. Figure 4.4 also shows the geographic distribution of coal generation, divided into80 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 4.4 Scale and Location of Fully Dispatched NGCC Potential (FDNP)and Coal Generation (MWh, 2008)Source: USREP, MITless and more efficient units where a “less Possible Contribution of NGCC Capacityefficient” unit is defined as one with a heat rate to a CO2 Reduction Goalover 10,000 Btu/kWh. Figure 4.4 represents only the average potentialIn many regions FDNP generation matches available over the course of the year, aggregatedwell with less efficient coal capacity, suggesting by state, therefore providing an upper limit ofopportunities for displacing emissions-intensive the substitution potential; it does not equateunits, while other locations show few such to “surplus” generation capacity. For thisopportunities. For example, Southeastern states discussion, “surplus” is defined as the amountsuch as Texas, Louisiana, Mississippi, Alabama of NGCC generation that can be used over theand Florida appear to have relatively larger course of one year to replace coal while respectingopportunities, while those in Midwestern states transmission limits, operation constraints andsuch as Illinois, Indiana and Ohio are relatively demand levels at any given time.smaller. Chapter 4: Electricity 81
  • To account for a number of system charac- teristics that may better identify the range of they import and export electricity from each opportunities for fuel substitution, we apply other, but have a relatively small amount of the ReEDS model (see Box 4.1). This model is NGCC surplus; well suited for examination of reliability and transmission constraints, demand fluctuations and reserve capacity margins that will limit but New England has relatively little coal these opportunities. Also, as noted, ReEDS generation, whereas Florida has a significant reports results by geographic regions.8 percentage of inefficient coal capacity that might be a candidate for displacement. This enables us to identify opportunities to change the fuel dispatch order nationwide, We analyze the potential for a version of and provides insights into five regions of the environmental dispatch by running the ReEDS country: the Electric Reliability Council of model for the year 2012 in three scenarios: CO2 Texas (ERCOT), Midwest Independent Trans- unconstrained, a 10% reduction in U.S. electric mission Operator (MISO), Pennsylvania- sector CO2 emissions, and a 20% reduction. Runs New Jersey-Maryland (PJM), New England for the year 2012 are used because the model does (ISO-NE) and Florida Reliability Coordinating not invest in new capacity in this time period; as Council (FRCC). Each region has different such, CO2 reductions are attributable to shift of generation costs, fuel mixes and ability generation among existing units. to trade electricity: Figure 4.5 illustrates the changes in generation by technology under the three scenarios. In the from the rest of the country; 20% CO2 reduction scenario, the NGCC fleet has an average capacity factor of 87%, displaces Figure 4.5 Generation by Technology under Various CO2 Constraints, U.S.9, 2012 2,500 2,000 Generation (TWh) 1,500 1,000 500 0 Base Case 10% CO2 Reduction 20% CO2 Reduction Coal Gas-CC Wind Nuclear Hydroelectric Other* Source: MIT Analysis82 MIT STUDY ON THE FUTURE OF NATURAL GAS 20.0
  • Figure 4.6 NGCC and Coal Generation in Select Regions under a 20% CO2 Constraint,U.S., 2012 600 500 400 Generation (TWh) 300 200 100 0 16: MISO 21: ERCOT 23: PJM 29: FRCC 32: ISO-NE Coal (base case) Gas Surplus Source: MIT Analysisabout one-third of 2012 coal generation represent the amount of additional NGCC(700 terawatt-hours (TWh)) and increases gas generation that is available for dispatch in theconsumption by 4 Tcf.9 current system after satisfying all system require- ments. This additional amount of generation is 20.0In Figure 4.5, as the carbon constraint calculated as the difference between the NGCCincreases, most of the electricity generation by generation dispatched in the base case and in thetechnology does not change. Coal and natural 20% CO2 reduction scenario. The largestgas are the exceptions: as the carbon constraint potential for substitution of NGCC for coalincreases, coal generation significantly declines, generation is in PJM, although in both PJMand NGCC proportionally increases. and MISO coal continues to dominate.Although NGCC displacement of coal genera- A closer look at how the imposition of a CO2tion is nearly one-for-one at the national level, limit would shift generation among units canthe change in generation and emissions is not be seen in the revised unit dispatch at differentuniform across regions. Figure 4.6 shows demand levels. For this analysis, we look atregional results, comparing coal generation in ERCOT, a system that is isolated from the restthe absence of a CO2 target to surplus NGCC of the U.S. and, in our re-dispatch scenarios,generation in a 20% reduction scenario. has regional percentage of CO2 reductions that tracks national reductions. Because of theseIn Figure 4.6, the left bars represent the amount similarities to the country, and because of theof regional coal generation absent carbon greater availability of operations informationconstraints, using ReEDS 2012 forecasts. This is from ERCOT, an analysis of ERCOT, usingthe “business as usual” scenario. The right bars Chapter 4: Electricity 83
  • Figure 4.7 Changes in Dispatch Order to Meet ERCOT’s 2012 Demand Profile, with and without a 20% CO2 Constraint MW 90,000 Oil-Gas-Steam Turbine Natural Gas – Combustion Turbine 80,000 Natural Gas – Combined Cycle Coal 70,000 Wind Hydro 60,000 Nuclear Annual load duration curve 50,000 40,000 30,000 20,000 10,000 0 Base Fuel Base Fuel Base Fuel Nameplate Switch Switch Switch capacity 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 Super peak demand – 40 hrs total Average annual dispatch Low demand – 736 hrs total Late summer afternoons profile – over 8,760 hrs Typical spring nights Source: MIT Analysis ReEDS provides additional details about fuel In Figure 4.7, the red line represents 17 time switching on a more granular timescale. periods of demand for the year, sorted from greatest to least demand. The bar graphs to the Bar graphs Figure 4.7 illustrates how existing capacity right of the nameplate capacity bar show the represent would be dispatched to meet 2012 projected dispatch profile in those time periods under dispatch pro demand for the highest peak, average and low two carbon scenarios: no reduction and 20% demand situations, with and without the CO2 reduction. target to force a change in unit dispatch.10 The figure shows an unconstrained base case and Not surprisingly, the results indicate that the a case with a 20% CO2 reduction. The average greatest opportunities for displacement of coal profile shows the generation dispatch for all generation exist during average and low technologies across an entire year (8,760 hours), demand periods. Figure 4.7 also shows that coal not a single time slice. generation is dispatched in every time period, indicating that not enough NGCC surplus exists in ERCOT to completely displace coal;84 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 4.2 National Emissions for CO2-Reduction Scenarios Base Case Case 1 – Case 2 – % Reduction % Reduction 10% CO2 20% CO2 from from Reduction Reduction Base Case Base Case for Case 1 for Case 2 CO2 (million metric tons) 2,100 1,890 1,680 SO2 (million tons) 5.66 5.66 5.46 — 4% NOx (million tons) 4.66 3.92 3.16 16% 32% Hg (tons) 48 40 32 17% 33%Source: MIT analysisconversely, surplus NGCC capacity exists and Changes to the dispatch order of generation,can displace some coal capacity in all demand from coal to gas, would lower prices in the SO2periods examined, even during the super peak, market, and might even yield a reduction inalthough the amount is small. national emissions below the CAIR limit, as shown with a 4% change in Case 2. Impor-Effect of System Re-Dispatch tantly, the reductions in NOx and Hg emissionson Criteria Pollutants could be substantial, by as much as one-third under the more stringent CO2 limit.The Clean Air Act (CAA) requires power plantcontrols on SO2, NOx, particulates and Hg. Table 4.3 shows the corresponding emissionsAccording to the EPA, “60% of the uncon- profiles by region for CO2 and Hg. (ReEDS doestrolled power plant units are 31 years or older, not provide adequate regional detail for SO2[some] lack advanced controls for SO2 and and NOx). Each region acts in its own bestNOx, and approximately 100 gigawatts (GW) economic interests under the given constraints.out of total of [more than 300] GW of coal are And, because of variation in generation costs,without SO2 scrubbers.” 11 installed capacity and transmission differences between regions, some regions have compara-Table 4.2 contains results from ReEDS under tive advantage dispatching less CO2 intensivethe three scenarios that indicate the potential generation. Depending on the regulatoryeffects of the CO2 constraint (also shown) on structure, regions with these advantages mayemissions of SO2, NOx and Hg. (The model produce more electricity, export it and/or selldoes not project particulate emissions, which credits (assuming a cap-and-trade approach);also would be reduced.) While ReEDS does not and regions which typically deploy technologiesfully model the trading markets for SO2 and that are more CO2 intensive take oppositeNOx, it makes a reasonable approximation by actions. This leads to uneven emissions effectscapping national emissions levels and making on individual regions.economically efficient dispatch decisions underthese constraints. In all three simulations the A 20% emissions reduction in electric sectorcap for SO2 emissions is based on the 2005 CO2 emissions through coal-to-gas displace-Clean Air Interstate Rule (CAIR) interpolated ment would represent mitigation of 8% of thefor 2012.12 U.S. total. The ReEDS model does not provide Chapter 4: Electricity 85
  • Table 4.3 Emissions of Select Regions Before and After Re-Dispatch, 2012 Base Case MISO ERCOT PJM FRCC ISO-NE CO2 (million metric tons) 543 153 446 67.2 19 Hg (tons) 13.4 2.77 11 1.32 0.138 Case 2 – 20% CO2 Reduction CO2 (million metric tons) 394 121 351 78.9 25.4 Hg (tons) 9.30 1.43 7.58 1.13 0.10 % Hg reduction 31% 48% 31% 14% 27% Source: MIT Analysis There is sufficient surplus NGCC capacity to in delivered coal prices and the decrease in displace roughly one-third of U.S. coal generation, delivered natural gas prices, combined with surplus capacity at highly efficient gas-fired reducing CO2 emissions from the power sector combined-cycle plants resulted in coal-to-gas by 20% and yielding a major contribution to fuel switching. Nationwide, coal-fired electric control of criteria pollutants. This would require power generation declined 11.6 percent from an incremental 4 Tcf per year of natural gas, which 2008 to 2009, bringing coal’s share of the corresponds to a cost of $16 per ton of CO2. electricity power output to 44.5 percent, the lowest level since 1978.” 15 an accurate estimate of the national economic In sum, there is sufficient surplus NGCC cost of this option, but an approximation can capacity to displace roughly one-third of U.S. be made by comparing the break-even CO2 coal generation, reducing CO2 emissions from price at which the cost of NGCC generation the power sector by 20% and yielding a major equals the cost of coal generation, given their contribution to control of criteria pollutants. different variable operations and maintenance This would require an incremental 4 Tcf per costs, heat rates and CO2 emissions rates.13 year of natural gas, which corresponds to a cost The result is an implicit cost of about $16 per of $16 per ton of CO2. Currently there is no ton CO2. national price on CO2, but there are both regional programs and federal regulatory More analysis is required to determine whether, activities underway. because of the geographic differences between NGCC and coal units, some new transmission R E CO M M E N D AT I O N infrastructure may be necessary. Nonetheless, The displacement of coal generation with a more complete analysis is very likely to prove the cost of this option to be low compared to NGCC generation should be pursued as most other mitigation options. For example, the most practical near-term option for one estimate of the per-ton CO2 emissions significantly reducing CO2 emissions from avoidance cost estimate to retrofit a typical power generation. sub-critical coal plant with post-combustion CSS is $74 per ton.14 It should also be noted that coal-to-gas fuel switching is already occurring. According to the Energy Information Agency (EIA), “The increase86 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • INTERMITTENT RENEWABLE ELECTRICITY Effects in the Short TermSOURCES AND NATURAL GAS DEMAND To elucidate the short-term effects, we use:In this section, we explore the impacts of theintroduction of significant amounts of inter-mittent wind and solar electricity generation base case, obtained from the ReEDS CO2on natural gas generation and overall natural Price-Based policy scenario (see Box 4.1); andgas demand.This analysis first explores the short-term effects daily dispatch patterns for ERCOT which, asof intermittent wind and solar generation on noted earlier, is an isolated system that can begas generation and demand, a scenario which studied without the complicating influence ofassumes that the capacity from technologies inter-regional transmission.— other than wind or solar — is fixed. SomeEuropean countries already approximate this With this 2030 generation portfolio as oursituation, where substantial volumes of wind reference point, we examine the daily dispatchor solar generation have been installed during patterns of all generation technologies, includ-the last few years. Also in some U.S. states, the ing natural gas, when greater or lesser levels ofproportion of intermittent generation exceeds wind or solar electricity generation are made10% and the dispatch of existing conventional available to be dispatched and the capacitiesgeneration units has had to adjust accordingly. of the other technologies are held constant.We then turn to longer-term impacts, where Wind generation. The results for varying levelsthe deployment of intermittent generation of wind generation are seen in:is assumed to take place gradually, possiblyin response to government policies that, for represen-example, set a mandatory target for renewable tative day for ERCOT;generation. Over time, capacity additions andretirements of other technologies are made as half thethe system adjusts to intermittent generation. amount of generation as in the base case; and twice the amount of generation as in the base case. Chapter 4: Electricity 87
  • Figure 4.8 Impact of Wind on a One-Day Dispatch Pattern for ERCOT 4.8a Wind Base Case 60,000 Hydro Solar 50,000 Natural Gas – Gas Turbine Natural Gas – Production (MW) 40,000 Combined Cycle Integrated 30,000 Gasification Combined Cycle with CCS 20,000 Biomass Coal 10,000 Wind Nuclear 0 h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 4.8b Wind 0.5 60,000 50,000 Production (MW) 40,000 30,000 20,000 Hydro 10,000 Solar Gas GT 0 Gas CCGT Gas CCGT CCS h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 Biomass Coal 4.8c Wind 2.0 Wind 60,000 Nuclear 50,000 Production (MW) 40,000 30,000 20,000 10,000 0 h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 Source: MIT Analysis88 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • In Figure 4.8a, the base case depicts the esti- The pattern with solar is somewhat differentmated existing contribution from wind in than for wind. The solar generation outputERCOT in 2030. The nighttime load (roughly basically coincides with the period of highhours 01 through 04) is met by nuclear and demand, roughly between hours 06 and 22.coal baseload plus wind generation. There is As seen in the base case Figure 4.9a, this is alsono appreciable output from gas between hours when NGCC capacity gets dispatched. The01 and 04 because it has higher variable costs natural gas plants are used more when solarthan nuclear and coal and it gets dispatched output is less (see Figure 4.9b). Conversely,last. Natural gas also has the flexibility to cycle. when solar is used more, less gas is dispatchedIn hours 05 through 23, when overall demand (see Figure 4.9c).increases during the early morning anddecreases in the late evening, NGCC generation The baseload plants are largely unaffected andadjusts to match the differences in demand. cycling is not a problem for them, since there is no intermittent solar-based generation duringAs depicted in Panel 4.8b, when less wind the low-demand night dispatched, the NGCC capacity is morefully employed to meet the demand, and the In sum, our short-term analysis shows that thecycling of these plants is significantly reduced. most significant impacts of a quick deploymentThe baseload plants continue to generate at of additional wind or solar at any given futurefull capacity. year will most likely be both a reduction in production from, and an increase in cycling of,In Panel 4.8c with twice as much wind as the gas-fueled NGCC plants; there is a less significantbase case, natural gas generation is reduced fall in production for the much-less-employed,significantly; the gas capacity that is actually single-cycle gas turbines and steam gas units.used is forced to cycle completely. Baseloadcoal plants are also forced to cycle because [In the short term]….the most significant impactsof the relatively low nighttime demand; coal of a quick deployment of additional wind or solar …plant cycling can increase CO2, SO2 and NOxemissions.16 will most likely be both a reduction in production from, and an increase in cycling of, gas-fueledSolar Generation. Like wind, for solar there NGCC plants….are figures depicting: a base case in ERCOT(Figure 4.9a); a case where solar provides half The displacement of gas is greater for solar thanthe amount of generation as the base case for wind, since solar production has a stronger(Figure 4.9b); and a case where solar provides correlation with demand than does windtwice the generation seen in the base case generation.(Figure 4.9c). Large wind penetrations may also displace some coal production and result in some cycling of these plants. No impact on nuclear production is expected with the average U.S. technology mix. Chapter 4: Electricity 89
  • Figure 4.9 Impact of Solar CSP (no storage) on One-day Dispatch Pattern for ERCOT 4.9a Solar Base Case 60,000 Solar Hydro 50,000 Biomass Gas – GT Production (MW) 40,000 Gas – CCGT CCS Gas – CCGT 30,000 Coal – IGCC CCS Coal – Old Biomass 20,000 Coal – Old NoScrub Wind 10,000 Nuclear 0 h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 4.9b Solar Base Case x 0.5 60,000 Solar Hydro 50,000 Biomass Gas – GT Production (MW) 40,000 Gas – CCGT CCS Gas – CCGT 30,000 Coal – IGCC CCS Coal – Old Biomass 20,000 Coal – Old NoScrub Hydro Wind 10,000 Solar Nuclear Gas GT 0 Gas CCGT h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 Gas CCGT CCS Biomass Coal 4.9c Solar Base Case x 2.0 Wind 60,000 Nuclear Solar Hydro 50,000 Biomass Gas – GT Production (MW) 40,000 Gas – CCGT CCS Gas – CCGT 30,000 Coal – IGCC CCS Coal – Old Biomass 20,000 Coal – Old NoScrub Hydro Wind 10,000 Solar Nuclear Gas GT 0 Gas CCGT h01 h02 h03 h04 h05 h06 h07 h08 h09 h10 h11 h12 h13 h14 h15 h16 h17 h18 h19 h20 h21 h22 h23 h24 Gas CCGT CCS Biomass Source: MIT Analysis Coal Wind Nuclear90 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Effects in the Long Term This combination of wind production and flexible generation capacity competes withTo explore the effects of the penetration of potential new nuclear capacity and also erodesintermittent generation over the long term, NGCC production. Wind impacts the preferredwe examine two policy scenarios, both with a new baseload generation technology, the onesystem expansion to 2050 and a target leading that is most economic but for which expansionto a 70% reduction of CO2 emissions in the is not subject to environmental or other limits.U.S. power sector. Late in the period, conventional coal production has been replaced, economically-competitiveWe look at two different versions of the 70% wind resources start becoming exhausted andreduction case because the means by which nuclear plus some solar penetration begins.the target is implemented — through differentmitigation policy instruments — has an effect CO2 Price-Based Case. In the CO2 Price-Basedon how the system responds to more or less case, the nature of the system adjustments inexpensive renewable generation. The two policy these simulations can be illustrated using aninstruments we examine are: example of the changes that would be brought about by lower-cost wind capacity. First, the CO2 price to achieve increased intermittent renewable generation the CO2 emissions reduction target; and needs to be accompanied by flexible back-up capacity, albeit with low utilization levels. In emissions constraint the U.S., spare capacity of gas-fueled plants is to achieve the same target. enough to meet this requirement initially, but eventually additional investment is neededWe then analyze how the electric system, and (gas turbines in these scenarios).gas use over time, would differ if the capitalcosts of solar or wind generation capacity were As this combination of new intermittent renew-higher or lower than the reference levels for the able and flexible electricity plants grows, ittwo base cases. Again the ReEDS model is starts to replace the expansion and utilizationemployed.17 of baseload generation technologies, nuclear or fossil generation with CCS (coal withoutIn the ReEDs simulations of both policy CCS has already been forced out of the systemscenarios, the generation mix evolves over time, by its CO2 emissions). However, these classicsimilar to that shown in Chapter 3, Figures 3.4a baseload technologies are not increasing;and 3.4b. During the early-to-middle decades therefore, the low-cost renewable capacity plusof the simulation period the dominant event is flexible generation increases in baseload andthe substitution of coal generation by NGCC even in mid-merit service, at the expense ofunits. At the same time, wind generators, with gas generation.gas turbine back-up, begin to be deployed asa baseload technology.18 Chapter 4: Electricity 91
  • This interaction can be illustrated with a As Figure 4.10 shows, increased cumulative summary of what happens in the base case wind generation, as a consequence of lower system for the ERCOT region when different wind investment costs, or an aggressive renew- renewable costs are simulated, therefore able portfolio standard, has a direct impact on changing the intermittent generation pene- the new investment and associated production tration levels. by natural gas, equal to almost one TWh of reduced natural gas generation for one TWh The results in Figure 4.10 are plotted to high- of wind output. This happens because NGCC light the way cumulative gas generation is the technology that is most vulnerable to changes with different assumptions about wind competition, both before and after coal wind-generation costs and the corresponding has been driven out of the market. It should wind-generation levels. The figure shows the also be noted that, while the cumulative total generation in TWh by type of generation generation of gas turbines (Gas-CT in Figure technology over the simulation period from 4.10) does not change enough to show in the 2005 to 2050 and assumes the underlying graph, gas turbine capacity actually increases emissions target is imposed by a CO2 price. substantially to support the additional wind It illustrates that the displacement of gas by contribution. wind takes place through changed patterns of investment and generation over many years. Figure 4.10 Cumulative Generation in ERCOT in the Period 2005–2050 for All Technologies Given Alternative Levels of Wind Penetration (TWh) 5,000 Wind (Gen) 4,500 Nuclear (Gen) Coal Old Scrubbed 4,000 Coal Old Unscrubbed 3,500 Oil-Gas Steam 3,000 Gas-CC TWh 2,500 Gas-CT 2,000 Hydro Utility PV 1,500 1,000 500 0 250 450 650 850 1,050 1,250 1,450 1,650 Source: MIT Analysis92 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • In Figure 4.10, the horizontal axis is cumulative generation does not require back-up fromwind output, the vertical axis is the cumulative flexible gas plants as much as wind does. In fact,output for all technologies, including wind solar can partially fulfill a peaking plant role.(if the two axes were plotted to the same scalethe function for wind would be a 45˚ line). The In summary, our analysis of gradual andbase-case level of wind generation is indicated sustained “long term” penetration of wind andwith a vertical line, so that output to the right solar shows that large-scale penetration of windof that point results from lower capital costs generation, when associated to flexible naturaland the output to the left results from higher gas plants, will assume a mostly baseload role,capital costs. and will reduce the need for other competing technologies such as nuclear, coal or evenFigure 4.10 also shows that the difference in gas-fueled combined cycles, if expansion withcumulative generation by the other technolo- coal and nuclear technologies does not take placegies is not much affected by changes in thecontribution of wind generation. It should be Our analysis shows that a gradual and sustainedrepeated that this is a result for ERCOT. The “long term” substantial penetration of wind, whendifferences in generation mix in other regions associated with flexible natural gas plants, willwill vary, though viewed at the national level assume a mostly baseload role, and will reduce the needthe pattern is very similar to that shown here.19 for other competing technologies such as nuclear, coalCO2 Cap Case. The result differs somewhat if or even gas-fueled combined cycles. This effect is lessemissions mitigation is accomplished by a CO2 pronounced in the case of solar.cap instead of a price. The fixed CO2 constraintimplies that an increment in wind output that because of economic, environmental or anydisplaces NGCC production and investment other reasons. This effect is less pronounced inalso reduces the need for other low-CO2 the case of the solar technology, because of itsbaseload capacity to reduce the emissions. characteristic daily production pattern.Cheaper wind creates slack under the emissions Although our analysis has been limited to a fewconstraint, which may be filled by whatever is alternative scenarios, we can observe a consis-the cheapest generation source. In some tent pattern for the impact of intermittentsimulations, this cheap generation comes from renewable generation: We see that an increaseotherwise almost-idle coal-fired plants. Thus, of wind or solar output systematically resultsas a minor perverse effect, under the CO2 in a proportionally significant reduction ofconstraint more wind can imply a small natural gas fueled production, while, at theincrement of additional coal production — same time, the total installed capacity of flexiblea condition that does not occur when coal generation (typically also natural gas fueledis burdened by a CO2 price. plants) is maintained or increased.The case of solar generation without storage Precise numerical estimations and any secondis similar to wind in many respects. However, order impacts are heavily dependent on thesince the production profile of solar has a high specific energy policy instruments and thelevel of coincidence with the daily demand and assumptions on the future costs of fuels andhas a more stable pattern, an increment in solar technologies. Chapter 4: Electricity 93
  • The detailed operational analysis of plausible R E CO M M E N D AT I O N future scenarios with large presence of wind In the event of a significant penetration of and solar generation reveals the increased need intermittent renewable production in the for natural gas capacity (notable for its cycling generation technology mix, policy and capability and lower capital cost) to provide regulatory measures should be developed reserve capacity margins. This does not how- ever necessarily translate into a sizeable utiliza- to facilitate adequate levels of investment tion of these gas plants. in natural gas generation capacity to ensure system reliability and efficiency. Additional Implications Although limited in scope, our analysis shows In deregulated wholesale markets with substan- the diversity and complexity of the impacts that tial penetration of renewables, the volatility of a significant penetration of intermittent marginal prices can be expected to increase. generation (mostly wind and solar, in practice) Also, mid-range technologies, of which NGCC have on the technology mix and the operation is the most likely candidate, will see their of any considered power system. The possible output reduced. The uncertainty regarding the future emergence of electricity storage options, adequate technology mix, and the economics of as well as enhanced demand responsiveness, such a mix under the anticipated future prices will also affect the need for flexible generation and operating conditions, raises concern about capacity, which is presently fueled by natural attracting sufficient investment in gas-fueled gas. The level and volatility of future energy plants under a competitive market regime. prices will determine the volume and nature of investment in future generation under market This issue is presently being addressed by conditions. Other regulatory frameworks several European countries with significant should also be considered. penetration of wind generation, where the patterns of production of NGCC and single These complicated implications and trade-offs cycle gas turbines and also of some baseload cannot be spelled out without the help of technologies, have already had major impacts. suitable computer models. The accuracy of Similar situations are developing in some parts the estimates of future fuel utilization and the of the U.S. Presently there is no consensus on adequate technology mix critically depends a suitable regulatory response to this situation, on the performance of these models. Unfor- which could include enhancements of any tunately, the state-of-the-art computer models capacity mechanisms such as those already that simulate and optimize the capacity expan- in place in most U.S. wholesale markets, new sion and the operation of power systems and categories of remunerated ancillary services electricity markets — such as ReEDS or Mem- or other instruments. phis — are still in a development phase and fall short of the requirements to incorporate intermittent generation, storage and demand response realistically, under a variety of energy policies and regulatory environments.94 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • R E CO M M E N D AT I O NA comprehensive appraisal of the economic, environmental and reliabilityimplications of different levels of significant penetration of renewablegeneration should be performed for power systems with different generationtechnology portfolios and under different energy policy scenarios.The information obtained from this appraisal should inform a central piecein the design of energy policies that contemplate mandating large amountsof solar or wind generation.Additional efforts should be made to expand or develop the sophisticatedcomputation electric system models that are needed for this task. Chapter 4: Electricity 95
  • NOTES 10 Although the trend for NGCC displacement of coal generation remains the same for this updated 1 Nameplate capacity is the nominal, maximum scenario, these results are numerically different instantaneous output of a power plant. than the results presented in the interim report. The interim report showed opportunities for coal 2 Absent other considerations, generation units are displacement in all time periods. The difference normally dispatched in economic merit order, i.e., stems from assumptions about how much NGCC those with lower variable operating costs first. capacity exists in ERCOT. The NGCC capacity 3 Channele Wirmin, EIA, private communication. numbers used for this 2012 simulation are more conservative, and projected forward from 2006 4 Steinhurst, W., The Electric Industry at a Glance, EIA capacity and generation data (2006 is the start Nuclear Regulatory Research Institute, 2009. year for ReEDS). 5 Electric Power Research Institute, A Primer on 11 Presentation, “Reducing Pollution from Power Electric Power Flow for Economists and Utility Plants,” Gina McCarthy, Assistant Administrator, Planners, 1995; Pérez-Arriaga, I., Rudnick H., U.S. EPA Office of Air and Radiation, October 29, Rivier, M., Chapter One: Electric Energy Systems, 2010. An Overview. 12 For a variety of reasons, deployment of required 6 Hydroelectric generation, shown in Figure 4.2 controls has been delayed, largely by court findings as constant over demand periods, will in fact tend of legal flaws in various rulemakings. The New to be concentrated in particular seasons and peak Transport Rule, which will replace Clean Air periods of the day. The MARKAL model does not Interstate Rule (CAIR) in place today, is expected represent this detail, though its inclusion would to be finalized in mid-2011 and will be imple- have only a small effect on the figure as it aggre- mented over time, with most coverage finalized gates all the national hydroelectric facilities. by 2014. The Transport Rule will cover SO2 and NOx. EPA released a proposed rule for mercury 7 The same change in unit dispatch could be emissions from coal and oil-fired power plants in approached using various forms of direct March, 2011 and plans to finalize the rule by the regulation, options not studied here. end of the year. A final rule on CO2 for power 8 The ReEDS model captures key characteristics of plants is expected sometime in 2012. the electricity network’s transmission constraints 13 This break-even price assumes a NGCC and reliability requirements by splitting the country variable O&M cost of $3.20/MWh, fuel price into 134 geographic partitions. Each partition of $5.38/mmBtu, heat rate of 6.04 mmBtu/MWh, balances demand and supply of electricity by and CO2 emissions of 0.053 tons/mmBtu. For coal, independently generating, importing, and the calculation assumes a variable O&M cost of exporting electricity. Collectively, subsets of these $4.30/MWh, fuel price of $2.09/mmBtu, heat balancing areas constitute the independent system rate of 10 mmBtu/MWh, and CO2 emissions of operators (ISOs) and regional transmission 0.098 tons/mmBtu. The cost of NGCC and coal organizations (RTOs). generation break-even when the sum of the variable 9 As noted in the introduction of this section, the O&M cost and price per ton CO2 multiplied by the expected maximum capacity factor for an NGCC amount of CO2 emitted are equal to each other, for plant is 85%. The EIA projects that this could the respective fuels. Start-up and shut-down costs, increase to 87% by 2016 ( ramp rates, associated changes in emissions, and oiaf/aeo/pdf/2016levelized_costs_aeo2010.pdf). other costs that have not been fully modeled are The average fleet capacity factor of 87% from not included in this calculation. ReEDS for the 20% CO2 reduction scenario 14 MIT Energy Initiative’s report on Retrofitting approaches the upper generation threshold of the of Coal-Fired Power Plants for CO2 Emissions country’s current NGCC fleet. Reductions, Cambridge, MA, 2009 ( edu/mitei/docs/reports/meeting-report.pdf). 15 EIA AEO 2010.96 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 16 See Bentek study, How Less Became More: Wind, Power and Unintended Consequences In the Colorado Energy Market, April 2010.17 See “Impact of intermittent renewable electricity generation on the technology mix and fuel consumption in the U.S. power system.” Yuan Yao, Ignacio J. Pérez-Arriaga. CEEPR (Center for Energy and Environmental Policy Research), MIT, May 2011.18 The ReEDS simulations of this level of mitigation show a greater penetration of renewable genera- tion than do the results of the EPPA model shown in Chapter 3, but the difference is not an impor- tant influence on the insights to be drawn from these calculations.19 Details of these cases are provided by Yao and Pérez-Arriaga, op cit. Chapter 4: Electricity 97
  • Chapter 5: DemandINTRODUCTIONNatural gas is attractive for a variety of end-useapplications because it is:Figure 5.1 Natural Gas End-Use Markets (2009) Residential/Commercial 3.5% 8.9% 46.5% Transportation 34.5% 40.8% 0.3% Total Natural Gas 2.9% 93.9% 2.6% Supply 3.4% (22.8 TCF) Industrial 0.1% 32.4% 34.9% 35.7% Coal 30.2% Natural Gas Petroleum Electric Power 6.3% 9.3% Renewable 13.8% Generation Electricity Chapter 5: Demand 99
  • Figure 5.2 Historical Trends in End-Use Consumption 12 Residential/Commercial 10 Industrial Transportation Consumption (Tcf) 8 Electric Power 6 4 2 0 1970 1975 1980 1985 1990 1995 2000 2005 Annual Energy Review 2009 CHAPTER OVERVIEW - energy services to buildings and institutions100 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • NATURAL GAS IN THE INDUSTRIALSECTOR 2Figure 5.3 Natural Gas Use by U.S. Manufacturing Industry Sector “All Other” includes: Food 11% All Other 19% Paper 8% Primary Petroleum Metals & Coal 11% Products 14% Nonmetallic Mineral Products 8% Chemicals 29% Chapter 5: Demand 101
  • Figure 5.4 Trends in U.S. Industrial Natural Gas Consumption and Intensity 10 2.5 9 8 2.0 Industrial Consumption (Tcf) and Natural Gas Intensity (MMcf/$) Industrial Price ($/MMBtu) 7 6 1.5 5 4 1.0 3 NG Consumption (Tcf) 2 0.5 NG Industrial Price ($/MM) 1 NG Intensity (MMcf/$) 0 0.0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Natural Gas Consumption and Efficiency Trends - quantity of natural gas used per dollar value of rising prices provide incentives for increased102 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 2 Principal Uses of Natural Gas as a Fuel and as a Chemical Feedstock - 3 3 - 2 - - Figure 5.5 U.S. Manufacturing Natural Gas Use by End-Use Application Feedstock 7% Other/Not Reported 7% Machine Drive 2% Conventional Boiler Use HVAC 22% 6% CHP/Cogen 14% Process Heating 42% Chapter 5: Demand 103
  • Natural Gas Use in Industrial Boilers 7 - - - relatively large for natural gas boilers but - - Modernization of the Natural Gas Industrial Boiler Fleet - 9 -104 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Replacement of Existing Coal Industrial Boilers with Efficient Natural Gas Boilers 2 - - - 2FINDINGReplacement of existing industrial naturalgas boilers with higher efficiency modelscould cost-effectively reduce natural gasdemand and reduce GHG emissions.R E CO M M E N D AT I O NThe DOE should review the current energyefficiency standards for commercial andindustrial natural gas boilers and assessthe feasibility of setting a more stringentstandard. - Chapter 5: Demand 105
  • - 2 - 2 2 reduction FINDING Replacement of existing industrial coal boilers and process heaters with new, efficient natural gas boilers could be a - cost-effective alternative for compliance with the EPA MACT Standards. Fuel switching has the potential to increase demand for natural gas while achieving substantial CO2 emissions reductions at a modest incremental cost.106 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Natural Gas Use in Manufacturing ProcessHeating - on industry - 2 - - Chapter 5: Demand 107
  • - - FINDING The potential for significant reductions in the use of natural gas for process heating lies in a shift to new manufacturing process technologies that use less energy-intensive processes and materials. RD&D Opportunities in Energy-Efficient - Technologies - 2 Additional opportunities for advances in - -108 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • FINDINGIndustrial energy efficiency RD&D programssupported by the DOE have historicallyled to significant improvements in energy-efficient technologies: technologies thatalso achieved significant reductions in CO2emissions while improving the economiccompetitiveness of manufacturing.R E CO M M E N D AT I O NThe DOE should continue to play a rolein accelerating the development of newtechnologies that can improve energy -efficiency. The DOE should again serveas a convener of industry technologyworking groups to develop roadmapsfor future energy-efficiency technologyimprovements. Based on these roadmaps,the DOE should develop a federally funded because it is easier to balance electricity genera-RD&D portfolio consisting of applied pre-competitive R&D as well as transformationalapproaches. The DOE RD&D portfolioshould encompass both industry-specifictechnologies in energy-intensive industriesand crosscutting technologies applicableacross a broad spectrum of manufacturingindustries.CHP Systems for Industrial Applications - - - Chapter 5: Demand 109
  • 23 2 - - -22 Chapter 5: Demand 111
  • Figure 5.6 Hourly Energy Load Profile for One Type of Residential Customer in New England – Sample forone day during Winter and Summer Winter Summer 25 25 4.0 4.0 Energy load [kWh per hour] Energy load [kWh per hour] 20 20 3.0 3.0 15 15 2.0 2.0 10 10 1.0 1.0 5 5 0 0 0.0 0.0 1 3 5 7 9 11 13 15 17 19 21 23 1 3 5 7 9 11 13 15 17 19 21 23 Hour Hour Electric load [kW-e] Heat load [kW-th] Electric load [kW-e] Heat load [kW-th] - FINDING Matching heat and power loads for residential application varied greatly depending residential and other small-scale applications poses a significant challenge to the feasibility of small-scale CHP systems - based on current technologies. NATURAL GAS DEMAND IN BUILDINGS - 27112 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • - -Comparing the Efficiency of SpaceConditioning and Hot Water Technologies 2Figure 5.7 2006 Breakdown of Building Energy Consumption in the Residential andCommercial Sectors 2006: Total Quads – Residential Sector [% of Sector Energy Consumption using 6 Total Electricity Primary Energy] (% of Sector Energy Consumption using 5 only Electricity Sales (e.g., Site Energy)) 4 [13.0%] [12.4%] [11.6%] [9.1%] [27.4%] Electricity (Balance of Primary Energy) (07.9%) (15.5%) (07.1%) (5.6%) (19.6%) Electricity (Retail Sales) 3 [26.5%] Renewable Fuels 2 (44.4%) LPG 1 Oil Natural Gas 0 Space Space Water Lighting Electronics Appliances Heating Cooling Heating 2006: Total Quads – Commercial Sector [% of Sector Energy Consumption using 6 Total Electricity Primary Energy] [26.1%] [15.8%] [15.9%] [6.2%] [24.8%] [11.2%] (31.7%) (% of Sector Energy Consumption using 5 (23.4%) (11.0%) (9.4%) (16.9%) (07.6%) only Electricity Sales (e.g., Site Energy)) 4 Electricity (Balance of Primary Energy) Electricity (Retail Sales) 3 Renewable Fuels 2 LPG 1 Oil Natural Gas 0 Space Space Water Lighting Electronics Other Heating Cooling Heating Equipment Buildings Energy Data Book Chapter 5: Demand 113
  • - 29 - - particular electricity versus natural gas and -114 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 5.1 Site vs. Source Energy Efficiency of Residential Heating, Cooling and Hot Water Systems Site Energy Efficiency (SCOP*) Source- Full-Fuel-Cycle Efficiency (FFC) Low Energy Best to-Site Low Energy Best Star Efficiency StarHeating System Type Electric Furnaces 0.95 — 0.99 0.32 0.31 — 0.32 Oil-Fired Furnaces 0.78 0.83 0.95 0.88 0.69 0.73 0.84 Gas-Fired Furnaces 0.78 0.90 0.98 0.92 0.72 0.83 0.90 Air Source Heat Pumps † 2.30 2.40 5.20 0.32 0.74 0.77 1.67 Ground Source Heat Pumps ‡ 2.50 3.30 4.80 0.32 0.80 1.06 1.54Cooling System Type Central AC † 3.81 4.25 6.74 0.32 1.22 1.37 2.17 Air Source Heat Pumps † 3.81 4.25 4.98 0.32 1.22 1.37 1.60 Ground Source Heat Pumps ‡ 2.55 4.13 6.57 0.32 0.82 1.33 2.11Hot Water System Type Electric Storage Tank 0.92 — 0.95 0.32 0.30 — 0.31 Oil-Fired Storage Tank 0.51 — 0.68 0.88 0.45 — 0.60 Gas-Fired Storage Tank 0.59 0.62 0.70 0.92 0.54 0.57 0.64 Electric Heat Pump Tank 0.92 2.00 2.35 0.32 0.30 0.64 0.76 Electric Instantaneous 0.93 — 0.99 0.32 0.30 — 0.32 Gas-Fired Instantaneous 0.54 0.82 0.94 0.92 0.50 0.75 0.87 † ‡ - - Chapter 5: Demand 115
  • Figure 5.8 Combined Source to Site Energy Efficiencies for Delivering Coal and Natural Gas-Fired Generation Versus Oil and Natural Gas to End-Use Customers Source to Site — Coal-Fired Generation Source to Site — Natural Gas-Fired Generation 100 100 98 97 96 97 94 93 New Gas 47 44 New Coal 36 33 Avg 39 37 Gas Avg 31 29 Coal Extraction Processing Transport Generation T&D Extraction Processing Transport Generation T&D Source to Site — Home Heating Oil Source to Site — Direct-Use Natural Gas 100 100 96 97 94 93 92 90 89 89 No No Generation Generation Extraction Processing Transport Equivalent Distribution Extraction Processing Transport Equivalent Distribution FINDING Source-to-site energy losses should be considered when choosing among energy options, especially ones that use different energy carriers. - effectiveness of direct fuel and electricity end - standards for appliances and space condition-116 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • R E CO M M E N D AT I O NImproved energy efficiency metrics thatprovide an FFC comparison of energyefficiency should be incorporated intonational standard setting activities.The improved metrics should includeboth FFC efficiency and cost-to-consumerfactors. -Looking Beyond Equipment EfficiencyStandards - FINDING - Energy efficiency metrics alone are not sufficient to inform consumers about the most energy efficient and cost-effective options for meeting household energy needs in different regions. - R E CO M M E N D AT I O N In addition to improved efficiency metrics for comparing appliances and building energy technologies, there is a need to inform consumers and developers as well as state and local regulators about the cost-effectiveness and suitability of various technologies, relative to local conditions. Chapter 5: Demand 117
  • FFC Efficiency and CO2 Emissions - both content of retail fuels is reasonably consistent 2 - 2 2 - 2 2 content of electricity 2 2Table 5.2 Retail Electricity and Fuel — Source-to-Site Efficiencies and CO2 EmissionsRegional Site-to-Source and CO2 Source-to- CO2 Emissions (lb CO2/MWh) ∆% fromEmissions Factors by NERC Region (2005) Site Efficiency Precomb. Generation T&D Combined US Avg. United States Average US 0.32 54 1,329 86 1,469 – Midwest Reliability Organization MRO 0.28 55 1,824 120 1,999 36.1 Southwest Power Pool SPP 0.30 63 1,751 114 1929 31.3 Reliability First Corporation RFC 0.31 38 1,427 94 1,559 6.1 Florida Reliability Coordinating Council FRCC 0.33 99 1,319 91 1,508 2.6 SERC Reliability Corporation SERC 0.31 45 1,369 90 1,504 2.4 Texas Regional Entity TRE 0.32 74 1,324 87 1,485 1.1 Western Electricity Coordinating Council WECC 0.38 51 1,033 57 1,142 -22.3 Northeast Power Coordinating Council NPCC 0.33 85 876 61 1,022 -30.4 Primary Residential Fuels Precomb. Distribution Combustion Combined Distillate Oil US 0.89 107 4 550 661 -55.0 Liquid Petroleum Gas US 0.89 74 4 476 553 -62.3 Natural Gas US 0.92 36 5 404 444 -69.8118 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 5.3 Combined Energy and Emissions Impacts of Using FFC Efficiency for Select NERC Regionsfor Energy Star Appliances Energy Consumption (MWh) Full Fuel Cycle CO2 Emissions (Ton CO2) Useful Site FFC National MRO NPCC SPP TRE Heating System Type Electric Furnaces 100 101.0 314.2 74 114 49 104 75 Oil-Fired Furnaces 100 120.5 136.7 45 Gas-Fired Furnaces 100 111.1 120.7 27 Air Source Heat Pumps † 100 41.7 129.6 31 47 20 43 31 Ground Source Heat Pumps ‡ 100 30.3 94.3 22 34 15 31 23 Cooling System Type Central AC† 100 23.5 73.2 17 27 11 24 17 Air Source Heat Pumps † 100 23.5 73.2 17 27 11 24 17 Ground Source Heat Pumps ‡ 100 24.2 75.3 18 27 12 25 18 Hot Water System Type Electric Storage Tank 100 105.3 327.4 77 119 51 109 78 Oil-Fired Storage Tank 100 147.1 166.9 55 Gas-Fired Storage Tank 100 161.3 175.2 39 Electric Heat Pump Tank 100 50.0 155.5 37 56 24 52 37 Electric Instantaneous 100 101.0 314.2 74 114 49 104 75 Gas-Fired Instantaneous 100 122.0 132.5 29 † ‡ 2 - - 2 2 2 - 2 - 2 2 2 Chapter 5: Demand 119
  • - 2 - FINDING - Use of equipment-based FFC efficiency and national average energy demand and CO2 emissions metrics alone are not sufficient to inform policy makers and consumers of the comparative cost and environmental benefits of competing appliances and building energy systems. R E CO M M E N D AT I O N DEMAND FOR NATURAL GAS AS A TRANSPORTATION FUEL More detailed and targeted approaches are needed to develop combined cost- - and emissions-effective strategies for meeting future energy and emissions goals on a local and regional basis. State 2 and Federal agencies should collaborate with the building industry and equipment - manufacturers to provide clear and accurate information to consumers. 2 burdens of different fuels including regional differences 2 Natural gas is garnering attention for its - sions-effectiveness of space conditioning and 2120 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • - CNG-Powered Vehicleslarge-scale liquid fuel infrastructure in place 2Global Natural Gas Vehicle Market 32 - 2 - 33 - Chapter 5: Demand 121
  • - FINDING At present gasoline-CNG fuel price spreads, U.S. heavy-duty vehicles used for short- - range operation (buses, garbage trucks, delivery trucks) have attractive payback times (around three years or less). Payback times for U.S. light-duty vehicles are attractive provided they are used in high-mileage operation (generally in fleets) and have a sufficiently low incremental cost — a representative number is around $5,000 for a payback time of three years or less. This condition is presently not met. FINDING Experience in other countries indicates the potential for substantial reduction of incremental costs for U.S. factory and aftermarket converted CNG vehicles. Table 5.4 Illustrative Payback Times in Years for CNG Light-Duty Vehicles for Average and High Mileage Use, Low and High Incremental Vehicle Cost and Fuel Price Spread between Gasoline and CNG on a Gallon of Gasoline Equivalent (gge) Basis. Assumes 30 miles per gallon. 12,000 mile per year 35,000 miles per year Incremental Cost $3,000 $10,000 $3,000 $10,000 Spread Price Fuel $0.50/gge 15 50 5.2 17 $1.50/gge 5 17 1.8 5.9122 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • 2 value of natural gas in reducing oil dependence R E CO M M E N D AT I O N The U.S. should consider revision to its current policies related to CNG vehicles, including how aftermarket CNG conversions are certified, with a view to reducing upfront costs and facilitating bi-fuel CNG- gasoline capability. LNG-Powered Long-Haul Trucks - 22 - Chapter 5: Demand 123
  • - LNG- - - - - - - - tional issues related to cryogenic fuel storage in - FINDING The deployment of LNG-powered, long- haul trucks presently faces operational limitations due to the use of onboard fuel storage at very low temperature (-162 C˚); the need for a new fueling infrastructure that ensures competitive pricing; a high incremental cost; and a likely lower resale value particularly in the important international market. These challenges are mitigated by use in the relatively modest - market of hub-to-hub transport.124 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Conversion to Liquid Fuels Our detailed analysis is for natural gas conver- - -liquid fuels depend on several criteria involving 39 energy loss during conversion of natural gas to 2 production cost of natural gas conversion to - -Figure 5.9 Conversion of Natural Gas to Liquid Fuels Natural Gas to Liquid Fuels Natural Gas Reformer Synthesis Gas Catalyst Methanol Mixed Alcohols Diesel DME Ethanol Gasoline Chapter 5: Demand 125
  • - - FINDING With deployment of plants using current technology, on an energy-equivalent basis, methanol could be produced from U.S. natural gas at a lower cost than gasoline at current oil prices. Table 5.5 Illustrative Methanol Production Costs, Relative to Gasoline (excluding taxes) at $2.30 per Gallon Natural Gas Price Methanol Production Cost Reduction Relative Cost, per gge to Gasoline, per gge $4/MMBtu $1.30 $1.00 $6/MMBtu $1.60 $0.70 $8/MMBtu $2.00 $0.30126 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • R E CO M M E N D AT I O N The U.S. government should implement an open fuel standard that requires automobile manufacturers to provide tri-flex-fuel operation in light-duty vehicles. It should also consider methanol fueling infrastructure subsidies similar to thoseFINDING given to the fueling infrastructure forMethanol could be used in tri-flex-fuel light- ethanol.duty vehicles with a modest incrementalvehicle cost (likely to be $100 to $200 morethan an ethanol-gasoline flex-fuel vehicle).It could also be used to power long-haultrucks in mixtures with gasoline, and couldprovide both vehicle and fuel cost savings.Barriers to methanol use include the lackof incentives for vehicle conversion andprovision of distribution infrastructure. An illus-and increase opportunities for reducing oil Chapter 5: Demand 127
  • R E CO M M E N D AT I O N The U.S. government should carry out a transparent comparative study of natural gas derived diesel, gasoline and methanol, and possibly natural gas derived ethanol, - mixed alcohol and DME, with each other and with oil-derived fuels and biofuels. The study should include cost analysis, vehicle requirements, infrastructure requirements and health and environmental issues. It also should include discussion of R&D needs for FINDING more efficient and lower-cost production. If the present oil to natural gas price spread is sustained, there will be materially increased opportunities for use of natural gas-based transportation fuels. FINDING The potential for natural gas to reduce oil dependence could be increased by conversion into room temperature liquid fuels that can be stored at atmospheric pressure. Of these fuels, methanol is the only one that has been produced for a long period at large industrial scale. Methanol has the lowest cost and lowest GHG emissions, but requires some infrastructure modification and faces substantial acceptance challenges. Natural gas derived gasoline and diesel have the advantage of being drop-in fuels, but carry a higher conversion cost.128 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • NOTES2 223 2379 27 29 32 Chapter 5: Demand 129
  • Chapter 6: InfrastructureIn the United States, the availability, reliability This chapter will describe and discuss:and price of natural gas are inextricably linkedto its production and delivery infrastructure.As seen in Figure 6.1, major components of the gas infrastructure;system include inter-state and intra-state trans-mission pipelines, storage facilities, liquefiednatural gas (LNG) regasification terminals and the natural gas infrastructure, with a focus ongas processing units, all of which establish the pipelines, LNG import terminals, processinglink between gas producers and consumers. and storage;This system is both mature and robust. affecting the natural gas infrastructure; and with the development of the Marcellus shale.Figure 6.1 Schematic of the U.S. Natural Gas InfrastructureImage modified from CHK Chapter 6: Infrastructure 131
  • TRENDS AFFECTING U.S. NATURAL GAS Population growth has been especially pro- INFRASTRUCTURE nounced in the Western U.S., where the popula- tion increased by 42% between 1980 and 2008. Several trends are altering the landscape of U.S. This growth, coupled with stricter air quality gas markets with implications for infrastructure regulations, has led to increased demand for gas needs and requirements. These include: chang- in the West, where gas consumption has ing production profiles; shifts in demand/ outpaced population growth, increasing by consumption patterns; and the growth of 68% in the last three decades. In the Northeast, LNG markets. environmental concerns and a shift away from oil in power generation and home heating has Changing Production Profiles led to increased gas consumption; between 1980 and 2008 the population in the Northeast As described in Chapter 2, production from U.S. increased by 19% but gas consumption large onshore shale basins is shifting the focus increased by 50%.2 of U.S. production from the Central and Western Gulf of Mexico (GOM), where it has These demand increases, largely for residential, been for the last two decades, back to onshore commercial and electricity uses, have been regions. While GOM production declined by accompanied by a reduction in demand from 42% between 2004 and 2008, onshore produc- industrial customers; this is illustrated by the tion in the lower 48 states (L-48) increased by relative decline in gas consumption in the 22% over the same time period.1 Southwest U.S., largely Texas, the only region of the country where gas consumption in absolute Areas with the most marked production terms and as a percentage of the U.S. total increases include the relatively immature Rocky actually dropped. This 15% decline in consump- Mountains, where production increased 103% tion over the last three decades can be attributed between 1998 and 2007; and parts of Eastern in part to high natural gas prices over the last Texas, where production increased by 177% several years which drove refineries, and ammo- over the same time period. This shift is nia and other chemical plants offshore.3 expected to be more pronounced as production increases from the Marcellus shale, concen- The U.S. and LNG Markets trated in New York and Pennsylvania, with additional production potential in Ohio and Growing gas demand and significant differ- West Virginia. ences in gas prices between global regions has increased the desirability of a global gas market. Shifts in Demand Patterns As seen in Chapter 3, gas prices are significantly lower under an Emissions Prediction and Policy There has also been a shift in U.S. gas demand Analysis (EPPA) scenario where there is a patterns over the last decades, associated in part relatively unconstrained global market in natural with relative population shifts to the South and gas compared to the current regionalized West from the Northeast and Midwest, the two market. While the U.S. represents around 24% regions in the country where population as a of global gas consumption, its engagement in percent of total U.S. population has declined. the development of a global LNG market is tempered by dramatic increases in the U.S. producible gas resource base, largely enabled by the affordable production of new unconven- tional gas resources.132 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Currently, the U.S. permits proprietary access activities. Of this total, 96 teragrams of CO2eto LNG suppliers for new regasification termi- were CH4 emissions; the remainder are fromnals; this would allow the developer of a non-combustion CO2. The draft EPA inventory,regasification facility to give preference to the released in late February 2011, doubled theimport of its own LNG or the LNG of its EPA’s estimates of methane emissions from gasaffiliates at that point of entry.4 This policy systems for 2008. A breakout of EPA’s estimateddecision was made to incentivize construction emissions from gas systems is seen in Figure 6.2of substantial import infrastructure in the U.S. (from EPA’s revised draft inventory estimatescreating opportunites for increased global also discussed in Appendix 1A).LNG trade. Methane leaks from gas systems, particularlyGHG EMISSIONS FROM THE NATURAL GAS at the levels indicated by the new EPA estimates,INFRASTRUCTURE could prompt efforts to capture those emissions for both environmental and business reasons.Natural gas is the cleanest burning fossil fuel, Reducing emissions from well completionsenhancing its desirability as a fuel option in a can, for example, create value for producerscarbon-constrained environment. As a fossil and can have a very short payback periodfuel, however, natural gas also emits greenhouse (3 to 8 months).5 While many larger producersgases (GHG), including CO2 emissions from and pipelines have already deployed relativelygas combustion and CO2 and methane emis- inexpensive methane detection and capturesions from the gas system, including produc- technologies and are able to realize profits fromtion, processing, transmission and distribution. use of these technologies, smaller producers may need new, more affordable technologiesAccording to EPA inventories released in 2010, to detect and capture methane 2008 GHG emissions from natural gassystems were 126 teragrams (one teragram is The EPA has also issued a final rule on manda-equivalent to one million metric tons) of CO2 tory reporting of GHG emissions from naturalequivalents (CO2e), less than 2% of total CO2 gas systems, after the Supreme Court deter-equivalent emissions from energy sources and mined the EPA could regulate GHGs as airFigure 6.2 Estimated CO2e Emissions from Natural Gas Systems Distribution 29.9 Transmission and Storage 43.4 Field Production 134.2 Processing 37.1 Source: EPA Draft GHG Emissions Chapter 6: Infrastructure 133
  • pollutants and the EPA issued an endangerment regulation of GHGs by the EPA implied in the rule in 2010, indicating that GHGs posed a promulgation of this rule could have major threat to public health and welfare. This rule impacts on gas system operations, particularly would require reporting from well pad equip- on production, transmission and storage, if the ment both onshore and offshore, gas process- estimates in Figure 6.2 are reasonably accurate. ing, pipelines, city gates, LNG import and EPA recently extended the deadline for applica- export facilities, underground storage and tion of best available monitoring methods for compressor stations. The rule covers annual gas systems. reporting of CO2, methane, and nitrous oxide emissions from facilities emitting 25,000 metric COMPONENTS OF THE NATURAL GAS tons of CO2e per year or more. The EPA INFRASTRUCTURE estimates the cost to the industry of imple- menting the rule to be $61 million for natural To move gas from production to demand gas and oil systems (the EPA does not separate centers over the next 20 years, it is estimated by gas from oil) and $20 million a year in subse- the Interstate Natural Gas Association of quent years in 2006 dollars. American (INGAA) that the U.S. and Canada will need approximately 28,900 to 61,900 miles The EPA has deferred direct emitter identifica- of additional transmission and distribution tion until confidentiality issues can be resolved. natural gas pipelines depending on assump- All other elements of the rule are now in effect.6 tions for gas demand — its base case identifies The EPA estimates that this will affect around almost 38,000 miles of pipelines with the 2,800 facilities. The EPA is careful to point out regional distribution depicted in Figure 6.3.7 that the 25,000 metric ton limit will exclude INGAA also projects a need for 371 to 598 small businesses from the requirements of the billion cubic feet (Bcf) of additional storage rule. It is unclear how many small producers capacity, a 15% to 20% increase over current would be exempt by the emissions limit. levels and consistent with the rate of additions Although the EPA recently postponed deadlines between 2005 and 2008.8 for mandatory emissions reporting, the ultimate Figure 6.3 U.S./Canada Pipeline Capacity Additions, 2009–2030 (in 1,000 of miles) West, 2.2 Arctic, 1.0 Canada, 4.7 Southwest, 8.6 Offshore, 2.2 Southeast, 4.6 Northeast, 2.7 Midwest, 3.3 Central, 8.4 Source: INGAA, 2009134 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 6.1 Total Expected Gas Pipeline, Midstream and LNG Expenditures, 2009–2030(billions $) Region Transmission Storage Gathering Processing LNG Total % Canada 33.0 0.4 1.2 1.0 - 35.5 17 Arctic 24 - 1.0 3.5 - 25.5 14 Southwest 27.6 1.3 4.2 7.5 0.4 41.1 20 Central 24.8 0.2 0.7 4.8 - 30.5 15 Southeast 15.4 1.4 0.4 2.3 1.3 20.8 10 Northeast 10.1 1.0 2.3 1.6 - 15.1 7 Midwest 12.9 0.4 0.2 - - 13.4 6 Western 8.7 0.5 0.1 1.0 - 10.4 5 Offshore 6.3 - 7.8 - - 14.1 7 Total 162.8 5.2 18.0 21.7 1.8 209.5 100 Percentage 78 2 9 10 1.0 100 Source: INGAA, 2009There will also be additional requirements for ment project meets federal environmentalgas processing, especially in light of the changes guidelines. The Coast Guard and Maritimein production patterns in the U.S. Investment Administration (MARAD) at the Departmentrequirements by sector for gas infrastructure of Homeland Security have responsibility forbetween now and 2030 are summarized in offshore LNG facilities. In addition to theseTable 6.1.9 Note that these figures assume federal agencies, there is a range of state entitiessuccess in bringing arctic gas to the L-48 from involved in the permitting process.Alaska and the Mackenzie delta; the Alaska gaspipeline has remained illusory for the last two The long lead times required to site and builddecades and its realization remains uncertain. gas infrastructure, driven in part by these complex regulatory decision-making structuresThere are several federal and state agencies for gas infrastructure siting, not only add to theinvolved in siting gas pipelines and other gas cost, but mean that many of the additions andinfrastructure. The Federal Energy Regulatory expansions we are seeing today were originallyCommission (FERC) regulates interstate pipe- contemplated as much as a decade ago. Thisline construction while states regulate intra-state highlights the ongoing tension between thepipeline construction. Other federal agencies needs of policy makers and regulators for moreplay significant roles in construction permitting, accurate data and information on supply andincluding the EPA, the Fish and Wildlife Service, demand trends and patterns, the associatedand the Office of Pipeline Safety (OPS) at the infrastructure needs, and the status of technol-Department of Transportation (DOT); the ogy development; and the inherent uncertain-OPS regulates the safety of pipeline operations ties and risks that accompany investment inover the infrastructure’s lifespan, starting with natural gas infrastructure across the supplyup-front safety certifications for permitting by chain.FERC. The EPA ensures that a pipeline develop- Chapter 6: Infrastructure 135
  • The U.S. Natural Gas Pipeline Network The largest single addition to the pipeline system between 2005 and 2008 was the Rocky Mountain The U.S. natural gas pipeline network includes: Express pipeline (REX) with a capacity of 1.8 Bcfd. This pipeline has effectively linked Western producer markets to Eastern consumer production sites; markets. Other notable additions include Gulf Crossing (1.4 Bcfd) and Midcontinent Express (1.2 Bcfd), both taking gas from the shale pipelines which move processed gas over long regions in Texas and Oklahoma to Alabama and distances from production sites to major Mississippi; and two expansions to move gas centers of demand; and into the Southeast U.S., the 1.6 Bcfd Gulf South Southeast Expansion; and the 1 Bcfd Southeast Supply header.11 which carry natural gas on to end users. The largest regional capacity increase in this time frame was from the Southwest region toMajor changes in U.S. gas markets have prompted the Southeast, where almost 6.7 Bcfd of pipe-significant additions to the U.S. pipeline network line capacity was added, in part to move shaleover the last several years. Between 2005 and 2008, supplies to markets. Capacity to move supplypipeline capacity additions totaled over 80 Bcfd, from the Midwest to the Northeast increased by 1.5 Bcfd, a 30% jump, followed by exports fromexceeding those from the previous four-year period the Central to Western U.S., at 1.4 almost 100%. West-to-East expansions are contributing to In this discussion, we focus largely on transmis- major changes in the general direction of sion pipelines additions, although safety, which pipeline flows in the U.S., which have histori- is briefly discussed, is also an important issue cally moved from south to north. 2030 forecasts for distribution pipelines and to some degree, suggest the need for an additional 20% of for gathering pipelines as well. interregional transport capacity.12 While forecasts and historical pipeline expansions Pipeline Additions. Major changes in U.S. gas offer a portrait of a robust and adequate markets have prompted significant additions to response to growth in gas demand, the poten- the country’s pipeline network over the last tial for large increases in gas-fired power several years. Between 2005 and 2008, for generation, either for fuel substitution from gas example, pipeline capacity additions totaled to coal or as firming power for intermittent over 80 billion cubic feet per day (Bcfd), renewable generation, could increase the need exceeding those from the previous four-year for gas pipeline infrastructure. period by almost 100%. Additions of 44.5 Bcfd in 2008 alone exceeded total additions in the Figure 6.4 depicts total pipeline capacity and five-year period between 1998 and 2002. The directional flows; the circled areas highlight rate of additions in 2009, while slower than in additions between 1998 and 2008, with volumes the previous several years, was still brisk with added and directions indicated by the key in 3,000 miles of pipelines added. Figure 6.410 the lower right-hand corner. highlights major inter-state pipeline additions over the 11-year period from 1998 to 2008.136 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 6.4 Major Additions to Natural Gas Transportation Capacity 1998–2008 Source: Presentation of James Tobin, EIA, Major Changes in Natural Gas Transportation Capacity, 1998–2008, November, 2011.West-to-East expansions are contributing Pipeline Safety. Recent gas pipeline explosionsto major changes in the general direction in California and Pennsylvania, which caused loss of life and property, underscore pipelineof pipeline flows in the U.S. safety as an ongoing issue. There is a range of reasons for pipeline accidents, from pipeline/ In Chapter 4 we discuss the need for increased construction defects to third-party accidents gas peaking units to firm intermittent renew- to corrosion. Figure 6.5 shows the number of able generation even though their capacity incidents by type of pipeline over the last 20 years. factors would most likely be very low. Similarly, According to statistics compiled by the DOT, recent analysis by the INGAA Foundation corrosion is the most common cause of leakage suggests that in the event of large-scale penetra- for transmission pipelines, and third-party tion of intermittent renewable generation, gas excavation incidents are the most common cause pipelines may need to dedicate firm capacity of leakage for distribution pipelines.15 Leakage to provide service to backup generators even is responsible for most serious incidents. though this capacity would be used infre- quently and the per-unit cost of the infrastruc- The DOT’s Pipeline and Hazardous Materials ture is likely to be very high.13 The INGAA study Safety Administration (PHMSA) has the pri- also forecasts an incremental delivery capacity mary federal responsibility for ensuring gas requirement of around 5 Bcfd of gas for new pipeline safety. In 2003, the PHMSA imple- firming generation though utilization would be mented a rule that required an integrity only around 15%, with implied transportation management program (IMP) for transmission costs that could be around six times more than full-rate utilization costs.14 Chapter 6: Infrastructure 137
  • Figure 6.5 Serious Gas Pipeline Incidents by Pipeline Type, 1991–2010 Transmission, 132 Gathering, 10 Distribution, 830 Source: PHMSA Existing pipeline safety research programs will reflect the different challenges of distribution within the federal government and within pipeline safety compared to transmission pipelines; they will likely be less prescriptive industry are small and the task of ensuring the and will also cover the operator’s entire area, integrity of the 306,000 miles of transmission compared to the requirements for transmission pipelines and 1.2 million miles of distribution pipelines to cover only “high consequence areas.” pipelines is enormous and essential. The DOT has noted the lack of incentives for pipelines. This rule required operators to test distribution pipeline operators to assess the transmission pipeline integrity in highly safety of distribution pipelines, writing that populated areas by 2012. Between 2003 and “…there are no robust market signals or 2009, after the implementation of the rule, incentives to prompt operators to thoroughly there were six total fatalities; tragically, there assess the condition of the pipelines or to were 10 fatalities in 2010 from the explosion implement integrity management programs.”16 and fire in San Bruno, California. Also, according to the U.S. Department of Energy’s (DOE) Office of Fossil Energy almost As noted, distribution pipelines are responsible one-quarter of U.S. gas pipelines are more than for the largest number of serious gas pipeline 50 years old.17 In addition, demand for natural safety incidents. Distribution pipelines also gas is expected to increase over the next couple pose more difficult problems for integrity of decades. Reduced management compared to transmission pipelines as they are much smaller inElectric Power diameter, Finally, existing pipeline safety research pro- Gas CCS Industrial Commercial Residential are shorter, include a significant amount of grams within the federal government are small Coal CCS Vehicles Lease, Plant and Pipeline plastic pipe, and have major branching of pipes and the task of ensuring the integrity of the Renew to serve end use customers. A PHMSA rule for 306,000 miles of transmission pipelines and Hydro distribution pipelines, which went into effect in 1.2 million miles of distribution pipelines is both large and essential. The PHMSA identifies Nuclear February 2010, requires IMPs to be implemented by August 2011. While plans are required, they $33.25 million in federal funding for pipeline Gas Oil Coal138 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 6.2 PHMSA Technology Research 2002–present (in millions of $) Category PHMSA Industry Total Damage Prevention $2.79 $2.33 $5.12 Pipeline Assessment and Leak $25.08 $32.77 $57.86 Detection Defect Characterization and $0.80 $1.20 $2.00 Mitigation Improved Design, Construction $4.58 $5.40 $9.98 and Materials Grand Totals: $33.25 $39.37 $72.62Source: PHMSA Web sitesafety technology development since 2002, around Pipelines and Regional Prices. With respect$4 million per year (Table 6.2). The PMHSA to pipelines and regional prices, in general, thealso identifies $16.94 million in “strengthening difference between daily prices at regional hubsstandards” research and $29.98 million in compared to Henry Hub prices (the market“knowledge document” research; the last two center in Louisiana that serves as the pricecategories could be characterized as “regulator’s point for New York Mercantile Exchangescience.” (NYMEX) futures contract) is the basis differ- ential or “basis.” The basis differentials are oftenIMPs are necessary but may not be sufficient to small, reflecting the short-run variable cost ofmeet safety needs. The gas industry noted the transporting gas or of displacing shipments ofneed for additional transmission and distribu- gas to one market center instead of another.tion R&D in a 2007 report.18 Specific focus Occasionally, when transportation bottlenecksareas could include: are long term, the basis differentials become large and reflect the different prices at which demand is being rationed in the different of system integrity; locations. A differential that greatly exceeds the cost of and reliability; transportation suggests system bottlenecks. According to FERC, Rockies tight gas and Marcellus shale will compete with traditional damage; supplies from the Gulf of Mexico. FERC anticipates that this new supply will help moderate severe basis spikes on peak demand construction, maintenance and repair; and days in the winter.20 system, operation, planning and regulatory acceptance and mitigating environmental issues.19 Chapter 6: Infrastructure 139
  • The relationship of the price differential to 2009, REX had the capacity to move 1.8 Bcfd infrastructure is observed in the basis differen- of natural gas from the Rockies to Ohio, then tials at the Cheyenne and Algonquin hubs to the Northeast. As noted, REX was the largest before and after the opening of the REX pipe- addition in the U.S. pipeline system between line, which is now moving gas supplies from the 2005 and 2008 and has effectively joined region to Eastern markets (Figure 6.6). These Western producer markets with eastern con- fairly dramatic changes demonstrate how sumer markets, a long-time goal of Rocky alleviating pipeline infrastructure bottlenecks Mountain producers. This pipeline has had a can incentivize production and lower consumer major impact on gas flows in the Midwest and prices overall. has reduced the basis differential at both the Algonquin and Cheyenne hubs. …alleviating pipeline infrastructure Natural Gas Processing bottlenecks can incentivize production and lower consumer prices overall. Each year in the U.S. some 530 natural gas processing plants process around 16 trillion Before the construction of the REX pipeline, cubic feet (Tcf) of raw natural gas. These natural gas transportation out of the Rockies facilities have an average capacity factor of region was very constrained, leading to lower around 68%. Natural gas often requires pro- gas prices than those at most of the other cessing because gas in its raw form can contain natural gas market centers. As of November impurities which may include sulfur, CO2, water Figure 6.6 Impacts of 2008 Pipeline Capacity Expansion on Regional Prices and Average Basis Source: Bentek, Beast in the East, 2010140 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • and other contaminants that need to be removed system, including power plants. Gas storagebefore transport through pipelines to demand is also used to hedge price variations.centers. Removing impurities such as sulfur,CO2 and water to produce pipeline-quality gas is There are around 400 storage facilities in thethe primary role of such processing facilities.21 L-48 owned by 80 corporate entities andUnderstandably, gas processing units are largely managed by 120 operators. Depleted reservoirslocated in gas-producing regions of the country. account for most storage facilities (82%),Currently, around 82% of gas-processing followed by aquifers (9%), with salt cavernscapacity is in six states: Louisiana, Texas, making up the remainder. Working gas storageWyoming, Kansas, New Mexico and Oklahoma. capacity nationwide in 2009 was around 4.2 Tcf, which represents about 20% of annual gasAs noted, gas production is increasing dramati- production. Over 53% of this capacity is foundcally and production patterns in the U.S. are in just five states: Michigan, Illinois, Louisiana,changing. The need for gas processing additions Pennsylvania and Texas.23is likely to be more pronounced in regionswhere gas production is relatively immature, There has been a great deal of interest in thesuch as in the Uinta Basin of Eastern Utah and relationship between storage and short-termthe Piceance Basin of Western Colorado. Gas price volatility. In 2005, the FERC chairmanprocessing is very limited in the Marcellus Shale noted that gas storage capacity had increasedBasin where, for example, Western Pennsylvania only 1.4% in almost two decades, while U.S.and Northern West Virginia combined have natural gas demand had risen by 24% over the530 million cubic feet (Mmcf) of processing same period, and speculated that there was acapacity, with 435 Mmcf of planned processing link to the record levels of price volatility thatadditions and a new 37,000 bpd fractionation were being experienced.24 In 2006, FERC issuedplant.22 Order 678 which, among other things, sought to incentivize the building of more storage byGas processing units also produce natural gas changing its regulations on market powerliquids (NGLs) from heavier hydrocarbons requirements for underground storage. Sincecontained in unprocessed “wet” gas. If there are the order was issued, total storage capacity hassufficient quantities of NGLs, the market increased by 169 Bcf, or 2% of overall storageconditions are right, and the processing facility capacity. This compares to a 1% increase in thehas the capacity to both treat and separate previous three-year period.NGLs from gas streams, consumer productscan be produced, including ethane, propane, There is also growing interest in high-deliver-butane and pentanes. These products can add ability gas storage. Storage facilities are classifiedvalue for gas producers, especially important in as either baseload or peakload facilities.a low gas price environment. In 2009, the U.S. Baseload storage facilities, most often in depletedgas industry produced 714 million barrels of reservoirs, typically support long-term seasonalNGLs, a 16% increase over the 2005 levels of requirements primarily for commercial, residen-production. tial and industrial customers. These facilities are large and are designed to provide steadyNatural Gas Storage supply over long periods of time; their injections (typically over 214 days, April to Oct) andNatural gas is stored in underground storage withdrawals (151 days, Nov to Mar) are slow.25facilities to help meet seasonal demand fluctua-tions, accommodate supply disruptions andprovide operational flexibility for the gas Chapter 6: Infrastructure 141
  • [The] growing relationship between the gas and 3% for all gas storage.27 More important thanpower infrastructures is highlighted by the increased capacity, however, is the withdrawal period. Table 6.3 highlights the much shorter, multi-need for high-deliverability gas storage to match the cycle capabilities of salt formation storagegrowth in gas-fired power generation associated with facilities compared to depleted reservoirs andfuel. The degree to which this interdependency aquifer storage.28stresses both the gas and power infrastructures andcreates conditions where the infrastructures and Salt caverns are typically located in the Gulf Coast region and are not found in many areasrelated contracting, legal and regulatory structures of increased gas demand, where geology limitsmay be inadequate is not fully understood. both baseload and peakload storage options; this is particularly true in the Northeast, the The operational characteristics of baseload West (areas of high gas demand for power storage may be inadequate as storage needed for generation) and parts of the desert Southwest. gas-fired power generation where gas demand varies greatly, not just by season but daily and The growing use of natural gas for power hourly. Managing this variability is especially generation, including the potential near-term important, for example, when, as seen under displacement of coal with Natural Gas Com- the carbon price scenario in Chapter 2, natural bined Cycle (NGCC) generation and increased gas becomes a more critical component of the penetration of intermittent renewables, dis- generation mix. Also, gas peaking units serve as cussed in detail in Chapter 4, underscores the backup for intermittent renewables which may growing interdependencies of the gas and have relatively low load. This type of demand electric infrastructures. This growing relation- also requires greater variability in storage with- ship between gas and power infrastructures is drawals than is found in baseload storage units. highlighted by the increased need for high- deliverability gas storage to match the growth High-deliverability storage provides an option in gas-fired power generation. The degree to for handling high-demand variability associated which this interdependency stresses both the with an increased role or natural gas in power gas and power infrastructures and creates generation.26 High-deliverability storage, conditions where the infrastructures and related typically in salt caverns, is only about 5% of contracting, legal and regulatory structures may overall gas storage, although capacity increased be inadequate is not fully understood. 36% between 2005 and 2008, compared to Table 6.3 Gas Storage Facility Operations Type Cushion Gas Injection Period Withdrawal Period (Days) (Days) Depleted Reservoir 50% 200–250 100–150 Aquifer Reservoir 50%–80% 200–250 100–150 Salt Cavern 20%–30% 20–40 10–20 Source: FERC Staff Report142 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • R E CO M M E N D AT I O N to seek authorization to re-export gas.31 On aA detailed analysis of the growing positive note, the large excess of import capac-interdependencies of the natural gas and ity provides options for supply diversity in thepower generation infrastructures should event of unexpected shortfalls in indigenous supply. Also, LNG supplies initially intended forbe conducted. This should include analysis U.S. markets have been diverted to otherof the system impacts of increased use countries, with European importers andof natural gas for power generation and consumers, including some key U.S. allies, asthe degree to which this stresses the the main beneficiaries.infrastructure or creates conditions wherestorage may be inadequate to meet power Federal Policy and LNG. During the lastgeneration needs. decade, federal policy facilitated the expansion of LNG import capacity. In 2002, as already noted, FERC issued the so-called HackberryLNG Infrastructure decision which aided investment in LNG import capacity by allowing LNG developersLNG regasification terminals are the last link in proprietary access to import facilities. Toa long supply chain that enables international address delays in LNG import terminal sitingtrade in natural gas and U.S. LNG imports. associated with jurisdictional conflicts, theIn 2000, the U.S. had four LNG regasification Energy Policy Act of 2005 granted FERCfacilities with a combined capacity of 2.3 Bcfd.29 exclusive jurisdiction over permitting ofHigh natural gas prices in the first decade of the onshore LNG regasification facilities, clarifying21st century, coupled with concerns about federal primacy in this process. Later that year,declines in domestic supplies and reserves, FERC, in an effort to expedite siting of LNGsparked a wave of construction of new LNG facilities, established mandatory pre-filingregasification terminals and expansions of procedures designed to help resolve NEPA andexisting ones. North America now has 22.8 Bcfd other community issues prior to the filing of aof LNG regasification-rated capacity either formal application with FERC by the developeroperating or under construction (with original to site a regasification facility.32 These statutoryplanning expectations of capacity factors of and regulatory actions helped enable thearound 50%), 89% of which is in the U.S. permitting of substantial additional regasifica- tion capacity in the U.S. Together with addi-These facilities are expensive. The EIA esti- tional volumes from Canada and Mexico,mated in 2003 that a typical new regasification 48.65 Bcfd was licensed to supply U.S. marketsterminal would cost $200 to $300 million for (but not all of this capacity was built).a sendout capacity from 183 to 365 Bcf (3.8 to7.7 million tons) per year of natural gas but These actions by FERC and other agenciesacknowledged a wide variation in cost, which illustrate a willingness on the part of the federalis very site specific. 30 government to expedite the building of energy infrastructure in order to achieve a policyIn 2009, U.S. consumption of imported LNG objective; in this instance, adequate and afford-was 1.2 Bcfd, leaving most of this new capacity able supplies of natural gas were deemed to beunused and the investment stranded. Demand in the public interest as it was widely believed atis, however, geographically uneven. The Everett the time that North American gas productionimport facility in Boston, for example, meets had peaked and that imports would be neces-around half of New England’s gas demand. sary to affordably meet demand.Gulf Coast terminals however have been forced Chapter 6: Infrastructure 143
  • This unused capacity has prompted facility Ohio, Pennsylvania and West Virginia, but it is owners and investors to explore opportunities also the least developed of major U.S. shale for using them as export as well as import basins. These Northeastern and Midwestern terminals; this would require the building of states are generally more densely populated and substantial new liquefaction infrastructure. less accustomed to natural gas production than Cheniere, the owner of the Sabine regasification Texas, Oklahoma, Arkansas and Louisiana, the facility for example, has entered into non-bind- locations of other major producing shale ing agreements with two potential purchasers basins. Production in these other basins will of LNG volumes, and is seeking funding to continue to alter U.S. gas supply forecasts build four LNG trains at the site. The U.S. DOE regardless of the development of the Marcellus. recently approved a permit for export of LNG Its sheer size, its under-development, its unique from this project to free trade agreement environmental issues and its proximity to countries only and FERC has initiated an major demand centers and the associated environmental review of the proposal. Others consumer benefits warrants a brief discussion such as Dominion at Cove Point are reviewing of some key infrastructure issues affecting the export opportunities as well. development of the Marcellus. INFRASTRUCTURE NEEDS AND THE The economics of shale production and the size DEVELOPMENT OF THE MARCELLUS of the Marcellus shale basin have created SHALE enormous interest in the development and production of this vast resource. The location As noted in Chapter 2, the natural gas produc- of Marcellus production in the Northeast, with tion profile of the U.S. has been altered by the the resulting lower transportation costs to this ability to produce natural gas from large U.S. market, could translate into lower gas prices for shale basins. The Marcellus shale may be the the region’s consumers, who have typically largest contiguous shale basin in the world, relied on LNG imports, and Canadian and underlying significant acreage in New York, GOM gas via pipeline. Figure 6.7 Average Transportation Costs to Northeast Markets ($ per Mmcf) 0.50 0.41 0.42 0.40 611 0.37 0.30 0.21 0.20 0.10 0.00 Marcellus Southeast/Gulf Canada Rockies Source: Bentek, Beast in the East, 2010144 MIT STUDY ON THE FUTURE OF NATURAL GAS Million Barrels
  • It could also shift GOM gas movements to for the Marcellus were to come on-line, onthe southeast, an attractive option for the schedule, the region would have 800 millionregion’s consumers who are on the high- cubic feet per day (Mmcfd) of gas processingpriced end of the Western coal supply chain. capacity by 2012. Also, two NGL pipelineFigure 6.7 shows the average and typical projects have been proposed from Pennsylvaniatransportation costs for producing regions to Chicago and Ontario which could ease thesupplying Northeast markets.33 pressure for NGL outlets. Planned pipeline expansions appear to be adequate.The Marcellus, however, needs substantialinfrastructure additions to move its gas to Minimizing flowback water, on-site treatmentmarkets. There are three transmission pipe-lines to serve the region either under con- options, water re-use, and new local and regionalstruction or certified for construction with water treatment facilities are all necessary ina combined capacity of over 1 Bcfd, and managing the environmental impacts of flowback andanother 4.8 Bcfd of planned additions to produced water, water transport, and the stress onexisting pipelines. These additions are essen- existing water treatment facilities in the region.tial: Marcellus producers estimated that, asof early 2010, less than half of the 1,100 wellsdrilled in the Pennsylvania Marcellus had Finally, of major interest and concern is thepipeline access.34 development of a water disposal infrastructure to mitigate the environmental impacts associ-It is expected that planned investments in ated with wastewater from drilling whichpipelines, which are in the several billion dollar includes flowback water and produced water.range, will also drive investments in under- Water disposal options in the Marcellus areground storage. This is critical for the region as limited. Strict regulations and complicatedthe geology of the Northeast precludes signifi- geology, particularly in Northeast Pennsylvania,cant storage in this key demand region, which limit the development of disposal wells close tocould create a storage bottleneck when moving drilling sites. There is extremely limited pre-gas from points West to Northeastern markets, treatment capacity in the region and theparticularly in the peak demand months in the climate is not conducive to evaporationwinter. options. Minimizing flowback water, on-site treatment options, water reuse, and new localThere is also wet gas in the Marcellus, particu- and regional water treatment facilities arelarly in Southwestern Pennsylvania. The needed to reduce the environmental impacts ofcondensate and NGLs from wet gas enhance flowback and produced water and waterthe economics of production, assuming favor- market conditions and adequate infra-structure to move NGL products to markets.A significant percentage of this wet gas in theMarcellus requires processing to providepipeline quality gas. The shortage of processingcapacity and outlets for wet gas products couldplace constraints on the production of pipelinequality gas, and could effectively shut-insignificant gas production in the Marcellus. Ifall planned gas processing capacity additions Chapter 6: Infrastructure 145
  • NOTES American Natural Gas Foundation study, Research 18 and Development in Natural Gas Transmission and 1 EIA, Table 5a, U.S. Gas Supply, Consumption and Distribution, March 2007. Inventories. Ibid. 19 2 EIA, U.S. Census data. FERC Northeast Natural Gas Market, Overview 20 3 Bernstein Research report, Natural Gas: Method in and Focal Points. the Madness, February, 2009. EIA report, Natural Gas Processing: the Crucial 21 4 CRS Report, Liquefied Natural Gas (LNG) in U.S. Link Between Natural Gas Production and Its Energy Policy: Issues and Implications, May 2004, Transportation Market, January, 2006. the so-called “Hackberry decision”, “…allowed Bentek, The Beast in the East: Energy Market 22 terminal developers to secure proprietary terminal Fundamentals Report, March 19th, 2010. access for corporate affiliates with investments in LNG supply.” Terminals that existed at the time of EIA Table 14, Underground Storage Capacity by 23 the ruling in 2002 were exempted. Congress State, December 2009. codified Hackberry in the 2005 Energy Policy Act. December 15, 2005, Statement by FERC chairman 24 5 EPA Methane to Markets presentation, Joe Kelliher on the Notice of Proposed Rulemaking International Workshop on Methane Emissions Announcement on Natural Gas Storage Pricing Reduction Technologies in the Oil and Gas Reform. Industry, Lake Louise, 14-16 September 2009. FERC Staff Report, Current State of and Issues 25 6 See EPA Web site, Petroleum and Natural Gas Concerning Underground Natural Gas Storage, 2004. Greenhouse Gas (GHG) Reporting Rule (40 CFR INGAA Foundation Web site notes that, 26 Part 98), EPA Climate Change Division. “additional conventional storage will be needed to 7, Dec 17 2009, meet growing seasonal demands and high 38,000 is base case for gas demand. deliverability storage will be required to serve fluctuating daily and hourly power plant loads.” 8, Dec 17 2009, ranges represent high and low cases in forecasts. EIA, Table Underground Natural Gas Storage by 27 Storage Type. 9 Ibid, high gas demand case. FERC Staff Report, Current State of and Issues 28 10 November 2008, Presentation of James Tobin, EIA, Concerning Underground Gas Storage, 2004. Major Changes in Natural Gas Transportation Capacity, 1998–2008. Gas Technology Institute, LNG Sourcebook, 2004. 29 11 Bentek, The Beast in the East: Energy Market EIA Report #:DOE/EIA-0637, December 2003. 30 Fundamentals Report, March 19th, 2010. FERC report, State of the Markets, 2009. 31 12 Ibid. FERC order 665’s discussion of pre-filing 32 13 INGAA Foundation study, Firming Renewable procedures noted that it is “desirable to maximize Electric power Generators: Opportunities and early public involvement to promote the wide- Challenges for Natural Gas Pipeline, March 16, spread dissemination of information about 2011. proposed projects and to reduce the amount of time required to issue an environmental impact 14 Ibid. statement (EIS).” 15 Serious incident is defined on PHMSA Web site as Bentek, The Beast in the East: Energy Market 33 an event involving a fatality or injury requiring Fundamentals Report, March 19th, 2010. hospitalization. Ibid. 34 16 PHMSA-Research and Special Programs Administration, U.S. Department of Transportation Web site, 2004-19854. 17 See DOE Fossil of Energy Web site, Transmission, Storage and Distribution program description, as of January 23, 2009.146 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Chapter 7: Markets and GeopoliticsAs we have seen in Chapter 3, there are substan- it readily transportable over long distances,tial economic benefits to a global natural gas by a variety of means, at moderate cost. This hasmarket. Geology, geography and historical allowed the development over time of a globalmarket and geopolitical arrangements have, oil market, where multiple supply sources servehowever, limited the development of a global multiple markets at transparent spot prices, withmarket that links supply centers to major price differences largely attributed to transporta-demand centers, which would have significant tion costs and oil quality. Notwithstandingenergy security ramifications. dependence on imports, the diversity and robustness of this marketplace adds significantlyAt present, trade is centered in three distinct to security of supply for consumers and toregional gas markets — North America, Europe security of markets for producers.(including Russia and North Africa) and Asiawith links to the Persian Gulf (see Figure 3.11). In comparison, natural gas markets are smallerEach has a different market structure resulting and less mature, and the physical characteristicsfrom the degree of market maturity, the sources of natural gas constrain transportation options.of supply, the dependence on imports and other Unlike oil, transportation costs — whether forgeographical and political factors. Importantly, pipeline gas or liquefied natural gas (LNG) —these regional markets set natural gas prices in constitute a significant fraction of the totaldifferent ways. In general, the U.S. has gas-on-gas delivered cost of natural gas. Also, because ofcompetition and open access to pipeline trans- the relative immaturity of natural gas markets,portation, and manages risk through spot and compared to oil, and the very high upfrontderivatives markets. The European market relies capital costs, long-term contracts have beenmore heavily on long-term contracts with price necessary to underwrite the cost of infrastruc-terms based on a mix of competing fuels, e.g., ture development and to ensure a market forfuel oil, and pipeline access is restricted. Asia uses the supplier.crude oil as a benchmark for natural gas pricesand favors long-term contracts; this structure has Pipeline gas accounts for almost 80% of today’skept LNG prices in Europe and Asia high relative interregional gas trade (a share that is expectedto other regions. These market features, along to decline as the LNG trade grows). Pipelineswith the availability of domestic natural gas may transit many countries. The number ofresources and geopolitical interests, establish the parties involved in a multi-national pipelineboundary conditions for the development of project can slow project development consider-global natural gas markets, at the same time that ably and political instability in host or transitsignificant price disparities between regions nations raises security of supply issues. Also,create greater interest in such a market. cross-border pipelines must invariably comply with multiple and dissimilar legal and regulatoryThis regionalized and varied structure of natural regimes, further complicating pipeline construc-gas markets stands in contrast to the global oil tion and operations. Finally, the strong mutualmarket, and it is instructive to understand the interests of buyers and sellers in cross-borderfundamentals of the difference between oil and pipeline projects are not fully shared by transitnatural gas markets. The physical characteristics nations, such as Ukraine for Russian supplyof oil — a very high energy density at normal to Western Europe.conditions of temperature and pressure — make Chapter 7: Geopolitics 147
  • Pipelines have a distinct economic advantage over LNG for shorter distances but LNG gains resources could diversify supply in strategic advantage over longer distances and is a key locations such as Europe and China, with enabler of a global gas market. LNG offers the mixed implications for market integration.The U.S. natural gas market functions well, with Of course, there are many unknowable factorsinfrastructure development more or less keeping pace that can impede market integration, including the geopolitical aims of current and futurewith changing market needs. natural gas exporters. potential for a greater diversity of suppliers and MARKET STRUCTURES markets, both key ingredients for increased reliability and energy security. Also, LNG is The U.S. Market generally contracted between a single buyer and seller, simplifying contract negotiations and The U.S. natural gas market is the most mature transport routes. However, the investment of the world’s three major regional markets. required for capacity expansions of each link in Significant exploitation of natural gas began in the LNG supply chain is considerable; since the latter half of the 19th century centered in minimizing investment risk is a fundamental Appalachia, with much larger production and driver for developing global LNG markets, consumption starting in the 1920s after discov- longer-term contracts are favored. eries in the Southwest. This expansion was aided by advances in pipeline technology, The geological realities of natural gas resources eventually creating a continent-wide, integrated are similar to those of oil in terms of the degree natural gas market. of concentration of conventional resources, with Russia, Iran and Qatar having the largest The regulatory institutions governing the conventional natural gas resource base (see natural gas markets in the U.S. have undergone Chapter 2). As with oil, at issue is the extent to their own historical evolution. New Deal which major resource holders, over time, will initiatives in the 1930s broke the control of the manipulate supply and prices to advance holding companies over local utilities and political and/or economic objectives in ways that established the Federal Power Commission as a are detrimental to the U.S. and its allies. Conse- regulator of the interstate sale and shipment of quently, the future structure of these markets natural gas. The Natural Gas Act of 1938 and its and the degree of integration that may develop subsequent amendments provided federal have both economic and security implications. eminent domain authority for the construction Several factors could lead to greater market of new interstate natural gas pipelines and integration and diversity of supply: natural gas storage. These policies facilitated the robust growth of a continent-wide network. can serve multiple major markets, such as the Initially, long-term contracts were the rule. Caspian; There was no single benchmark price for natural gas in the U.S. This changed with the passage of the Natural Gas Policy Act of 1978, of a market in which cargoes seek favorable which gradually led to the removal of price prices, a trend that has been seen in the controls on the interstate sale of natural gas in Atlantic basin; and the U.S. Starting in 1985, ceilings were removed148 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • on the sale of new natural gas and the Federal U.S. Oil and Natural Gas PricesEnergy Regulatory Commission (FERC) issueda series of Orders between 1985 and 1993 that There have been long-running discussionsserved to create an open and transparent about the relationship between oil and naturalcontinent-wide market in natural gas. This gas prices; these have intensified as the ratio ofmarket-based focus was extended to natural oil to natural gas prices reached historic highsgas storage in the Energy Policy Act of 2005. over the last year. This growing spread could have enormous implications for U.S. naturalA robust spot market has developed in the U.S. gas markets and is especially critical for gasand Canada, with prices set by the forces of producers, industrial gas users and the use ofsupply and demand. Contracts continue to play natural gas as a transportation fuel. For CNGa role, albeit diminished, in the market, where or LNG vehicles, a low natural gas price relativeprice clauses typically reference the spot to oil is essential for a reasonable paybackmarket. This expansion has been supported by period because the vehicle capital cost isan expanded pipeline network and associated appreciably higher (see Chapter 5). In thismidstream gas facilities. The U.S. natural gas chapter, we explore the history of these pricesmarket functions well, with infrastructure and price movements in the U.S. market duringdevelopment more or less keeping pace with the preceding decades.changing market needs (see Chapter 6). Oil prices have hovered around $100/barrel (bbl)At present, North America is largely self- for much of the last year while the U.S. Henrysufficient in natural gas, and this situation is Hub (HH) price has been consistently belowlikely to continue for some time, as indicated $5/MMBtu, for a ratio at or above 20. (Wein Chapter 3. The substantial surplus of LNG caution the reader that this ratio involves twoimport capacity, discussed in Chapter 6, different quantities; it is normally stated ineffectively provides backup capacity in the event terms of the price for a barrel of oil, aboutof unanticipated supply shortfalls or high prices. 6 MMBtu, in relation to the price for a 1 MMBtu of natural gas because these are the benchmarksIt should also be noted that the U.S. exports in commodity markets.) A common assump-natural gas. LNG exports from Alaska to Japan tion is that opportunities for substituting oil forhave been in place for 40 years, but are likely natural gas, and vice versa, will equilibrate theto face additional competition in the Asian prices. A simple energy equivalency argumentmarket, particularly as the Cook Inlet produc- would pin the price of a barrel of oil at abouttion tapers off. Part of this competition may six times the natural gas price, but this simplecome from Canada, which has a large shale gas energy-equivalence argument is unlikely to beresource. The Department of Energy (DOE) accurate because oil and natural gas undergohas approved an application to export LNG different processing, distribution and storagefrom a Gulf of Mexico (GOM) facility. The U.S. for different end uses. A number of “rules ofalso exports natural gas by pipeline to Mexico thumb” have emerged. An empirical rule thatand Canada, although with a significant net is often invoked sets the crude oil/gas priceimport from Canada. Especially since passage ratio at 10. Others are based on the competitionof the North American Free Trade Agreement between natural gas and distillate fuel oil or(NAFTA), there has been increased North between natural gas and residual fuel oil, usingAmerican energy market integration. typical ratios of fuel oil and crude oil prices.The large Canadian shale gas resource adds tothe diversity of supply within the functioningNorth American market. Chapter 7: Geopolitics 149
  • Figure 7.1 Log Values of the Natural Gas and Oil Spot Prices, 1991–2010 (2010 dollars) 3.0 5.0 Natural Gas Oil 2.5 4.5 2.0 4.0 Log Natural Gas Price Log Oil Price 1.5 3.5 1.0 3.0 0.5 2.5 0.0 2.0 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Figure 7.1 shows the (natural) logarithm of the Figure 7.2 shows the data of Figure 7.1 as a set HH natural gas price and the West Texas of WTI and HH price pairs along with the Intermediate (WTI) crude oil price (the simple rules of thumb indicated above.2 Over logarithms are used so that the same percentage this time period, the oil and natural gas prices change in price appears the same irrespective of each spanned a wide range, and the ratio of the the price) over the period 1991 to 2010. It is WTI and HH prices ranged from about 5 to 20. clear in Figure 7.1 that no simple rule of thumb None of the simple rules of thumb reproduce can fully capture the relationship between the the principal trends over the full range of oil natural gas and oil prices. The natural gas price prices. However, it is interesting that, during the is approximately twice as volatile as the oil period 1991 to 2010, the oil/natural gas price price, and short-run swings in both prices are ratio consistently exceeded 10, sometimes overlayed on top of whatever long-run relation- substantially, when the WTI price was above ship may exist. A more detailed statistical analy- $80/bbl. As already noted, the ratio is close to sis by Ramberg and Parsons confirms this point 20 in the first half of 2011. Should these price even after incorporating key exogenous factors ratios persist at high oil prices, the opportuni- affecting the natural gas price, such as seasonal- ties for opening up the transportation fuels ity, storage levels, shut-in production and the market to natural gas would be enhanced. vagaries of weather.1 Nevertheless, they also find that it is possible to identify a statistically significant relationship between the two price series.150 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 7.2 Price Benchmarks Versus Observed Prices 1991–2010 30 Burner Tip Distillate Energy Content Burner Tip Resid 25 10-to-1 Natural Gas Price, 2010 $/MMBtu Best Fit Observed Prices 20 15 10 5 0 0 25 50 75 100 125 150 Oil Price, 2010$/bbl Figure 7.2 charts natural gas prices as a function of oil prices. The four straight lines show the four pricing rules-of- thumb. Using the ordering of the lines at the right of the figure, the top line is the burner-tip parity rule based on natural gas competing with distillate fuel oil, the second line is the energy-content equivalence rule, the third line is the burner-tip parity rule based on natural gas competing with residual fuel oil, and the fourth line is the 10-to-1 rule. The slightly curved line is the best-fit line calculated from a statistical analysis incorporating a number of additional variables and dynamics. The scatterplot of data points are the actual price combinations observed over the 1991 to 2010 period. All observed prices are quoted in real terms in 2010 dollars.Using a relationship that is linear in the loga- During the period 1991 to 2010, the oil/gas price ratiorithm of prices, and accounting for a number consistently exceeded 10, sometimes substantially,of additional variables moving the natural gasprice, Ramberg and Parsons derive a best-fit when the WTI price was above $80/barrel…that can be approximated as Should these price ratios persist at high oil prices, the opportunities for opening up the transportationPWTI PWTI = IO fuels market to natural gas would be enhanced.PHH 70with the WTI and HH prices in dollars. This European and Asian Marketsrelationship is also shown in Figure 7.2 (solidline, labeled “best fit” relationship) and captures, The European natural gas market developedto some extent, the increasing price ratio with later than that in the U.S. The initial impetusincreasing oil price. However, their analysis also came with the discovery of the Groningen fieldsconfirms that the “best fit” relationship has in the Netherlands starting in 1959. In the earlyshifted towards higher oil/gas price ratios in 1960s, Algeria began LNG shipments to therecent years.2 U.K., then to France. Small quantities of Chapter 7: Geopolitics 151
  • natural gas from the Soviet Union flowed into This does not appear likely to change in the the other countries of Europe beginning with near term. With limited indigenous conven- Austria in 1968. tional natural gas resources, industrialized Asia and the emerging economies in that region are The current structure of Europe’s natural gas almost totally dependent on imported LNG markets is shaped by the 1973 Organization from Southeast Asia, Australia and the Middle of the Petroleum Exporting Countries (OPEC) East. This dependence places a high premium oil embargo. The European reaction was to on security of supply, which is reflected in the explicitly tie the delivered price of natural gas region’s dependence on long-term, relatively to the price of crude oil or crude products. This high-priced contracts indexed to oil. limits the development of a deep and liquid spot natural gas market in Europe. The indexation of natural gas contract prices to the oil price was a necessary innovation to Currently, almost half the natural gas for enable long lead-time contracts to partially Organization for Economic Cooperation and accommodate fluctuating energy prices. But oil Development (OECD) Europe is imported, is an imperfect index for natural gas, as seen mostly by pipeline from Russia and North in our discussion of U.S. prices. Since the spot Africa, sometimes traversing other countries. market oil and natural gas price relationship LNG also supplies parts of Europe and is does not match any simple formula, an oil- especially important to Spain and Portugal, indexed contract price cannot mimic very well which are on the far end of the Russian pipeline the spot natural gas price; oil indexed prices are system. out of sync with the value of marginal deliveries of natural gas, sometimes being too high and The long supply chains into Europe, the other times too low. Therefore they cannot give prevalence of pipeline gas and the relative the right signals for consumption of natural inflexibility of the markets create much more gas, inhibiting efficient use of the resource. In significant security of supply concerns than are order for both buyers and sellers to capture the experienced in North America. Diversification full value of natural gas resources, it is essential of supply is a high priority. However, even for long-term contracts to reflect the specific though the U.S. is not significantly dependent supply and demand conditions of natural gas, on imports, American security interests can be meaning a liquid market in gas spot deliveries. strongly affected by the energy supply concerns Absent this, buyers and sellers have not been of its allies. able to do better than index contracts to the liquid oil price. Encouragement of the expan- There have been moves in the EU to liberalize sion of a liquid market in spot natural gas gas markets, starting with the U.K. in 1986. As deliveries in Asia is in the interest of buyers part of a larger energy market liberalization and sellers and other parties in the value chain. effort, the EU in 1998 sought to create common As the use of natural gas grows throughout rules for an internal natural gas market. The industrialized Asia and Europe, the opportunity result has been the development of a small spot is ripe to realize the establishment of a spot market on the European continent. Ultimate market. This would make it possible to switch success will depend upon the future course of long-term contracts from a price linked to spot the EU’s regulatory reform. Progress is slow. oil markets to a price linked to spot natural gas markets. In turn this will create the opportunity Industrialized Asia led the way in setting LNG for the expanded use of natural gas and improve prices through oil-indexed long-term contracts the possibility for international linkage. Never- and remains bound to this market structure. theless, the path to a spot market is likely to be152 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • complex and slow, and long-term contracts Clearly other scenarios could result fromoperating side by side with the spot market will changes in resource estimates or from geopo-be necessitated by the capital requirements of litical realities.very long pipelines and LNG infrastructure. Extrapolating from the lessons learned from theFinally, we note that domestic markets in some North American market, an interconnected deliverymajor supplier countries, such as Russia,operate with very large subsidies. This leads to system combined with price competition areinefficient use that impacts volumes of natural essential features of a “liquid” market.gas available for export. In addition, a functioning integrated marketLong-term contracts operating can help overcome disruptions, whetherside by side with the spot market political in origin or caused by natural disas- ters. An example of this was seen in the U.S. oilwill be necessitated by the capital markets, which recovered quickly following therequirements of very long pipelines 2005 hurricanes in no small part because ofand LNG infrastructure. international market adjustments.IMPLICATIONS OF MARKET INTEGRATION Overall, a “liquid” global natural gas market would be beneficial to U.S. and global economicExtrapolating from the lessons learned from interests and, at the same time, it wouldthe North American market, an interconnected advance security interests through diversitydelivery system combined with price competi- of supply and resilience to disruption. Thesetion are essential features of a “liquid” market. factors moderate security concerns aboutThis system would include a major expansion import dependence.of LNG trade with a significant fraction of thecargoes arbitraged on a spot market, similar to DIVERSITY OF SUPPLYtoday’s oil markets. As already noted, the distribution of conven-As described in Chapter 3, the Emissions tional natural gas resources is highly concen-Prediction and Policy Analysis (EPPA) model trated, with Russia, Iran and Qatar being thewas used to investigate the consequences of largest resource holders. Indeed the globalglobal natural gas prices differentiated only by market scenario of Chapter 3, referenced abovetransportation costs (which are appreciable for with regard to U.S. import possibilities, showslong distances between buyer and seller). We Russia and the Middle East becoming majoremphasize that this is not a prediction that such suppliers to all three of the major regionala market will emerge, but rather an exploration natural gas markets — the U.S., Europe andof the implications of global market integra- industrialized and emerging Asia. The recenttion. For the U.S., with the median expectations experience of Europe (curtailment of Russianfor both North American and global gas natural gas) and the uncertain political futureresources, the U.S. becomes a substantial net in the Middle East are a cause of concern,importer of gas in future decades in an inte- especially in Europe and Asia because ofgrated market and long-term domestic prices their large demand and limited or decliningare lower than in the regionalized market production.structure. Also, greater diversity of supply isseen for all the major markets in this scenario. Chapter 7: Geopolitics 153
  • As has already happened in the U.S., unconven- near future. Even with this restriction, the tional resources could change the picture estimate is for 5,760 Trillion cubic feet (Tcf), dramatically. The Energy Information Admin- which is a substantial fraction of the approxi- istration (EIA) recently released “World Shale mately 16,000 Tcf mean estimate of global Gas Resources: An Initial Assessment”.3 This resources discussed in Chapter 2. None of these report, prepared by Advanced Resources shale resources was included in the global estimate or in the trade models of Chapter 3.The scale of the global shale gas resource is a ARI acknowledges that the estimates may havepotential game-changer…the trade flows in a global considerable uncertainty at this time, and will be refined over time as the shale resources aremarket could be affected substantially…and the investigated by an increasing number ofleverage of MRHs to follow politically motivated industry players.strategies would presumably be diminished. The distribution of these shale resources is also International (ARI), presents estimates for interesting. Figure 7.3 shows some of the potential shale gas development in 48 basins in results 3 along with the current annual natural 32 countries outside the U.S. It does not include gas use in those countries. Pertinent to the regions with large conventional resources, such discussion above, France and Poland are each as Russia and the Middle East, since these seem estimated to have around 180 Tcf, and China unlikely to develop the shale resource in the over 1,200 Tcf.Figure 7.3 Global Shale Opportunities: Technically Recoverable Shale Reserves and 2009 Consumption (Tcf) Source: : EIA/ARI, 2011154 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • These resources dwarf annual use and therefore large coal conversion to liquid fuel programs.present the possibility of exports that signifi- For natural gas, the end use with the most dif-cantly affect import requirements for their ficulty for adjustment to a sudden disruptionregional natural gas markets. How this plays is space heating. This was seen in January 2009out remains uncertain; for example, while when Russian natural gas to Europe was cut offPoland intends to pursue production aggres- because of a dispute with Ukraine, a key pipelinesively, France has declared a moratorium transit country from Russia to Europe. Althoughbecause of concerns about environmental the U.S. is not at risk of natural gas supplyimpact. Nevertheless, the trade flows in a global disruptions because of the large North Ameri-market could be affected substantially if the can resource and production infrastructure,global shale gas resource is developed at scale the vulnerability of key allies is itself a securityover the next decade or so, and the leverage of concern. Furthermore, the opportunity toMRHs to follow politically motivated strategies substitute natural gas for oil as a transportationwould presumably be diminished. fuel feedstock improves resilience to “oil shocks.”Conventional natural gas finds, even if not on Transparent markets with diverse supply,the scale of the apparent shale resource, can whether global in reach or within large regionsalso impact diversity and security of supply that encompass both major suppliers and largewhen they occur in strategic locations. A recent demand centers, do much to alleviate securityexample (2009 and 2010) is the large offshore risks. Nevertheless, the anticipated growth infinds in the eastern Mediterranean Levantine gas use, combined with the geological realitiesbasin. The expectation is for more than 25 Tcf of conventional gas resources, inevitably willof resource in the Israeli economic zone. produce continuing concerns, such as:Inevitably there will be issues to be resolvedinvolving the maritime borders of Israel, 1. Natural gas dependence could constrainLebanon, Gaza and Cyprus. Nevertheless, it U.S. foreign policy options. U.S. freedomappears that the security of supply for Israel, of action in foreign policy is tied to globalwhich currently uses about 0.2 Tcf of natural energy supply. Iran, for example, presentsgas per year, has been transformed by the off- many security challenges in the Middle Eastshore natural gas finds. In particular, it offers and is in confrontation with the West overthe possibility of greatly reduced oil dependence a developing nuclear weapons capability.through direct or indirect use in transportation. However its oil exports and its potential for natural gas exports set up conflictingNATURAL GAS SECURITY CONCERNS objectives for the U.S. and its allies: alteringAND RESPONSES Iran’s behavior, yet not risking supply interruptions of the oil and (eventually)Energy supply generates security concerns natural gas markets. Such situations threatenwhen an economy is exposed to sudden disrup- allied cohesion in foreign policy.tions that cannot be addressed by substitutionof alternative primary energy sources. It should Specifically, the U.S., with its unique interna-be noted that any source can be replaced with tional security responsibilities, can besufficient time and investment. For example, constrained in pursuing collective action ifsecurity concerns led France to make a strategic its allies are limited by energy securitydecision to base its electricity supply on nuclear vulnerabilities.4 The natural gas cutoff topower. Restricted access to oil led World War II-era Europe demonstrated Russia’s market powerGermany and Apartheid-era South Africa to Chapter 7: Geopolitics 155
  • in a situation where key allies have inad- 3. Competition for control of natural gas equate alternative supplies and insufficient pipelines and pipeline routes is intense in short-term substitution possibilities in a key key regions. Control of pipeline routes gives sector. Russia has argued that the Ukraine natural gas suppliers tremendous leverage dispute was commercial, that Ukraine over consuming nations, and competition should not have blocked transshipments for these routes is often a “high stakes game.” and that it is a reliable supplier. However, The landlocked Caspian region, which the fact that they were selectively moving possesses large oil and gas resources, provides towards market prices in some Former an important example of the geopolitical Soviet Union states and not others suggested complexity that can develop. Decades ago, political motivations for the disruption. the Caspian was surrounded by only the In any event, security implies removing or USSR and Iran, and the legacy natural gas minimizing vulnerabilities, so U.S. support pipeline infrastructure is entirely through and encouragement of shale gas develop- Russia. The Russia-Ukraine-Europe natural ment, alternative pipeline supplies (e.g., gas delivery cutoff of 2009 spurred Europe to from the Caspian region) and transparent further its intentions to explore pipeline LNGs markets with a robust LNG infra- routes out of the Caspian Sea region to structure should be viewed as favoring Europe while avoiding Russia. This mirrors U.S. security interests. the earlier construction of the Baku-Tbilisi- Ceyhan (BTC) oil pipeline that took anA global “liquid” natural gas market is beneficial to East-West route from Azerbaijan to GeorgiaU.S. and global economic interests and, at the same to Turkey, but the gas pipeline is more complicated precisely because of the physicaltime, advances security interests through diversity characteristics of oil and natural gas and theof supply and resilience to disruption. resulting transportation options. The BTC oil pipeline can use ships to cross the Caspian 2. New market players could introduce for supply from Kazakhstan and ships to impediments to the development of trans- export the oil from Turkey. On the other parent markets. The new large consuming hand, the proposed Nabucco pipeline from economies, such as China and India, are Baku to Austria is thousands of kilometers increasingly seeking bilateral arrangements long and crosses Romania, Bulgaria and that include non-market concessions. Such Hungary just from Turkey to the Austrian arrangements have the potential to influence hub. Furthermore, supply from the long-term political alignments, move away Eastern side of the Caspian, particularly from open, transparent natural gas markets Turkmenistan, is crucial for supplying and work against the interests of consuming sufficient natural gas volumes, but a subsea nations as a whole. Major natural gas pipeline to Baku faces complications because producers have shown some interest in of unresolved seabed jurisdictional disputes. forming a cartel to control supply, but this Yet another complication is competition for movement is not yet very advanced.5 Global Turkmen natural gas from China, which has shale gas developments would make such a already begun supply through a very long cartel very difficult to implement effectively. pipeline to Shanghai. Not surprisingly, the competition and competing political pres- sures on the governments in Central Asia and the Caspian region over pipelines out of the region is intense. It is unclear how this will be resolved.156 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • While the Caspian presents a particularly 2. A federal multi-agency coordinating body complex situation, long pipelines crossing should be established to better integrate multiple countries inherently raise trans- domestic and international implications shipment concerns. Another example is the of natural gas market developments with proposed Iran-Pakistan-India pipeline. For a foreign and security policy. summary, see “Natural Gas and Geopolitics: From 1970 to 2040”.6 Numerous agencies (Energy, State, Treasury, Defense, Commerce, etc.) have a major stake4. Longer supply chains increase the vulner- in this integration, so the Executive Office of ability of the natural gas infrastructure. As the President must exercise the necessary supply chains multiply and lengthen, these convening power and leadership. To be infrastructures have become increasingly successful, strong energy policy support for vulnerable to both malevolent attacks and the coordinating group must be established natural disasters. Pipelines, processing facilities, LNG terminals and tankers are A federal multi-agency coordinating body should “soft targets,” i.e., easy to locate and destroy, be established to better integrate domestic and usually undefended and vulnerable to attacks, including cyber attacks. international implications of natural gas market developments with foreign and security policy.As the use and trade of natural gas grow overthe coming decades, with an uncertain global in the Department of Energy. This is inmarket structure, U.S. policy makers must be accord with the recommendation for awell informed and manage the interrelationship Quadrennial Energy Review issued by thebetween natural gas markets, both domestic President’s Council of Advisors on Scienceand international, and security in order to limit and Technology.7adverse effects on U.S. and allied foreign policy. 3. The IEA should be supported in its effortsR E CO M M E N D AT I O N S to place greater emphasis on natural gas and security concerns.1. The U.S. should sustain North American energy market integration and support To do so meaningfully, it must bring the development of a global “liquid” natural gas large emerging natural gas-consuming market with diversity of supply. A corollary economies (such as China, India, Brazil) is that the U.S. should not erect barriers to into the IEA process as integral participants. natural gas imports or exports. The process should promote open and transparent energy markets, including the Robust global LNG trade and progress natural gas market. toward spot pricing of cargoes, especially in Asia, are necessary for establishment A global natural gas market may lead, as in of a global natural gas market. the U.S., to lower natural gas prices relative to oil. If this in turn stimulates more substi- tution of natural gas for oil in the transpor- tation fuels market, IEA’s core mission of advancing energy security will be advanced. Chapter 7: Geopolitics 157
  • 4. The U.S. should continue to provide natural gas resources. The GSGI is led by diplomatic and security support for the the Department of State, with support from siting, construction and operation of global the Departments of Interior, Energy and natural gas pipelines and LNG facilities that Commerce and from the Environmental promote its strategic interests in diversity Protection Agency. It provides assistance and security of supply and global gas as requested on resource assessments; market development. production and investment potential; and business and regulatory issues. China, 5. The U.S. government, in concert with the India, Jordan and Poland are working with private sector, should seek to share the GSGI. experience in the characterization and development of global unconventional The experience of states in regulating natural gas resources in strategic locations. environmental performance of shale gas This includes strengthening the Global production should also be brought to bear Shale Gas Initiative (GSGI). through the GSGI. Global shale gas resources at the several 6. The U.S. should take the lead in international thousand Tcf scale have the potential to be cooperation to reduce the vulnerability of game-changers with regard to the market natural gas infrastructure; help set security and security issues discussed in this chapter. standards for facilities and operations; and The U.S. has a strong interest in seeing this provide technical assistance for sharing development and, to date, has been by far threat information, joint planning and the leader in exploiting unconventional exercises for responding to incidents. NOTES What is the Gas Exporting Country Forum (GECF) 5 and what is its objective?, EIA 2009; 1 David J. Ramberg and John E. Parsons, MIT Center for Energy and Environmental Policy Research Natural Gas and Geopolitics: From 1970 to 2040, 6 report 10-017, November 2010. D. Victor, A. Jaffe, and M. Hayes, editors, 2 ibid. Cambridge University Press, 2006. 3 World Shale Gas Resources: An Initial Assessment, Report to the President on Accelerating the Pace 7 prepared by ARI for the U.S. EIA, April 2011, of Change in Energy Technologies through an Integrated Federal Energy Policy, President’s Council of Advisors in Science and Technology, 4 National Security Consequences of U.S. Oil November 2010, Dependency; J. Deutch and J. Schlesinger, chairs, D. Victor, project director; Council of Foreign Relations Independent Task Force Report No. 58 (2006).158 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Chapter 8: Analysis, Research, Developmentand DemonstrationNatural gas is well positioned, with current In addition to prudence with regard to green-technology, to play an increasingly important role house gas (GHG) emissions, another importantin serving society’s clean energy needs over the energy policy driver is reduced oil decades, assuming a policy “level playing The analysis of Chapter 5 presented multiplefield.” As seen in the analysis of Chapter 3, this pathways for natural gas substitution for oil inis especially so in a carbon-constrained world, the transportation sector. Once again, thewherein the pathway to significant CO2 emissions research challenges are to lower costs andreductions has three major components: increase flexibility of use. FINDING demand reduction relative to business-as-usual, There are numerous RD&D opportunities to including reductions arising from more address key objectives for natural gas supply, efficient buildings, industrial processes and delivery and use: transportation technologies; natural gas as an extended “bridge” to a very resource development as an important low carbon future, principally by displacing the more carbon-intensive fossil fuels — coal contributor to the public good; and oil; natural gas production, delivery and use; in the longer term, “zero-carbon” technologies as the dominant energy supply, which may include fossil fuel combustion with CO2 applications for public policy purposes, capture and sequestration. such as emissions reductions andContinuing research, development and demon- diminished oil dependence;stration (RD&D) will play an important role indetermining the interplay of these componentsover time, especially as RD&D affects the relative gas infrastructure;costs of various technologies and fuels. Whilesuch cost reduction requirements are particularlyacute for the zero-carbon technologies, RD&Dthat lowers cost and minimizes environmental resource most effectively.impact is important for all three components.Indeed such technological progress can facilitate The fact that natural gas serves multiple sectorspolicy implementation that accelerates CO2 in competition with other primary fuels impliesemissions reduction, just as policy and regula- that many end-use efficiency RD&D programstion can stimulate technology development. will not be specific to natural gas (e.g., technol- ogy development for improving overall building energy efficiency). Similarly, there are many common elements of the technology base both Chapter 8: RD&D 159
  • for oil and gas exploration and production, supported research and characterization work such as advanced drilling technologies (e.g., for unconventional natural gas reservoirs, and nanoparticle drilling fluids) and for CO2 this provided an important foundation for sequestration following fossil fuel combustion subsequent RD&D and development of the (e.g., the science of CO2 sequestration and unconventional natural gas industry (a point monitoring, novel capture technologies and to be discussed later in the chapter). However, hydrogen-rich operation of combustion the DOE focus on natural gas RD&D was not turbines). Robust RD&D programs in all of sustained for a variety of reasons, including these areas are very important for the future of a fairly robust public-private partnership natural gas and should be supported by public (the Gas Research Institute (GRI)) that was and private funding, but our discussion in this dedicated to natural gas RD&D across the value chapter will be confined to areas that are chain. The Royalty Trust Fund (RTF) indicated uniquely tailored to production and use of the in Table 8.1 is an example of a more recent natural gas resource and that promise to have public-private partnership dedicated specifically significant impact. to exploration and production, with public funding legislatively mandated as a very small It is worth reiterating that, while we focus on fraction of Federal royalties on oil and gas natural gas-specific technologies, the overall production. Administration proposals to publicly-funded energy RD&D program should eliminate even this funding, made by both the have a strong portfolio dedicated to the first previous and current Administrations, high- and third components identified above: light the lack of agreement on the need for and demand reduction and zero emissions tech- role of publicly-funded natural gas RD&D. nologies. Notwithstanding the overall desira- bility of a level playing field, and in anticipation Our perspective is rooted in the importance to of a carbon emissions charge, support should be society of wise use of the major unconventional provided through RD&D and targeted subsidies natural gas resource that has been fully appreci- of limited duration, for very low-emission ated only recently. This resource is important technologies that have the prospect of comple- both for addressing GHG emission challenges menting and competing with natural gas in the and for energy security, and the public has an longer term. This would include efficiency, interest in its effective and responsible produc- renewables, CO2 sequestration for both coal tion and its efficient use. Clearly, the increasingly and natural gas generation and nuclear power. prominent role of natural gas in the energy mix creates an impetus for increased private sector NATURAL GAS RESEARCH NEEDS RD&D, when the benefits of such activities can AND OPPORTUNITIES be readily appropriated. This is happening to some degree for the upstream as the major oil Relative to the role of natural gas in the energy and oil service companies move more strongly sector, the Department of Energy (DOE), the into unconventional resources. Nevertheless, lead government funder of energy RD&D, has there will be a need for public and public-private historically had very small programs dedicated funding of research with longer and/or more to natural gas exploration, production, trans- uncertain payback periods than will attract portation and use. This is evident in Table 8.1, private funding. In addition, there are important which shows Congressionally-appropriated and research needs for natural gas transportation Administration-requested amounts in recent and end-use in addition to production. Priority years. In the early years of the DOE, in response RD&D areas specific to natural gas follow. to the oil shocks of the 1970s, the agency160 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Table 8.1 DOE Gas Technologies RD&D Program Funding ($ Million) FY08 FY09 FY10 FY11 (Req) FY12 (Req) Gas Hydrate Technologies 14.9 14.6 15.0 17.51 10.0 Effective Environmental 5.0 4.9 2.8 0.0 0.0 Protection Total Natural Gas Technologies 19.8 19.4 17.8 0.0 0.0 Royalty Trust Fund 50.0 50.0 50.0 0.0 0.0 Total Government Spending 69.8 69.4 67.8 17.5 10.0Source: FY 2009 – 2012 DOE Budget Request to Congress.2Improving the Economics of Resource Methane Hydrates. The Chapter 2 discussionDevelopment also indicates the potential for major methane resources from economic hydrates production.Analysis and Simulation of Gas Shale More basic research issues need to be resolvedReservoirs. Our discussion of supply in for methane hydrates than for other natural gasChapter 2 demonstrates the importance of sources. RD&D might usefully focus on: theshale gas to the overall supply curves but also systematic remote detection of highly concen-noted the potential for substantially higher trated deposits; long-term production tests,resource production. DOE R&D funding should particularly in permafrost-associated hydrates;be aimed at the basic science that governs shale and geo-hazard modeling to determine theformations. Such a program could help develop impact of extracting free natural gas on thea better understanding of the physics that stability of associated hydrate-bearing sediments.underlies fluid flow and storage in gas shales;facilitate the development of more accurate The longest production test to produce naturalreservoir models and simulation tools; and gas from forced dissociation of methanedevelop imaging tools and models for charac- hydrate deposits had only a six-day durationterizing the geologic, geochemical and geophys- due to the nature of the experiment, financialical shale rock properties. The models should concerns and other issues. The technology andbe able to predict the short-term and long-term expertise to conduct a long-term productionbehavior of induced and natural fractures in an test exist today. Financial and logistical barriersintegrated fashion. Practical 3-D models can have been the major impediments to complet-improve reservoir management. Better resource ing such a test in permafrost-associatedcharacterization will enable assessment of hydrates. Determining the degree of safety andresource play potential and well performance environmental risk associated with productionbased on petrophysical measurements. from natural gas hydrates will require that appropriate data be collected during and afterImproved microseismic formation mapping long-term production tests that are conductedwill advance optimization of real-time fracture over the next few years. Many of the safety andtreatments. At the macroscopic scale, new environmental issues will have to be addressedseismic techniques should be developed to by modeling that takes into account a rangeidentify “sweet spots” and natural fracture of potential risks, including blowouts;orientation. Publicly funded research in these co-production of CO2, water and gasses;areas will promote transparency into theeffective use of the critical shale resource. Chapter 8: RD&D 161
  • borehole, formation and/or seafloor destabili- Natural Gas Combined-Cycle with CCS. zation; and warming and potential thawing Chapter 4 highlighted the importance of of permafrost. natural gas in an electricity system with large amounts of variable and intermittent sources. Methane hydrates are a good candidate, some- If CO2 emission constraints are severe enough time in the future, for another public-private to require the capture of CO2 from natural gas success story of the type illustrated in Box 8.1 as well as coal plants, it will be important to for coalbed methane (CBM), i.e., a combina- understand the cycling characteristics and tion of government funding for resource possibilities for natural gas power plants with characterization, public-private partnership CCS. This need will be ameliorated if inexpen- for technology transfer and synergistic, time- sive large-scale storage solutions are developed, limited financial incentives to advance com- but a research program to understand cycling mercial deployment. As the majors move into capabilities at different time scales for natural today’s unconventional resources and apply gas generation would be prudent. their research capacity, methane hydrates could be thought of as “tomorrow’s unconventional Fugitive Emissions. Methane emissions in resource.” natural gas production, transportation and use are not well understood. Research is needed Reducing the Environmental Footprint of for developing technologies and methodologies Natural Gas Production, Delivery and Use for reliably detecting and measuring such emissions. This may have significant monetary Water. As discussed in Chapter 2, a comprehen- consequences in a world where CO2 emissions sive program is needed to address issues of are priced. Furthermore, the economic value of water use and backflow and produced water the methane implies that capture of the natural in unconventional gas production. Such a gas emissions for beneficial use merits develop- program could lead to: improved treatment, ment of improved technologies and methods. handling, re-use and disposal of fluids; more sustainable and beneficial use of produced The DOE and EPA should co-lead a new effort water; and more effective stimulation tech- to review, and update as appropriate, the niques that require less water and other fluids methane and other GHG emission factors to be injected into the subsurface. Nearly associated with fossil fuel production, transpor- complete recycle of flowback frac water is tation, storage, distribution and end-use. These an important goal. Some of the key water results are important for overall energy policy, treatment needs include removal of polymers, as discussed in Chapter 1. control of suspended solids and scale control. Basic research on novel approaches is appropri- ate for public support.162 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Box 8.1 Unconventional Gas: Public/Private Partnerships and Tax IncentivesThe interplay of early DOE funding, industry-matched GRI applied RD&D and synergisticpolicy incentives had a material impact on U.S. unconventional natural gas development.This is illustrated in Figures 8.1 and 8.2 for CBM and shale, respectively.3 The DOE fundingwas focused on reservoir characterization and basic science. GRI implemented industry-ledtechnology roadmaps leading to demonstration. This overlapped with a time-limited taxcredit put in place for wells drilled from 1980 to 1992, with their production eligible for thecredit through 2002. The results of this multi-pronged approach to public-private RD&Dand deployment are particularly striking for CBM. For shale, the program is credited withlaying a foundation by developing new logging techniques, reservoir models and stimulationtechnologies. See Appendix 8.A.Figure 8.1 CBM RD&D Spending and Supporting Policy Mechanisms 14 2.50 (Millions of dollars in 1999 dollars) 12 Annual CBM Production (Tcf)/ Value of Tax Credits ($/Mcf) 2.00 10 Program Budget 1.50 8 6 1.00 4 0.50 2 0 0.00 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Year CBM Production Eligible Gas DOE Spending GRI Spending Tax CreditsFigure 8.2 Shale Gas RD&D Spending and Supporting Policy Mechanisms 40 3.50 Annual Shale Gas Production (Tcf)/ (Millions of dollars in 1999 dollars) 35 3.00 Annual Program Budget 30 2.50 Tax Credits ($/Mcf) 25 2.00 20 1.50 15 1.00 10 5 0.50 0 0.00 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Year Shale Gas Production DOE Spending GRI Spending Tax Credits Chapter 8: RD&D 163
  • Expanding Current Use and Creating through direct use in combustion engines or Alternate Applications for Natural Gas through conversion to a liquid fuel. Power Generation. As seen in Chapter 3, For light-duty vehicles, extensive simulations of natural gas use in the power sector is expected the safety and environmental performance of to increase substantially. Growth will be vehicles retrofit for CNG operation should be especially important under CO2 emissions carried out with a view to streamlining regula- constraints, since natural gas substitution for tions and lowering cost, to bring U.S. condi- coal is, along with demand reduction, the least tions more in line with the certified retrofit costly response in the near-to-intermediate costs elsewhere. term. We also saw in Chapter 4 that natural gas capacity is likely to increase substantially in There are multiple pathways to natural gas- response to a greater deployment of wind and derived liquid transportation fuels (methanol, solar, and transmission constraints and natural ethanol, mixed alcohols, DME, diesel, gasoline, gas infrastructure are both central consider- etc). Various fuels and fuel combinations can be ations for such a development (Chapter 6). used in appropriately modified internal com- Advanced analysis and simulation tools are bustion engines, including optimization for needed for the converged electricity and gas increasing efficiency by use of alcohol fuels and sectors. Such tools will be invaluable for DME. Different fuels will have different fueling informing technically-grounded energy policies infrastructure requirements. The DOE should and regulations. The model/simulation tools support a comprehensive end-to-end analysis, need to incorporate several features, including: supported by engineering data, of the multiple pathways. The analysis would include an assessment of costs; vehicle requirements; integrating power sector top-down and environment, health and safety effects; and bottom-up approaches; technology development needs. This informa- tion will be important for guiding energy policy and the introduction of oil alternatives. operation and natural gas distribution requirements with large penetration of Improving Conversion Processes intermittent sources, distributed generation, and smart grids; Industry has often been at the forefront of energy-efficiency improvements because of the direct impact on the bottom line, but signifi- for electricity and gas capacity planning and cant additional opportunities lie at the nexus infrastructure development. of energy efficiency, environmental quality and economic competitiveness. Some process Mobility. As noted in Chapter 5, natural gas improvements may require substantial changes currently plays a very small role in transporta- in manufacturing, such as novel membranes for tion. In the U.S., it is used almost exclusively for separations, more selective catalysts-by-design fleets with high mileage and small geographical for synthesis or improved systems integration area driving requirements. However, the strong for reduced process heating requirements. In desire to reduce oil dependence, together with the chemicals industry, the promise of biomass today’s historically large spread between oil and feedstocks and new bioprocessing technologies natural gas prices, has led to an examination of is attracting considerable interest and needs natural gas as a material alternative transporta- further RD&D. Yet another opportunity would tion fuel. This can be accomplished either be development of new process technologies164 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • for low-temperature separation methods. Suchdevelopments can substantially reduce natural and repair;gas requirements and improve industrialcompetitiveness.The potential for significant reductions in theuse of natural gas for industrial process heatinglies in a shift to new manufacturing process In addition, the DOE should support noveltechnologies that require less process heat or concepts focusing on in-line inspections,utilize new, less energy-intensive materials corrosion prevention and protection and(Chapter 5). anticipatory maintenance.The DOE should support pre-competitive Modeling and simulation tools should beresearch in these areas and also use its conven- developed in the public domain for analysis ofing power to bring together energy-intensive the growing interdependency of the natural gasindustry sectors to identify opportunities for and power generation infrastructures. These arelowering energy needs, emissions and costs. needed to support analysis of the systemRoadmaps for future energy-efficiency tech- impacts of increased use of natural gas fornology improvements would be developed power generation and associated infrastructurethrough this public-private collaboration. This stresses and vulnerabilities, particularly withis essentially the role played in the past by the respect to changes in storage and deliverabilityIndustries of the Future Program, and some- requirements.thing like it should be re-created. Crosscuttingtechnologies applicable across a broad spec- Improving the Efficiency of Natural Gas Usetrum of manufacturing industries (such asmaterials for extreme environments and We saw in Chapter 5 that, in addition to powerseparation technologies) would also be identi- generation and industrial use, the other majorfied and should be included in a new DOE use of natural gas is for space conditioningprogram. and appliances in residential and commercial buildings. Lower-cost, gas-fired, instantaneousImproving Safety and Operations hot water heaters are an example of an appli-of Natural Gas Infrastructure ance improvement that can significantly reduce natural gas consumption. Similarly, lower-costPipeline safety, discussed in Chapter 6, is an high-efficiency heat pumps for appropriateincreasingly critical issue because of the age climates can economize on natural gas usedof much of the natural gas transmission and for space heating. Advances in these and otherdistribution system. There is a strong public building energy technologies are a good targetinterest in this area, but the federal program for public-private small. Public-private partnership is appro-priate for: Combined heat and power was seen in Chapter 5 to offer significant system efficiency, emissions and economic benefits, especially for larger of system integrity; installations (Megawatt scale). This should be encouraged. However, micro-CHP (kilowatt scale) will need a substantial breakthrough to become economic. Micro-CHP technologies with low heat-to-power ratios will yield greater damage; benefits for many regions, and this suggests Chapter 8: RD&D 165
  • sustained research into kW-scale, high- Apart from the funding increase to support the temperature, natural gas fuel cells. Basic cost-shared advanced turbine development, the research into new nano-structured materials program has averaged about $24 million/year. will be central to such programs. This low funding level must be viewed in the FUNDING AND MANAGEMENT context of parallel public-private approaches to OF NATURAL GAS RD&D natural gas research funding and management. The Federal Energy Regulatory Commission Given the importance of natural gas in a (FERC) exercised an authority to require a carbon-constrained world, and the opportuni- surcharge on interstate pipeline gas volumes ties indicated above for improved utilization of to support consumer-focused RD&D for the the resource, an increase is in order in the level natural gas industry. The FERC-approved of public and public-private RD&D funding surcharge in 1978 was equal to 0.12 cents per indicated in Table 8.1. However, the budgetary Mcf, rising to 1.51 cents per Mcf a decade later.4 pressures facing the Administration and This led to a research fund in excess of $200 Congress dim the prospects for additional million/year for an extended period, yielding appropriations in the next several years. To over $3 billion over the life of the surcharge. discuss an alternate path forward, it is impor- tant to understand the history that led to the The GRI was established in 1976 as a private current low level of research support. A more non-profit research organization charged detailed description of natural gas RD&D fund- with managing the funds. It was required to ing is given in Appendix 8.A. have a Board of Directors representing the natural gas industry, industrial consumers and The DOE natural gas research funding history the public and to submit a research plan is summarized in Figure 8.3. Between 1978 annually for FERC approval. Important features and 2010, the total expenditure was just over of this approach were applied research and $1 billion. Major elements have included: development closely connected to industry operational and technology needs, a broad RD&D portfolio from production to end-use, natural gas resources (especially shale) in the and the ability to make long-term commitments early years of DOE operations; and attract cost-sharing based on an assured funding stream. GRI programs leveraged substantial industry matching funds. on environmental protection; Clearly, the GRI funding was substantially greater than the DOE’s. Joint portfolio planning focusing on advanced drilling, completion was performed regularly to ensure that the and stimulation; programs were complementary. Box 8.1 shows the interplay between the early DOE support for unconventional natural gas RD&D, the efficiency, low NOx gas turbines in collabora- sustained GRI effort to work with industry in tion with industry during the 1990s, with developing and demonstrating unconventional nearly $300 million of DOE support (see natural gas production technology and a Appendix 8.B); synergistic time-limited tax credit for uncon- ventional production. There has been a consid- erable and continuing return on a relatively decade. modest RD&D investment.166 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • Figure 8.3 DOE Natural Gas Research Funding History 80 10 Spending (Millions of Nominal $) 8 NG Prices ($/Mcf Nominal) 60 6 40 4 20 2 0 0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Year Eastern Gas Shales Project Western Gas Sands Project Methane from Coal Geopressured Aquifers Gas from Oil Shale Exploration and Production Infrastructure Advanced Turbine Systems Environmental and Advanced Research Utilization Capital Equipment Gas Hydrates Infrastructure Congressionally-directed Projects Natural Gas Prices Source: DOE Office of Fossil Energy.However, in the wake of pipeline deregulation, of the FERC-approved program and todaythe surcharge was ended. In a regulated envi- serves as a research-performing non-profitronment, the surcharge was easily passed on organization. Its budget is substantially lessby the pipeline companies to ratepayers. After than that of GRI.pipelines became common carriers in 1992,large gas consumers could contract directly The Energy Policy Act of 2005 established thewith natural gas producers. In this new market- Royalty Trust Fund (RTF) to support a 10-yearplace, the surcharge, although small, became a $500 million research program (see Table 8.1)competitive issue. The combination of “bottom with a narrower research scope than had beenline” pressures associated with competitive the case for GRI: the Ultra-Deepwater andmarkets, the tendency of state regulators to Unconventional Natural Gas and Other Petro-eschew rate increases in competitive markets leum Resources Research Program. It is focusedand a number of “free riders” (primarily exclusively on exploration and production,intrastate pipelines in Texas that did not pay including associated environmental impacts.the surcharge) resulted in phaseout of the The RTF draws its funding from a small fractionsurcharge between 2000 and 2004. The GRI of royalties paid to the Federal governmentended as a research management organization for oil and gas production leases of Federalthrough a merger, in 2000, with the Institute on-shore and off-shore tracts. The programfor Gas Technology to form the Gas Technology structure has many similarities to that of GRI:Institute (GTI). The GTI managed the phaseout 75% of the funds are managed by a non-profit Chapter 8: RD&D 167
  • research management organization, the Recently, the President’s Council of Advisors on Research Partnership to Secure Energy for Science and Technology (PCAST) put forward America (RPSEA); an annual program plan is a set of recommendations for federal energy approved by the Federal government, in this research and policy that draws upon this case the DOE; there are specific industry experience.5 The PCAST first recommends an cost-sharing requirements; in principle, the overall annual funding level for energy research mandatory funding allows for long-term stable programs of around $16 billion, an increase of funding of projects in collaboration with $10 to $11 billion over the DOE funding level. industry. Unfortunately, the advantages of To be effective, PCAST observed that the stable funding have been more difficult to funding must be “long-term, stable and have capture in this case since, as seen in Table 8.1, broad enough bipartisan support to survive there have been persistent attempts to changes of Administration” and, recognizing terminate the program. the intense pressures on the annual domestic discretionary budget, recommended further FINDING that the additional funding be found largely through “new revenue streams,” analogous to the FERC surcharge or the RTF. The PCAST RD&D program was not compensated further suggested that there is value in the by increased DOE appropriations or external management of a portion of these funds, with strong industry input particularly private funding for natural gas research for the development and demonstration phases, is down substantially from its peak and allowing the DOE to focus on its core strengths is more limited in scope, even as natural of funding basic and translational research and gas takes a more prominent role in to serve an oversight role for the externally managed funds. These recommendations would extend the alternative models for funding and managing natural gas research to the entire energy RD&D portfolio and carry a certain The GRI and the RTF research models highlight degree of irony given the demise of GRI the value of federally-sanctioned alternative stimulated by deregulation and the continuing research models, with industry-led portfolios pressures on the RTF. and dedicated multi-year funding mechanisms, in those cases specifically for natural gas RD&D. This value is derived primarily from: consistent funding over time; significant opportunities for industry input in program development and technical project reviews; and active collaboration between government, industry, academic institutions, the national labs and non-governmental organizations. GRI also had a significant analytical unit, used widely by industry and policy makers until it was eliminated in 2001, as the surcharge funding started phasing out. Such a role is not easily incorporated into the DOE applied energy offices.168 MIT STUDY ON THE FUTURE OF NATURAL GAS
  • R E CO M M E N D AT I O NThe Administration and Congress shouldsupport RD&D focused on environmentallyresponsible, domestic natural gas supply.This should entail both a renewed DOEprogram, weighted towards basic research,program, weighted towards applied RD&D,that is funded through an assured fundingstream tied to energy production, deliveryand use. In particular, the RTF should becontinued and increased in its allocationcommensurate with the promise andchallenges of unconventional natural gas.Furthermore, consideration should be givenresearch model for natural gas Chapter 8: RD&D 169
  • NOTES Process Gas Consumers Group, Petitioner, v. 4 Federal Regulatory Energy Commission, 1 In FY 2011, a new methane hydrates program will Respondent. American Gas Association, Interstate be initiated by the DOE Office of Basic Energy Natural Gas Association of America, Fertilizer Sciences under the Geosciences Research program. Institute, Gas Research Institute, Georgia Industrial Group, Intervenors. No. 88-1109. United States 2 FY 2009 – FY 2012 DOE Budget Request to Court of Appeals, District of Columbia. Congress. President’s Council of Advisors on Science and 5 3 DOE Office of Budget. FY 1978 to FY 1996, DOE Technology, Report to the President on Accelerating Budget Requests to Congress; Gas Research the Pace of Change in Energy Technologies Institute 1979–1983 to 1994–1998, Research and Through an Integrated Federal Energy Policy, Development Plans. Chicago, Ill., Gas Research November, 2010. Institute.170 MIT STUDY ON THE FUTURE OF NATURAL GAS