THE OXFORD INSTITUTE FOR ENERGY STUDIES a recognized independent centre of the university of oxford liil UNIVERSITY OF OXFORD Oil and Gas Value Chain James Henderson December 2017 The Oil & Gas Supply Chain Production Transportation Transmission Market Upstream Gas to Liquids Oil and Gas Field Life Cycle Access esource Assessment asin Evaluation oarse Seismic Distribution of Resource Potential Field Abandonment 1 Field Production Field life typically 5 to 25 years. Field will be in decline phase for last 30% to 50% of its life. I Field Development Production well drilling, platform & processing facilities, pipeline to shore & grid. Hundreds of million to several billion dollars. Rate at which prospective areas are made available by Host Governments for exploration activity by Oil & Gas companies Exploration Evaluation Potential Fields identified from Seismic data Risked exploration economics (success rate 1 in 8) Cost of exploration well $5 - $50 million Viability Reserves Assessment A \ Determined by Prospectivity, Fiscal Terms, oil & gas price outlook, rig day rates and availability. -a Iii Viability Determined by Fiscal terms & price outlook, contractor rates, financing and transportation infrastructure. Exploration & Appraisal Drilling Drilling exploration & subsequent appraisal wells to establish presence of hydrocarbons and range of field reserves The Origins of Oil and Gas Oil within the Reservoir Sandstone Grain Oil in 'pore' between grains 7ÜK Oil Production Drive Exploration Activity • Oil and Gas is an extractive industry. Companies aim to replace current production with new finds. • Companies often explore in many different regions under differing fiscal regimes, onshore and offshore. • Success rates for exploration wells may be as low as 1 in 5. • Need to take a portfolio approach and a systematic means of evaluating and selecting exploration investments. ©M, Finding Oil and Gas - Seismic Survey Finding Oil and Gas - Exploration Drilling Three Fundamental Questions: - Is there hydrocarbon in the target structure ? - If there is, is it oil or gas ? - If there is, how much is there ? I o Exploration Drilling CO |ects a ma of sand, water and CfWfYfcC4fct mto Bie A©!l Natural ©a* Bows out ot we*. Recovered water is stored *» open p*s. then taken to a treatment Storage Natural gas is piped tanks to market •'000 31000 4000 xooo 6000 7.000 Marcelius Shale Wafer Hydraulic Fracturing Hydraulic fracturing, or *|racing." involves tho injection of moro than a rruaon gallons of water, sand and chemicals at high pressure down and across into horizontally drifted woiis as far as 10.000 foot below the surface. The pressurised mature causes the rock layer, in this case the Marcoiius Shale, lo crack. These frssures are hekJ open by the sand particles so that natural gas from tho shalo can flow up tho wei. Well turns horizontal The shale tt fractured by the pressure nsxJe tho well Graph* by Al Granberg Source: EIA Shale resources remain the dominant source of U.S. natural gas production growth 40 35 30 25 20 15 10 U.S. DRY NATURAL GAS PRODUCTION TRILLION CUBIC FEET History BILLION CUBIC FEET PER DAY 2013 Projections 1990 1995 2000 2005 2010 2015 2030 2035 Source: EIA, Annual Energy Outlook 2015 Reference case ©M, 7ÚK Homing into the Sweet-spots Once You Know Where to Look - You Need to Define the Core ■ Defining the core of a shale play post-development drilling is relatively easy - it is a statistical exercise based on mapping Initial Production rates for standardized completions e.g. Barnett ■ Defining the core pre-drill is much harder - shale plays tend to be gradational in nature, so defining the core relies on mapping optimal convergence of various technical attributes Source: Kimmeridge Energy, http://730926bdeaea 1361e79-997641d029b6764b67dd905fd3aab10cj8.cf2jackcdn.com/2-%20Finding%20 71K Specific Challenges for Shale • Shale gas reservoirs show much more production variability than conventional gas reservoirs. Shale gas wells within a single field, completed using identical drilling and fracture stimulation programs frequently show a 2-5x variation in initial rate and/or recovery factor. • Production 'sweet spots' are very real and can change rapidly between adjacent well locations - or even between adjacent frack stages in the same horizontal well. When exploring for a new shale gas reservoir, this variation means that a number of test wells need to be drilled before a decision can be made about the commercial viability of that reservoir. • This means that a significant portion of the development wells will be uneconomic or only marginally economic. • There is no single explanation for these production sweet spots. Source: D. Cooke, University of Adelaide, Australia. E E (0 o Shale Gas Well Decline Curves u (0 0 (/) 0 GC 0) (0 O (0 70,000 60,000 EnCana Horizontal Barnett Wells Decline Data >P25 P50 P75 ^—MEDIAN MEAN 10,000 -I-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1- 1 2 3 4 5 & 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Months of Production • 420 Barnett Shale wells suggest considerable variance in type-curve methodology. • Mean over-predicts EUR by 10-15%. Labyrinth Consulting Services, Inc. Houston SIPES Slid* West Virginia Shale Gas Pad - Drilling Phase Production Phase - Same Location || Shale Gas - Summary • The US shale gas phenomenon reversed the decline trend of US Gas production in the early 2000s. US became an LNG exporter in 2016. • US shale gas has been successful in terms of production growth due to: • Multiple, extensive, highly prospective plays. • Regulatory system evolved during 100+ years of continuous conventional oil and gas activity. • Landowner mineral rights. • Many competing players in exploration & production and hightech service sector. • Wide open spaces. • To date industry has failed to replicate this model in Poland, China and UK. • As much about population density, public opinion, regulatory style (and speed) and local industry dynamis®!^ as geology. Gas Sector Commercial Value Chain Development and rz) Transport rz) Storage ^ Processing ^ Distribution production LPG & Condensate Local Storage 1 Gas Strategies Urban Industry || Bringing Gas to Market - Infrastructure • Challenges: - Low energy density as a gas - Expensive to transport and store - High confidence of both reliable supply and demand needed prior to infrastructure investment. • Long Distance (high pressure) pipelines - Supply and Market (initially) physically 'locked'. - Subsequent network developments and amortised initial investment invites governments and regulatory bodies to enforce competition: • Third party access to pipeline and storage capacity • Removal of gas destination restrictions • Liquefied Natural Gas (LNG) ^jp; Long Distance Pipelines and LNG bcm 1,600 60% 1,400 2016 2017 2018 2019 2020 2021 2022 2023 1014 1015 101b 1011 101S 1019 2030 2031 1011 2033 2034 2035 2036 2037 2038 2039 10A0 LNG trade ^^^B Pipeline trade ^^^»LNG share Source: GECF Global Gas Outlook 2017 E E (0 O) o i| Gas Processing Facility E E (0 o Gas Processing - Function u (0 0 (/) 0 GC 0) (0 O (0 Generalized Natural Gas Processing Schematic Loaso Operations Gas Reservoir Oil Reservoir Dty Gas (to Pipeline) Dry (Residue) ff> Gas Fraelionalor <'° P«P«*«>») * Optional Step, depending upon me source and type of gas stream. •Source. Energy Informaiion Administration. Office of Ol end Gas. Natural Gas Division Natural Gas liquids (NGLs) □nunc Dopane Biüanfi ternary Extract valuable Condensate (light oil, propane, butane and some ethane. Remove water & nitrogen Remove C02 and H2S Must meet grid calorific value range and Wobbe index (calorific value divided by sqare root of density) - which determines flame stability. Yemen Liquefaction Facility 0 E E (0 o (0 0 0) 0 oc 0) (0 o (0 Liquefaction -161°C GAS GAS Treatment and Purification z • Removes condensate, C02/ Mercury, and H2S •Causes dehydration Refrigerant Loop Storage a J J L I Compression" Purified gas is cooled to minus 161 C at which temperature it becomes a liquid at atmospheric pressure. Volume reduced by a factor of 600 compared to gas atmospheric pressure. Source: Katherine D'Ambrosio The Gas into Power value chain Gas Components of Chain Gas Flow Production Transportation & Treatment I Pipeline Tariff I Power Components of Chain _Electricity Flow_ ^ ^^TBlectlcity WTW Transmission & Distribution 17 Price paid to Gas Distributor Price paid to Gas Producer Price paid to Gas Distributor Revenue Flow Electricity Price paid to Generators Electricity Consumers Electricity Price paid by Consumers 1 Gas Strategies 39 7ÜK E E (0 O) p Gas Fired Generation -Combined Cycle Gas Turbine Kent, UK Transporting Gas 'mk - From Production Source to Market - Summary As demand for gas has grown and in some cases nearby production sources have declined or not kept pace with consumption growth: • Long distance pipelines have been constructed; notably: • From Norway to the UK and North Europe. • From Russia to Northwest, Eastern and South East Europe. • From Algeria and Libya to Spain and Italy. • Throughout US, Canada and Mexico. Less prominently in: • South America • Asia • Africa • LNG was a key channel of gas supply in Asia (Japan, Korea, Taiwan & more recently India and China) and is becoming more widespread: • European periphery (UK, Spain, France, Italy, Turkey) • New markets for LNG are emerging with some frequency. • The growing volumes of LNG which are not constrained in terms of destination by contractual terms represent a powerful force for price arbitrage between regional markets. OP^ WZ Investment Economics • Risk versus Reward - Geological - Political/Fiscal - Technological - Market (demand) and Price • Time value of money - High up-front (risk) investments, long field life, multi-year payback period. - Access to finance - cashflow, debt, equity • Competing Opportunities - Global portfolios - Oil.Gas, (Tarsands), (Gas to Liquids) ®K The DCF Calculation as a foundation - companies' must earn an adequate return on investment Time value of money Prvsvnt Vilue o I- + Future Value 3 -4- Years -1 ■■ 510,000 + interest Option B 510,000 - Interest -* &10/Q-D0 Provided money can earn interest, any amount of money is worth more the sooner it is received Money available at the present time is worth more than the same amount at a future time because of its earning potential The DCF Calculation as a foundation - WACC concept Weighted average cost of capital is corporate "interest rate" Where; E = market value of equity D = market value of debt rt = CQ£tof equity rd = cost of debt t = corporate ta* rate WACC is the cost to a company of financing the capital for a project, including debt and equity Cost of debt = average interest rate for company Cost of equity is theoretical return to investors in the company Cost of Equity = Risk free rate +Beta*(Market return - Risk free rate) Essentially, how much return would an investor expect relative to putting his money with US Treasury stock, or in the stock market The DCF Calculation as a foundation - WACC Calculation Cost of Debt = 5% Cost of Equity Risk Free Rate - 4% Market Return - 8% Company Beta - 1.2 Calculation = 4%+(1.2*(8%-4%) Cost of Equity = 4%+4.8%=8.8% WACC Share of Equity - 50% Share of Debt - 50% Corporate tax rate - 20% Calculation = (8.8%*0.5)+[(5%*.5)*.8] WACC = 4.4%+(2.5%*.8)=6.4% Cashflow Analysis - Revenue Less Costs Cashflow = Revenue less: transport costs, royalty, state tax, federal tax, operating costs, capital costs, abandonment costs. ©M, 7ÜK DCF - The Sum of Future Annual Discounted Cashflows DCF--L- +-...+-^ (1 + r)1 (1+r)2 (l + 0n CF = Cash Flow r = discount rate (WACC) A typical spreadsheet summary of a cashflow model OCF Valuation Protected Free Cash Flow Calendar Years erxj.nq Decern oer 31 Yean Year 2 Year 3 Year 4 Year 5 Year 6 (S tn thousands) EBITDA $8,954 S9898 $10,941 $12093 $13,367 $13,367 Less D&A 1.112 1222 1.343 1.476 1.623 1.623 EBfT 7.842 8.676 9.598 10.617 11.745 11.745 Less Cash Taxes (35%) (2.745) (3,037) (3.359) (3.716) (4.111) (4.111) Tax adjusted EBIT 5.097 5.639 6.239 6.901 7.634 7.634 Pluss D«\A 1.112 1222 1 343 1.476 1.623 1.623 Less Capital Expenditures (1.750) (1.750) (1 750) (1.750) (1.750) (1.750) Less Change in Net Working investment (318) (350) (384) (423) (465) (465) Unlevered Free Cash f low S4.141 S4.762 $5,447 $6,205 $7,042 $7,042 $19,845 - $4,141 $4,762 $5.44/ $6,205 / $7,042 + (1*.11)2 (1 ♦ .11|3 (1 ♦ -11)* (1 ♦ .11)* ©M, 7ÜK Analysis to Support the Decision to drill an exploration well Geologists/Geophysicists: - Interpret Seismic data and assess reservoir size probability distribution. - Assess the probability of source, reservoir and trap. Reservoir Engineer: - Assess the recoverable reserves and reservoir properties for the 90%,50% and 10% cases. - Assess the number of production wells required. - Develop annual production profile for the life of the field. Facilities Engineer: - Creates conceptual design for min, mean and max cases with costing and cost phasing. Petroleum Economist: - Models the cashflow of the three reserve cases including tax or Production sharing effects. Derives the Net Present Value of Cashflows, the Internal rate of return and other metrics. - Integrates the NPV's over the reserve distribution range to derive the Expected Present value. - Performs decision tree analysis based on the probability of the exploration well being successful. - Presents the investment case to management. Create a theoretical cashflow based on assumptions known to date Monte Car o reserve simulation: results and input parameter summary f. W V a Recoverable hydrocarbon Volumetric parameters 0WC7GWC depth (m) thickness (w) Reservoir GRV area (km1) (10* m3} Petrophylslcal parameters Are» N/G PVT parameters Reservoir Pressure (MPa) Kesurvuir Tenpuralu* Factor (Srr'/Rrri') Hi II (I ev«lupiii«nt Ii.II.um-1«: ■ r. Recovery fsctor M11-1 Preliminary results Minimum 78.13 2000.01 18.26 8.002 148.12 9.62 20.16 i.i.i.:m» 1.00 46.08 9/.00 322.00 0.601 M■ I I il ■ I W4 00 'KM III 25.29 8.0/0 224.8'j 12.23 30.15 i.'-.H', t.oo lt.. .inj t/.OO 322.00 »..'.il 5000 GAS ■ i ■ a I ■ I < - M.i i im i Ml 338.45 204$ M 39.// 11.1/1 412.92 14.09 39./0 Z9.85 t.00 I'.i.l.iK m/.00 322.00 0.849 124.60 P5o 165.48 P10 223.34 2804.81. 2824.61 ill I,. 21./9 2/.01 34.13 8.15« I 19322 •317 245.14 10.192 316.0«; 10.66 24.56 12.02 29.9/ 13.19 ! 35.48 64.52 /0.03 Z5.45 1.00 1.00 1.00 46.08 46.08 46.08 9/.00 9/.00 9/.00 322.00 322.00 322.00 0.660 0./14 0790 V ( f fl ■ K I a » #t k «j is a r r ftl 1^ »i et W o -f I« <ö Recoveiolil« lly.lioc.iilioii (l>cl MMIil.ll JO n O 3 £ o too 95 90 85 80 75 n 65 M 56 Ml il 40 30 25 20 15 10 5 ii 78.0 1 F f nun tleiisay MOSI I II- -1', (Model O l'i ii.-i-ii il —•— Pi nli.ilil- 500 0 EPV = $1,870 mm @10% Discount rate 1-^ -1 0 10 20 30 40 50 60 70 80 90 100 % Probability Reserves are less than I\Lb_If thp fiplH is viahlp n\/pr thp pntirp rangp thpn asRiimft thp NPV nf thp 50% case equals the EPV Decision Tree Analysis Cost of Exploration Well =$50 mm Probability of finding hydrocarbon = 20% 0.2 *(1870-50)< = 364 Probability of oil case =100% EPV $1,870m Probability of gas case = 0% Oil Case Min NPV $800mm MeanNPV $1,870mm Max NPV $2,900mm EPV $1,870mm Gas Case (not evaluated) 0.8 *(- 50) = -40 $324 mm Probability of not finding hydrocarbon = 80% Dry Well This is called the Expected Monetary Value (EMV) at the discount rate used. 7K Risked Rate of Return EMVvs Discount rate 3,000 2,500 2,000 1 1,500 I 1,000 LU 500 0 -500 EMV @ 10% Discount Rate = $ 324 mm ^ / Risked Rate of F (eturn = 15% ^^^^^^^^^^^^^^^^^ i i i i i i i i i i i i i i i 0 5 10 15 20 Discount Rate % Exploration Proposal 'It is recommend that the company drill an exploration well on the prospect at a cost of $50mm. The probability of discovering oil is 20% (in in 5). The mean discovery case has a recoverable reserves level of 900 million barrels of oil and a NPV @ 10% discount rate of $l,900mm. Risked exploration economics indicate an Expected Monetary value of $324mm @ 10% discount rate and a Risked Rate of Return of 15%/

100 n o Q. CO CM c re 4-i (/) c o o 80 60 40 20 Excess i| Supply rkJL Syrian Conflict OPEC PdVSA _ cuts worker's production strike in Venezuela I I I I i i i i i SSSSSSSSSS00000000CDa)CD0000tD(J)0)a)0>0)(J)0)m0)O)OOOOOOOOOOT-T-rT-T-OOOOOO0>OOOOOCDOOOOCnOC>OOOOOOOOOOOOOOOOOOO T-T-T-^T-^T-T-T-^T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-T-(\|(NCNCNCNCN|(N(NOJCS|NCS|CN|OJCN OPEC formed in 1960s to break the power of the "Seven Sisters" First attempt at intervention was in 1967 during the Arab-Israeli conflict The oil market has been significantly out of balance Figure 3.3 Global demand / supply balance 102 100 98 96 94 92 90 88 86 84 82 J_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_L 2.5 2.0 Implied Stock - i Ch.&Misc to i. j Bal (RHS) 1.0 0.5 ^ Oil Demand 0.0 -Ü.5 -1.0 -1.5 Oil Supply -2.0 - -2.5 2004 MS 2006 2007 2003 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Supply and demand have seen a significant mismatch over the past three years, mainly sue to rising supply The change in stocks is a critical issue - if they are rising then there is too much oil in the market ®Mi At present stocks are close to record highs Significant Non-OPEC supply potential exists, especially in the US Figure 2.6 Selected sources of non-OPEC supply changes, 2015-21 • The rise of US shale is the most important factor in the oil market at present • The flexibility of output, and its responsiveness to price, is a very new phenomenon • Other producers with longer-term investment horizons are struggling to ^yy^ react OPEC manoeuvres since 2014 _Q _Q 160.00 140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00 Saudi decides to compete with US shale Signs of US production decline OPEC meeting fails to reach agreement in Doha Oil price collapses to $27- calls for OPEC meeting OPEC and non-OPEC renew agreement for 2018 OPEC agrees to cut output and co-operate with z\ non-OPEC > # ^ A* A# A* A* The rise of US shale has raised questions about the continuing relevance of OPEC Saudi Arabia decided to compete for market share, to force out higher-cost producers However, the strategy was not very successful - OPEC + Russia have been forced to curb production to encourage and oil price recovery Falling oil price = lower cashflow = lower investment Capital expenditure declines slowed and cash from operations increased from the second quarter of 2016 as crude oil prices stabilized cash flow items and Brent price billion 2016S; Brent in 2016 $/b 180 160 140 120 100 80 60 40 20 C cash from operations Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 201J 2012 2013 2014 2015 2016 Source. U.S. Energy Information Administration, Evaluate Energy, Bloomberg Note: b-barrel Companies have dramatically cut back investment in oil exploration and development over the past two years This will inevitably lead to a slowdown in supply - a classic commodity cycle The key question is whether there will be a supply crunch and a price spike, and what impact this might have for the longer term ^(J|j^ Oil products and refining capacity are also important Figure 4.1 Changes in regional demand and refining capacity Demand Capacity OECD Arnikas OECD Europe OECD Asia FSU Oceania China Other Asia Mon-OECD Middle East Africa Americas • Lower oil prices encourage higher refining margins as well as demand growth • Refining capacity expansion is focused on developing markets in Asia and the Middle East • Oil product prices move in tandem with crude prices, but tend to provide e^tiia profit when oil prices are low Downstream Oil Value Chain GASOLINE DISTRIBUTION SYSTEM AND VALUATION FLOWS ^*-| npTB CRUDE CHL The downstream oil business Refining margins (US$/bbl) USGC MeoLm Sour Coking i NWE Light Sweet Cracking Sngapore Mcoum Sour Hydrocracking Refining margins have risen as crude prices have fallen 06 07 09 10 Refining utilisation (% capacity Refinery utilisation is a critical factor in oil economics -below 80% is a bad sign North America I S. & Cent. America I Europe I CIS The Gas Commercial Chain - Pricing & Risks Regulated/ Market Price Production Contract Price Physical Flow - Volume Risk ■^nys Revenue Flow - Price Risk Quality / Credit / Contract Risks 1 Gas Strategies 74 7m Gas Market Evolution - Away from long-term contracts to market-based pricing ü c CD UJ H—' CD CO c C/D CO CD o Non-competitive market Merchant pipes "Strategic" relationships No consumer choice Supply security Competitive market Mature market Stage of market development Intensive growth Initial growth Competitive supply Regulated transport Consumer choice Security from portfolios & futures markets Basis-priced transportation Storage, load balancing & services competitive © Long-term contracts © Time Short-term contracts Spot/forward deals Pricing mechanism's development stages: Futures trading (7) - cost-plus or market related based on alternative fuel prices (2) - escalation formulas, based on either alternative fuel prices or gas markets (?) - based on traded prices and futures prices (commodities markets) 1 Gas Strategies Source: Gas Strategies 75 7ÜK Historically regional pricing has been prevalent US Henry Hub Average German Import Price cif UKNBP Japan LNG cif Fukushima disaster 99 00 01 02 03 04 05 06 07 08 10 11 12 13 14 15 16 0 For many years prices in different regions were close, despite limited interconnectivity A supply-demand imbalance from 2010 saw a huge disparity emerge, with Asia paying a significant premium Global gas prices since 2012 20.00 Global gas prices have started to converge for four key reasons: - Less demand growth than expected in Europe (decline) and Asia (slower growth) - Increasing prevalence of LNG, which connects markets - A growing oversupply of gas - The availability of US LNG exports, which has introduced a new market-based pricing mechanism 7m E E (0 o u (0 0 (/) 0 DC 0) (0 o (0 Global LNG Supply 2008 - 2030 Existing, Under Construction & FID'd 600 500 400 R 300 u CO 200 100 2008 2010 Source: Author's Assumptions 2015 2020 2025 2030 > □ ©1^ European Balance - Low Asian LNG & European Gas JIK Demand Case 2015 - 2030 u CO 800 700 600 500 400 300 200 100 New LNG LNG 'Glut' Supply Needed 5 Years 2015 -100 -200 LNG Available for Europe ■ Russian Pipeline Gas 50 bcma minimum) Other Pipeline Ga Domestic Production (including Norway) 2020 2025 T-1 2030 LNG 'Glut' cleared by: • Additional coal to gas switching in Europe. • 'Induced' spot demand in Asia. • Reduced US LNG send-out ©M, mz Indicative Price Paths -Low Asian Demand Scenarios 12 Brent 2 0 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 — — Henry Hub (Forward Curve) _ — NBP (Forward Curve) « « Russia Contract Price — « Russia Contract Price (with Concessions) Japanese Average Import Price Henry Hub Europe does not need Russian Gas above 150 bcma until 2023. System needs new LNG beyond current supply under development in 2027, so prices rise to LRMC by then. Gazprom's pipeline supplies to Europe are a significant competitive threat to LNG producers Queflwenlwf.glBcncHixhcrehen AUTIV • Gazprom has surplus production potential in West Siberia • It has a very low delivered cost in Europe • Russia is essentially the Saudi Arabia of the gas market - its actions can determine price and volume for competitors Coal prices (US$/t) show what happens when a fuel is in decline Northwest Europe marker price US Central Appalachian coal spot price index Japan steam spot cif price China Qinhuangdao spot price 16 0 Coal prices have collapsed in the face of increasing environmental challenges In particular US coal producers have been put under pressure by shale gas Elsewhere, countries are questioning how much coal they can afford to burn. Unfortunately, a lower prices also stimulated demand The Gas versus Coal dilemma in Europe 14.00 12.00 3 10.00 "I 8.00 I 6.00 => 4.00 2.00 0.00 (N (N m m m m LT) LT) ~ Q. C >~ Q. C >~ Q. C >~ Q. C >~ Q. C >~ Q. 03 03 cu 03 03 CD 03 03 cu 03 03 cu 03 03 cu 03 03 cu 00 00