national - npc · · 2006-02-28national petroleum council phone: (202) 393-6100 fax: (202)...

NATIONAL PETROLEUM COUNCIL

Phone: (202) 393-6100Fax: (202) 331-8539

An Oil and Natural Gas Advisory Committee to the Secretary of Energy

1625 K Street, N.W.Washington, D.C. 20006-1656

September 25, 2003

The HonorableE. Spencer AbrahamSecretary of EnergyWashington, D.C. 20585

Dear Mr. Secretary:

On behalf of the members of the National Petroleum Council, I am pleased to submit to you theCouncil’s report on natural gas: Balancing Natural Gas Policy – Fueling the Demands of a Growing Economy.This report was prepared at your request to provide insights on energy market dynamics as well as advice onactions that can be taken by industry and government to ensure adequate and reliable supplies of energy forAmerican consumers. We further recognize the importance of your request and the urgency of ourrecommendations, given the heightened sensitivity among consumers to energy costs and reliability.

Natural gas continues to be a vital source of energy and raw material, and will play an important role in achieving the nation’s economic and environmental quality goals. The Council finds that recentfundamental shifts in North American natural gas markets have led to the current market conditions ofhigher gas prices and increased price volatility. This situation will likely persist and could deteriorate unlesspublic policy makers act now to reduce the conflicts that are inherent in current public policies.

Clearly, the recent tightening of the natural gas supply/demand balance places greater urgency onaddressing the future of this important energy source and resolving conflicting policies that favor natural gasusage, but hinder its supply. The Council has reached out to hundreds of experts in the public and privatesectors, representing both suppliers and consumers, to analyze future supply, demand, and infrastructurerequirements, in order to advance recommendations that we believe will equip local, state, and nationalpolicy makers to make sound and balanced decisions for the future.

The Council recommends a balanced portfolio of actions by industry and government that includes:

• Encouraging conservation and efficiency

• Improving demand flexibility and efficiency

• Increasing supply diversity

• Sustaining and enhancing infrastructure

• Promoting efficiency of markets.

The Hon. E. Spencer AbrahamSeptember 25, 2003Page Two

Policies most likely to have an immediate impact are actions to promote consumer conservation and energy efficiency. Actions to increase supply diversity and demand flexibility should also be takenimmediately because several years will elapse before their full impacts will be felt. The Council'srecommended approach includes action in all of these interdependent areas, because neither increasingsupplies nor improving the efficiency of gas consumption would alone be sufficient to achieve the country’sgoals. It is vital to accomplish both.

The Council further recommends that the Department of Energy schedule a series of workshopsdesigned to review steps taken to implement the report’s recommendations, and to monitor the implicationsof ongoing changes in market conditions.

The Council is available to discuss further the results of this report and to aid in the implementationof its recommendations.

Respectfully submitted,

Bobby S. ShackoulsChair


Bobby S. Shackouls, ChairLee R. Raymond, Vice Chair

Marshall W. Nichols, Executive Director

U.S. DEPARTMENT OF ENERGY

Spencer Abraham, Secretary

The National Petroleum Council is a federaladvisory committee to the Secretary of Energy.

The sole purpose of the National Petroleum Council is to advise, inform, and make recommendations

to the Secretary of Energy on any matter requested by the Secretary

relating to oil and natural gas or to the oil and gas industries.

All Rights ReservedLibrary of Congress Control Number: 2003113298

© National Petroleum Council 2003Printed in the United States of America

INTEGRATED REPORT – TABLE OF CONTENTS i

TABLE OF CONTENTSINTEGRATED REPORT

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Study Request . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Study Organization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Study Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Study Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Retrospectives on 1999 Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Chapter One: Integrated Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Natural Gas Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Natural Gas Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Infrastructure Expansions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Potential Price Ranges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Capital Expenditures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Chapter Two: Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Role of Natural Gas in the Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Natural Gas Market History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Natural Gas Prices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Major Assumptions/Methodologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Macroeconomic Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Residential and Commercial Demand . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Industrial Gas Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Electric Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Transportation and Storage Infrastructure . . . . . . . . . . . . . . . . . . . . . . . 25

Chapter Three: Natural Gas Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Study Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Demand Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

INTEGRATED REPORT – TABLE OF CONTENTSii

Energy, Natural Gas, and the Economy . . . . . . . . . . . . . . . . . . . . . . . . . 28

Gas Demand for Electric Power Generation . . . . . . . . . . . . . . . . . . . . . 28

Industrial Natural Gas Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Residential and Commercial Natural Gas Demand . . . . . . . . . . . . . . . . 32

Macroeconomic Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Natural Gas Price and Oil Substitution . . . . . . . . . . . . . . . . . . . . . . . . . 35

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Industrial Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Value of Gas to Industrial Consumers . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Industrial Natural Gas Use Drivers and Trends . . . . . . . . . . . . . . . . . . . 39

Overview of the Modeling Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Projection of Future Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Major Gas-Intensive Industries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Petroleum Refining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Primary Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Stone, Clay, and Glass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Food and Beverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Other Primary Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Other Industries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Additional Policy Issues of Industrial Consumers . . . . . . . . . . . . . . . . . 71

Residential and Commercial Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Residential Consumers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Commercial Consumers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Natural Gas Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Regional Considerations in Residential and Commercial Demand . . . . 79

Efficiency in Residential and Commercial Consumption . . . . . . . . . . . 79

Summary of Residential and Commercial Demand . . . . . . . . . . . . . . . 81

Electric Power Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

Study Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Electric Power Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Generation Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Alternate Fuel Capability and Fuel Substitution . . . . . . . . . . . . . . . . . . 90

Assumptions and Case Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Reactive Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Balanced Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Detailed Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

New Build Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Environmental Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Sensitivities and Scenarios Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Other Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Power Markets and Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Renewable Power Generation Technologies . . . . . . . . . . . . . . . . . . . 98

Conclusions, Recommendations, and Uncertainty Discussion . . . . . . . 99

Comparison with 1999 Study Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

Recommendations Related to Natural Gas Demand . . . . . . . . . . . . . . . . . . 102

Encourage Increased Efficiency and Conservation through Market-Oriented Initiatives and Consumer Education . . . . . . . . . . 102

Increase Industrial and Power Generation Capability to Use Alternate Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Additional Demand Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Chapter Four: Natural Gas Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Study Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Summary of Supply Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Overall Supply Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Resource Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Production Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Development Costs and Technology Improvements . . . . . . . . . . . . . . 115

Commercial Resource . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

North American Production Outlook . . . . . . . . . . . . . . . . . . . . . . . . . 118

U.S. Lower-48 Production Outlook . . . . . . . . . . . . . . . . . . . . . . . . 119

Canadian Production Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

Drilling Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

Reserves Development and Resource Replacement . . . . . . . . . . . . . . . 124

Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

Arctic Gas and LNG Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

Resource Base Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Resource Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Assessment Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Technical Resource Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Technical Resources of the United States, Canada, and Mexico . . . . . 137

United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Resource Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Proved . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Undiscovered . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Main Conclusions from Super-Region Comparison . . . . . . . . . . . . . . 143

Cost Methodology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Gulf of Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Lower-48 Onshore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Alaska – Onshore and Offshore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Atlantic Offshore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Pacific Offshore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Western Canada Onshore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

INTEGRATED REPORT – TABLE OF CONTENTS iii

Canada Offshore and Onshore Other . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Nonconventional Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Rig Fleet Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Production Performance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

U.S. and Canadian Production Overview . . . . . . . . . . . . . . . . . . . . . . . 150

U.S. Lower-48: Activity vs. Production . . . . . . . . . . . . . . . . . . . . . . . . . 150

Western Canada: Activity vs. Production . . . . . . . . . . . . . . . . . . . . . . . 154

Individual Gas Well Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Estimated Ultimate Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Initial Production Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Initial Decline Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

Base Decline Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

Proved Reserves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Regional Production Summaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Declining Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Holding Steady/Slightly Increasing Regions . . . . . . . . . . . . . . . . . 163

Increasing Production Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

2000-2001 Drilling Campaign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Model Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

Technology Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Technology Subgroup Process for the Study . . . . . . . . . . . . . . . . . . . . 171

Historical Perspective of Technology Contributions . . . . . . . . . . . . . . 172

Projected Technology Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . 172

Issues and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Key Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

Natural Gas Hydrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Synthetic Gas/Coal Gasification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Access Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Expert Teams for Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Rocky Mountain Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Prior Reports – Federal Lease Stipulations . . . . . . . . . . . . . . . . . . 178

Conditions of Approval – Post Leasing . . . . . . . . . . . . . . . . . . . . . 178

Methodology – Green River, Uinta/Piceance, Powder River,San Juan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

Analysis Extension – Wyoming Thrust Belt . . . . . . . . . . . . . . . . . . 182

Permitting Pace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Offshore United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Access Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Sensitivity Case Summaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Sensitivity Modeling Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Sensitivity Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Summation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

INTEGRATED REPORT – TABLE OF CONTENTSiv

LNG Imports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

LNG Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

LNG Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

LNG Global Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

LNG Global Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

Model Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

Case Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

Reactive Path Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194

Balanced Future Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

Low Sensitivity Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

Controlling Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

Case Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197


Arctic Developments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200

Canadian Arctic Gas Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200

Resource . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200

Current Status of Project Development . . . . . . . . . . . . . . . . . . . . . 201

Risks and Hurdles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

Alaska Arctic Gas Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

Resource . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

Attempts to Commercialize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

Current Status of Project Development . . . . . . . . . . . . . . . . . . . . . 202

Risks and Hurdles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

Arctic Supply Assumptions for NPC Study . . . . . . . . . . . . . . . . . . . . . 205

Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206


Comparison to Other Supply Outlooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Energy Information Administration Preliminary Annual Energy Outlook 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Lower-48 Total Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Technical Resource Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Reserve Additions per Gas Well . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Drilling Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

Lower-48 Regional Production Response . . . . . . . . . . . . . . . . . . . 213

LNG, Arctic Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

1999 NPC Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

Total Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

Technical Resource Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

Recovery per Well . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

Drilling Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

INTEGRATED REPORT – TABLE OF CONTENTS v

Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

Lower-48 Regional Production Response . . . . . . . . . . . . . . . . . . . 225

Supply Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Onshore – Increase Access (Excluding Wilderness Areas and National Parks) and Reduce Permitting Costs/Delays 50% over Five Years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Offshore – Lift Moratoria on Selected Areas of the Federal OCS by 2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Arctic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

LNG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

Additional Supply Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Chapter Five: Transmission, Distribution, and Storage Infrastructure . . . . . 231

Study Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

Summary of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

Results from the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

Future Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

Scenarios and Sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

Challenges to Building and Maintaining the Required Transmission Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242

Contracting Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242

Contracting New Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

Construction Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247

Operations and Maintenance of Existing Infrastructure . . . . . . . . . . . 249

Operational Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Maintenance Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

Distribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252

Distribution Infrastructure Investment . . . . . . . . . . . . . . . . . . . . . . . . 253

Challenges to Building and Maintaining the Required Distribution Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

Reliable Gas Service in a Changing Market . . . . . . . . . . . . . . . . . . . . . 256

Productivity Improvements Require R&D Investments . . . . . . . . . . . . 257

Storage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258

Depleted Oil and Gas Reservoirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

Aquifers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260

Salt Caverns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260

LNG Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260

Propane-Air Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260

Historical Background and Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . 261

Results from the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

The Outlook for Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

Challenges to Building and Maintaining the Required Storage Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

INTEGRATED REPORT – TABLE OF CONTENTSvi

Comparison to Other Transportation Outlooks . . . . . . . . . . . . . . . . . . . . . 266

Transmission, Distribution, and Storage Recommendations . . . . . . . . . . . 267

Sustain and Enhance Natural Gas Infrastructure . . . . . . . . . . . . . . . . . 267

Chapter Six: Mexico Supply/Demand Outlook . . . . . . . . . . . . . . . . . . . . . . . . 269

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

Economic Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270

Natural Gas Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

Natural Gas Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

Current Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

Resource and Reserves Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

Natural Gas Import/Export Balance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

Potential LNG Import Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276

Pipeline Interconnection Capacities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

LNG Import Terminals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

Strategic Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278

Chapter Seven: Natural Gas Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Market-Related Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Price Transparency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Physical Market Liquidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Financial Market Liquidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Natural Gas Price Volatility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Price Volatility in the North American Market . . . . . . . . . . . . . . . . . . 284

Volatility Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284

Historical Natural Gas Prices and Volatility . . . . . . . . . . . . . . . . . . . . . 285

Comparison of Natural Gas Price Volatility vs.Crude Oil and Electricity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

Key Drivers of Natural Gas Price Volatility . . . . . . . . . . . . . . . . . . . . . 289

Supply and Demand Elasticity Effects . . . . . . . . . . . . . . . . . . . . . . 289

Pipeline Capacity and Operational Factors . . . . . . . . . . . . . . . . . . 290

Lack of Timely, Reliable Information . . . . . . . . . . . . . . . . . . . . . . . 291

Alternate Fuel Price Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

Factors that Mitigate Gas Price Volatility . . . . . . . . . . . . . . . . . . . . . . . 292

Exposure to Price Volatility and Its Effects . . . . . . . . . . . . . . . . . . . . . . 292

Retail and Commercial Customers . . . . . . . . . . . . . . . . . . . . . . . . 292

Industrial Customers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

Electric Power Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

Natural Gas Producers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

Chapter Eight: Capital Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

Capital Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

Financial Assumptions, Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 295

Exploration and Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

Natural Gas Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

INTEGRATED REPORT – TABLE OF CONTENTS vii

INTEGRATED REPORT – TABLE OF CONTENTSviii

Chapter Nine: Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299

Factors Related to the Economic Environment . . . . . . . . . . . . . . . . . . . . . . 301

Demand-Related Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301

Electricity Sales as a Function of Economic Growth . . . . . . . . . . . . . . 301

Industrial/Power Generation Fuel Switching . . . . . . . . . . . . . . . . . . . . 301

Existing Fossil-Fuel Generating Capacity . . . . . . . . . . . . . . . . . . . . . . . 301

New Fossil-Fuel Generating Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . 301

Nuclear Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Renewable Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Industrial Production and Energy Intensity . . . . . . . . . . . . . . . . . . . . . 302

Residential/Commercial Energy Efficiency . . . . . . . . . . . . . . . . . . . . . . 302

Domestic Supply-Related Factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Resource Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Resource Base Sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Upstream Technological Advancements . . . . . . . . . . . . . . . . . . . . . . . . 303

LNG Assumptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

Assumptions for Pipeline Construction. . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

Construction of the NPC Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304

Summary Results of the NPC Cases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305

Further Information on Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310

Chapter Ten: Modeling Methodologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

The EEA Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

Gas Market Data and Forecasting System . . . . . . . . . . . . . . . . . . . . . . 312

Hydrocarbon Supply Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312

The Altos Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313

The NARG Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313

The NARE Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

Modeling Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

Resources and Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

Pipelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

Further Altos Modeling Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

Appendices

Appendix A: Request Letter and Description of the NPC . . . . . . . . . . . . . . A-1

Appendix B: Study Group Rosters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1

Acronyms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AC-1

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GL-1

Study Request

By letter dated March 13, 2002, Secretary of EnergySpencer Abraham requested the National PetroleumCouncil (NPC) to undertake a new study on naturalgas in the United States in the 21st Century.Specifically, the Secretary stated:

Such a study should examine the potential impli-cations of new supplies, new technologies, newperceptions of risk, and other evolving marketconditions that may affect the potential for natu-ral gas demand, supplies, and delivery through2025. It should also provide insights on energymarket dynamics, including price volatility andfuture fuel choice, and an outlook on the longer-term sustainability of natural gas supplies. Ofparticular interest is the Council’s advice onactions that can be taken by industry andGovernment to increase the productivity and effi-ciency of North American natural gas marketsand to ensure adequate and reliable supplies ofenergy for consumers.

In making his request, the Secretary made referenceto the 1992 and 1999 NPC natural gas studies, andnoted the considerable changes in natural gas marketssince 1999. These included “new concerns over nation-al security, a changed near-term outlook for the econ-omy, and turbulence in energy markets based onperceived risk, price volatility, fuel-switching capabili-ties, and the availability of other fuels.” Further, theSecretary pointed to the projected growth in thenation’s reliance on natural gas and noted that thefuture availability of gas supplies could be affected by“the availability of investment capital and infrastruc-

ture, the pace of technology progress, access to theNation’s resource base, and new sources of suppliesfrom Alaska, Canada, liquefied natural gas imports,and unconventional resources.” (Appendix A containsthe complete text of the Secretary’s request letter and adescription of the NPC.)

Study Organization

In response to the Secretary’s request, the Councilestablished a Committee on Natural Gas to undertakea new study on this topic and to supervise the prepara-tion of a draft report for the Council’s consideration.The Council also established a CoordinatingSubcommittee and three Task Groups – on Demand,Supply, and Transmission & Distribution – to assist theCommittee in conducting the study.

Bobby S. Shackouls, Chairman, President and ChiefExecutive Officer, Burlington Resources Inc., chairedthe Committee1, and Robert G. Card, Under Secretaryof Energy, served as the Committee’s GovernmentCochair. Robert B. Catell, Chairman and ChiefExecutive Officer, KeySpan Corporation; Lee R.Raymond, Chairman and Chief Executive Officer,Exxon Mobil Corporation; and Richard D. Kinder,Chairman and Chief Executive Officer, Kinder MorganEnergy Partners, L.P., served as the Committee’s ViceChairs of Demand, Supply, and Transmission &Distribution, respectively. Jerry J. Langdon, ExecutiveVice President and Chief Administrative Officer,

INTEGRATED REPORT - PREFACE 1

PREFACEINTEGRATED REPORT

1 William A. Wise, Retired President and Chief Executive Officer, El Paso Energy Corp., served as Chair of the Committee until May 16, 2003.

Reliant Resources, Inc., chaired the CoordinatingSubcommittee, and Carl Michael Smith, AssistantSecretary, Fossil Energy, U.S. Department of Energy,served as Government Cochair.

The members of the various study groups weredrawn from the NPC members’ organizations as wellas from many other industries, non-governmentalorganizations, and government organizations. Thesestudy participants represented broad and diverse inter-ests including large and small producers, transporters,service providers, financers, regulators, local distribu-tion companies, power generators, and industrial con-sumers of natural gas. Appendix B contains rosters ofthe study’s Committee, Coordinating Subcommittee,three Task Groups and their subgroups. In addition tothe participants listed in Appendix B, many more peo-ple were involved in regional and sector-specific work-shops in the United States and Canada.

Study Approach

The study benefited from an unprecedented degreeof support, involvement, and commitment from thegas industry. The breadth of support was based ongrowing concerns about the adequacy of natural gassupplies to meet the continuing strong demand forgas, particularly in view of the role of gas as an envi-ronmentally preferred fuel. The study addresses boththe short-term and long-term outlooks (through2025) for North America, defined in this study as con-sisting of Canada, Mexico, and the United States. Thereader should recognize that this is a natural gas study,and not a comprehensive analysis of all energy sourcessuch as oil, coal, nuclear, and renewables. However,this study does address and make assumptions regard-ing these competing energy sources in order to assessthe factors that may influence the future of natural gasuse in North America. The analytical portion of thisstudy was conducted over a 12-month period begin-ning in August 2002 under the auspices of theCoordinating Subcommittee and three primary TaskGroups.

The Demand Task Group developed a comprehen-sive sector-by-sector demand outlook. This analysiswas done by four subgroups (Power Generation,Industrial Utilization, Residential and Commercial, andEconomics and Demographics). The task of eachgroup was to try to understand the economic and envi-ronmental determinants of gas consumption and toanalyze how the various sectors might respond to dif-

ferent gas price regimes. The Demand Task Group wascomposed of representatives from a broad cross-sectionof the power industry as well as industrial consumersfrom gas-intensive industries. It drew on expertisefrom the power industry to develop a broad under-standing of the role of alternative sources for generatingelectric power based on renewables, nuclear, coal-fired,oil-fired, or hydroelectric generating technology. It alsoconducted an outreach program to draw upon theexpertise of power generators and industrial consumersin both the United States and Canada.

The Supply Task Group developed a basin-by-basinsupply picture, and analyzed potential new sources ofsupply such as liquefied natural gas (LNG) and Arcticgas. The Supply Task Group worked through five sub-groups: Resource, Technology, LNG, Arctic, andEnvironmental/Regulatory/Access. Over 100 peopleparticipated. These people were drawn from majorand independent producers, service companies, con-sultants, and government agencies. These workinggroups conducted 13 workshops across the UnitedStates and Canada to assess the potential resourcesavailable for exploration and development.Workshops were also held to examine the potentialimpact on gas production from advancing technology.Particular emphasis was placed on the commercialpotential of the technical resource base and the knowl-edge gained from analysis of North American produc-tion performance history.

The Transmission & Distribution Task Group ana-lyzed existing and potential new infrastructure. Theiranalysis was based on the work of three subgroups(Transmission, Distribution, and Storage). Industryparticipants undertook an extensive review of existingand planned infrastructure capacity in NorthAmerica. Their review emphasized, among otherthings, the need to maintain the current infrastructureand to ensure its reliability. Participants in theTransmission & Distribution Task Group includedrepresentatives from U.S. and Canadian pipeline, stor-age, marketing, and local distribution companies aswell as from the producing community, the FederalEnergy Regulatory Commission, and the EnergyInformation Administration.

Separately, two other groups also provided guidanceon key issues that crossed the boundaries of the primarytask groups. An ad hoc financial team looked at capitalrequirements and capital formation. Another teamexamined the issue of increased gas price volatility.

INTEGRATED REPORT - PREFACE2

Due to similarities between the Canadian and U.S.economies and, especially, the highly interdependentcharacter of trade in natural gas, the evaluation ofnatural gas supply and demand in Canada and theUnited States were completely integrated. The studyincluded Canadian participants, and many other par-ticipating companies have operations in both theUnited States and Canada. For Mexico, the evaluationof natural gas supply and demand for the internalmarket was less detailed, mainly due to time limita-tions. Instead, the analysis focused on the net gastrade balances and their impact on North Americanmarkets.

As in the 1992 and 1999 studies, econometric mod-els of North American energy markets and other ana-lytical tools were used to support the analyses.Significant computer modeling and data support wereobtained from outside contractors; and an internalNPC study modeling team was established to takedirect responsibility for some of the modeling work.The Coordinating Subcommittee and its Task Groupsmade all decisions on model input data and assump-tions, directed or implemented appropriate modifica-tions to model architecture, and reviewed all output.Energy and Environmental Analysis, Inc. (EEA) ofArlington, Virginia, supplied the principal energy mar-ket models used in this study, and supplemental analy-ses were conducted with models from AltosManagement of Los Altos, California.

The use of these models was designed to give quan-tified estimates of potential outcomes of natural gasdemand, supply, price and investment over the studytime horizon, with a particular emphasis on illustrat-ing the impacts of policy choices on natural gas mar-kets. The results produced by the models are criticallydependent on many factors, including the structureand architecture of the models, the level of detail ofthe markets portrayed in the models, the mathemati-cal algorithms used, and the input assumptions speci-fied by the NPC Study Task Groups. As such, theresults produced by the models and portrayed in theNPC report should not be viewed as forecasts or asprecise point estimates of any future level of supply,demand, or price. Rather, they should be used as indi-cators of trends and ranges of likely outcomes stem-ming from the particular assumptions made. Inparticular, the model results are indicative of the like-ly directional impacts of pursuing particular publicpolicy choices relative to North American natural gasmarkets.

This study built on the knowledge gained andprocesses developed in previous NPC studies,enhanced those processes, created new analyticalapproaches and tools, and identified opportunities forimprovement in future studies. Specific improvementsincluded the following elements developed by theSupply Task Group:

� A detailed play-based approach to assessment ofthe North American natural gas resource base,using regional workshops to bring together indus-try experts to update existing assessments. Thiswas used in two detailed descriptive models, onebased on 72 producing regions in the United Statesand Canada, and the other based on 230 supplypoints in the United States, Canada, and Mexico.Both models distinguished between conventionaland nonconventional gas and between provedreserves, reserve growth, and undiscoveredresource.

� Cost of supply curves, including discovery processmodels, were used to determine the economicallyoptimal pace of development of North Americannatural gas resources.

� An extensive analysis of recent production perform-ance history, including virtually every gas well drilledin the United States and Canada between 1990 and2002. This analysis clearly identified basins that arematuring and those where production growth poten-tial remains, and helped establish forward-lookingassumptions in the models.

� A model to assess the impact of permitting in areascurrently subject to conditions of approval.

� A first-ever detailed NPC view and analysis of LNGand Arctic gas potential.

The Demand Task Group also achieved significantimprovements over previous study methods. Theseimprovements include the following:

� Regional power workshops and sector-specificindustrial workshops to obtain direct input on con-suming trends and the likely impact of changing gasprices.

� Ongoing detailed support from the power industryfor technology and cost factors associated with cur-rent and future electric power generation.

� Development of a model of industrial demandfocusing on the most gas-intensive industries andprocesses.


Study Report

Results of this 2003 NPC study are presented in amulti-volume report as follows:

� Volume I, Summary of Findings and Recommenda-tions, provides insights on energy market dynamicsas well as advice on actions that can be taken byindustry and government to ensure adequate andreliable supplies of energy for American consumers.It includes an Executive Summary of the report andan overview of the study's analyses and recommen-dations.

� Volume II, Integrated Report, contains discussions ofthe results of the analyses conducted by the threeTask Groups: Demand, Supply, and Transmission &Distribution. This volume provides further sup-porting data and analyses for the findings and rec-ommendations presented in Volume I. It addressesthe potential implications of new supplies, newtechnologies, new perceptions of risk, and otherevolving market conditions that may affect thepotential for natural gas demand, supplies, anddelivery through 2025. It provides insights on ener-gy market dynamics, including price volatility andfuture fuel choice, and an outlook on the longer-term sustainability of natural gas supplies. It alsoexpands on the study’s recommended policy actions.This volume presents an integrated outlook for nat-ural gas demand, supply, and transmission in theUnited States, Canada, and Mexico under two pri-mary scenarios and a number of sensitivity cases.

The demand analysis provides an understanding ofthe economic and environmental determinants ofnatural gas consumption to estimate how the indus-trial, residential/commercial, and electric power sec-tors may respond under different conditions. Thesupply analysis develops basin-by-basin resourceand cost estimates, presents an analysis of recentproduction performance, examines potential tech-nology improvements, addresses resource accessissues, and examines potential supplies from tradi-tional areas as well as potential new sources of sup-ply such as liquefied natural gas and Arctic gas. Thetransmission, distribution, and storage analysis pro-vides an extensive review of existing and plannedinfrastructure in North America emphasizing,among other things, the need to maintain the cur-rent infrastructure and to ensure its reliability.

� Task Group Report Volumes and Appendices includethe detailed data and analyses prepared by the

Demand, Supply, and Transmission & DistributionTask Groups and their Subgroups, which formed thebasis for the development of Volumes I and II. Theoutput of the study's computer modeling activities isalso included. The Council believes that these mate-rials will be of interest to the readers of the reportand will help them better understand the results.The members of the National Petroleum Councilwere not asked to endorse or approve all of the state-ments and conclusions contained in these docu-ments but, rather, to approve the publication ofthese materials as part of the study process.

Included with the Task Group Reports is a CD-ROMcontaining further model output on a regional basis.The CD also contains digitized maps, which wereused in assessing the potential impact of conditionsof approval for access to key Rocky Mountainresource areas.

A form for ordering additional copies of the reportvolumes can be downloaded from the NPC website,http://www.npc.org. Pdf copies of Volumes I and IIalso can be viewed and downloaded from the NPCwebsite.

Retrospectives on 1999 Study

In requesting the current study, the Secretarynoted that natural gas markets had changed substan-tially since the Council’s 1999 study. These changeswere the reasons why the 2003 study needed to be acomprehensive analysis of natural gas supply,demand, and infrastructure issues. By way of back-ground, the 1999 study was designed to test the capa-bility of the supply and delivery systems to meet thethen-public forecasts of an annual U.S. marketdemand of 30+ trillion cubic feet early in this centu-ry. The approach taken in 1999 was to review theresource base estimates of the 1992 study and makeany needed modifications based on performancesince the publication of that study. This assessmentof the natural gas industry’s ability to convert thenation’s resource base into available supply alsoincluded the first major analytical attempt to quanti-fy the effects of access restrictions in the UnitedStates, and specifically the Rocky Mountain area.Numerous government agencies used this work as astarting point to attempt to inventory various restric-tions to development. This access work has been fur-ther expanded upon in the current study. Furtherdiscussions of the 1999 analyses are contained in theTask Group reports.

INTEGRATED REPORT - PREFACE4

The 1999 report stated that growing future demandscould be met if government would address several crit-ical factors. The report envisioned an impending ten-sion between supply and demand that has sincebecome reality in spite of lower economic growth overthe intervening time period. On the demand side, gov-ernment policy at all levels continues to encourage useof natural gas. In particular, this has led to largeincreases in natural gas-fired power generation capaci-ty. The 1999 study assumed 144 gigawatts of newcapacity through 2015, while the actual new capacity isexpected to exceed 200 gigawatts by 2005. On the sup-ply side, limits on access to resources and other restric-tive policies continue to discourage the development ofnatural gas supplies. Examples of this are the 75%reduction in the Minerals Management Service’sEastern Gulf Lease Sale 181 and the federal govern-ment’s “buying back” of the Destin Dome leases off thecoast of Florida.

The maturity of the resource base in the tradition-al supply basins in North America is another signifi-

cant consideration. In the four years leading up to thepublication of this study, North America has experi-enced two periods of sustained high natural gasprices. Although the gas-directed rig count didincrease significantly between 1999 and 2001, theresult was only minor increases in production. Evenmore sobering is the fact that the late 1990s was atime when weather conditions were milder than nor-mal, masking the growing tension between supplyand demand.

In looking forward, the Council believes that thefindings and recommendations of this study are amplysupported by the analyses conducted by the studygroups. Further, the Council wishes to emphasize thesignificant challenges facing natural gas markets and tostress the need for all market participants (consumers,industry, and government) to work cooperatively todevelop the natural gas resources, infrastructure, ener-gy efficiency, and demand flexibility necessary to sus-tain the nation’s economic growth and meetenvironmental goals.


CHAPTER 1 - INTEGRATED OUTLOOK 7

Natural gas is a critical source of energy and rawmaterial, and will play a vital role in achievingthe nation’s economic and environmental

goals. Current higher gas prices are the result of a fun-damental shift in the supply and demand balance.North America is moving to a period in its history inwhich it will no longer be self-reliant in meeting itsgrowing natural gas needs; production from tradi-tional U.S. and Canadian basins has plateaued.Government policy encourages the use of natural gasbut does not address the corresponding need for addi-tional natural gas supplies. A status quo approach tothese conflicting policies will result in undesirableimpacts to consumers and the economy, if notaddressed. Further, a continuation of incremental pol-icy reactions to market events will likely lead to higherenergy costs and increase the potential for economicdislocations of North American industries. The solu-tion is a balanced portfolio of actions that includesincreased energy efficiency and conservation; alternateenergy sources for industrial consumers and powergenerators, including renewables; gas resources frompreviously inaccessible areas of the United States; liq-uefied natural gas (LNG) imports; and gas from theArctic.

While there is considerable uncertainty in any projec-tion, the NPC arrived at this view through fundamentalanalysis of the basic components that make up the bal-ance of supply and demand. Thorough study was con-ducted of the North American indigenous natural gasresource base, the production history of mature NorthAmerican basins, and likely advances in upstream tech-nology, to arrive at an overall view of indigenous supply.This was complemented by a comprehensive review ofthe potential for LNG imports and Arctic gas to supple-

ment that supply. Analyses of demand were similarlyundertaken with particular attention paid to the poten-tial for demand growth for power generation, and fordemand impacts on key industrial, residential, and com-mercial sectors in response to higher gas prices. Thecapability of existing transmission, distribution, andstorage infrastructure as well as requirements for newinfrastructure were also projected based on the outlooksfor supply and demand.

Scenarios

A status quo approach to natural gas policy yieldsundesirable outcomes because it discourages econom-ic fuel choice, new supplies from traditional basins andAlaska, and new LNG terminal capacity. The NPCdeveloped two scenarios of future supply and demandthat move beyond the status quo. Both require signif-icant actions by policy makers and industry stakehold-ers to effect change. These scenarios, “Reactive Path”and “Balanced Future,” are discussed below.

These scenarios were developed by a range of mar-ket participants, including representatives of produc-ers, pipelines, local distribution companies, industrialconsumers, power companies, and government agen-cies. These scenarios bring together the data and

INTEGRATED OUTLOOKCHAPTER 1

The current policy direction – unaltered –will likely lead to difficult conditions fornatural gas, but industries, government, andconsumers will react. Therefore, this studyassumes action by all these parties beyondthe status quo.

analyses of North American supply, demand, andinfrastructure in internally consistent frameworks foranalyzing choices open to the principal stakeholders inNorth American gas over the study time period. Thus,they are not forecasts, per se, and reflect in some areasthe offsetting and/or complementary effects of actionsby suppliers and consumers. For example, certaincombinations of actions may lead to lower demand ina lower-price environment; conversely, the lack ofthose actions could foster higher natural gas demand,despite a higher-price environment.

Each of the two scenarios has different assumptionsregarding key variables related to supply and demand inresponse to public policy choices. These key variablesincluded degrees of access to gas resources, greaterenergy efficiency and conservation, and increased flexi-bility to use fuels other than gas for industry and powergeneration. The two scenarios result in contrastingdemand, supply, infrastructure, and price profiles. Eachscenario assumes a continuation of current standardsfor environmental compliance.

“Reactive Path” assumes continued conflict betweennatural gas supply and demand policies that supportnatural gas use, but tend to discourage supply develop-ment. However, in addition to these broad policies, theassumptions built into this case acknowledge thatresultant higher natural gas prices will likely be reflect-ed in significant societal pressure to allow reasonable,

economically driven choices to occur on both the con-suming and producing segments of the natural gasindustry. In essence, market participants, includingpublic policy makers, “react” to the current situationwhile inherent conflicts continue. The supply responseassumes a considerable amount of success and devia-tion from past trends, evidenced by a major expansionof LNG facilities, construction of Arctic pipelines, anda significant response in lower-48 production fromaccessible areas. The resulting demand level is lowerthan other outlooks including the EIA, with lessupward pressure on the supply/demand balance. Evenwith uncertainty surrounding air quality regulations,there is potential for construction of new, state of theart, fully compliant coal-based generation plants at lev-els that approach the prior coal boom years in the1970s. Together, this scenario implies a degree of suc-cess in supply and demand responses significantlybeyond what has been demonstrated over recent years.

The Reactive Path scenario results in continuedtightness in supply and demand leading to higher nat-ural gas prices and price volatility over the study peri-od. Federal Reserve Board Chairman Alan Greenspanprovided the best characterization of the conflictbetween policy choices in his testimony to the UnitedStates Senate Committee on Energy and NaturalResources: “We have been struggling to reach an agree-able tradeoff between environmental and energy con-cerns for decades. I do not doubt we will continue tofine-tune our areas of consensus. But it is essential that


Reactive Path Scenario

� Public policies remain in conflict, with actionstaken in a reactive mode

� Siting industrial facilities and powerplants con-tinues to favor natural gas due to investmentand regulatory uncertainties

� No additional alternative fuel backup to exist-ing facilities

� Significant new generation capacity includingcoal with firm environmental control, andrenewables

� Access/permitting restrictions to lower-48 pro-duction persist

� Two-year LNG regasification plant permitting;seven new terminals during the study period

� Arctic pipelines built

Balanced Future Scenario

� Public policy more symmetrical, proactive

� Siting new plants is emission performance ori-ented, and more fuel neutral

� Clean air goals met with time, technology, andmarket-based mechanisms; emissions tradingand fuel-switching ability expanded

� Additional new generation capacity includingcoal with firm environmental control, andrenewables

� Access to lower-48 supplies enhanced

� LNG permit timing improved; nine new termi-nals

� Arctic pipelines built


our policies be consistent. For example, we cannot, on theone hand, encourage the use of environmentally desirablenatural gas in this country while being conflicted on larg-er imports of LNG. Such contradictions are resolved onlyby debilitating spikes in price.”

Alternatively, “Balanced Future” is a scenario inwhich government policies are focused on eliminatingbarriers to market efficiencies. This scenario enablesnatural gas markets to develop in a manner in whichimproved economic and environmental choices can bemade by both producers and consumers. On thedemand side, opportunities for conservation, energyefficiency, and fuel flexibility are both authorized andencouraged while adhering to current environmentalstandards. On the supply side, barriers to development

of new natural gas sources are progressively lowered,both for domestic and imported natural gas. Theresult is a market with lower gas prices and volatilitydue to enhanced supply and more flexible demand.This scenario results in a better outcome for NorthAmerican consumers than the “Reactive Path.”

It would be possible to construct many different sce-narios or visions of the future to illustrate the NPCanalysis. For example, neither the Reactive Path nor theBalanced Future scenario reflects the effect of not devel-oping major new LNG import facilities or the Arctic gaspipelines; neither scenario reflects actions that mightseverely limit CO2 emissions or the permitted carboncontent of fuels; and neither scenario attempts to spec-ulate on ground-breaking new technology that could

Demand

Greater energy efficiency and conservation arevital near-term and long-term mechanisms formoderating price levels and reducing volatility.

Power generators and industrial consumers aremore dependent on gas-fired equipment and lessable to respond to higher gas prices by utilizingalternate sources of energy.

Gas consumption will grow, but such growth willbe moderated as the most price-sensitive indus-tries become less competitive, causing someindustries and associated jobs to relocate outsideNorth America.

Infrastructure

Pipeline and distribution investments will average$8 billion per year, with an increasing sharerequired to sustain the reliability of existing infra-structure

Regulatory barriers to long-term contracts fortransportation and storage impair infrastructureinvestment.

Supply

Traditional North American producingareas will provide 75% of long-term U.S. gasneeds, but will be unable to meet projecteddemand.

Increased access to U.S. resources (excludingdesignated wilderness areas and national parks)could save consumers $300 billion in naturalgas costs over the next 20 years.

New, large-scale resources such as LNG andArctic gas are available and could meet 20-25% of demand, but are higher-cost, havelonger lead times, and face major barriers todevelopment.

Markets

Price volatility is a fundamental aspect of a freemarket, reflecting the variable nature ofdemand and supply; physical and risk manage-ment tools allow many market participants tomoderate the effects of volatility.

There has been a fundamental shift in the natural gas supply/demand balance that has resulted in high-er prices and volatility in recent years. This situation is expected to continue, but can be moderated.

A balanced future that includes increased energy efficiency, immediate development of new resources,and flexibility in fuel choice, could save $1 trillion in U.S. natural gas costs over the next 20 years. Publicpolicy must support these objectives.

Findings


fundamentally alter demand patterns or supply poten-tial. The NPC did not consider such possibilities asbeing likely enough to be integrated into the base sce-narios. However, each scenario was tested against vari-abilities in these and other major underlyingassumptions through the use of sensitivity analyses.Major assumptions tested included weather patterns,economic growth, the price of competing fuels, the sizeof the domestic gas resource base, timing of infrastruc-ture implementation, and the role of other electric gen-eration technologies such as nuclear and hydroelectricplants. These sensitivity analyses provide additionalinsight to the conclusions reached from the base sce-narios and reinforce the study findings and recommen-dations.

In either scenario, it is clear that North Americannatural gas supplies from traditional basins will beinsufficient to meet projected demand; choices must bemade immediately to determine how the nation’s nat-ural gas needs will be met in the future. The best solu-tion to these issues requires actions on multiple paths.

Flexibility in fuel use must be encouraged, diverse sup-ply sources must be developed, and infrastructuremust be made to be as reliable as possible. Policychoices must consider domestic and foreign sources ofsupply, large and small increments of production, andthe use of other fuels as well as gas for power genera-tion. All choices face obstacles, but all must be sup-ported if we are to achieve robust competition amongenergy alternatives and the lowest cost for consumersand the nation. The benefits of the Balanced Futurescenario to the economy and environment unfold overtime; but it is important that these policy changes beimplemented now; otherwise their benefits will bepushed that much farther into the future, and theuneasy supply/demand balance we are experiencingwill continue.

Findings

National Petroleum Council projections of futuredemand and supply are illustrated in Figures 1-1and 1-2.

� Natural gas demand for power generation increases, reflecting future utilization of recent, significant additions of natural gas-fired generation.

TR

ILL

ION

CU

BIC

FE

ET

YEAR

0

5

10

15

20

25

30

35

1990 1995 2000 2005 2010 2015 2020 2025

CANADA

U.S. POWER

U.S. COGENERATION

U.S. INDUSTRIAL

U.S. RESIDENTIAL/COMMERCIAL

OTHER*

*Includes net Mexico exports, lease/plant/pipeline fuel, and net storage.

� Natural gas use in the industrial sector erodes, illustrating projected losses in industrial capacity in the most gas-intensive industries.

Figure 1-1. Natural Gas Demand History and Outlook


Natural Gas Supply

Abundant natural gas resources exist in NorthAmerica and worldwide. Historically, the producingregions of the Gulf Coast and Gulf of Mexico, theMidcontinent, and West Texas have provided themajority of the U.S. natural gas supply as shown inFigure 1-3. In recent years, the Rockies have become asignificant source of supply, primarily from noncon-ventional tight gas and coal bed methane production.Natural gas imports via pipeline from Canada havealso played an expanding role in the U.S. supply pictureand now account for 14% of U.S. consumption. Of theother sources of imports, LNG currently accounts foronly one percent of U.S. supply and Mexico is a smallnet importer of gas from the United States.

A thorough study was conducted to assess theremaining potential of traditional North Americannatural gas producing basins, as well as the potentialfor growth from new supply sources. The resultingoutlook is that indigenous North American produc-tion will remain relatively flat, as increasing production

from the Rockies and deepwater Gulf of Mexico offsetsdeclining production from the maturing traditionalbasins. Supply growth will be met by LNG importsand new Arctic developments.

Natural Gas Demand

Government policies have encouraged industrialconsumers and power generators to become increas-ingly reliant upon natural gas-based technologies tomeet their energy requirements and to satisfy more-stringent air quality standards. Recent major growth innatural gas-fired generation capacity and continuedresidential customer connections create the potentialfor even greater natural gas consumption. Continuedenergy conservation and more efficient use of existingequipment can ease short-term market pressures, andwill continue to have a significant impact on futureenergy consumption. However, this alone will notsolve the problem; additional supplies are required.Figure 1-4 shows regional demand growth patterns forNorth America.

• Production from traditional basins remains strong but has plateaued; Rockies and deepwater

Gulf of Mexico offset declines in other areas.

0

5

10

15

20

25

30

35

1990 1995 2000 2005 2010 2015 2020 2025

TR

ILL

ION

CU

BIC

FE

ET

OTHER LOWER-48*

NON-ARCTIC CANADA

ALASKA

LNG

GULF OF MEXICO SHELF

ROCKIES

MACKENZIE

DELTA

GULF OF MEXICO DEEPWATER

YEAR

* Includes lower-48 production, ethane rejection, and supplemental gas.

• Growth is driven by LNG imports and Arctic supply.

Figure 1-2. Natural Gas Supply History and Outlook


MACKENZIE

WESTERN CANADA

ALASKA

EASTERN

CANADA

EASTERN

INTERIOR

MIDCONTINENTWEST COAST

ROCKIES

GULF COAST

ONSHORE

WEST TEXAS/

PERMIAN

MEXICO

LNG TOTAL

TR

ILLIO

N C

UB

IC F

EE

T / Y

EA

R

2002 2025

GULF OF MEXICO

SLOPE


1

2

3

4

0

Figure 1-3. Regional Natural Gas Supply


Figure 1-4. Regional Natural Gas Demand

EAST NORTHCENTRAL

SOUTHATLANTIC

EAST SOUTHCENTRAL

PACIFIC

WEST NORTHCENTRAL

MOUNTAIN

MEXICO

BRITISHCOLUMBIA

ALASKA

ALBERTA

SASKATCHEWAN

MANITOBA

ONTARIOQUEBEC

MARITIMES

WEST SOUTHCENTRAL

NEWENGLAND

MID-ATLANTIC

1

2

3

4

0

TR

ILLIO

N C

UB

IC F

EE

T / Y

EA

R

2002 2025

Infrastructure Expansions

Significant new infrastructure will be neededthrough 2013, then the projected need for capital fornew infrastructure will decrease while sustaining capi-tal becomes an increasing percentage of total capitalrequirements. Regulatory policy that supports long-term contracts and regulatory certainty will berequired to develop new infrastructure and maintainthe same level of reliability.

Potential Price Ranges

The NPC-projected price ranges for the alternatescenarios are illustrated in Figure 1-5. Supply anddemand are projected to balance at higher price rangesthan historical levels. The price ranges will be prima-rily determined by demand response throughincreased efficiency, conservation, and alternate fueluse, the ability to increase conventional and noncon-ventional supplies from North America including theArctic, and increasing access to world resourcesthrough LNG imports. As has previously beendescribed, the scenarios that drive these outlooks movebeyond the status quo. They require significant initia-tive by policy makers and industry stakeholders toimplement the recommendations of this report and

take market-driven actions. Each scenario has poten-tial additional price variability due to external factorsincluding weather, lower-48 supply response to higherprices, timing of infrastructure development, potentialbreakthrough technologies, and use of competingfuels. These projected ranges are not intended to beprecise estimates of future prices but are provided toprovide insights to the effects of government policy.Additionally, sensitivity analyses (detailed later in thisreport) were performed to test the effects of changes inkey assumptions.

Capital Expenditures

Over $1.4 trillion (2002 dollars) in capital expendi-tures will be required to fund the U.S. and Canadiangas upstream and infrastructure industry from 2003 to2025. Eighty-five percent will be spent in the explo-ration and production sector ($1.2 trillion), with theremaining 15% ($0.2 trillion) spent on pipelines, stor-age, and distribution, as shown in Figures 1-6 and 1-7.These expenditures represent a significant increaseover the 1990-2000 period for the exploration and pro-duction sector. Expenditures for the pipeline, storage,and distribution sector are expected to remain relative-ly constant, considering increasing needs for “sustain-ing capital” to meet reliability requirements.


2005 2010 2015 2020 2025

1

2

3

4

5

6

7

8

2000

YEAR

DO

LLA

RS

PE

R M

ILLI

ON

BT

U (

2002

DO

LLA

RS

)

0

BALANCED FUTURE

REACTIVE PATH

1995

Figure 1-5. Average Annual Henry Hub Prices


Figure 1-6. North American Upstream Expenditures – Balanced Future Scenario

Figure 1-7. North American Infrastructure Expenditures – Balanced Future Scenario

0

10

20

30

40

50

60

70

80

1990 1995 2000 2005 2010 2015 2020 2025

YEAR

BIL

LIO

NS

OF

200

2 D

OLL

AR

S

0

10

20

BIL

LIO

NS

OF

2002 D

OLLA

RS STORAGE

PIPELINES

LDC

1990 1995 2000 2005 2010 2015 2020 2025

YEAR

Note: This figure is a revision of Figure 53 in Volume I of this report.

Sensitivities

There are many factors that will affect future gasmarkets in North America. The NPC considered thesefactors and created base assumptions for the BalancedFuture and Reactive Path scenarios. Because of theuncertainty inherent in making such assumptions for20+ years into the future, the NPC often developedalternative sets of assumptions to test how the naturalgas market might evolve under different circumstances.Those alternative assumptions addressed the effects of adifferent economic environment, government policies,natural resource size, upstream technological trends,weather, end-use efficiency improvements, and otherfactors. Figure 1-8 shows the price and volume effect ofdiffering assumptions. Table 1-1 lists the assumptionchange for each case. The factors have widely varyingprobabilities.

Recommendations

The findings of the National Petroleum Councildescribed in this report represent the conclusions of theCouncil from the detailed analysis undertaken over thecourse of this study. They provide the clear motivationfor the recommendations that follow. Collectively andindividually, policy makers will make decisions affect-ing the future of natural gas in the economy. Thesechoices will have significant effects on resource avail-ability, on natural gas production, on the cost-effectiveuse of natural gas, on the capacity of infrastructure toserve markets, and on prices and price volatility.Prompt implementation of the NPC’s recommenda-tions will reduce the conflicts in current public policyand benefit both consumers and the environment.Further details on the recommendations can be foundlater in this report.


Table 1-1. Sensitivity Analysis

Sensitivity Assumption

High Resource Base (10% probability)

+35% of base, +570 TCF

Fuel Flexibility Industrial switching increased from 5% to 28% by 2025, fuel backup included in 25% of new gas-fired generation

High Supply Technology Increase exploration success, cost reduction, well recoveries from ~ 1% to 1.5%/year

Low Economic Growth 2.7%/year GDP growth (vs. 3%/year base)

Increased Access to Resources 10%/year improvement in access to new resources for five years

High LNG Imports 15 BCF/D imports (vs. 12.5 BCF/D Reactive Path)

Less Access Continued trend toward more restrictive permitting, higher costs to develop resources

No Alaska Pipeline 4 BCF/D pipeline not built

High Electricity Growth Limited energy efficiency; GDP-to-electricity elasticity fixed at 72%

High Economic Growth 3.3%/year GDP growth (vs. 3%/year base)

West Texas Intermediate $28 Oil Price

+$8/bbl vs. Reactive Path and Balanced Future Scenarios

Low LNG Import 6 BCF/D (vs. 12.5 BCF/D Reactive Path)

Static Supply Technology No improvement in exploration success, costs, recoveries

Low Resource Base (90% probability)

-30% of base; -490 TCF


Improve Demand Flexibility and Efficiency

Encourage increased efficiency and conservationthrough market-oriented initiatives and con-sumer education.

Increase industrial and power generation capa-bility to utilize alternate fuels.

Sustain and Enhance Infrastructure

Provide regulatory certainty by maintaining aconsistent cost-recovery and contracting envi-ronment and removing regulatory barriers tolong-term capacity contracting and cost recoveryof collaborative research.

Permit projects within a one-year period utiliz-ing a Joint Agency Review Process.

Increase Supply Diversity

Increase access and reduce permitting impedi-ments to development of lower-48 natural gasresources.

Enact enabling legislation in 2003 for an Alaskagas pipeline.

Process LNG project permit applications withinone year.

Promote Efficiency of Markets

Improve transparency of price reporting.

Expand and enhance natural gas market data col-lection and reporting.

Recommendations

Note: Values shown are averages for the 2011 to 2025 period.

-2.00 0.0 2.00 4.00 -4,000 -2,000 0 2,000

CHANGE IN VOLUMES(BILLION CUBIC FEET PER YEAR)

VS. REACTIVE PATH

CHANGE IN PRICE(2002 DOLLARS)

VS. REACTIVE PATH

HIGH RESOURCE BASE P10

FUEL FLEXIBILITY

HIGH SUPPLY TECHNOLOGY

LOW ECONOMIC GROWTH

INCREASED ACCESS

HIGH LNG IMPORTS

LESS ACCESS

HIGH ELECTRICITY GROWTH

HIGH ECONOMIC GROWTH

WTI $28 OIL PRICE

NO ALASKA PIPELINE

LOW LNG IMPORTS

STATIC SUPPLY TECHNOLOGY

LOW RESOURCE BASE P90

Figure 1-8. Price and Volume Impacts of Selected Sensitivities

CHAPTER 2 - BACKGROUND 19

Role of Natural Gas in the Economy

Natural gas is a critical source of energy and rawmaterial, permeating virtually all sectors of the econo-my. Today natural gas provides nearly one-quarter ofU.S. energy requirements1 and is an environmentallysuperior fuel, thereby contributing significantly toreduced levels of air pollutants. It provides about 19%of electric power generation and is a clean fuel forheating and cooking in over 60 million U.S. house-holds. U.S. industries get over 40% of all primary ener-gy from natural gas. Figure 2-1 illustrates thecontribution of natural gas to U.S. energy needs, andFigure 2-2 shows gas use by sector.

North America’s natural gas exploration and pro-duction industry has been successful in efficiently find-ing and developing the continent’s indigenousresources, and an extensive infrastructure has beendeveloped to efficiently transport natural gas from itsdiverse sources to its multiple markets. Technologyadvances throughout the supply chain have increasedsupply, reduced costs, and minimized environmentaleffects. Effective mechanisms for the sale, purchase,and pricing of natural gas have evolved, and there hasbeen a progressive reliance in recent years on competi-tion and open markets at each point along the naturalgas supply value chain.

Today, many regulations and policies affecting natu-ral gas are in conflict. Public policies promote the useof natural gas as an efficient and environmentallyattractive fuel. These policies have led to restrictions

on fuels other than natural gas for the siting of powergeneration and industrial facilities, restrictions on fuelswitching, and fuel choice limitations. Other laws andregulations have been enacted that limit access to gas-prone areas – areas where gas can be explored for andproduced in an efficient and environmentally friendlymanner – and there are outright bans to drilling in cer-tain regions. There are laws and regulations thatunnecessarily hinder pipeline and infrastructure sitingor interfere with the functionality of the market inways that lead to inefficiencies. Overall, these conflict-ing policies have contributed to today’s tight

BACKGROUNDCHAPTER 2

Source: Energy Information Administration.

RENEWABLES 7%

COAL

23%

NATURAL GAS

24%

PETROLEUM

38%

NUCLEAR 8%

Figure 2-1. Average Annual Energy Use, 1997-200197 Trillion Cubic Feet per Year (Equivalent)


1 Data from Energy Information Administration,Monthly Energy Review, April 2003.

supply/demand balance, with higher and volatile gasprices. The beneficial effects of additional gas use canbe achieved more efficiently and at a lower cost withpolicies that eliminate the current conflicts.

Natural Gas Market History

Natural gas use in the United States has reflectedchanges in the economy, in natural gas infrastructure,in natural gas-based technologies, in the regulation ofnatural gas, and in environmental initiatives.Regulation of the natural gas industry in the UnitedStates has had profound effects on natural gas supplyand demand and has at times led to stresses on theeconomy, including the natural gas shortages experi-enced in the 1970s. In 1938, Congress passed theNatural Gas Act, representing the first major federalinvolvement in natural gas sales and distribution, andgave the Federal Power Commission (the forerunner ofthe Federal Energy Regulatory Commission) jurisdic-tion over interstate natural gas sales and the ratescharged for interstate natural gas delivery. TheSupreme Court later determined that the FPC alsoshould regulate the prices of natural gas sold in theinterstate market. This decision had a complicated andfar-reaching effect on the natural gas industry and cre-

ated significant administrative difficulties for the FPCin trying to set prices for a large number of natural gasproducers.

Ultimately, the regulated prices were lower than themarket value of the natural gas. These relatively lowprices prompted a surge in demand, but failed toencourage additional production, thus leading to ashortage of supply. In addition, since the intrastatemarket was unregulated, producers sold as much oftheir production as possible into the intrastate market,thereby exacerbating the shortage in the interstatemarket. In turn, the natural gas shortages led the FPCto impose priority systems whereby scarce natural gaswas allocated to certain customers, with deliveries cur-tailed to certain customers, such as industrial con-sumers, who were deemed “low priority.”

U.S. gas consumption grew for many years beforepeaking at 23 trillion cubic feet (TCF) per year in 1972as shown in Figure 2-3. Post-1972 declines in gas con-sumption and production were the result of both eco-nomic and regulatory forces. For example, gasutilization by industrial consumers and power genera-tors was severely limited by the Power Plant andIndustrial Fuel Use Act; gas demand was further damp-ened by the attendant effects of two economic reces-


0 10 20 30 40

POWER GENERATION

INDUSTRIAL

TRANSPORTATION

RESIDENTIAL

COMMERCIAL

QUADRILLION BTU

GAS OTHER


Figure 2-2. U.S. Energy Use by Sector, Year 2002

sions. During this period, price-induced conservationreduced residential gas use by 16%. At the low point in1986, both gas consumption and production had fallenin excess of 25% from the 1972 peak.

Factors that contributed to a rebound in gas con-sumption and production in the second half of the1980s continue to shape today’s natural gas market-place. These factors include repeal of the Power Plantand Industrial Fuel Use Act in 1987, the elimination ofgas price controls (1989), a restructuring of interstatepipeline contracts, implementation of FERC Orders436 and 636, passage of the Clean Air Act Amendments(1990), and robust economic growth. Generally fallinggas prices after the mid-1980s also stimulated gasdemand. Gas became a commodity, one where therewas a competitive market and a financial frameworkfor gas buyers and sellers to make short- and longer-term decisions.

U.S. gas consumption increased from 16.2 TCF in1986 to 23.5 TCF in 2000, surpassing the 1972 peak.Residential and commercial gas growth returned aslimits on new hook-ups were lifted, and continued asgas was competitively priced and environmentally pre-ferred. Industrial consumers found gas an attractivelypriced fuel for process heat, feedstock, and cogenera-

tion, as well as an effective method to meet air qualitystandards. Electric generators also found gas pricesattractive; especially with new, highly efficient, low-emission and low-cost gas-fired turbines providing anenvironmentally acceptable and economically efficientway to satisfy electricity demand growth. Falling realgas prices during the 1990s made gas attractive as a fueland feedstock.

In response to demand growth, U.S. gas productionrose after 1986, utilizing excess well deliverability andapplying advances in production technology. Much ofthis gain was a result of the industry’s transition froma regulated market where most gas was sold bypipelines and was bundled with transmission to onewhere gas was sold in a competitive market. This tran-sition resulted in a temporary “gas bubble” where bothproductive capacity (supply) and deliverability (trans-mission and distribution) exceeded gas demand. Bythe mid-1990s this excess capacity dissipated due todemand growth and limited increases in productivecapacity, as illustrated in Figure 2-4.

Canadian gas imports increased five-fold between1986 and 1999, filling a growing gap between U.S. con-sumption and lower-48 production. Canadian pro-duction growth occurred from increased drilling


0

5

10

15

20

25

30

1970 1975 1980 1985 1990 1995 2000

TR

ILLIO

N C

UB

IC F

EE

T

U.S. POWER

U.S. COGENERATION*

U.S. INDUSTRIAL

U.S. COMMERCIAL/RESIDENTIAL

U.S. PIPELINE, LEASE, & PLANT FUEL

* Energy Information Administration reports cogeneration beginning 1989.

Sources: Energy Information Administration and National Energy Board of Canada.

CANADA

YEAR

Figure 2-3. U.S. and Canadian Natural Gas Demand

activity and pipeline capacity expansions to U.S. mar-kets. In addition, small volumes of liquefied naturalgas (LNG) were imported to terminals inMassachusetts and Louisiana.

During the 1990s, gas prices were generally flat todeclining. Although gas supply and demand weremore closely balanced, mild weather and newCanadian import projects masked any pronouncedtightness. However, by the late 1990s, gas prices beganto reflect the value of natural gas in its more diverseapplications where gas was becoming a competitivealternative. These applications were often electric gen-erating plants or industrial boilers capable of burningoil or oil-derivative fuels. Increasingly, gas prices werebecoming correlated with oil prices.

Another driving factor in gas demand growth dur-ing the 1990s was environmental regulations. TheClean Air Act Amendments of 1990 were primarilyfocused on reducing sulfur dioxide (SO2) and nitro-gen oxide (NOx) emissions from electric powerplantsand, to a lesser extent, from other industrial andtransportation sources. To comply with the mandatesof both the first (1995-1999) and second (2000+)phase of the Act, generators and industry turnedincreasingly to natural gas, either in the form of fuel-

switching or investments in new, gas-only equipment.While this strategy has been highly effective in reduc-ing emissions, it has limited the ability to switch toother fuels when the natural gas market becomesstressed.

Natural Gas Prices

Natural gas prices remained in the $1.50 to $3.00 permillion Btu (MMBtu) range through the 1990s. Withthe tightening supply demand balance in the late1990s, prices have risen and become more volatile, asshown in Figure 2-5. In late 2000, gas prices rose sig-nificantly with a short peak at nearly $10.00/MMBtu.Low gas storage levels at the start of the 2000-2001


0

45

50

55

GAS PRODUCTION

PRODUCTIVE CAPACITY

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

YEAR

1995 1996 1997 1998 1999 2000 2001 2002

Source: Energy and Environmental Analysis, Inc.

2003

Figure 2-4. Lower-48 Dry Gas Production vs. Dry Gas Productive Capacity

Finding: There has been a fundamentalshift in the natural gas supply/demandbalance that has resulted in higher pricesand volatility in recent years. This trend isexpected to continue, but can be moderatedthrough policy actions.

winter, record cold temperatures in November andDecember, and record storage withdrawals promptedfears of gas shortages. Gas consumers responded tothese price signals by reducing gas consumptionthrough a variety of measures that included fuelswitching, factory closings, and conservation.

Producers responded to these record prices in late2000-2001 by embarking on an intensive drilling pro-gram. Gas production rose moderately in 2001 asdrilling peaked. With rising prices, producers pursuedtheir available inventory of higher-cost, lower-produc-tivity prospects. Supply available to U.S. marketsreached its highest level in 20 years as a result of thesetransitory factors

Gas prices on the New York Mercantile Exchange(NYMEX) fell from $9.00/MMBtu at the start ofJanuary 2001 to under $2.00/MMBtu in October 2001.The transitory factors that contributed to this pricevariability were: mild weather from late winter of 2000through the summer of 2001; reduced gas demand dueto weak GDP growth and falling industrial output; anddeclining oil prices (West Texas Intermediate (WTI)crude oil falling from $30/bbl in January 2001 to under$20/bbl by year-end).

In 2002, another mild winter and stagnant manufac-turing activity depressed gas demand. Gas storage lev-els were at record-high levels throughout most of 2002.Gas prices fell significantly from the 2000-2001 peaks.The early onset of cold temperatures in the 2002-2003winter and falling natural gas production depletedNorth American storage inventories, which fell torecord lows in early spring contributing to the run-upin prices in 2003.

Major Assumptions/Methodologies

Macroeconomic Assumptions

Throughout history, economic growth has beenassociated with higher energy consumption. In thepast 25 years, energy use grew 1.0% annually whilegross domestic product (GDP) increased at a 3.0%annual rate. While energy use is pervasive in the U.S.economy, there have been steady gains in efficiencywith a downward trend in the amount of energy neces-sary to produce one dollar of GDP. U.S. GDP growthwas assumed to average 3.0% per year from 2005 to2025, similar to the historical trend. The impact ofchanges in natural gas prices on GDP growth was ana-lyzed, but analysis conducted for the NPC by Global


0

2

4

6

8

10N

OM

INA

L D

OLLA

RS

PE

R M

ILLIO

N B

TU

Source: Platt's Inside FERC Monthly Price. YEAR

1995 1996 1997 1998 1999 2000 2001 2002 2003

Figure 2-5. Henry Hub Monthly Index Prices

Insight suggests that GDP growth rates would not besignificantly impacted by higher natural gas prices.

U.S. industrial production was assumed to grow ata rate of 3% per year, which is below the 1995-2002historical growth rate of 3.4% per year. The lower-than-historical growth rate was chosen to reflect theeffect of higher-than-historical natural gas prices ongas-intensive industries.

Prices for alternative fossil fuels were assumed with-in historical ranges. WTI crude oil was assumed to fol-low the forward curve for the first few years and thensettle at a long-term equilibrium price of $20/bbl inconstant 2002 dollars. This is similar to the averagereal price in the 1990s and consistent with the financialcommunity view. Coal prices were assumed to followthe historical trend of decline from a level of$1.25/MMBtu in real terms. This decline reflects con-tinued productivity improvements and competition.

Macroeconomic and other assumptions used in theReactive Path and Balanced Future scenarios are sum-marized in Table 2-1. Additional assumptions used inthe Reactive Path scenario are highlighted below.

Residential and Commercial Demand

� Residential and commercial demand for natural gasis driven primarily by population growth and demo-graphic patterns.

� Trends in conservation/efficiency improvementscontinue.

� Natural gas demand in this sector is fairly insensitiveto natural gas prices.

Industrial Gas Demand

� This NPC study included more detailed modeling ofthe gas-intensive industries than the 1992 and 1999studies because of the recent higher-than-historicalgas price environment. This study also incorporatesfeedback obtained from gas-intensive industry rep-resentatives in several workshops.

� Natural gas-intensive industrial production isassumed to grow at a 1.0% to 1.5% average annualrate.

� The chemical and refining sectors represent 50%of the total output from gas-intensive industries.The sectors most impacted by higher natural gasprices are chemicals, ammonia, methanol, andmetals.

� Capacity idled for at least two years is assumed toshut down permanently.

� Local, state, and federal government environmentalregulations are assumed to continue to limit fuelswitching from natural gas to oil.


Macroeconomic Assumptions Average Annual Growth Rate

U.S. GDP Growth 2.8% 2002-2005; 3.0% thereafter

Canadian GDP Growth 2.4% 2002-2005; 2.6% thereafter

Inflation Rate (GDP Deflator) 2.5%

Other Assumptions

Weather Historical averages

Oil Price WTI at $20/bbl (2002$) 2005-2025

Refiner Acquisition Cost of Crude Oil (RACC) 90% of WTI

Residual Fuel Oil Price 84% of RACC by 2004

Distillate 140% of RACC

Coal Price $1.25/MMBtu and declining in real terms at 1.0%/year to $1/MMBtu by 2025

Table 2-1. Macroeconomic and Other Assumptions

Electric Power

� Electricity Demand Growth – Growth rates werebased on income elasticity that declines from 0.72 in2003 to 0.62 in 2025 to reflect a continued shift inthe U.S. economy away from energy-intensive indus-tries, and some power conservation as a result of anextended period of higher-than-historical naturalgas prices.

� Technology – Power generation technology isassumed to continue to improve incrementally.

� Coal – Mercury emission regulations are assumed toresult in the closure of approximately 20 gigawatts ofcoal capacity by 2015, although the financial struc-ture of the industry is assumed to support someenvironmental retrofits of existing coal-fired power-plants. A limited number of new coal plants arebuilt in selected locations.

� Nuclear – Nuclear powerplant utilization is assumedto rise to 90-92% after 2006. No episodic or perma-nent nuclear plant shutdowns are assumed.However, no new nuclear plants are built due to thelong lead-time and high capital costs.

� Oil and Gas Steam – Some existing oil and oil/gassteam units are retired before 2010 andregional/local government limitations on oil usageare maintained.

� Renewables – Renewable capacity will be aggressive-ly constructed, such that over 70 gigawatts of renew-able generation will be constructed by 2025.

� The remainder of new generation capacity requiredis assumed to be gas fired.

Supply

� The study used U.S. Geological Survey, MineralsManagement Service, and Canadian Gas PotentialCommittee play-level assessments and conductedindustry/government workshops to review theNorth American resource base.

� A detailed 72-region supply model was used to eval-uate supply, with detailed finding rates and costs.

� Production performance was analyzed over the pastten years, with assessments of initial productionrates, production decline rates, and total well recov-eries for each major producing basin.

� Cost-of-supply curves were used to estimate marketclearing prices that satisfy demand.

� Mackenzie Delta gas was assumed to start up in 2009at a production rate of 1 billion cubic feet per day(BCF/D) and then ramp up to 1.5 BCF/D beyond2015. The Alaskan gas pipeline was assumed to startup at 2.5 BCF/D in 2013 and then rise to 4 BCF/D in2014 and thereafter.

� LNG imports were evaluated and growth to 12.5BCF/D is projected by 2025 in the Reactive Pathscenario; 15 BCF/D in the Balanced Future sce-nario.

� Annual improvement factors were applied to reflectadvancing technologies that will lower drilling andinfrastructure costs, improve exploration success,and increase well recoveries over the model time-frame.

Transportation and Storage Infrastructure

� A detailed 115-node model with 317 differentpipeline corridors was used to determine thepipeline and storage infrastructure.

� Known or expected pipeline and storage projectscoming on stream during the next five years werespecified. Beyond the first five years pipeline capac-ity was assumed to come on stream two and a halfyears after justified by price basis differential, withsome capacity added more quickly in areas with sig-nificant supply development.

� For large projects such as Arctic, LNG, and deep off-shore, infrastructure was assumed to be built just intime.

� Working gas capacity of storage was increased overthe study period by about 700 BCF from the 2003level, based on the daily load analysis.

� Substantial additions to transmission capacity areprojected to link new storage capacity to markets inthe Northeast.

� The historical storage injection seasonal pattern waschanged to reflect the emerging July-August peakuse of gas-fired generation.

� Distribution facilities are demographically driven,based on growth in customers.


CHAPTER 3 - NATURAL GAS DEMAND 27

This chapter describes the methods used by theDemand Task Group to develop an outlook fornatural gas demand and describes the results of

these analyses. The demand-related aspects of thisstudy’s findings and recommendations are based onthese analyses. The demand outlook is discussedbelow, along with additional details for each of themajor areas driving the demand for natural gas. Thefull demand study documentation is found in theDemand Task Group Report and its appendices.

Study Approach

The analysis of natural gas demand focused on theprimary factors affecting current natural gas consump-tion and evaluated variables that are likely to affectlong-term usage. This analysis consisted of the follow-ing elements:

� An assessment of historical and expected macroeco-nomic and demographic factors affecting thedemand for natural gas.

� A detailed evaluation of installed and likely addi-tions to future power generation capacity within theregions and sub-regions of the North AmericanElectric Reliability Council, including the manner inwhich this capacity will likely be used. This analysisalso assessed the recent, massive buildup in naturalgas-based generation.

� An assessment of natural gas utilization in the mostenergy-intensive industries, including estimates ofshort-term demand elasticity and the potential forshort and longer term demand destruction.

� An assessment of future trends for residential andcommercial gas consumption.

� Assessments of the effects of energy efficiency andtechnology advancement on natural gas demand.

The study of demand was undertaken by four work-ing groups: Economics and Demographics, led byShell Trading Gas and Power; Power Generation, ledby American Electric Power; Industrial Utilization,led by Process Gas Consumers; and Residentialand Commercial, led by KeySpan Corporation.KeySpan provided overall coordination of theDemand Task Group.

The Economics and Demographics Subgroup devel-oped critical assumptions necessary to run economet-ric and other analyses for the Demand, Supply, andTransmission & Distribution Task Groups. Theseassumptions included major North American eco-nomic growth parameters and alternate fuel prices,mainly U.S. coal and oil prices. Model outputs wereadditionally vetted to see if key assumptions needed tobe altered.

The Power Generation Subgroup focused its effortson understanding the factors that are likely to drivecapacity and utilization decisions of existing and newgas-fired generation. A variety of electric power gener-ators from various regions was represented or consult-ed in this analysis, and workshops were held in NewOrleans, Phoenix, and Baltimore. A suite of cost fac-tors were developed for the construction and utiliza-tion, or dispatch, of generation capacity by fuel type.The team performed extensive analysis on investmentcriteria; likely technology advances; outlooks for coal,nuclear, and hydroelectric capacity; the potentialeffects of Regional Transmission Organizations; theeffects of state, provincial, and local regulations and

NATURAL GAS DEMANDCHAPTER 3

standards; and practices governing the flexibility ofpower generators to substitute fuels.

The Industrial Utilization Subgroup attempted to ana-lyze the recent changes observed in industrial natural gasdemand. Emphasis was focused on critical gas-intensiveindustries such as chemicals, primary metals, and paper.The subgroup received considerable support and infor-mation from these and other key energy-intensive indus-tries. A series of outreach sessions was held to evaluatetrends and to analyze factors influencing gas demand bysector and process within the sectors.

To analyze future trends in residential and commer-cial gas consumption, the Residential and CommercialSubgroup used econometric models and capital stockmodels. These models included the effects of weather,demographic trends, population growth, residentialhousing stock, capital stock efficiency, commercialfloor space, penetration of gas-based technology, andgas prices as determinants of gas consumption.

All of the subgroups placed particular emphasis onunderstanding the historical and potential role of ener-gy efficiency. Similarly, the impact of environmentallaws and regulations was modeled to ascertain past andanticipated effects on natural gas demand. Finally, therole of energy market mechanisms – which either facil-itate or impede efficient natural gas utilization – wasassessed within each demand sector.

Demand Outlook

The Demand Task Group analyzed the demand-related aspects of scenarios on natural gas supply,demand, and infrastructure through 2025. This analy-sis was largely performed using a series of econometricand other models developed by Energy andEnvironmental Analysis, Inc. (EEA). Although the useof EEA’s models was similar to the 1992 and 1999 NPCstudies, the EEA models were augmented by extensivework on the industrial segment to create a greaterdegree of granularity in the price elasticity and switch-ing behavior of this demand segment.

Natural gas demand in 2002, by major consumingand geographic segment, is depicted in Figure 3-1.Industrial demand and gas for electric power have beenhistorically concentrated in the primary gas-producingregions, while the large population centers in colderclimates show larger residential and commercialdemands.

Figure 3-2 depicts the results of the Reactive Pathscenario for North American demand. In this scenario,North American natural gas demand grows at anapproximate 1% annual average rate through 2025.Today’s largest consuming sector, industrial gas, showsrelatively little growth, while gas for power generationgrows at a faster pace than any of the other sectors.Residential and commercial gas consumption growthslows to less than 1% per year.

Energy, Natural Gas, and the Economy

Natural gas has played a critical role in the UnitedStates’ energy picture during the last 50 years. Naturalgas was about 16% of the total energy consumption inthe early 1950s. As it became a widespread fuel forhome heating, and a significant fuel and feedstock inindustrial applications, its share of total energy grew tonearly 32% in the early 1970s.

During the early 1980s, however, natural gas use as apercentage of total energy consumption dropped toabout 23%. Gas consumption declined due, in part, toa period of relatively high gas prices, gas shortages, andcurtailments that led to government policies discour-aging the use of gas for certain applications. Since themid-1980s, natural gas has maintained approximatelya 25% share of total energy consumption while its usehas grown by about 2.1% per year. Figure 3-3 showsnatural gas in relation to the other primary sources ofenergy for the United States.

Gas is a major source of energy in every sector of theeconomy except the transportation sector, as depictedin Figure 3-4. It has become a highly desirable fuel ineach of the sectors where it enjoys significant marketshare due to its ease of use, historical competitive costs,and most recently its desirable environmental impactcharacteristics of low emissions.

Gas Demand for Electric Power Generation

Over the past decade, power generation has beenincreasing its demand for natural gas. The drivers forthis growth include increasing electricity demand, therapid buildup of gas-fired generating capacity, andmore stringent environmental policies. Figure 3-5shows historical and projected power generationcapacity. The large quantity of natural gas-fired gen-eration capacity installed between 1998 and 2005underpins the outlook for significant gas demandgrowth over much of the 2005-2025 period, as depict-ed in Figure 3-6.



ALASKA

BIL

LIO

N C

UB

IC F

EE

T

PE

R Y

EA

R

1,000

500

0


BRITISH

COLUMBIA

PACIFIC

MOUNTAIN

ALBERTA

SASKATCHEWAN

WEST NORTH

CENTRAL

MANITOBA

WEST SOUTH

CENTRAL

MEXICO

(2001 DATA)

ONTARIO

QUEBEC

MARITIMES

NEW

ENGLAND

SOUTH

ATLANTIC

MID-ATLANTIC

EAST NORTH

CENTRAL

EAST SOUTH

CENTRAL

RESIDENTIAL/COMMERCIAL

INDUSTRIAL

ELECTRIC POWER

OTHER

Figure 3-1. North American Natural Gas Demand by Sector and Region, 2002

Figures 3-5 and 3-6 also suggest that coal will con-tinue to provide a significant share of total power gen-eration. However, permitting and siting a new coalfacility will remain a formidable challenge. The analy-sis expects that new technologies for lowering air emis-sions of sulfur oxides (SOx) and nitrogen oxides (NOx)will make it possible to build coal-fired powerplants inthe regions that traditionally have the highest percent-age of coal-fired capacity. It was assumed that no coalplant would be successfully sited in the non-attainmentareas of the U.S. east or west coasts during the periodof this study. Figure 3-7 shows the non-attainmentareas and provides a geographic distribution of thenew gas-fired capacity that is included in the NPC’soutlook. Except in the Pacific Northwest and upperMidwest, over 50% of the ultimate projected gas-firedcapacity is in place or under construction.

Industrial Natural Gas Demand

The industrial sector is currently the largest demandsegment for natural gas. It fueled much of the growthin demand from the mid-1980s until late in the 1990s.As shown in Figure 3-2, an important element of thatgrowth was the use of natural gas in cogenerationapplications producing combined heat and power for


0

5

10

15

20

25

30

35

1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025

TR

ILLIO

N C

UB

IC F

EE

T

CANADA

U.S. POWER

U.S. COGENERATION

U.S. INDUSTRIAL

U.S. COMMERCIAL/RESIDENTIAL

U.S. PIPELINE, LEASE, & PLANT FUEL

HISTORICAL PROJECTED

YEAR

Figure 3-2. Historical Gas Demand and Reactive Path Projection


NUCLEAR 8% RENEWABLES 7%

COAL

23%

NATURAL GAS

24%

PETROLEUM

38%

Figure 3-3. Average Annual Energy Use, 1997-200197 Trillion Cubic Feet per Year (Equivalent)


0 10 20 30 40

POWER GENERATION

INDUSTRIAL

TRANSPORTATION

RESIDENTIAL

COMMERCIAL

QUADRILLION BTU

GAS OTHER


Figure 3-4. U.S. Energy Use by Sector, 2002

0

200

400

600

800

1,000

1,200

1,400

1995 2000 2005 2010 2015 2020 2025

YEAR

HYDROELECTRIC

RENEWABLES

NUCLEAR

COAL

GAS/OIL DUAL-FUEL

OIL

GAS

GIG

AW

AT

TS

Figure 3-5. U.S. Electric Power Generation Capacity by Fuel Type

the industrial facilities and optimizing the efficiency ofproducing steam and electricity. The six most gas-intensive industrial groups are:

� Chemicals

� Petroleum Refining

� Primary Metals

� Food and Beverage

� Paper

� Stone, Clay, and Glass.

Figure 3-8 shows the outlook for natural gasdemand by energy-intensive industrial segments.Higher natural gas prices for feedstock and fuel com-bined with global competition results in a more sub-dued outlook for demand growth in the industrialsector. The chemical industry, in particular, is project-ed to reduce gas demand through 2025. Figure 3-9shows the outlook for gas demand by type of end-useapplication within the industrial sector. Envi-ronmental emission characteristics, ease of use, andcontinued efficiency gains may allow natural gas tomaintain its share of these end-use applications energyrequirements.

The NPC found that industrial consumers are undermore severe pressure from higher natural gas pricesthan other demand sectors. Further, the investmentdecisions of industrial gas consumers are predicated ona wide range of factors, many of which pertain to inter-national competition and geopolitical considerations.Therefore, the potential for additional demanddestruction exists in all gas-intensive manufacturing

Residential and Commercial Natural Gas Demand

Residential and commercial customers primarily usenatural gas for space heating, water heating, and cook-ing. Over 60 million households in the United Statesuse natural gas for some application. Natural gas has ahigh saturation rate for space heating, particularly inthe population centers in the northern climates. Thisextensive use as a heating fuel contributes to the sea-sonal nature of the natural gas industry. Consumptionof natural gas in the United States has a distinct annu-al cycle. Average monthly gas demand in the winterpeaks at 35-40% above the annual average.

Residential and commercial natural gas demand isexpected to increase more slowly over the 2005-2025


YEAR

0

1

2

3

4

5

6

1995 2000 2005 2010 2015 2020 2025

TE

RA

WA

TT

HO

UR

S (

TH

OU

SA

ND

S)

HYDROELECTRIC

NUCLEAR

COAL

GAS

RENEWABLES

OIL

Figure 3-6. U.S. Electricity Generated by Fuel Type

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 33

0 5 5 6 107

NEW YORK

115 16 18 20 22

FRCC

7

27 28 33 3747

ERCOT

SERC

8

3748

5940

70

5

37 38 40 43 47

SPP

417 18 18 19 20

MAIN

1 4 4 5 7 9

MAPP

011 11 13 15 18

WECC CA/NV

313 13 15 16 18

NEW ENGLAND

4

27 28 31 35 40

ECAR

0 6 8 1320 24

WECC NW

115 21 24

15 18

WECC ROCKIES

112 12 16 21 25

MAAC

2000 2005 2010 2015 2020 2025

Notes: Units are measured in gigawatts. Regions are based on North American Electric Reliability Council (NERC) regions.

Shaded areas are the EPA non-attainment areas.

Figure 3-7. EPA Non-attainment Areas and New Gas-Fired Capacity


0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1995 2000 2005 2010 2015 2020 2025

YEAR

TR

ILLIO

N C

UB

IC F

EE

T

FOODPAPER

REFINING

CHEMICALS

STONE, CLAY, & GLASS PRIMARY METALS

OTHER INDUSTRIES

Figure 3-8. U.S. Industrial Gas Consumption by Industry

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1995 2000 2005 2010 2015 2020 2025

YEAR

TR

ILLIO

N C

UB

IC F

EE

T

PROCESS HEAT

OTHER/COGENERATION

FEEDSTOCK

BOILERS

Note: New cogeneration after 1998 not included.

Figure 3-9. U.S. Industrial Gas Demand Breakdown

period. Efficiency gains and demographic shifts arethe primary factors in this decline. In the ReactivePath scenario, growth in this sector is slightly less than1.0% annually.

Macroeconomic Factors

The primary macroeconomic and exogenous priceassumptions that were used in the demand projectionsare listed in Table 3-1. The same primary macroeco-nomic assumptions were used in both the ReactivePath and Balanced Future cases.

Natural Gas Price and Oil Substitution

The reduction in fuel substitution capability recent-ly observed in both the industrial and electric powersectors is due primarily to actions taken by consumersin response to environmental requirements. In theReactive Path scenario, this trend continues through2025. However, this trend has implications for the rela-tionship between natural gas and competing fuelsand/or feedstocks. Natural gas has historically tendedto sell at a discount to crude oil on an annual averagebasis. In general, this pricing relationship reflected theextent to which petroleum products could be substi-tuted for natural gas, including relative costs of trans-portation and storage. Figure 3-10 shows that naturalgas has approached price parity with crude oil in recentyears, reflecting the increasing investment in either gas-only facilities or facilities that are restricted to usingcleaner petroleum products.

In both the Reactive Path and Balanced Future sce-narios, the price of crude oil is assumed to decline from

year 2003 levels to its historical average of approximate-ly $20/bbl for West Texas Intermediate crude oil in the2005-2025 period. The results of each scenario suggestthat natural gas would tend to sell at a premium tocrude oil in the future using this oil price assumptionbecause of the limited ability of market participants tosubstitute oil and its derivative products for natural gas.However, the Balanced Future scenario projects naturalgas to be priced closer to crude oil parity.

Conclusions

This NPC analysis suggests that natural gas demandis likely to remain at recent levels for several years, withpower generation demand continuing to grow whileindustrial demand continues to erode. Demand is thenlikely to increase slowly due to increasing needs forpower generation, the energy requirements of a grow-ing economy, and the environmental imperatives inboth the United States and Canada. This analysis alsosuggests considerably higher natural gas prices will benecessary to balance supply and demand, resulting inlower demand levels than projected in the 1999 NPCstudy, as well as in other U.S. government forecasts.


0

1

2

1970 1980 1990

YEARB

TU

EQ

UIV

AL

EN

T B

AS

IS

2000

GAS AT PREMIUM TO

CRUDE OIL

GAS AT DISCOUNT TO

CRUDE OIL

Figure 3-10. Ratio of Wellhead Natural Gas Price to West Texas Intermediate Crude Oil Price

Table 3-1. Primary Macroeconomic Assumptions,2005-2025

U.S. GDP Annual Growth 3.0%

U.S. Industrial Production Annual Growth 3.0%

U.S. GDP Deflator Annual Change 2.5%

Canadian GDP Annual Growth 2.6%

Crude Oil, WTI (2002$) $20.00

U.S. Coal Price Annual Changes (2002$) -1.0%

Industrial Demand

This section provides details of natural gas use in theindustrial sector. The historical determinants of indus-trial gas use and the factors that will affect its future useare reviewed and analyzed. Finally, projections ofindustrial gas use through 2025 are summarized.

Industrial consumers used 7.2 trillion cubic feet(TCF) or about 32% of total U.S. gas consumption in2002. Industrial businesses use natural gas for energyand as a raw material or feedstock. Figure 3-11 illus-trates regional energy use for U.S. industrial consumersin 2002. Figure 3-12 provides a basic description of theuse of natural gas as a raw material. Natural gas use inthe industrial sector has developed over many years.Over the last 60 years, industrial consumers have madeconsiderable investment in capital equipment for pref-erential use of natural gas. Except for a few periods,natural gas has been a widely available, cost-effectivefuel and feedstock. Since 2000, however, the price ofnatural gas has risen significantly, attendant with newconcerns about its availability. The high price for nat-ural gas alone changes the competitive environmentfor many industrial consumers.

The price of natural gas relative to other fuels is a keyvariable in future industrial gas demand. A key vari-able to project the future use of natural gas in thisstudy is industrial production. Industrial productionmeasures changes in the output of production versus abaseline year. In contrast to past NPC studies, empha-sis was placed in this analysis on relating gas price fore-casts to future industrial production.

Value of Gas to Industrial Consumers

Natural gas has become an increasingly importantfuel for industrial consumers. Natural gas in industri-al applications offers flexibility, controllability, and lowemissions; relative to other fuels, it has been cost-competitive. Table 3-2 summarizes the characteristicsof natural gas and competing fuels. Historically, natu-ral gas on a heat content (dollars per Btu) basis hasbeen less expensive than all other fuels except for coal.The operational characteristics of natural gas are asgood or better than more-expensive energy sources(e.g., distillate or electricity). Natural gas is widelyavailable, easy to transport, and requires no on-sitestorage. Natural gas can be used in a wide variety ofapplications to provide a high degree of control with-out negatively affecting product quality; for example,in contact heating and drying processes.


PACIFIC 1

MOUNTAIN 1 WEST NORTH

CENTRAL

EAST NORTH

CENTRAL

MID-ATLANTIC

NEW ENGLAND

PACIFIC 2

MOUNTAIN 2

WEST SOUTH

CENTRAL

EAST SOUTH

CENTRAL

SOUTH

ATLANTIC

TR

ILLIO

N B

TU

1,500

500

1,000

0

GAS

OIL

COAL

ELECTRIC

Figure 3-11. U.S. Industrial Energy Use in 2002


Figure 3-12. Simplified Diagram of Natural Gas Use as a Raw Material

• SIDING, PIPE, FLOORING

• SOLVENTS, METAL CLEANING,

• ELECTRONICS, POLYMERS

• PIPE, SHOWER CURTAINS

• DRY CLEANING, METAL

• CLEANING, DEGREASING

• COATINGS, ADHESIVES,

• PRINTED CIRCUIT BOARD

• TOOTHPASTE, COSMETICS,

• FOOD

• FOOD PACKAGING, FILM, • TRASH BAGS, DIAPERS,

• TOYS, HOUSEWARES

• CRATES, DRUMS, BOTTLES,

• FOOD CONTAINERS

• SIDING, WINDOW FRAMES,

• SWIMMING POOL LINERS, PIPES

• AUTOMOTIVE ANTIFREEZE

• PANTYHOSE, CARPETS,

• CLOTHING

• BOTTLES, FILM

• INSULATION, CUPS, MODELS

• INSTRUMENT LENSES

• TIRES, FOOTWEAR, SEALANTS

• CARPET BACKING,

• PAPER COATINGS

• FERTILIZER, FEEDS,

• EXPLOSIVES, CHEMICALS

• CARPETS, CLOTHING,

• HOME FURNISHINGS

• GASOLINE

• PLYWOOD, PARTICLE BOARD,

• INSULATION

• LATEX, PAINTS, OTHER COATINGS,

• ADHESIVES, TEXTILE FINISHING

• ELECTRONICS, METAL CLEANING,

• PAINT REMOVER, SILICONES,

• INSULATION

• GLAZING, SIGNS,

• OTHER ACRYLICS

Chlor-AlkaliCHLOR-ALKALI

Ethane Based

Ethylene

ETHANE-BASED

ETHYLENE

Hydrogen

(non-refinery)

HYDROGEN

(NON-REFINERY)

MethanolMETHANOL

• CLEANER GASOLINE/

• HEATING OIL

• FUEL CELLS

AmmoniaAMMONIA

Sources: American Chemistry Council; Dow Chemical.

NATURAL

GAS

STEAM

POWER

FEEDSTOCK

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 38

Coal Residual Oil Distillate Oil Electricity Natural Gas

Cost Low Low High Mid High Low Mid

Transportation Difficult Difficult Medium Easy Easy

Storage Difficult Difficult Medium NA NA

Combustion Difficult Difficult Medium Easy Easy

Controllability Poor Poor Good Very good Very Good

Direct Contact No No Many Yes Yes

Emissions High High Medium "Zero" Low

Historical Price $1-2/MMBtu $3-5/MMBtu $4-5/MMBtu $12-14/MMBtu $2-4/MMBtu

Major Uses Large boilers,boiler

cogeneration,cement

calcining

Large boilers,refinery heaters,

lime calcining

Diesel fuel fortransportation.Backup fuel for

many small- andmid-sized boilers,

many processheat applications,

primary fuel foronly a few

Electric ArcFurnace, lighting,

machine drive,many drying,

heating, melting,and curing

applications

Boilers, cogeneration(boiler and turbine),all kinds of processheat, largest includechemical, refining,

primary metals, glassmelting

Table 3-2. Characteristics of Industrial Fuels


The importance of gas in the industrial sector rela-tive to other energy sources is shown in Figure 3-13.Gas is the primary fuel for boilers, cogeneration, andprocess heating. Gas is also an important feedstock.However, gas is not the primary fuel in electric appli-cations (e.g., lighting and motors) and transportation,where diesel fuel predominates.

The use of alternate fuels has been important forindustrial consumers. For example, periodically whengas prices were high relative to alternate fuels, facilitiescapable of switching to another less-expensive fuel hadimportant competitive advantages. That role hasdiminished during the last decade, however, as fuel-switching capability has dwindled due to the combi-nation of regulatory and operational factors.Government data on alternate fuel use and capabilityare not current. The NPC modeling was based uponinformation from extensive outreach efforts among theindustrial community.

Natural gas is used in practically every part of theindustrial sector. Figure 3-14 shows that 72% of indus-trial energy and 80% of industrial natural gas is con-sumed in six of the most gas-intensive industries:Chemicals; Petroleum Refining; Primary Metals; Foodand Beverage; Paper; and Stone, Clay, and Glass. Theperformance of these key industries has been used todetermine the overall industrial demand for gas.

Industrial Natural Gas Use Drivers and Trends

Key drivers for industrial energy and gas uses are:

� Production Growth. “Industrial Production” is akey factor in describing the driving forces for totalU.S. energy and gas demand. Many factors deter-mine production growth, including growth in theU.S. and global economies and the competitivenessof U.S. industry in global markets. High relativeenergy prices reduce industry competitiveness,potentially lowering the demand for energy and gas.

� Industry Mix. Long-term trends in industry miximpact energy demand. Over the last 30 years, manytraditional manufacturing industries have declinedin the United States due to foreign competition orother factors. The United States has evolved to moretechnology and service industries, which consumeless energy per gross domestic product (GDP) dol-lar. Some basic energy-intensive industries, such asprimary metals and fertilizer, have experienced tem-porary and permanent declines.

STAND-ALONE BOILER

COGENERATION

PROCESS HEAT

HVAC

OTHER ELECTRIC GENERATION

LEASE AND PLANT

TRANSPORTATION

MACHINE DRIVE

OTHER ELECTRIC

PROCESS COOLING

FEEDSTOCKS

GAS

OIL

COAL

ELECTRIC

0 2,000 4,000 6,000 8,000

TRILLION BTU

Figure 3-13. U.S. Industrial Energy Consumptionby End Use in 2001



� Energy Efficiency and Process Change. In a con-tinuous effort to improve their cost structures, mostindustries focus on energy efficiency. As old equip-ment is replaced and upgraded, the industry stayscompetitive. Process changes and technologyimprovements create major reductions in energyuse. Specifically, the increased use of recycled mate-rials, increased recovery of waste heat and fuels,development of more efficient processes and tech-nologies, and increased penetration of cogenerationsystems have resulted in greater energy efficiencyand increased industrial productivity.

� Fuel Switching. Some industrial applications aredesigned to substitute fuels depending on econom-ics. Short-term fuel switching facilitates alternatefuel use for periods of hours to weeks. For example,gas boilers may switch to residual fuel oil as a sec-ondary fuel when gas prices exceed fuel oil prices ona dollars-per-Btu basis. The total consumption ofthe secondary fuel may not be large, but this switch-ing capability serves an important role in industrycompetitiveness and in temporarily reducing gasdemand. Long-term fuel switching stems from aprocess change to use alternate fuels in response toeconomics or supply concerns, and usually entails alarge capital investment.

� Price Response and Demand Curtailment. Theprospect of a protracted price increase can stimulateinvestments in higher efficiency equipment, fuel-switching equipment, or facility shutdowns. A facil-ity shutdown reduces energy demand but it alsoreduces production capacity, and may lead to loss ofjobs and other negative economic outcomes.

� Changes in Raw Materials. Some changes in rawmaterial actually increase energy consumption. Inpetroleum refining, the crude oil quality has beendeclining, increasing the need for more processing.More complex operations increase energy and natu-ral gas use. New requirements for low-sulfur trans-portation fuels are expected to increase these effects.

� Environmental and Other Regulation. More strin-gent emissions requirements have increased indus-trial natural gas use. Gas is preferred because itlowers emissions more than other fossil fuels andcan meet emission limits at a lower cost. Other reg-ulation has encouraged gas use. For example, thePublic Utility Regulatory Policies Act of 1978encouraged new gas-fired industrial cogeneration

0 2,000 4,000 6,000 8,000

AGRICULTURE

MINING

CONSTRUCTION

FOOD AND BEVERAGE

PAPER

CHEMICALS

PETROLEUM REFINING

STONE, CLAY, & GLASS

PRIMARY METALS

RUBBER

METAL DURABLES

OTHER MANUFACTURING

TRILLION BTU

GAS

OIL

COAL

ELECTRIC

Figure 3-14. U.S. Industrial Energy Consumption by Sector in 2001


facilities during the 1980s and 1990s because theygreatly increased industrial cost efficiencies.

As shown in Figure 3-15, industrial natural gas con-sumption grew steadily up to the early 1970s andpeaked at 8.7 TCF in 1973. After the “oil shock” of thatyear, industrial gas demand dropped more or less con-tinuously to a low of 5.6 TCF in 1983.

Many factors contributed to the decline in naturalgas demand beginning in the early 1970s, as listed here:

� General economic downturns during the period

� Foreign competition

� Evolution toward technology and service industries

� Major increases in efficiency and implementation ofnew technologies.

As the overall economy improved in the late 1980s,industrial production grew. Many industries becamemore competitive and increased production. Adaptingto new environmental regulations, companies becamemore efficient by employing gas technologies.Cogeneration grew very rapidly during this period. Asa result of these trends, industrial gas consumptionrebounded to 8.9 TCF by 1996.

Industrial gas demand started to decline again in1997. Overall industrial production since 1997 hasbeen much slower than earlier in the 1990s. Moreimportantly, the energy-intensive industries have beenimpacted particularly and reported lower productionlevels relative to the non-energy-intensive industries.Weak economic performance by these industries cou-pled with higher gas prices has resulted in the recentdeclining use of natural gas in the sector.

Overview of the Modeling Approach

The industrial sector comprises a diverse mix of cus-tomers, many of which have different gas consumptionneeds, drivers, fuel-switching capability, and price elas-ticity dynamics. An accurate representation of the sec-tor required a “bottom-up” approach to modeling forthe NPC study. The model was developed to forecastU.S. industrial demand for 26 industries, 11 regions,and 4 end-use categories (boilers, process heat, feed-stocks, and other) reflecting economic growth assump-tions and a range of natural gas prices. Because of itssize, complexity, and importance to gas-consumptiontrends, the modeling of the chemical industry was fur-ther disaggregated into ammonia, methanol, hydrogen,and other chemical industry products.

The model was designed to explicitly capturechanges and improvements in technology includingimprovements in energy efficiency, short- and long-term fuel switching, and global competition. In orderto develop input parameters and to validate the results,outreach seminars were conducted with representa-tives of key gas intensive industries to capture emerg-ing trends and key drivers.

The model was used also to test various demandsensitivities and policy choices. Adjustments for com-petitive and price elasticity effects were made to finetune demand in each sector. Particular focus wasplaced on gas price elasticity dynamics because modelprice outputs were on the upper end of historicalnorms, and there was little data to calibrate sustaineddemand response to higher prices and greater globalcompetition.

Figures 3-16 and 3-17 illustrate the analysis processfor non-chemical and chemical industry demand,respectively. These figures show that the projection ofgas demand for each sector is made in a multi-stepprocess. Historical energy consumption and industri-al production data are used to calculate historical “gas


TR

ILLI

ON

CU

BIC

FE

ET


0

2

4

6

8

10

1960 1970 1980YEAR

1990 2000

Figure 3-15. Industrial Natural Gas Consumption


Figure 3-16. Industrial Demand Analysis – Non-Chemical Uses

BASE INTENSITY TRENDS (STEAM

REQUIREMENT/UNIT OF OUTPUT)1

NATURAL GAS CONSUMPTION

IN BOILERS

BASE INTENSITY TRENDS

(GAS USE/UNIT OF OUTPUT)


IN PROCESS HEAT/OTHER USES

1 Steam unit efficiencies are assumed to improve 0.3% per year.

2 Gas price elasticity factors are from Industrial Sector Technology Use Model, Energy and Environment Analysis, Inc.

BOILER GAS USE PROCESS HEAT &

OTHER GAS USES

GAS PRICE ELASTICITY2

BASE GAS USE IN PROCESSES

AND OTHER CATEGORIES

BOILER FUEL SWITCHING

AND EFFICIENCY EFFECTS

BASE GAS USE IN BOILERS

Figure 3-17. Industrial Demand Analysis – Chemical Uses

PRODUCT DEMAND GROWTH

(AMMONIA, METHANOL,

ETHYLENE)


FOR FEEDSTOCK/RAW MATERIAL

GROSS DOMESTIC PRODUCT

PRODUCTION INDEX

BASE INTENSITY TRENDS (GAS

USE/UNIT OF PRODUCTION)

BASE GAS USE TO PRODUCE

OTHER CHEMICALS


TO PRODUCE OTHER CHEMICALS

FEEDSTOCK/RAW MATERIAL OTHER CHEMICALSHYDROGEN

(NON-REFINERY)

NON-REFINERY

PRODUCED

HYDROGEN

GROWTH RATES

NATURAL GAS

CONSUMPTION

FOR

NON-REFINERY

HYDROGEN

IMPORT SHARE (BASED ON

IMPORTED PRODUCT PRICES)

PRODUCTION COSTING

MODEL (DOMESTIC COST

OF PRODUCTION)

GAS PRICE ELASTICITY


energy intensity” values for each sector. Future gasenergy intensity values are projected reflecting:

� Trends in long-term technology and efficiency effects

� Fuel-switching effects for alternate fuels (subject toknown limits)

� Price elasticity that results in additional fuel-switching capability or investment in efficiencyimprovements.

The forecasted values of gas energy intensity in eachsector are applied to projected industrial production.The model outputs reflect recent trends in the compo-sition of the industrial sector of the economy and theassumptions of overall economic growth. In addition,the projected industrial production reflects globalcompetition from countries that have lower natural gasand energy costs and where energy cost differentialsconstitute a significant competitive advantage relativeto product transportation costs.

Projection of Future Demand

The NPC projection of industrial gas demandaddressed key factors affecting gas-intensive industries.These include:

� Industrial production growth

� Overall efficiency trends

� Fuel switching, both short- and long-term

� Demand elasticity

� Effects of global competition on commodity chem-icals.

The EEA model employed by the NPC study fore-cast 26 industries in 11 North American regions with4 end-use categories (boilers, process heat, feedstocks,and other). The model incorporated economicgrowth assumptions and a range of future natural gasprices. Because of its size, complexity, and impor-tance to gas consumption trends, the modeling of thechemical industry was further disaggregated intoammonia, methanol, hydrogen, and other chemicalproducts.

The model was designed to explicitly capture tech-nology improvements in energy efficiency, short- andlong-term fuel switching, and global competition. Inorder to develop input parameters and to validate the

results, outreach seminars were conducted with repre-sentatives of key industry segments to capture emerg-ing trends and key drivers.

Industrial production growth is the key driver of gasconsumption. Table 3-3 lists the growth factors used inthe projections compared to recent historical data(1992-1998). Industrial production for gas-intensiveindustries grew at a slower rate than for other indus-tries. Often, energy consumption grew at a slower ratethan production due to better energy efficiency.During the 1992-1998 period, overall energy con-sumption grew at only 1.3% per year.

Compared to other fuels, gas consumption grew at afaster rate, particularly from growth in gas-intensiveprocesses. New cogeneration during this period alsocontributed to the increase. Gas consumption grew by2.4% per year when cogeneration was excluded.

Industrial production growth was strong during the1990s. This high growth rate is not forecast in thisstudy. During the forecast period of 2001-2025, indus-trial production is projected to increase by only 1.1%per year and gas consumption is expected to decreaseby 0.4% per year. The decline in gas consumption isdue to the overall lower projection of industrial pro-duction, continued efficiency improvements, processchange, and the overall effects of higher natural gasprices. Some increased fuel switching away from gas isprojected towards the end of the forecast, as gas pricestrend higher.

The principal differences between the Reactive Pathand the Balanced Future scenarios with regard toindustrial consumers are assumptions for fuel-switching capability, both short-term and long-term,and greater efficiency improvement in the BalancedFuture scenario.

To model fuel-switching behavior of industrial con-sumers, boiler-switching relationships were developedfor each region of the United States and Canada; theseare contained in the Demand Task Group Report. Anexample of these relationships is shown by Figures 3-18 and 3-19, the boiler-switching capability modeledin the Reactive Path and Balanced Future scenarios,respectively, for the West South Central region of theUnited States. In the Balanced Future scenario, thepercentage of industrial boilers that would be able tofuel switch was increased from a low in 2003 of 2% to8%, depending on the region, to a high of 28% in all


1992-1998 2001-2030

IndustrialProduction

GasUse

Gas NoCogen

IndustrialProduction

GasUse

Gas-Intensive Industries 2.4 2.9 4.3 1.1 -0.6

Food and Beverage 1.8 3.8 4.0 1.1 -0.4

Paper 0.4 3.5 4.6 0.0 -1.3

Petroleum Refining 1.2 6.7 8.2 1.0 -1.2

Chemicals* 0.6 1.3 0.4 0.8 -0.1

Stone, Clay, and Glass 3.8 2.8 2.8 2.8 0.8

Primary Metals 3.5 1.8 0.3 -0.2 -2.7

Other Industries 5.2 1.9 2.0 2.6 0.1

*Industrial production growth rate for 1992 to 1998 is for the Organic Chemicals industry, overall industry growthwas much higher but includes less gas-intensive processes. Industrial production growth rate for 2001 to 2030uses the model results’ average of the growth rates of gas feedstocks and non-gas-intensive chemical industryproduction.

Table 3-3. Growth Factors (Percent)

Figure 3-18. Industrial Boiler Switching Curve Used for West South Central Region of United States in Reactive Path Scenario

MAXIMUM PERCENTAGE OF RESIDUAL FUEL USE REACHED

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0% 5% 10% 15% 20% 25% 30%

PR

ICE

DIF

FE

RE

NC

E (

GA

S M

INU

S R

ES

IDU

AL

FU

EL O

IL, IN

2001 D

OLLA

RS

)

PERCENT OF BOILERS THAT CAN SWITCH TO RESIDUAL FUEL OIL

2004-2025

Note: Switching curve is constant throughout forecast.

regions by 2025. Since the switchable boilers cannotoperate 100% on oil due to operational constraints, themaximum oil percentage for the switching curves wasvaried to account for the differences in boiler capabili-ties by region.

The aggregate results modeled for industrial boilerswitching and the attendant fuel utilization are illus-trated in Figures 3-20 and 3-21 for the Reactive Pathscenario, and in Figures 3-22 and 3-23 for the BalancedFuture scenario.

To further apply the process described by Figures 3-16 and 3-17 (“Industrial Demand Analysis”), theNPC developed price elasticity relationships. Baseelasticity trends were taken from the Industrial SectorTechnology Use Model, developed by EEA; these weremodified to reflect the major industry groupings ana-lyzed by the NPC Study group, and are shown in Table3-4. Further, energy intensity price elasticity factorswere taken from the Industrial Sector Technology UseModel, and modified to reflect the major industrygroupings, and then developed for both the ReactivePath and Balanced Future scenarios; Tables 3-5 and 3-6 contain these factors. The Demand Task Group

Report provides details of energy intensity price elas-ticity for each industry sector. An example of theserelationships is shown by Figures 3-24 and 3-25 for thePetroleum Refining industry, as modeled in theReactive Path and Balanced Future scenarios, respec-tively.

The resulting industrial gas demand projected forthe Reactive Path scenario is shown in Figure 3-26.This suggests a continuation of the current decline ingas demand from the 1997 high down to about 7 TCFper year in about 2007. Industrial gas demand in bothscenarios is forecast to be relatively flat to 2013, afterwhich a small increase begins.

Figure 3-27 shows the historical and projected gasdemand by industry, illustrating both the trajectoryand overall magnitude of gas consumption in eachindustry. The chemicals industry is projected toremain the largest industrial gas consumer, althoughits consumption drops significantly from recent levels.This decline is largely due to loss of market share toglobal competition at the projected gas prices. Naturalgas feedstock is projected to drop for ammonia as isethane used for ethylene production.


Figure 3-19. Industrial Boiler Switching Curve Used for West South Central Region of United States in Balanced Future Scenario

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0% 5% 10% 15% 20% 25% 30%

PERCENT OF BOILERS THAT CAN SWITCH TO RESIDUAL FUEL OIL

PR

ICE

DIF

FE

RE

NC

E (

GA

S M

INU

S R

ES

IDU

AL

FU

EL O

IL,

IN 2

00

1 D

OL

LA

RS

) 2004 2010 2025

Note: Switching curve shifts over time, allowing more switching.


Figure 3-20. Industrial Boiler Switching in Reactive Path Scenario

0

20

40

60

80

100T

RIL

LIO

N B

TU

OF

RE

SID

UA

L F

UE

L O

IL

0

20

40

60

80

100

PE

RC

EN

T O

F S

WIT

CH

ING

CA

PA

BIL

ITY

US

ED

2001 2005 2010YEAR

2015 2020 2025

MAXIMUM BOILER OIL DEMAND POTENTIAL (TBTU)

BOILER RESID DEMAND (TBTU)RESID SWITCHING CAPABILITY USED (%)

Figure 3-21. Industrial Boiler Fuel Use in Reactive Path Scenario

0

500

1,000

2001 2005 2010

YEAR

2015 2020 2025

1,500

2,000

2,500

3,000

3,500

TR

ILLIO

N B

TU

OF

FU

EL C

ON

SU

ME

D

BOILER GAS DEMAND

BOILER RESIDUAL FUEL OIL DEMAND


Figure 3-22. Industrial Boiler Switching in Balanced Future Scenario

0

100

200

300

400

500

600

700

800T

RIL

LIO

N B

TU

OF

RE

SID

UA

L F

UE

L O

IL

0

10

20

30

40

50

60

70

80

90

100

PE

RC

EN

T O

F S

WIT

CH

ING

CA

PA

BIL

ITY

US

EDMAXIMUM BOILER OIL DEMAND POTENTIAL (TBTU)

BOILER RESID DEMAND (TBTU)RESID SWITCHING CAPABILITY USED (%)

2000 2005 2010

YEAR

2015 2020 2025

Figure 3-23. Industrial Boiler Fuel Use in Balanced Future Scenario

0

500

1,000

1,500

2,000

2,500

3,000

3,500

2001 2005 2010 2015 2020 2025

YEAR

TR

ILLIO

N B

TU

OF

FU

EL C

ON

SU

ME

D

BOILER GAS DEMAND

BOILER RESIDUAL FUEL OIL DEMAND

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 48

YearFood &

Beverage PaperPetroleumRefining Chemicals

Stone,Clay, &Glass

Iron &Steel

PrimaryAluminum

OtherPrimaryMetals

OtherManufacturing

Non-Manufacturing

2001 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2002 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2003 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2004 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2005 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

2006 0.996 0.968 0.991 0.974 0.982 0.967 0.953 0.992 1.004 1.012

2007 0.993 0.936 0.983 0.949 0.963 0.933 0.906 0.983 1.009 1.023

2008 0.989 0.903 0.974 0.923 0.945 0.900 0.859 0.975 1.013 1.035

2009 0.985 0.871 0.965 0.897 0.927 0.866 0.812 0.967 1.017 1.046

2010 0.982 0.839 0.957 0.872 0.909 0.833 0.765 0.959 1.022 1.058

2011 0.979 0.816 0.951 0.867 0.900 0.819 0.734 0.954 1.023 1.065

2012 0.976 0.793 0.944 0.863 0.892 0.805 0.703 0.949 1.024 1.071

2013 0.974 0.770 0.938 0.858 0.883 0.791 0.671 0.944 1.025 1.078

2014 0.971 0.748 0.932 0.854 0.875 0.777 0.640 0.939 1.026 1.085

2015 0.968 0.725 0.926 0.850 0.866 0.763 0.609 0.934 1.027 1.091

2016 0.967 0.712 0.923 0.854 0.862 0.757 0.594 0.930 1.025 1.099

2017 0.965 0.700 0.920 0.858 0.858 0.750 0.580 0.927 1.024 1.107

2018 0.964 0.687 0.916 0.863 0.854 0.743 0.565 0.924 1.022 1.115

2019 0.962 0.675 0.913 0.867 0.849 0.737 0.551 0.920 1.020 1.123

2020 0.961 0.662 0.910 0.871 0.845 0.730 0.536 0.917 1.019 1.131

2021 0.959 0.653 0.907 0.875 0.842 0.724 0.527 0.913 1.015 1.136

2022 0.958 0.644 0.905 0.878 0.838 0.718 0.519 0.909 1.012 1.141

2023 0.956 0.635 0.903 0.882 0.834 0.712 0.510 0.905 1.008 1.146

2024 0.955 0.626 0.900 0.885 0.830 0.706 0.501 0.901 1.005 1.151

2025 0.953 0.617 0.898 0.889 0.827 0.700 0.492 0.898 1.001 1.156

Table 3-4. Base Energy Intensity Trends (Indexed, 2001 = 1.000)

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 49

YearFood &


Stone,Clay, &Glass

Iron &Steel

PrimaryAluminum

OtherPrimaryMetals

OtherManufacturing

Non-Manufacturing

2000 (0.600) (0.050) (1.100) (0.130) (0.500) - (0.400) (0.500) (0.600) (0.350)

2001 (0.610) (0.050) (1.103) (0.130) (0.507) (0.014) (0.407) (0.523) (0.617) (0.365)

2002 (0.620) (0.050) (1.107) (0.130) (0.513) (0.028) (0.413) (0.547) (0.633) (0.380)

2003 (0.630) (0.050) (1.110) (0.130) (0.520) (0.042) (0.420) (0.570) (0.650) (0.395)

2004 (0.640) (0.050) (1.113) (0.130) (0.527) (0.056) (0.427) (0.593) (0.667) (0.410)

2005 (0.650) (0.050) (1.117) (0.130) (0.533) (0.070) (0.433) (0.617) (0.683) (0.425)

2006 (0.660) (0.050) (1.120) (0.130) (0.540) (0.084) (0.440) (0.640) (0.700) (0.440)

2007 (0.670) (0.050) (1.123) (0.130) (0.547) (0.098) (0.447) (0.663) (0.717) (0.455)

2008 (0.680) (0.050) (1.127) (0.130) (0.553) (0.112) (0.453) (0.687) (0.733) (0.470)

2009 (0.690) (0.050) (1.130) (0.130) (0.560) (0.126) (0.460) (0.710) (0.750) (0.485)

2010 (0.700) (0.050) (1.133) (0.130) (0.567) (0.140) (0.467) (0.733) (0.767) (0.500)

2011 (0.710) (0.050) (1.137) (0.130) (0.573) (0.154) (0.473) (0.757) (0.783) (0.515)

2012 (0.720) (0.050) (1.140) (0.130) (0.580) (0.168) (0.480) (0.780) (0.800) (0.530)

2013 (0.730) (0.050) (1.143) (0.130) (0.587) (0.182) (0.487) (0.803) (0.817) (0.545)

2014 (0.740) (0.050) (1.147) (0.130) (0.593) (0.196) (0.493) (0.827) (0.833) (0.560)

2015 (0.750) (0.050) (1.150) (0.130) (0.600) (0.210) (0.500) (0.850) (0.850) (0.575)

2016 (0.760) (0.050) (1.153) (0.130) (0.607) (0.224) (0.507) (0.873) (0.867) (0.590)

2017 (0.770) (0.050) (1.157) (0.130) (0.613) (0.238) (0.513) (0.897) (0.883) (0.605)

2018 (0.780) (0.050) (1.160) (0.130) (0.620) (0.252) (0.520) (0.920) (0.900) (0.620)

2019 (0.790) (0.050) (1.163) (0.130) (0.627) (0.266) (0.527) (0.943) (0.917) (0.635)

2020 (0.800) (0.050) (1.167) (0.130) (0.633) (0.280) (0.533) (0.967) (0.933) (0.650)

2021 (0.810) (0.050) (1.170) (0.130) (0.640) (0.294) (0.540) (0.990) (0.950) (0.665)

2022 (0.820) (0.050) (1.173) (0.130) (0.647) (0.308) (0.547) (1.013) (0.967) (0.680)

2023 (0.830) (0.050) (1.177) (0.130) (0.653) (0.322) (0.553) (1.037) (0.983) (0.695)

2024 (0.840) (0.050) (1.180) (0.130) (0.660) (0.336) (0.560) (1.060) (1.000) (0.710)

2025 (0.850) (0.050) (1.183) (0.130) (0.667) (0.350) (0.567) (1.083) (1.017) (0.725)

Table 3-5. Energy Intensity Price Elasticity for Reactive Path Scenario (Change in Energy Intensity per Change in Gas Price)

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 50

YearFood &


Stone,Clay, &Glass

Iron &Steel

PrimaryAluminum

OtherPrimaryMetals

OtherManufacturing

Non-Manufacturing

2000 (0.600) (0.050) (1.100) (0.130) (0.500) - (0.400) (0.500) (0.600) (0.350)

2001 (0.610) (0.050) (1.103) (0.130) (0.507) (0.014) (0.407) (0.523) (0.617) (0.365)

2002 (0.620) (0.050) (1.107) (0.130) (0.513) (0.028) (0.413) (0.547) (0.633) (0.380)

2003 (0.630) (0.050) (1.110) (0.130) (0.520) (0.042) (0.420) (0.570) (0.650) (0.395)

2004 (0.679) (0.052) (1.167) (0.136) (0.557) (0.072) (0.452) (0.643) (0.713) (0.443)

2005 (0.727) (0.055) (1.224) (0.142) (0.594) (0.102) (0.485) (0.715) (0.776) (0.491)

2006 (0.776) (0.057) (1.281) (0.148) (0.631) (0.132) (0.517) (0.788) (0.839) (0.539)

2007 (0.825) (0.059) (1.338) (0.154) (0.668) (0.162) (0.550) (0.860) (0.902) (0.587)

2008 (0.873) (0.061) (1.396) (0.160) (0.705) (0.192) (0.582) (0.933) (0.964) (0.635)

2009 (0.922) (0.064) (1.453) (0.165) (0.742) (0.221) (0.615) (1.005) (1.027) (0.683)

2010 (0.970) (0.066) (1.510) (0.171) (0.779) (0.251) (0.647) (1.078) (1.090) (0.731)

2011 (1.019) (0.068) (1.567) (0.177) (0.816) (0.281) (0.679) (1.151) (1.153) (0.779)

2012 (1.068) (0.070) (1.624) (0.183) (0.853) (0.311) (0.712) (1.223) (1.216) (0.827)

2013 (1.116) (0.073) (1.681) (0.189) (0.890) (0.341) (0.744) (1.296) (1.279) (0.875)

2014 (1.165) (0.075) (1.738) (0.195) (0.927) (0.371) (0.777) (1.368) (1.342) (0.923)

2015 (1.214) (0.077) (1.795) (0.201) (0.964) (0.401) (0.809) (1.441) (1.405) (0.970)

2016 (1.262) (0.080) (1.853) (0.207) (1.001) (0.431) (0.842) (1.513) (1.467) (1.018)

2017 (1.311) (0.082) (1.910) (0.213) (1.038) (0.461) (0.874) (1.586) (1.530) (1.066)

2018 (1.360) (0.084) (1.967) (0.219) (1.075) (0.491) (0.906) (1.659) (1.593) (1.114)

2019 (1.408) (0.086) (2.024) (0.225) (1.112) (0.521) (0.939) (1.731) (1.656) (1.162)

2020 (1.457) (0.089) (2.081) (0.230) (1.148) (0.550) (0.971) (1.804) (1.719) (1.210)

2021 (1.505) (0.091) (2.138) (0.236) (1.185) (0.580) (1.004) (1.876) (1.782) (1.258)

2022 (1.554) (0.093) (2.195) (0.242) (1.222) (0.610) (1.036) (1.949) (1.845) (1.306)

2023 (1.603) (0.095) (2.252) (0.248) (1.259) (0.640) (1.068) (2.022) (1.908) (1.354)

2024 (1.651) (0.098) (2.310) (0.254) (1.296) (0.670) (1.101) (2.094) (1.970) (1.402)

2025 (1.700) (0.100) (2.367) (0.260) (1.333) (0.700) (1.133) (2.167) (2.033) (1.450)

Table 3-6. Energy Intensity Price Elasticity for Balanced Future Scenario (Change in Energy Intensity per Change in Gas Price)


Figure 3-24. Change in Energy Intensity for Petroleum Refining Sector in Reactive Path Scenario

0

0.5

1.0

1.5

2.0

2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

GAS PRICE (2001 DOLLARS)

GA

S U

SE

PE

R U

NIT

OU

TP

UT

(IN

DE

XE

D,

2003 =

1.0

)200320102025

Figure 3-25. Change in Energy Intensity for Petroleum Refining Sector in Balanced Future Scenario

0

0.5

1.0

1.5

2.0

2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

GAS PRICE (2001 DOLLARS)

GA

S U

SE

PE

R U

NIT

OU

TP

UT

(IN

DE

XE

D, 2003 =

1.0

)

200320102025

The petroleum refining industry is projected toremain the second largest gas consumer. Gas con-sumption is projected to decline at a lower rate compared to the chemicals industry. The other gas-intensive industries project fairly flat gas demand, withthe exception of primary metals. The primary metalsindustry has the most significant and sustained declinein gas consumption of all of the industries due to con-tinuing process change and global competition. The“other” industry group is the only segment to show anincrease in gas demand in the forecast, although it doesnot grow above historical levels. Each industry isaddressed in more detail in the Demand Task GroupReport.

Figure 3-28 shows the Balanced Future forecastof overall industrial gas demand. Figure 3-29 pro-vides this information for each industry segmentmodeled by the NPC. These projections show adecline from current levels, though not quite asmuch as in the Reactive Path scenario. The projec-tion fluctuates around the 7 TCF per year level overmost of the forecast period. Although more fuel-switching capability is in this scenario, gas con-sumption is actually higher. Fuel-switching


Figure 3-26. U.S. Industrial Gas Demand in Reactive Path Scenario

0

1995 2005

YEAR

2015 2025

6

7

8

9

10T

RIL

LIO

N C

UB

IC F

EE

T


Figure 3-27. U.S. Industrial Gas Consumption by Industry in Reactive Path Scenario

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1995 2000 2005 2010 2015 2020 2025

YEAR

TR

ILLIO

N C

UB

IC F

EE

T

FOODPAPER

REFINING

CHEMICALS

STONE, CLAY, & GLASS PRIMARY METALS

OTHER INDUSTRIES

reduces peak demands, and thus price volatility,without large effects on annual gas load. Theincreased flexibility in the Balanced Future sce-nario lowers gas prices and allows industry to relyon gas to a greater degree.

The Balanced Future scenario shows a decline inthe chemicals industry. The gas consumption in theBalanced Future for industry stays generally above2.5 TCF per year whereas it is significantly lower inthe Reactive Path. Higher gas use stems from lowerprices in the Balanced Future scenario and increasedcompetitiveness of the industry.

Major Gas-Intensive Industries

Chemicals

Natural gas is used in the chemical industry as a fueland as a raw material. Natural gas liquids, includingethane, propane, and butane, are major petrochemicalfeedstocks contributing to the production of a host ofother consumer goods, such as plastics, pharmaceuti-cals, and electronic materials.


Figure 3-28. U.S. Industrial Gas Demand in Balanced Future Scenario

0

1995 2005

YEAR

2015 2025

6

7

8

9

10T

RIL

LIO

N C

UB

IC F

EE

T


Figure 3-29. U.S. Industrial Gas Consumption by Industry in Balanced Future Scenario

0.5

0

1.0

1995 2000 2005

YEAR

2010 2015 2020 2025

1.5

2.0

2.5

3.0

3.5

TR

ILL

ION

CU

BIC

FE

ET

CHEMICALS

PRIMARY METALS

FOOD

REFINING

OTHER INDUSTRIES

STONE, CLAY, & GLASS

PAPER

Role of Chemicals in the U.S. Economy. The U.S.chemical industry (SIC 28 or NAICS 325)1 represents10,000 companies operating 13,500 manufacturingfacilities across all 50 states. In 2001, the chemicalindustry contributed $163.5 billion to U.S. GDP, near-ly 2% of total GDP, more than any other manufactur-ing industry. The industry directly employs over onemillion people. According to the American ChemistryCouncil, the industry has a 5:1 multiplier effect in theeconomy, such that its direct employment of one mil-lion people creates another 5 million jobs. This meansthat, in total, the chemical industry is responsible forabout 6.1 million jobs in the United States, or about5% of the total U.S. workforce.

The U.S. chemical industry has developed into thelargest chemical segment in the world, in part fromaccess to low-cost energy and feedstock in the form ofnatural gas. The U.S. chemical industry accounts formore than a quarter of total world production ofchemical products. The industry is the nation’s topexporter. In 2002, the industry exported $81.1 billionof goods and services, more than agriculture, aero-space, or motor vehicles. While chemicals are thelargest exporting industry in the United States,imports have grown in recent years such that the bal-ance of trade in chemicals declined from a favorable$20.5 billion surplus in 1995, to the first-ever tradedeficit of $5 billion in 2002.

Intensity of Natural Gas Use in the ChemicalIndustry. Substantial improvements in energy effi-ciency have taken place in the energy-intensive chemi-cal industry. Since 1974, the industry has reduced itsenergy use for fuel and power consumption per unit ofoutput by nearly 40%, as illustrated in Figure 3-30.One of the principal sources of efficiency gains hasbeen implementation of cogeneration technologies.These applications create two forms of energy (electricpower and steam) with the same amount of fuel, andare often twice as efficient as older utility generationfacilities.

Chemical Industry Demand Outlook. The chemi-cal industry uses about 2.5 TCF of gas annually and isthe largest single industrial user of natural gas (about35%), accounting for 12% of all U.S. natural gas con-

sumption. The industry uses 76% of its natural gasconsumption for fuel and power. The majority ofsteam boilers and cogeneration in chemical industryfacilities are fueled by natural gas. The remaining 24%of natural gas consumption is directly used as feed-stock, primarily in the manufacture of hydrogen,ammonia, and methanol.

The average operating margin (a measure of prof-itability) for basic chemical companies was 6.8% in1999, when the price of natural gas averaged $2.27 permillion Btu (MMBtu). In 2001, when the price of nat-ural gas averaged $3.97 per MMBtu, operating margindropped to 0.6%. This decrease in operating marginsled many chemical companies to evaluate whether tocontinue operations in the United States. During thewinter 2000-2001 natural gas price spike, some chemi-cal operations were idled: about 50% of the methanolcapacity, 40% of the ammonia capacity, and 15% of theethylene capacity simply shut down and furloughedtheir workforce.

In recent years, the chemical industry experienced aprotracted inventory correction, impacts of the highvalue of the dollar, higher natural gas prices, as well as a global recession. Both the Reactive Path andBalanced Future scenarios incorporate a recovery in


1 The North American Industry Classification System(NAICS) has replaced the U.S. Standard IndustrialClassification (SIC) system.

Figure 3-30. Energy Intensity for Fuel and Powerin the U.S. Chemical Industry

0

20

19801975 19901985

YEAR

1995 2000

40

60

80

100

120

IND

EX

(1

97

4 =

10

0)

Source: American Chemistry Council.

overall industrial production, consistent with overallGDP growth of 3%.

Figure 3-31 shows the NPC Reactive Path projectionwith a breakdown of gas demand for boilers, processheat, other processes, and feedstocks. Production inthe gas-intensive organic chemicals industry grew byonly 0.6% annually from 1992 to 1998 and gas usegrew by 1.3% during the same period. Gas use in boil-ers fell during the historical period but this was offsetby increases in gas use for process heaters, feedstocks,and other processes including cogeneration. Gas usewould have increased at only 0.4% per year withoutnew cogeneration.

Gas use for feedstocks and other processes decline inthe projection due to higher gas prices. The mostaffected industries are petrochemical and basic chemi-cal sub-industries that use gas as a feedstock, e.g.,ammonia production. The decline would be offset ifthere is even higher hydrogen production than cur-rently anticipated for use in the refinery sector to man-ufacture low-sulfur transportation fuel. Gas use forboilers and process heat increases because of growth inother segments of the industry, including drugs, andsoaps and detergent manufacture. New cogeneration is

accounted for in the power sector but would con-tribute to increased gas consumption in the chemicalindustry.

Figure 3-32 shows a projection of gas use in thechemical industry for the Balanced Future scenario.All of the components are higher than in the ReactivePath scenario due to the lower gas prices.

Petroleum Refining

The U.S. refining industry (SIC 291 or NAICS32411) transforms crude oil into transportation fuels,such as gasoline and diesel fuel; lubricants; industrialfuels; and chemical plant feedstocks. The industry ishighly capital intensive with an infrastructure replace-ment value of about $300 billion. Each day, U.S.refineries process more than 16 million barrels ofcrude oil to produce 350 million gallons of gasoline,210 million gallons of distillate products, and 125 mil-lion gallons of other finished products. The refiningindustry employs about 100,000 people including con-tract workers.

Intense competition has resulted in a low return oncapital for the refining industry over the last 20 years.


Figure 3-31. U.S. Chemical Industry Gas Demand in Reactive Path Scenario

0

200

400

1995 2000 2005

YEAR

2010 2015 2020 2025

600

800

1,000

1,200

1,400

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

FEEDSTOCK

During this period, the refining industry has investedheavily to meet environmental regulations. The pend-ing clean fuels regulations will require an additional$17 billion in capital investment. The heavy invest-ment burden has led to the closure of over half of U.S.refineries since the mid-1970s. The last new refinerywas built in 1976. The remaining refineries have keptpace with the growing product demand throughimproved utilization and incremental investment.Product imports have remained essentially constant atless than 1 million barrels per day.

Refineries are the second largest industrial consumerof natural gas, representing 14% of total industrialconsumption. Natural gas usage at U.S. refineries willbe in flux from 2003 to 2025. Increasing demand forpetroleum products, net efficiency gains, and cleanfuels regulations are key factors impacting future natu-ral gas demand from U.S. refineries. In general, effi-ciency gains in the industry will be offset by capacityexpansions and increased processing complexity toproduce cleaner fuels from lower quality feedstocks.The price of natural gas also affects refinery demandbecause fuel-switching alternatives are readily availableto most refiners.

Refinery Energy Fundamentals. Petroleum refiner-ies are energy intensive, consuming over 500 thousandBtu for every barrel of crude oil processed. Much ofthe energy required is derived from crude oil as it isconverted into finished gasoline and diesel fuel. Abreakdown of energy used in the refining process isshown in Figure 3-33.

Refineries purchase and produce energy during therefining process. Refinery fuel gas, a byproduct prima-rily from catalytic cracking of heavy crudes and fromprocess heaters, augments purchased energy.Purchased energy consists primarily of natural gas,electricity, and steam. Utility companies or cogenera-tion facilities supply electricity to refineries. Refineriespurchase steam from third parties, typically from thecogeneration units.

Natural gas is used in three principal processes:

� To supplement refinery fuel gas system

� Feed gas for hydrogen generation units

� Fuel for gas turbines used to generate power or drivelarge rotating equipment.


Figure 3-32. U.S. Chemical Industry Gas Demand in Balanced Future Scenario

0

200

400

1995 2000 2005 2010

YEAR

2015 2020 2025

600

800

1,000

1,200

1,400B

ILLIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

FEEDSTOCK

The refinery fuel gas system is complex. Mostprocess units produce fuel gas but it is consumed invarious processes. By altering process conditions, morefuel gas can be produced but at the expense of liquidproducts, particularly propane, butane, and gasoline.

As natural gas prices on a Btu-basis exceed those ofliquid products, overall refinery economics can favoradditional fuel gas production to effectively reduce theneed for natural gas. With efficiency improvements,less natural gas is required to supplement refinery fuelgas. Propane and butane typically account for about1% of refinery energy needs. Many refineries are ableto vaporize propane or butane into the fuel gas systemas an alternate to natural gas. This capability is limit-ed, however, by infrastructure and other technical fac-tors (e.g., re-condensation of the propane and butane).Heavy fuel oil currently accounts for about 1% ofrefinery energy. Historically, refineries burned heavyfuel oil in some heaters and boilers thereby reducingthe need for natural gas. However, environmental reg-ulations requiring reduction of sulfur and particulateemissions have largely eliminated oil burning.

Cogeneration of electric power and steam atrefineries significantly improves overall efficiency.Natural gas, as opposed to refinery fuel, is typicallyburned in gas turbine generators for reliability andwarranty concerns. Sustained higher natural gasprices would provide an incentive for refiners toswitch to distillate fuel use. Higher natural gas pricesrelative to purchased power prices may discouragenew cogeneration projects.

Hydrogen is used in the desulfurization process ofgasoline and distillate fuels. Natural gas is used typi-cally as feed gas for hydrogen generators. The hydro-gen unit converts natural gas to hydrogen and carbondioxide (CO2), which is most often vented to theatmosphere. Refinery fuel gas, propane, butane, andlight naphtha could be substituted for natural gasfeedstock to hydrogen plants with modifications toequipment. These alternate feedstocks, however,increase CO2 venting compared to natural gas.

Projections of Future Natural Gas Needs. Therefinery industry has made a projection of future gasdemand, assuming demand growth and prevailing nat-ural gas prices. Refineries expect to expand capacity byabout 1.5% a year to keep pace with product demand.This growth rate is less than half the projected indus-trial growth rate and includes an allowance forimproved fuel efficiency of motor vehicles. Whilesomewhat higher than growth rates in the immediatepast, the projected growth rate is close to actual growthover the last 25 years.

Since the late 1970s, more than half of all U.S.refineries have shut down. The remaining refineriesproduce more total products and have invested inimprovements to increase capacity. Imports have beenrelatively constant over the past decade. The NPCanalysis assumes refinery capacity growth of 1% a yearwith product imports supplying any shortfall.

The major integrated oil companies have commit-ted to the U.S. government to improve refinery ener-gy efficiency by 10% from 2002 to 2012. As refineriesbecome more efficient, energy usage will increase asadditional processing and new units meet morestringent clean fuels requirements. Over the past 10years, refinery efficiency improvements have aver-aged about 1.5% per year, while overall energy useper barrel of crude oil has dropped by only about


Figure 3-33. Breakdown of Energy Used in U.S. Petroleum Refining Processes

REFINING

FUEL GAS

43%

FCC

COKE

22%

ELECTRICITY

10%

NATURAL

GAS, STEAM,

AND OTHER

25%

PURCHASEDPRODUCED &

CONSUMED

Source: ConocoPhillips.

0.5% per year. In addition, a point of diminishingreturns on energy improvement is expected as effi-ciency improves.

In recent years, the most efficient refineries havereached an efficiency plateau while less efficientrefineries continue to improve. For the NPC analysis,energy efficiency is expected to improve 1% per yearfrom 2002 to 2012, 0.5% per year for 2013 to 2022, and0.25% per year thereafter through 2030. Refinery flarelosses are assumed to reduce by 50% compared to cur-rent levels, with a corresponding reduction in naturalgas demand.

The clean fuels regulations will require additionaldesulfurization capacity at most U.S. refineries. Manydesulfurizers can be revamped to achieve lower sulfurlevels, but new units will be required. About 100 newunits will be required to produce clean gasoline, and 90new units to produce clean diesel.

The net energy requirements for these new desul-furizers are modest, adding about 37 billion cubicfeet (BCF) per year, or 5% of current refining naturalgas demand. However, the hydrogen required for thedesulfurization of gasoline and diesel fuel willrequire a significant net increase in natural gasdemand, even after efforts are made to fully utilizeavailable hydrogen. The NPC analysis assumes that20% of the additional hydrogen required comes fromimproved management of existing hydrogen systems.The additional natural gas demand is estimated to be118 BCF per year by 2010, or about 65% more thancurrent natural gas usage for hydrogen production.

Demand growth, net efficiency gains, and regulatoryimpacts were projected for the NPC study based on theassumptions described above. Assuming natural gascontinues to be the most economic incremental fuelfor refineries, the resulting demand projections showan increase in natural gas demand of about 0.9% peryear for the period from 2003 to 2030. Overall energyuse drops from 536 thousand Btu per barrel to 495thousand Btu per barrel during the period. The high-er level of efficiency assumed for the 2003-2012 periodis more than offset by increased hydrogen productionfor clean fuels. During the 2013-2022 period, efficien-cy gains keep pace with capacity growth and naturalgas demand is relatively flat. From 2023 to 2030, natu-ral gas demand increases as efficiency gains fall short ofcapacity growth.

As natural gas prices exceed those of alternate refin-ery fuels (i.e., propane, butane and gasoline), refinerswill adjust operating conditions to increase fuel gasproduction and reduce natural gas consumption. It isestimated that refiners could reduce natural gasdemand in the short-term by about 45% through oper-ational changes. Sustained higher natural gas priceswould provide an incentive for refiners to invest infuel-switching capabilities on large heaters, boilers, andgas turbines. Ultimately, natural gas for refinery fuelcould be limited to the volume required to balanceswings in the refinery fuel gas system. This minimumvolume is estimated at 20% of the natural gas current-ly used for refinery fuel. However, reducing natural gasdemand to these levels would require sustained capitalinvestments.

Figure 3-34 illustrates the major elements of naturalgas demand in petroleum refining. Higher natural gasprices could affect natural gas feedstock to hydrogenplants. By modifying equipment and/or the catalyst,refiners could switch hydrogen plants to fuel gas,propane, butane, or naphtha. Assuming half of refin-ery hydrogen plants are switched to alternate fuels,overall natural gas demand for hydrogen productionwould be reduced by an equivalent amount.Combined with the reduction in natural gas for refin-ery fuel, overall natural gas projections could bereduced by two-thirds of the levels shown in Figure 3-34 – one-third through short-term operationalchanges, and an additional one-third over time as fuel-switching projects are implemented.

In summary, natural gas demand at U.S. refineriesis expected to increase by about 33% over current lev-els by 2030, assuming that natural gas remains theincremental fuel of choice. Efficiency improvementswill be more than offset by the need for additionalrefinery fuel to meet demand growth and to producehydrogen to make clean fuels. If higher natural gasprices fundamentally alter relative economics versusreadily available alternatives, these demand projec-tions could be reduced by up to one-third in the shortterm and by up to two-thirds in the long term asrefiners optimize the trade-offs between natural gascosts and product value.

Petroleum refineries experienced productiongrowth of 1.2% per year during 1992-1998 and gas useincreased 6.7% per year as consumption grew from 959 BCF in 1992 to 1,416 BCF in 1998. Figure 3-35



Figure 3-34. U.S. Major Elements of Natural Gas Demand in Petroleum Refining

0

200

400

600

800

1,000

1,200

1,400

2000 2005 2010 2015

YEAR

2020 2025

TR

ILLIO

N B

TU

TOTAL NATURAL GAS

TOTAL FUEL GAS

NATURAL GAS – CHEMICAL HYDROGEN

NATURAL GAS – FUEL

Figure 3-35. U.S. Petroleum Refining Gas Demand in Reactive Path Scenario

0

200

1995 2000 2005 2010

YEAR

2020 20252015

400

600

800

1,000

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

illustrates the refinery trends in gas use for boilers,process heat, and other processes in the Reactive Pathscenario. Figure 3-36 provides this information for theBalanced Future scenario. Most of the increased gasuse occurred in process heaters where low-priced nat-ural gas displaced crude oil-derived fuels, especiallyfuel gas. Cogeneration with combustion turbines wasanother growth area because refineries preferred to usenatural gas in the turbines for reliability and warrantyreasons.

In the Reactive Path scenario, refinery production isexpected to grow at 1% per year during the forecastperiod of 2001-2030 and gas use is expected to decreaseby 0.7% per year. Annual gas consumption is expectedto decline from 1,365 BCF in 2001 to 1,109 BCF in2030. Gas use decreases while production increases forseveral reasons. Although substantial gains in energyefficiency have already been accomplished, theseimprovements are expected to continue. Capacityexpansion at refineries is expected to focus on down-stream processes where steam demand is less. Also,higher gas prices are expected to result in a shift backto fuel gas and other oil-derived fuels in some processheaters. Refiners have greater fuel-switching capabili-

ties than other industries and are likely to react quick-ly when the price ratio of natural gas to crude oil priceincreases. Finally, new cogeneration in the forecastperiod is included in the power sector analysis, andsuch growth contributed much to the growth in gas useduring the 1990s. These factors are offset to someextent by increased energy intensity required to pro-duce lower-sulfur transportation fuels. Much of theincreased hydrogen production required for theseclean fuels is assumed to be produced by merchantplants in the chemical industry rather than in the refin-ing industry itself.

The Balanced Future forecast shows slightly highergas use due to higher consumption for process heat,indicating less fuel switching.

Primary Metals

Aluminum. The U.S. aluminum industry (SIC333/5 or NAICS 3313) is the world’s largest, producingabout $39 billion in products and exports in 2000 andaccounting for 17% of the world’s primary aluminumproduction in 1997. Aluminum products are used intransportation, construction, packaging, consumer


Figure 3-36. U.S. Petroleum Refining Gas Demand in Balanced Future Scenario

0

1995 2000 2005 2010

YEAR

2015 2020 2025

200

400

600

800

1,000

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

durables, and electrical industries. As a lightweight,high-strength, recyclable, and structural material, alu-minum will continue to play an increasingly importantrole in the U.S. economy as applications are extendedinto infrastructure, aerospace, and defense industries.According to the Census Bureau, during 2000, the U.S.aluminum industry employed 141,000 people, operat-ing over 300 plants in 35 states and impacting commu-nities throughout the country, either through physicalplants and facilities, recycling, heavy industry, or theconsumption of consumer goods.

Aluminum metal is classified as primary aluminumif it is produced from ore and as secondary aluminumif it is produced predominantly from recycled scrap. In2001, exports of aluminum products accounted for11.3% of total shipments, while imports accounted for38.3% of supply.

Global primary aluminum production has grown at2.2% annually over the last ten years and demand forthe product continues to rise as new applications aredeveloped. The primary production of aluminumrequires the availability of skilled labor, proximity toconsumer markets, a highly developed infrastructureand, especially, low cost and reliable energy. Importedaluminum is the fastest growing source of U.S. supplyand new primary aluminum facilities increasingly arebeing located outside of the United States, near sourcesof low-cost electricity.

Aluminum remains one of the most energy-inten-sive materials to produce. Only paper, gasoline, steel,and ethylene manufacturing consume more total ener-gy in the United States than aluminum. Aluminumproduction is the largest consumer of energy on a per-weight basis and is the largest electric energy consumerof all industries. Electricity accounts for nearly 98% ofthe energy used in primary aluminum production,accounting for one-third of the cost.

Recycled aluminum requires only about 6% of theenergy needed for primary aluminum production. In2000, more than 48% of the aluminum produced byU.S. industry came from recycled material; 40 yearsago, recycled material was used to generate less than18% of U.S.-produced aluminum. Recycling is thelargest contributor to the reduction of the energyintensity of aluminum produced in the United States.

Production variations of aluminum in the UnitedStates are more reflective of the costs to produce alu-

minum than of domestic demand. This factor makesenergy efficiency and energy management primeindustry objectives.

The large electricity demands of the aluminumindustry are relevant when assessing the environmen-tal impact of production and the sensitivity of theindustry to fluctuations in the electricity market. TheU.S. primary aluminum industry has more than half ofits capacity sited in regions where lower cost hydro-electric power is generated.

Although the aluminum industry uses natural gas asenergy input for various steps in the productionprocess, the price fluctuations of natural gas have theirlargest impact in terms of the price of electricity. Whilethe aluminum industry is a large consumer of bothnatural gas and electricity, its annual expenditure forelectricity is over $2 billion. A key determinant of theindustry’s viability in the United States is access to low-cost reliable energy and the development of energy-efficient production processes.

Iron and Steel. The United States is the largest steelproducer in the world, producing 107 million tons ofraw steel in 1998, nearly 13% of total world produc-tion. The iron and steel industry (SIC 33 or NAICS331) provides about 5% of the total U.S. manufactur-ing GDP, employing more than 150,000 productionworkers in jobs paying 50% above the average for allU.S. manufacturing. Steel is used in a diverse range ofapplications ranging from shipbuilding, nationaldefense and construction, to food storage and trans-portation.

A steel import surge that began in 1998 placed sig-nificant financial pressure on the industry. Large levelsof imports brought about by world steel overcapacity(from economic downturns in Asia and theCommonwealth of Independent States) drove pricesdown to unprecedented levels. As a result, 35 steelcompanies, representing 40% of total U.S. steel pro-duction, entered into bankruptcy or liquidation.American steel producers are currently engaged in amajor restructuring and consolidation in response tothis crisis.

Two processes are used for making steel in theUnited States. About 53% is made by integrated steelmakers using the Basic Oxygen Furnace (BOF) process.The BOF process is used to produce steel needed forpackaging, car bodies, appliances, and steel framing; it


uses about 70 to 80% of molten iron and 20 to 30%recycled scrap. The Electric Arc Furnace (EAF) processaccounts for about 47% of raw steel production in theUnited States and is used to produce steel shapes suchas railroad ties and bridge spans. EAFs use electricityas the primary source of energy to melt charged mate-rials, which typically consist of nearly 100% recycledsteel or scrap.

The steel industry is highly energy-intensive.Energy costs account for 12 to 15% of the cost ofmanufacturing steel, on the order of $50 per ton.Steel making requires energy both to supply heat andpower for plant operations and as a raw material forthe production of blast furnace coke. Its aggregatedaverage energy consumption of about 19 MMBtu perton of steel shipped represents approximately 2 to 3%of the energy consumed in the United States and over10% of the energy use in the industrial sector.Natural gas accounts for nearly 20% of the steel’senergy consumption and electricity accounts for mostof the balance.

Over the past 20 years, the iron and steel industryhas invested nearly $7 billion in environmental controlequipment. Through a combination of technologicalinnovation and operating practice changes, the indus-try has reduced its process energy intensity by about45% since 1975. The industry’s overall recycling rate isnearly 71%; over 70 million tons of scrap were recycledin 2002. According to an Environmental ProtectionAgency (EPA) estimate, the energy savings associatedwith the use of recycled iron units, rather than iron ore,is equivalent to the annual electricity needed to power18 million homes.

As part of the R&D effort of the U.S. steel industry,steelmakers are increasingly interested in replacingother energy sources with natural gas. The industry isstimulated by the possibility that concerns with climatechange and greenhouse gas emissions, as well as otherenvironmental considerations, might ultimatelyrequire greater fuel switching to gas.

Gas Demand Projection for Primary Metals. Theprimary metals industries (essentially iron and steel)have seen dramatic changes in recent decades. Largeintegrated steel mills have been replaced by scrap-based steelmaking in minimills and all metal producershave become subject to aggressive global competition.Surprisingly, the primary metals industry grew by3.5% per year from 1992 to 1998. This was a fairly pos-

itive period for the sector, driven largely by a healthydemand from auto manufacturing and other metal-using sectors. Gas consumption grew at a slower rate,1.8% per year during this period and would havegrown at only 0.3% without coincident growth incogeneration. Annual gas consumption grew from 692 BCF in 1992 to 769 BCF in 1998.

Figure 3-37 shows the trends for the Reactive Pathscenario in gas use for boilers, process heat, and otherprocesses in primary metals. New cogeneration proj-ects, classified as other processes, contributed toincreased gas use in the historical period. Gas use forprocess heat grew with production until 1996 whenconsumption began a long-term decline that continuesthroughout the forecast period. Gas use for boilersdeclines throughout the historical and forecast periods.These declines in gas use by the steel industry reflectthe shift from large, integrated mills to minimills,which are less energy-intensive and use electricity asthe major source of energy. Significant improvementsin energy efficiency and process changes continue toreduce the amount of gas used in the metals sectors.Intense global competition also has made the primarymetals industry very aggressive about reducing costssuch as for natural gas in heat treatment furnaces,where oxyfuel burners and electric thermal technolo-gies can reduce or replace gas load. Global competitionhas had a strong negative effect on U.S. metals produc-tion, despite these advances by U.S. manufacturers.

In the forecast period, gas consumption is expectedto decline 2.7% per year while production declines0.2% per year. The forecast assumes trends towardmore efficient production technologies to continuewith most of the reductions in gas consumption com-ing from process heat applications.

Figure 3-38 shows the Balanced Future scenario forprimary metals. The results are very similar to theReactive Path scenario because the changes are duelargely to industry trends rather than gas price issues.

Paper

The U.S. forest products industry (SIC 26 or NAICS322) employs 1.5 million people and ranks among thetop ten manufacturing employers in 42 states, with anestimated payroll of $51 billion. The United States isthe world’s largest producer of forest products, withtotal annual sales that exceed $250 billion.



Figure 3-37. U.S. Primary Metals Gas Demand in Reactive Path Scenario

100

0

200

1995 2000 2005 2010

YEAR

2015 2020 2025

300

400

500

600B

ILLIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

Figure 3-38. U.S. Primary Metals Gas Demand in Balanced Future Scenario

100

1995 2000 2005 2010

YEAR

2015 2020 2025

0

200

300

400

500

600

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

Products from America’s forest and paper industryrepresent more than 8% of U.S. manufacturing output,making it the sixth largest domestic manufacturingsector. The annual increase in per-capita consumptionaveraged 1.8% from 1960 to 1980, 1.6% from 1980 to1993, and has been projected at 0.6% from 1990 to2040. The United States produced 88 million tons ofpaper and paperboard in 1999, over 700 pounds forevery person.

Wood for pulping represents the largest cost amongmaterial inputs to the pulp and paper industry,accounting for an average of 21% of total material andenergy costs. The industry is the third largest industri-al consumer of energy. Pulp, paper, and paperboardmills account for about 12% of total manufacturingenergy use in the United States. Energy intensity in thepaper and allied products industry in 1991 was 20thousand Btu per dollar value of shipments, ranking itas the second most energy-intensive industry group inthe manufacturing sector. The forest products indus-try spent more than $7.6 billion on purchased fuelsand electricity in 1998, or just less than 3% of the valueof its shipments that year.

Energy costs traditionally have been in the top fivecost categories for the industry; additionally energycosts as a proportion of operating costs have increaseddramatically. Many paper mills that have closedrecently cited rising energy costs as a main or con-tributing factor in the shutdown.

Since 1972, the industry has reduced its use of fossilfuels and purchased energy by about 2%, yet increasedits total production by nearly 64%. Over that period,the industry also reduced average total energy use by30% (per ton of product produced). The industry’sgains in energy efficiency are in part due to the successof onsite electricity generation at mills throughout thecountry. The forest products industry currently meetsnearly 63% of its energy needs through self-generatedelectricity. The industry leads all other manufacturingsectors in on-site electricity generation, producingnearly 43% of U.S. self-generated electricity. This ener-gy is harnessed primarily through efficient cogenera-tion with use of woody waste products and otherrenewable sources for biomass fuel (bark, wood, pulp-ing liquor).

Cogeneration processes turn waste materials into arenewable energy source that diverts waste from land-fills, reduces reliance on fossil fuels and offsets green-

house gas emissions by substituting carbon-neutralbiomass for fossil fuels. The industry’s use of renew-able fuels represents the equivalent of about 20 millionbarrels of oil per year and offsets the CO2 emissions ofabout 16 million automobiles annually.

Although the industry is more than 60% energy self-sufficient, it relies on natural gas, coal, fuel oil, and pur-chased electricity to meet the balance of its energyneeds. Increases in oil and gas prices threaten the com-petitiveness of many mills. High energy prices haveforced some companies to cease operations and sellelectric capacity instead of making a product, makingthem vulnerable to foreign producers.

Workshops with industry representatives indicatedthat efforts to hold down energy costs are often coun-tered by regulations that discourage fuel flexibility andthe development of energy efficient operating proc-esses. Implementation of New Source Review require-ments was cited as a mechanism to effectively forcecompanies to continue using fuels associated withinstalled equipment while discouraging new invest-ment in energy-efficient technologies and processes.These workshops also found that in a volatile naturalgas price market, forest products manufacturers wouldbenefit from the flexibility to substitute lower-costalternative fuels – coal, biomass, and shredded tires – tofuel boilers.

Gas Demand Projection. Gas consumption for theproduction of paper and allied products grew by 3.5%per year during 1992-1998 while production grew byonly 0.4% per year. Annual gas use by the industrygrew from 505 BCF to 622 BCF over the period. Figure3-39 shows the gas use for boilers, process heat, andother processes in the Reactive Path scenario, and it isapparent that gas use for boilers is the major factorcontributing to this increase. The major driver for theincrease was strong growth in the energy-intensivecomponents of the industry (from paper, pulp, andpaperboard mills).

In both the Reactive Path and the Balanced Futureoutlooks, the recent rapid growth in gas consumptionwould not continue. Between 2001 and 2025, no netgrowth in production of paper products is projectedand gas consumption is expected to decline by 0.7%per year. Figure 3-40 shows the Balanced Future pro-jection for the paper industry. The scenario showslower gas consumption in the boiler sector due toincreased flexibility to switch fuels.



Figure 3-39. U.S. Paper Industry Gas Demand in Reactive Path Scenario

1995 2000 2005 2010

YEAR

2015 2020 2025

0

100

200

300

400

500B

ILLIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

Figure 3-40. U.S. Paper Industry Gas Demand in Balanced Future Scenario

0

100

1995 2000 2005 2010YEAR

2015 2020 2025

200

300

400

500

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

Stone, Clay, and Glass

The stone, clay, and glass industries (SIC 32 orNAICS 327) produce cut stone products and clay prod-ucts, including bricks, glass, concrete, gypsum, andlime. The industry employed 508,000 people and sold$93 billion in products in 1998. The industry producesthe cement, bricks, and glass products used in homesand infrastructure as well as containers for many foodand consumer products. Unlike some sectors, thisindustry does not compete to a great extent withimports or export. Further, production is often locat-ed within the same region as the target market since itsproducts are heavy, with a low price-to-weight ratio,thereby making transportation costs prohibitive forlong-distance shipping. The highest concentration ofenergy use is in the East North Central and Mid-Atlantic regions where 36% of this industry’s energy isconsumed. After mined raw materials, energy is one ofthe most important inputs in production.

The major drivers for energy consumption in thestone, clay, and glass industry are demand in the con-struction industry and new combustion technologies.Global competition is less of a factor and fuel pur-chases represent less than 5% of the value of most fin-ished products. The proliferation of oxy-fuel burnerson many natural gas-fired kilns and furnaces improvesenergy efficiency, raises throughput capacities, andlowers emissions.

Industry natural gas use increased from 329 BCF to388 BCF between 1992 and 1998. The annual growthin gas use was 2.8%, which is slightly less than the pro-duction growth rate of 3.8% during this period.Unlike chemicals, refining, or primary metals, boilersplay a small part in this sector and cogeneration con-tributed little to increased gas use during the period.More than 80% of the natural gas is used for processheat, especially for the production of clay products(bricks), glass, and gypsum. The industry is usingmore-efficient equipment, including electric boostersand oxy-fuel furnaces, to reduce energy costs andincrease production.

Gas consumption within the industry (SIC 32) arefor production of glass (155 BCF), brick (47 BCF),gypsum (42 BCF), and cement (26 BCF). Natural gasprovides most of the thermal energy used in the glassindustry. Gypsum-producing kilns and brick-firingfurnaces generally use natural gas, while cement andlime are most often produced in coal-fired kilns.

Recent increases in gas price present a challenge formanufacturers with limited options to fuel switchbecause the obvious alternatives (propane and distil-late oil) may have little cost advantage in the longerterm. Energy conservation has taken the form ofincremental improvements (heat recovery and betterinstrumentation in kilns) and shifting from wetprocess kilns to dry and vertical process kilns.

Demand Outlook for Stone, Clay, and Glass.Production is expected to grow 2.8% per year duringthe forecast period, which is the highest rate of the sixmajor gas-intensive industries. Gas use would grow ata modest 0.8% per year in the Reactive Path scenariowith annual consumption rising from 361 BCF in 2001to 458 BCF in 2030. The slower growth of gas usereflects a continuation of the trends seen in the 1990s,slower growth in flat glass production, and improvedprocess efficiencies. In addition, the NPC model showsthat higher natural gas prices will lead to fuel switchingto oil in the glass and brick industries and to coal andwaste fuels in the cement industry.

Production in the stone, clay, and glass industry isprojected to grow about 1.1% per year in the longterm, although incremental energy efficiency improve-ments will hold the growth in energy consumption toabout 0.6% per year. Natural gas consumption willgrow at a slower rate, about 0.5% per year, althoughthis industry derived estimate was based on lower gasprice projections than now expected.

Figure 3-41 shows the projection of major naturalgas uses in the Reactive Path scenario. Figure 3-42shows the projection in the Balanced Future scenario.In this scenario, lower gas prices increase gas use forprocess heat relative to the Reactive Path scenario.

Food and Beverage

The food and beverage industry (SIC 20 or NAICS311 and 3121) is one of the most diverse and disaggre-gated of the six energy-intensive industries, producingnumerous products from sausage to milk to frozendinners to beer. The food industry has seen rapid con-solidation in recent years but its operations remainquite varied. Food processors continually develop newand more products to meet the market demand. Theindustry employed 1,640,000 people and sold $491 bil-lion of products in 1998.

The food and beverage industry spent $6.1 billion in1998 to purchase 1,150 trillion Btu of energy. Natural



Figure 3-41. U.S. Stone, Clay, and Glass Gas Demand in Reactive Path Scenario

0

100

1995 2000 2005 2010

YEAR

2015 2020 2025

200

300

400B

ILLI

ON

CU

BIC

FE

ET

BOILERS

PROCESS HEAT

OTHER PROCESSES

Figure 3-42. U.S. Stone, Clay, and Glass Gas Demand in Balanced Future Scenario

0

100

1995 2000 2005 2010

YEAR

2015 2020 2025

200

300

400

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

gas is the largest single source of purchased energy forthe industry, which consumed about 655 BCF in 1998.Electricity (68 million megawatt hours) and coal (7 million tons) were the other major energy sources.The industry produces high value-added finishedproducts; overall energy costs are just 1.4% of the valueof total shipments.

Natural gas use in the food industry increased from524 BCF to 655 BCF between 1992 and 1998, respec-tively. Gas use grew by 3.8% per year during this peri-od, which is one of the highest rates in themanufacturing sector. Increase in gas use outpacedproduction during this period from rapid growth ofthe energy-intensive sub-industries (the wet corn mill,beet sugar, soybean oil mill, and malt industries), rela-tive to the less energy-intensive sub-industries.

The food industry will likely continue improve heat-ing and cooling technologies that achieve better prod-uct quality. In outreach sessions between the NPC andindustry representatives, energy was not highlighted asa primary reason for new technology development.Energy-focused projects, however, have been successfulin highly integrated operations such as wet cornmilling and malt beverage production.

An important innovation in grain processing is drycorn milling. Wet corn milling is one of the most energy-intensive processes in the food industry. Drycorn milling focuses on ethanol production and sim-plifies the process by eliminating the initial soakingand oil separation steps. Dry milling facilities, oftenowned by farming cooperatives, are smaller and lessexpensive than wet milling plants. Generally, drymilling plants use natural gas in boilers instead of coal.Because of capital constraints, dry corn milling opera-tions may not be optimized and may be good candi-dates for cogeneration projects.

The food and beverage industry relies on natural gasfor a variety of process and boiler fuel uses. Boilers forsteam production consume more than half of the ener-gy used in the food and beverage industry, includingabout 398 BCF of natural gas. Companies prefer nat-ural gas because gas boilers are cleaner and more flexi-ble than the main alternative, coal boilers. Cleanlinessis of primary importance at a food manufacturingfacility and natural gas avoids the problems in storage,dust and air pollution control associated with coal.Also, many food plants are too small and/or do not

operate on a sufficiently continuous schedule to makecoal boilers practical.

Steam is used for food drying, cooking, and con-centration processes. Direct process heating process-es such as food drying, baking, and frying accountedfor 188 BCF. Other processes, specifically space heat-ing, used 69 BCF. Cogeneration steam boilers, tur-bines, and engines used 122 BCF according to theNPC estimates. The major gas consuming parts ofthe food industry are animal slaughtering and pro-cessing (115 BCF), wet corn milling (84 BCF), fruitand vegetable preserving (64 BCF), and dairy pro-cessing (49 BCF).

Expansion in the food companies is driven largely bypopulation growth and is shaped by changing con-sumer preferences. For several decades, Americanshave continued to spend about 15% of their income onfood. As household wealth increases, people buy moreprepared foods and dine in restaurants more frequent-ly. The food industry has responded with productsthat expand and complicate the manufacturingprocess. For instance, supermarkets now stock hun-dreds of frozen food products. Product innovation,product quality, and labor costs are of greater impor-tance to food producers than energy.

The manufacture of some product lines is concen-trated in certain regions; for example, milk in the EastNorth Central or orange juice in the South Atlantic.Other products, e.g., baked goods, generally are pro-duced locally. Overall, about 43% of the food and bev-erage industry’s energy consumption was in the EastNorth Central and West North Central regions.Imports and exports are growing but remain less of afactor than in other industries.

Demand Outlook for the Food Industry. Thefood industry is projected grow at a moderate paceand natural gas sales will grow at a slightly slowerrate, about 1% per year. Coal consumption, which isconcentrated in the wet corn milling and malt bever-age industries, will remain steady or slightly decline asdry corn milling gains market share and sales of maltbeverages flatten.

Figure 3-43 shows the gas demand for boilers,process heat, and other processes for the Reactive Pathscenario. Process heat and other processes accountedfor most of the increase in gas consumption during thehistorical period when gas demand grew by 3.8% per



Figure 3-43. U.S. Food Industry Gas Demand in Reactive Path Scenario

50

1995 2000 2005 2010

YEAR

2015 2020 2025

0

100

150

200

250

300B

ILLIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

Figure 3-44. U.S. Food Industry Gas Demand in Balanced Future Scenario

50

0

100

1995 2000 2005 2010

YEAR

2015 2020 2025

150

200

250

300

BIL

LIO

N C

UB

IC F

EE

T

BOILERS

PROCESS HEAT

OTHER PROCESSES

year, which is greater than the growth in production(2.4%).

In the Reactive Path scenario, production increasesby 1.1% annually after 2001 while gas use decreases by0.1% per year. Gas load in 2030 is projected to be 593 BCF. Figure 3-43 reveals that gas use for boilerscontinues to grow but gas use for process heat andother applications is flat or declining. Two reasons forthe lack of growth for these end uses are improved effi-ciency and slower growth in the energy-intensive sub-industries. New cogeneration is not included in theforecast. Figure 3-44 shows gas demand in theBalanced Future scenario. In this scenario, gas use forboilers is relatively flat, indicating greater potential forfuel switching in boilers.

Other Primary Metals

The major gas-consuming components of the pri-mary metals industry are in iron and steel, and alu-minum. “Other primary metals” comprises theremainder of the sector and consumes about 196 BCFper year or one-third of the gas consumed in the pri-mary metals industries. This industry group includesmetal foundries, primary and secondary smelters ofnon-ferrous metals (except aluminum), and establish-ments engaged in rolling, drawing, and extruding ofmetal products. In 2001, this segment accounted foralmost half of the value in shipments from the primarymetals industry. During the same year, the industryemployed almost 330,000 workers, or over 60% of totalemployment in the primary metals industry.Approximately one-third of the industry’s outputcomes from the East North Central region and anoth-er 15% comes from the Mid-Atlantic region. Theseregions have a heavy presence of metal fabricators (e.g.,automobile manufacturers), which is the major marketfor the industry’s products.

The major uses of natural gas in this industry aremetal melting, heat-treating, and space heating.Process heat accounts for 60% of natural gas consumedin the industry, and the rest is for space heating(through direct heat or steam space heaters). The rel-atively high employment level and location of theplants in the Midwest and Mid-Atlantic states drive thedemand for space heating.

Industrial production in this segment has grown atabout 0.5% per year in recent years. This relativelyhigh growth rate reflects a healthy period in the auto-

mobile industry and other markets for products. TheNPC forecast assumes continued growth at this rate.

Other Industries

The primary focus for the industrial sector analysisis the six key industries (chemicals; petroleum refining;primary metals; paper; stone, clay, and glass; and foodand beverage), which accounted for 80% of the indus-trial natural gas consumption. However, 19 otherindustries comprise the industrial sector (SIC cate-gories 1-39), including (by SIC):

� 1 Crops

� 2 Livestock

� 10,14 Non-Energy Mining

� 11,12,13 Energy Mining

� 15 Construction

� 21 Tobacco Products

� 22 Textile Mill Products

� 23 Apparel & Textiles

� 24 Lumber & Wood

� 25 Furniture & Fixtures

� 27 Printing & Publishing

� 30 Rubber & Misc. Plastics

� 31 Leather & Products

� 34 Fabricated Metals

� 35 Non-Electric Machinery

� 36 Electric Equipment

� 37 Transportation Equipment

� 38 Instruments

� 39 Miscellaneous

Gas consumption for these 19 industries in 2001 was alittle less than 1,500 BCF, about 20% of total industrialgas consumption and 7% of total U.S. consumption.The gas consumption in many of the individual indus-tries was quite small. Energy consumption in generaland gas consumption specifically is a much smallerpercentage of the value added for these industries thanfor the energy-intensive industries. Although all costsare important to the profitability of any enterprise, thegas component of cost for these industries is typicallyless than 1%. For this reason, less effort was spent on


detailed modeling of the individual industries and theywere treated as a group.

Some distinctions among these industries are war-ranted. Tobacco and leather products are declining inthe United States and are not large energy or gas con-sumers. The electronics industry is growing quicklybut is not gas intensive. A few of these industries dohave some significant gas consumption.

The SIC industries 1 through 15 are non-manufac-turing and include agricultural, construction, andmining. The total gas consumption of these sectors in2001 was about 600 BCF. Of this, about 480 BCF wasin the mining industry and much of this was for gas-fired cogeneration related to enhanced oil recovery incentral California. The remainder of the gas was usedfor other on-site generation, space heating and processheating.

SIC industries 20 through 39 are manufacturing. Themanufacturing SICs other than the six major gas-con-suming sectors consumed about 900 BCF. This con-sumption can be divided into three primary groupings:

� Rubber and Miscellaneous Plastics (SIC 30) – 79 BCF.

� Metal Durables (SICs 34-38) – 492 BCF.

� All Remaining SICs – 329 BCF.

The rubber and miscellaneous plastics category hasbeen a fast-growing sector, pushed by increaseddemand for plastic in consumer goods and electronicitems. Gas is used for steam generation and processheat. However, it is not a gas-intensive sector and gasuse has lagged the production growth.

The metal durables category includes appliances,automobiles, and electronic equipment. As such, it wasby far the fastest growing sector of the economy duringthe 1990s, growing by 15% per year. This sector is nota gas-intensive sector and it saw major decreases inenergy intensity during the 1990s. The metal durablessector consumed only 1.8 thousand Btu of energy perdollar of output in 2000 compared to 40 thousand toover 100 thousand Btu per dollar of output in the ener-gy intensive industries. In recent years, the growth inthis sector has dropped substantially due to the tech-nology downturn. The uses of gas in this segmentinclude space heating, process heating, and somecogeneration.

Industrial production in the Other Industries grewby 5.2% per year during the 1990s. The NPC projec-tion assumes a lower rate of 2.6% per year based onmore recent industry performance. Natural gas con-sumption grows by only 0.1% per year in this sectordue to the low and decreasing gas intensity.

Additional Policy Issues ofIndustrial Consumers

The NPC found that industrial consumers broadlysupport energy and environmental policies that lead tolower prices and improve the competitiveness ofenergy-intensive industrial operations. In particular,industrial consumers support policies designed to fos-ter the development of adequate and reliable suppliesof natural gas and other energy sources. Furthermore,industrial consumers also support policies to removecurrent impediments to fuel switching and to encour-age increased fuel flexibility by allowing industrial con-sumers to make economically rational fuel choices.

The NPC found that industrial consumers broadlysupport policy initiatives that allow robust competi-tion among energy alternatives and the lowest cost forconsumers to be achieved. The industrial consumersindicated broad support for national energy policiesthat simultaneously pursue multiple policies related tosupply, infrastructure, and demand, including:

� Continued efficiency improvements and increasedconservation measures

� Rational environmental policies

� Increased access to public lands for exploration andproduction

� Streamlined permitting for LNG facilities

� Streamlined permitting for natural gas pipelineprojects

� Enabling legislation for the Alaska pipeline project.

Some gas-intensive industrial consumers have beenadversely impacted by relatively higher natural gasprices as well as price volatility. For some industrialconsumers, higher energy costs have led to plant shut-downs and decisions by some to move operations over-seas. This obviously impacts employment and thecommunities in which these industries are located.Despite industrial consumers’ strong desires for lowernatural gas prices, the NPC found that industrial con-sumers consistently reject wellhead price controls as a


policy option to address higher natural gas prices asprice controls would only aggravate the current tightsupply/demand balance.

In addition to supporting the broad energy policiesdiscussed above, the following issues (not ranked inorder of importance) are of particular importance toindustrial consumers:

� Fuel switching

� Technological innovations to increase conservationand efficiency

� Combined heat and power

� Biomass and renewables

� FERC policies governing pipeline tariffs and serviceofferings.

Fuel Switching. Industrial end-users of natural gascannot respond to changes in natural gas prices asreadily as they were able to in the past by switching toalternative fuels. Increasingly stringent environmentalstandards and compliance actions have created theincentive for many industrial consumers to rely moreheavily on natural gas as their sole energy source. Dueto environmental restrictions on burning other fuelssuch as coal and fuel oil, some industrial consumershave eliminated the capability to switch fuels in orderto receive necessary approvals for plant expansions.Industrial consumers believe they should better com-municate with local, state/provincial, and federal offi-cials regarding the effects of mandating single-fuelcapability in power generation and industrial facilities,and work to allow dual-fuel switching criteria in rele-vant codes and standards.

Energy Efficiency and Conservation. The NPCfound that industrial consumers have made significantstrides in recent years to control energy costs, boththrough the use of more efficient technology, as well asthrough conservation measures. However, manyindustrial consumers remain heavily dependent onnatural gas both as a fuel and as a raw material. Givenrecent experience with higher energy costs, industrialconsumers reported that they are motivated to contin-ue to pursue research and development efforts to fur-ther improve efficiency and increase conservation,including collaborative research with the Departmentof Energy (DOE) where such programs exist.

Combined Heat and Power. Combined Heat andPower (CHP, i.e., cogeneration) plants are as much as

twice as efficient as traditional utility power plants andare generally located at or near the demand site, whichfurther improves efficiency by reducing energy lostthrough transmission line losses. Many industrial con-sumers believe that federal statutes, such as the PublicUtility Regulatory Policies Act (PURPA) that enablesthe use of CHP systems in monopoly utility regions,should not be repealed or should not be repealed untilcompetitive electricity markets are available to theCHP facility or until other mechanisms are in place toprotect investments in CHP systems as the electricitymarkets evolve. The NPC takes no position on main-taining, eliminating, or modifying PURPA.

Biomass and Renewables. Renewable “biomass”fuels such as bark and pulping byproducts are usedin the forest products industry as an integral part ofmany operations. In some operations, these fuelshave a significant impact on overall fuel use, particu-larly that of natural gas. The forests products indus-try, in conjunction with DOE, has been developinggasification technologies to turn black liquor (abyproduct of a chemical process that turns wood intopaper) and other biomass into fuels. If fully devel-oped, the U.S. forest products industry believes thesetechnologies could allow the industry to be energyself-sufficient, as well as provide surplus power to thegrid. In addition, the paper industry indicates thatcarbon reductions from black liquor gasificationcould transform the industry facilities from an emit-ter of 24 million tons of carbon each year to a carbonsink capable of absorbing 18 million tons or more ofgreenhouse gases. Given the potential to reduce nat-ural gas demand through increased reliance on blackliquor gasification, the pulp and paper industry sup-ports further collaborative research efforts to developthese technologies.

FERC Policies. As end-users of natural gas, indus-trial consumers depend on pipeline and distributionsystems to deliver natural gas supplies to their plantsand facilities. Thus, industrial consumers are affectedby the rates charged and the regulatory policies gov-erning the transportation and delivery of natural gas.In particular, industrial consumers participate in avariety of FERC proceedings and policy matters. Thesegenerally involve FERC’s efforts to open up the inter-state transportation grid to provide industrials andother shippers increased service options, pipeline tar-iffed services and associated rates, and pipeline credit-worthiness standards.



Residential and Commercial Demand

Natural gas is used by over 60 million U.S. house-holds and supplies over 40% of commercial energyrequirements. The residential and commercial sectorsaccounted for over one-third of U.S. natural gas con-sumption in 2002. Since 1997, residential and com-mercial natural gas use has remained relativelyconstant. Figures 3-45 and 3-46 illustrate the growthin number of residential and commercial customers,respectively; Figures 3-45 and 3-46 also show naturalgas demand in these sectors since 1990.

Natural gas demand growth in the residential andcommercial sectors is related primarily to populationgrowth, economic growth, and the costs of using gasversus other fuels for space heating and similar appli-cations. Residential and commercial demand alsoreflects demographic shifts, penetration of gas-basedtechnologies, growth in floor space, and levels of effi-ciency of gas burning appliances. Weather, measuredin terms of heating degree-days, has an importantshort-term impact on both residential and commercialgas consumption.

To analyze future trends for residential and com-mercial gas consumption, the NPC used econometricmodels and capital stock models. These models incor-porated weather, demographic trends, populationgrowth, residential housing stock, capital stock effi-ciency, commercial floor space, penetration of gas-based technology, and gas prices as determinants of gasconsumption.

The primary residential sector uses of natural gas arespace heating, water heating, cooking, and clothes dry-ing (see Table 3-7). Other uses include natural gas fire-places, barbeques, swimming pool heaters, andoutdoor lighting. The primary commercial sector nat-ural gas uses are space heating, space cooling, andwater heating.

Residential Consumers

In 1997, the average household consumed 91.2MMBtu and spent $603 on natural gas. In 2001, aver-age household consumption fell to 79.3 MMBtu butannual expenditures for natural gas rose to $750. Thisreflects a 13% decrease in consumption and a 24%increase in expenditures. During this period the aver-age price increased 43% from $6.62 in 1997 to $9.45per MMBtu in 2001.1

In 2002 there were approximately 119 million hous-ing units in the United States. Total natural gas con-sumption by households equaled almost 5 TCF or 22%of total U.S. gas consumption. Slightly more than 62%of U.S. housing units used natural gas in 2002 (seeFigure 3-47). Although newer houses are larger, aver-age gas use per household is declining because of bet-ter insulation and more energy efficient equipment.

As a space heating fuel, natural gas competes withelectricity, fuel oil, and propane. Over the decades,natural gas has become the dominant space heatingfuel in the United States (see Figure 3-48).

The use of fuels for space heating differs regionally,as shown in Table 3-8. Electricity is the major compet-ing heating source in all regions except the Mid-Atlantic and New England. Electricity is a strongcompetitor in the South Atlantic and East SouthCentral regions where electric heat pumps providespace heating for 28% and 22% of households, respec-tively. The major alternative to natural gas space heat-ing in the Mid-Atlantic and New England is fuel oil. In2001, fuel oil was the main heating fuel in 25% ofhouseholds in the Mid-Atlantic and 50% in NewEngland.

New residential construction is heavily weightedtoward natural gas heating. In recent years, approxi-mately 70% of newly completed single-family homesinstalled gas heat.2 In addition, the percentage of nat-ural gas heating in new multi-family constructionincreased slightly. By comparison, 27% of new hous-ing units installed electric heat, predominantly in theSouthern states. Fuel oil is losing its market share as aheating fuel nationwide. Table 3-9 provides summaryinformation on the application of natural gas andcompeting fuels in new housing in 2001.

Water heating is the second most important residen-tial use of natural gas. Unlike space heating, waterheating is not weather sensitive. The market penetra-tion of natural gas water heating is similar to that ofspace heating. Regionally, the greatest penetration is inthe Midwest followed closely by the West. Penetrationwas lowest in the South, where electric heating is thegreatest. The Northeast region experienced significant

1 American Gas Association, 2002 Gas Facts: A StatisticalRecord of the Gas Industry, 2001 Data, 2003, pg. 59.

2 Ibid, pg. 72.


1990 1995

YEAR

2000

CU

ST

OM

ER

S (

MIL

LIO

NS

)

0

2

4

6

DE

MA

ND

(T

RIL

LIO

N C

UB

IC F

EE

T)

CUSTOMERS DEMAND

0

20

40

60

80 8

Figure 3-45. U.S. Residential Customers and U.S. Residential Demand

0

2

4

6

1990 1995

YEAR

2000

CU

ST

OM

ER

S (

MIL

LIO

NS

)

0

2

4

6

DE

MA

ND

(T

RIL

LIO

N C

UB

IC F

EE

T)

CUSTOMERS DEMAND

Figure 3-46. U.S. Commercial Customers and U.S. Commercial Demand




growth in gas water heating while growth in the otherregions was marginal.

Gas cooking is the third most important residentialuse of natural gas. Nevertheless, both oven and rangesectors are dominated by electric appliances. All

regions showed a slight decline in natural gas appliancepenetration with the exception of the South, whichexhibited a modest increase in natural gas ranges.Most of the decline in the percentage of natural gasappliances is attributed to growing penetration of elec-tric appliances in new houses. In addition, the wide-spread use of microwave ovens appears to havedecreased gas use in cooking.

Natural gas consumption and expenditures are pos-itively correlated with household income: the higherthe household income, the more a household con-sumes and spends on energy. This higher use andrelated expenditures reflects in the typically largerhomes owned by higher-income families, requiringmore heating. However, the cost of fuel is, on average,a higher proportion of household income for low-income families. The average residential energy costsin 2001 (including heating, cooling and all other ener-gy uses in the home) for U.S. households in 2001 was$1,537 per household, or 7.0% of income. Low-income households spent an average of $1,311 onenergy, representing 14.0% of household income; forhouseholds qualifying for Low Income Home EnergyAssistance Program (LIHEAP) funding – two-thirds ofwhich have incomes less than $8,000 per year – the

Consump-tion

(MCF)

ApplianceMarketShare

(Percent)

Space heating 69.7 52

Water heating 34.1 51

Cooking 11.7 35

Clothes drying 3.7 22

Gas Fireplaces 9.7 NA

Source: American Gas Association, 2002 Gas Facts:A Statistical Record of the Gas Industry, 2001 Data.

Table 3-7. U.S. Residential Market 2001 Annual Natural Gas Consumption per Appliance

20011997

0

20

40

60

80

100

NATURAL

GAS

FUEL OIL LPG SOLARELECTRIC WOOD KEROSENE

PE

RC

EN

T

Source: Energy Information Administration, Residential Energy Consumption Surveys, 1997 and 2001.

Note: Households may use more than one fuel.

Figure 3-47. Fuel Used in U.S. Housing Units


percentage of income represented by energy expendi-tures was 17.2%.3

The LIHEAP program commenced in 1982 with theobjective of assisting low- and fixed-income house-

holds in paying their fuel and utility bills, includingwinter heating bills and summer cooling bills. Theprogram was designed to be a targeted assistance pro-gram with government funding, rather than a utilityprogram where low-income assistance was built intorates and spread among a larger number of consumers.Between 1981 and 2000, LIHEAP funding increased22%. The funding stands at $1.7 billion for FY 2003.However, for the same period the number of federallyeligible households rose over 49%.

Commercial Consumers

The commercial sector accounts for about 14% oftotal U.S. gas consumption. This sector is more diversethan the residential market, consisting of businessestablishments and service organizations such as retailand wholesale facilities, hotels and motels, restaurants,and hospitals. The commercial sector also includespublic and private schools, correctional institutions,and religious and fraternal organizations. The end-usemarkets in the commercial sector are less seasonal thanresidential customers. Commercial customers con-sume about 7.5 times more gas, on a per customerbasis, than customers in the residential sector.3 The LIHEAP Home Energy Notebook for FY 2001.

0

10

20

30

40

50

60

NATURAL

GAS

FUEL OIL LPG OTHERELECTRIC WOOD KEROSENE

20011997P

ER

CE

NT

Figure 3-48. U.S. Primary Household Heating Fuel

Region 2001 1997

Total U.S. 54.8% 52.7%

Northeast 52.4% 46.0%

Midwest 76.8% 75.0%

South 39.5% 38.0%

West 59.5% 58.0%

Source: American Gas Association, 2002 Gas Facts:A Statistical Record of the Gas Industry, 2001 Data.

Table 3-8. Percentage of U.S. Households with Natural Gas as the Main Heating Fuel

Source: Energy Information Administration, Residential Energy Consumption Surveys, 1997 and 2001.

Commercial natural gas consumption grew at anaverage annual rate of approximately 2.6% between1990 and 1997, compared to 1.6% for residential con-sumption. Although total consumption was rising, useper customer was reduced. Between 1990 and 1997,the average annual consumption per commercial cus-tomer declined by 0.7%.

The average growth rate for commercial gas con-sumption was -0.5% between 1997 and 2002. Asshown in Figure 3-49, some of this variation was due to

weather – a cold year (1997) followed by a warm year(1998). The number of commercial natural gas cus-tomers increased approximately 6% between 1997 and2001, from 4.6 million to 4.9 million.

Figure 3-50 illustrates the growth in the number ofcommercial natural gas customers. The commercialmarket fluctuates, and the upward trend in the numberof customers does not necessarily reflect the amount offloor space served by natural gas. New commercialbuildings are constructed and older buildings are


Single-Family Multi-Family Combined

Natural Gas 70% 47% 65%

Electric 27% 53% 32%

Fuel Oil 3%Less than500 units

2%

Other 1% – 1%

Total 100% 100% 100%

Source: American Gas Association, Residential Natural Gas Market Survey, 2001 Data.

Table 3-9. U.S. Private Housing Completions in 2001 by Heating Fuel

2.5

3.0

3.5

TR

ILLIO

N C

UB

IC F

EE

T

2.0

0

1990 1995

YEAR

2000


Figure 3-49. Natural Gas Delivered to U.S. Commercial Consumers

0.0

3.5

1990 1995

YEAR

2000

4.0

4.5

5.0

5.5

CU

ST

OM

ER

S (

MIL

LIO

NS

)

Figure 3-50. Number of U.S. CommercialNatural Gas Customers


converted to commercial uses. At the same time, oldercommercial buildings may be razed or converted tononcommercial uses. Commercial floor space mayalso fluctuate concurrently. Commercial floor spacemay be converted to non-commercial uses, which willimpact commercial demand. Minimal growth in com-mercial demand is due in part to efficiency in buildingdesign and natural gas appliances and equipment.

Most of the natural gas consumed by the commer-cial sector is used for space heating and water heating,and there has been a strong trend for customers tochoose gas for these applications where gas is available.Other uses such as cooling, cooking, drying, desiccantdehumidification, and cogeneration applications com-prise a smaller share of natural gas applications. Whilespace and water heating usage have become increasing-ly efficient, alternate uses of natural gas have continuedto make up a larger share of total commercial gas use.

Natural gas has been losing market share amongcommercial customers to electricity in most end-usesexcept cooking. The loss has been the greatest in cool-ing and space heating.

Commercial customers normally operate under afirm utility rate, paying a premium compared to aninterruptible rate. Many commercial consumers arecapable to installing dual-fuel applications. Theseapplications are designed to use either natural gas oroil as a fuel. To take advantage of dual-fuel capabilities,some commercial consumers typically elect interrupt-ible utility service at a lower rate. Consequently, unlikethe residential sector, energy prices and weather mayencourage fuel switching for some end-uses in com-mercial markets.

One of the newest and most promising growth driv-er in commercial gas use is on-site power generation.In order to provide backup capability and to limitpower use during peak periods, many commercial cus-tomers have installed on-site generators to supporttheir buildings’ electrical needs. In certain capacity-constrained regions customers with on-site generationreceive capacity payments where this product is part ofthe region’s electricity market design. Natural gaspowered reciprocating engines, turbines, and fuel cellsare used in many commercial settings to generate elec-tricity. This type of on-site generation is also referredto as distributed generation and allows commercialbuildings to be more independent from the utility grid

and the possibility of power disruption and inconsis-tent high-quality electricity. It also provides commer-cial building managers with more control over theirpower supply.

Cogeneration has slowly penetrated certain commer-cial markets in recent years. Like distributed genera-tion, cogeneration can be an alternative source ofelectric power during peak periods for power demand.Electricity is generated with a natural gas generator andis the co-production of electrical and thermal energy,also called combined heat and power (CHP). Becausethe thermal energy that is produced during electric pro-duction is used to provide heating, the energy conver-sion efficiency of cogeneration facilities can be as highas 70%, allowing for substantial savings in fuel com-modity costs for the building owner. Hospitals, air-ports, and other establishments that cannot afford to besubject to brownouts or blackouts use cogeneration.

Natural Gas Vehicles

For purposes of this study, natural gas vehicle(NGV) usage was assessed within the commercial sec-tor. In recent years NGVs have penetrated fleet vehicleand urban transit bus markets. There are almost60,000 natural gas vehicles in the U.S., according to theNatural Gas Vehicle Coalition. The U.S. Postal Servicecurrently operates the nation’s largest fleet of naturalgas vehicles and United Parcel Service (UPS) operatesthe largest private fleet. Furthermore, utilities, airportshuttle services, taxi companies, police departments,school districts, police departments, and ice rinks(Zambonis) also operate large fleets of natural gasvehicles. A prominent off-road application of NGVs isforklifts in warehouse operations.

There were approximately 6,200 natural gas transitbuses operating in the United States at the end of 2001.Natural gas buses represented approximately 11% ofall transit buses and 97% of all alternatively fueledtransit buses. At the beginning of 2002, an additional1,313 natural gas transit buses were on order. Almost21% of all transit buses on order are natural gas pow-ered. Nearly 28% of 2002-2005 “potential bus orders”of 11,195 are powered all or in part by natural gas.

The main attraction for most NGV purchasers is thefavorable environmental characteristics. Many of thecompanies and governmental agencies that are con-verting their fleets to natural gas are doing so to com-ply with air quality regulations.


Regional Considerations in Residential andCommercial Demand

Current natural gas consumption is affected by his-torical accessibility to natural gas. The Northeast,which includes New England and the Mid-Atlanticstates, has been more gas-limited than other areas ofthe country. New England, in particular, has the low-est rates of natural gas penetration due to limitedaccess to natural gas for most of the 20th Century.Consequently, households in the Northeast have his-torically tended to use oil for heating because of itswider availability.

The Northeast markets are relatively distant fromtraditional major natural gas supply areas in theSouthwest and in western Canada, and the regionreceives the vast majority of its natural gas suppliesthrough pipelines from these regions. A recently com-pleted pipeline from Canada’s Sable Island gas fields toNew England and expansions and/or other projects areexpected to help meet the growing demand for naturalgas in the Mid-Atlantic and New England regions. TheMaritimes and Northeast Pipeline and Portland GasTransmission System projects, which will transportCanadian gas to the New England area, provided morethan half of new capacity in 1999. Those two projectsincreased overall pipeline capacity into the Northeastregion by 5%.

Over the past 20 years, residential natural gas use hasincreased in the Northeast as new natural gas pipelineshave been built. Newly constructed and existinghomes were able to choose natural gas instead of heat-ing oil. As new infrastructure is integrated into thecurrent system allowing new supplies to reach the NewEngland and Mid-Atlantic areas, and regional utilitiesto expand their distribution system accordingly, totaldemand for this region should show growth. For thoseareas in New England where natural gas is available,LNG supplements supply but is used only for shortdurations.

The South Atlantic and East South Central regionsare other areas with unique space heating profiles.There has been a significant decline in the percentageof consumers that use natural gas as the main heatingfuel. In these regions, the dominant residential spaceheating fuel has become electricity. Space heating ispredominantly from built-in electric units, electriccentral warm-air furnaces, or heat pumps. The heatpump has become increasing popular in these regions.

The evolution of the heat pump is a reflection ofchanges in the construction of residential structures,particularly multi-family housing units, where ductwork and vents are replacing pipes and radiators aswell as new heating equipment and technology. TheAmerican Gas Association reported that electric utili-ties in these areas encouraged consumer to add heatpumps and maintain gas furnaces as backup systems.Consequently, the percentage of household demandfor natural gas for heating is low in these two regions.

Current natural gas consumption is also an outcomeof historical accessibility to natural gas in urban andrural locations. Sub-regional profiles of householdswith natural gas service may differ from the regionalprofile. Households with gas service are predominant-ly in the more urban areas, while the percentage ofhouseholds with gas service in rural areas is muchlower. Figures 3-51 and 3-52 illustrate this trend.

Efficiency in Residential and CommercialConsumption

One of the most significant energy efficiency andconservation measures for the natural gas industry wasthe adoption of efficiency standards for commercialappliances in the Energy Policy and Conservation Actsof 1975 and 1978. The 1975 legislation established anenergy conservation program for major householdappliances, many of which used natural gas. The 1978legislation broadened the mandate of the 1975 act toinclude commercial building heating and air condi-tioning equipment as well as water heaters. In 1987,additional measures were put into place with theNational Appliance Energy Conservation Act, whichset energy efficiency standards for appliances accordingto a statutory time schedule stretching into 21stCentury.

The U.S. Environmental Protection Agency intro-duced ENERGY STAR in 1992 as a voluntary labelingprogram designed to identify and promote energy-efficient products. The ENERGY STAR label is nowon major appliances, office equipment, lighting, homeelectronics, and more. The EPA has also extended thelabel to cover new homes, and commercial and indus-trial buildings. The ENERGY STAR program deliverstechnical information and tools that consumers needin order to choose energy-efficient solutions and bestmanagement practices. Energy efficiency can result inthe delivery of the same (or more) services for lessenergy. Energy efficiency helps the economy by saving


0

20

40

60

80

100

TOTAL U.S. CITY TOWN SUBURBS RURAL

AVAILABLE UNAVAILABLE

Source: Energy Information Administration, 2001 Residential Energy Consumption Survey.

PE

RC

EN

T

Figure 3-51. Percentage of U.S. Housing Units with Natural Gas Available in Neighborhood in 2001

0

20

40

60

80

TOTAL U.S. CITY TOWN SUBURBS RURAL

Note: More than one use may apply.

Source: Energy Information Administration, 2001 Residential Energy Consumption Survey.

PE

RC

EN

T

Figure 3-52. Natural Gas Used in U.S. by Type of Location in 2001

consumers and businesses millions of dollars in ener-gy costs. Energy-efficient solutions can reduce theenergy bill for many homeowners and businesses by20% to 30%.

In the residential sector, newer housing stock is, onaverage, 18% larger than the existing housing.However, energy use per square foot is lower for newconstruction.4 This trend will likely continue, withnewer houses being tighter as a result of more stringentbuilding codes, better insulation, tighter window treat-ment, and tighter building design. Ongoing structuralefficiencies will continue to reduce the demand fornatural gas per customer.

Newer housing units are equipped with more effi-cient natural gas heating equipment. In the 1970s, nat-ural gas furnaces averaged annual fuel utilizationefficiency (AFUE) of about 65%. New furnace ship-ments in 2001 averaged an AFUE of 86%. Currently,all installed natural gas furnaces in 2001 averaged anAFUE of 77%. According to the American GasAssociation, technological enhancements in furnaceefficiency resulted in an average 4% fall in gas spaceheating use per customer nationwide between 1997and 2001.

In the commercial sector, use per customer declinedby 18% from 1979 to 1999. The decline in consump-tion can be attributed to the gained efficienciesbrought about by legislation and building codes.Another measure of customer conservation is con-sumption intensity (use per square foot of space). Anexamination of natural gas use per square foot con-firms that the average commercial building uses lessgas compared to 1979 levels. This measure fell rough-ly 40% over the past two decades.

Summary of Residential and CommercialDemand

Demand in the residential and commercial sectorswas analyzed for both the Reactive Path and BalancedFuture scenarios. Residential and commercial natu-ral gas demand is expected to increase in both sce-narios due to the combined effects of penetration ofgas-based technology, population growth, andgrowth in floor space, offset by energy efficiencygains. The 2000 to 2025 annual growth rate in the

Reactive Path scenario is slightly less than 1.0% inboth the residential and commercial markets. In theBalanced Future scenario, residential demandincreases by approximately 0.5% annually, while theannual average growth rate of commercial demand ishigher at 1.0%.

The Balanced Future scenario assumes significantlygreater efficiency gains in residential appliances, com-mercial equipment, and building standards. TheBalanced Future scenario demonstrates that policychanges such as expanding and diversifying natural gassupplies, increasing energy efficiency and fuel flexibili-ty, improving energy market efficiency and sustainingand expanding natural gas infrastructure can lowerprices and dampen the demand for natural gas in theresidential sector.

Residential demand in the Balanced Future scenariois lower than in the Reactive Path scenario primarilydue to increased efficiency in space and water heatingper household. Figure 3-53 depicts projections fortotal U.S. residential natural gas demand in the twoscenarios. Table 3-10 compares the difference in


4 Energy Information Administration, Annual EnergyOutlook 2003, pg. 57.

0

5,000

5,500

BIL

LIO

N C

UB

IC F

EE

T

6,000

6,500

2000 2010

YEAR

2020

REACTIVE PATH

BALANCED FUTURE

Figure 3-53. Total U.S. Residential Natural Gas Demand

demand and the average annual growth rate of the sce-narios. Additionally, this table shows the effects of dif-ferent economic growth rates modeled in sensitivityanalyses, comparing higher and lower GDP growth tothe Reactive Path scenario. Figure 3-54 illustrates theeffects of energy efficiency modeled in the ReactivePath and Balanced Future scenarios.

Unlike the residential sector, the commercial sec-tor experiences higher demand growth in the

Balanced Future scenario than in the Reactive Pathscenario. The conservation assumptions reflectedfewer opportunities for additional conservation thanin the residential sector, and the lower prices in theBalanced Future scenario were modeled as stimulat-ing additional commercial gas consumption. Figure3-55 depicts projections for total U.S. commercialnatural gas demand in the two scenarios. The figureindicates that gas consumption in the BalancedFuture scenario rises above that of the Reactive Path


Consumption:2000(BCF)

Consumption:2025(BCF)

Annual PercentChange:

2000-2025

Reactive Path 5,116 6,167 0.75

Balanced Future 5,116 5,817 0.51

High Economic Growth 5,116 6,252 0.81

Low Economic Growth 5,116 6,091 0.70

Table 3-10. U.S. Residential Natural Gas Consumption

0

200

400

600

800

1,000

1,200

1,400

1,600

2010 2020

YEAR

2025

BIL

LIO

N C

UB

IC F

EE

T

REACTIVE PATH

BALANCED FUTURE

Figure 3-54. Cumulative Energy Efficiency Effects in U.S. Residential and Commercial Sectors

scenario, especially after 2020. Table 3-11 comparesthe difference in commercial demand and the aver-age annual growth rate of the Reactive Path andBalanced Future scenarios, as well as sensitivityanalyses assessing higher and lower economicgrowth.

Although increased efficiency for space heating,water heating, and space cooling per square foot wasbuilt into the Balanced Future scenario, the impact of

lower gas prices was greater, resulting in an overallincrease in gas consumption.

As noted earlier, consumption is also a function ofthe growth rate of the economy. This study analyzedresidential and commercial consumption under lowand high GDP growth assumptions. Demand growthwill be mitigated by efficiency gains as old, inefficientequipment is replaced and houses are renovated and become more energy efficient. In addition, high


Consumption:

2000(BCF)

Consumption:

2025(BCF)

Annual Percent

Change:2000-2025

Reactive Path 3,346 4,093 0.81

Balanced Future 3,346 4,180 0.89

High Economic Growth 3,346 4,153 0.87

Low Economic Growth 3,346 4,043 0.76

Table 3-11. U.S. Commercial Natural Gas Consumption

3,000

3,500

4,000

4,500

0

BIL

LIO

N C

UB

IC F

EE

T

2000 2005 2010

YEAR

2015 2020 2025

REACTIVE PATH

BALANCED FUTURE

Figure 3-55. Total U.S. Commercial Natural Gas Demand

natural gas prices will likely provide a catalyst for resi-dential and commercial consumers to consume lessnatural gas by reducing the amount of energy servicesthey consume. The most immediate means to reduceenergy consumption is to adjust thermostat settingsand use more energy-efficient natural gas equipment.Table 3-12 illustrates demand reduction results froman aggressive response scenario that includes improvedefficiency, lower gas market shares, and permanentthermostat turn-back of 2°F – down in winter, up insummer.

Figure 3-56 illustrates the projections for regionalgrowth in residential and commercial natural gasdemand for the Reactive Path scenario. The largestimpact projected was in the South Atlantic, East SouthCentral, and West South Central regions. The Mid-Atlantic, New England, and the East North Centralregions exhibited the smallest percentage change inconsumption.


2000 2005 2010 2015 2020 2025

2,000

0

1,000

0

1,000 0

1,000

2,000

0

3,000

1,000

4,000

2,000

0

Figure 3-56. U.S. Residential and Commercial Natural Gas Demand in Reactive Path Scenario (Billion Cubic Feet per Year)

Regions Decrease

New England 9.2%

Mid-Atlantic 9.7%

East North Central 9.3%

West North Central 9.1%

South Atlantic 31.9%

East South Central 27.6%

West South Central 27.0%

Mountain 10.4%

Pacific 19.9%

United States 15.1%

Table 3-12. U.S. Residential and CommercialSensitivity – Decrease in Gas Consumption in 2025


Electric Power Sector

The electric power industry is the largest consumerof primary energy in the United States, as shown inFigure 3-57. It converts fossil fuels like coal, naturalgas, and oil plus nuclear, hydropower, and renewableenergy into electricity, which is then transmitted anddistributed to end-use customers.

The electric power industry is comprised of manystakeholders including vertically integrated utilities,municipalities, rural cooperatives, governmentalauthorities, independent power producers, fuel suppli-ers, federal and state regulators, and the all-importantconsumer. Each stakeholder has a compelling interestin the efficiency, environmental impact, cost, and reli-ability of the power industry.

The United States and Canada have organized theirtransmission grids into three frequency synchronousregions. These regions operate at the same alternatingcurrent frequency and are only interconnected withdirect current transmission ties. These regions havebeen organized in this manner to strive for highest

degree of reliability. The overall organization respon-sible for the guidelines and operating procedures is theNorth American Electric Reliability Council (NERC),which is comprised of the regional councils as shownon Figure 3-58. The three synchronous regions are:ERCOT, WECC, and the Eastern Interconnect (com-prised of all remaining regional councils – MAPP,MAIN, ECAR, NPCC, MAAC, SPP, SERC, andFRCC).1

The power industry has been in significant transi-tion, which is expected to continue for several moreyears. Wholesale power markets have been opened tomore competitors and prices are market based. Somestates have initiated retail competition with mixeddegrees of success. Merchant generators have built oracquired significant quantities of generation capacityand constitute a large percentage of the ownership oftotal capacity. Industry restructuring has impacted thefinancial strength and credit quality of numerous mar-ket participants. Independent system operators oftransmission have grown in importance. Envi-ronmental regulations continue to be promulgatedpiecemeal with uncertainty over timing, flexibility, andstringency. Partially as a result of these factors, naturalgas is becoming a more important fuel source for theindustry.

In the time period of this study, the electric powergeneration segment is anticipated to be the primarydriver of increases in overall natural gas demand.Power generation becomes the primary driver of natu-ral gas demand growth due to the increased percentageof gas-fired generation capacity in the United States’generation fleet coupled to the close correlationbetween electric power demand growth and the UnitedStates economic growth. A correlation betweengrowth of electric power and the U.S. economy isexpected to continue into the future.

Approximately 200,000 megawatts of gas-fired gen-eration will have been added to the generation fleet bythe end of 2005, representing a 31% increase of totalgeneration capacity and a 290% increase in the gas-fired only generating capacity, measured from the endof 1998. This new fleet is a combination of combus-tion turbines, used mainly for peaking power, and

1 See North American Electric Reliability Council website(www.nerc.com) for full names and details of the region-al councils.

RESIDENTIAL

7COMMERCIAL

4

INDUSTRIAL

23

TRANSPORTATION

26

ELECTRIC

POWER

37

Figure 3-57. Annual Average Primary EnergyDemand by Sector, 1997-2001 (Quadrillion Btu)


combustion turbines combined with a steam genera-tor, commonly referred to as a combined cycle plant.The generation fleet includes the portion of industrialsector’s CHP facilities that provide power, and excludesthe portion that produces steam for process use. Thecombined gas fleet can consume large quantities of gasat any given time depending upon weather-drivendemand and the relative competitiveness of other fuels,normally either residual fuel oil or distillate oil, avail-able to meet peak loads.

A comprehensive understanding of power genera-tion’s role in natural gas demand requires insight intothe demand for electricity, the economics and mechan-ics in power generation dispatch, regional differencesin generation fleet composition, transmission gridcapabilities, power pool dispatch operations, and keydrivers to investment decisions and fuel choices fornew generation construction.

Study Approach

The NPC evaluated electric power supply (capacity)and demand regionally using a model that solves formonthly electricity demand, power generation by typeof fuel, generating capacity additions, and fuel use.New capacity builds were determined in a separatemodel using logic parameters provided by study par-ticipants. Wide ranges of potential generation tech-nologies were considered whenever the model logiccalled for new capacity to be built. The study partici-pants imposed some constraints on new builds fueledby coal and residual fuel oil, but the general approachwas to allow economically rational choices to be madein both the Reactive Path and Balanced Future scenar-ios. Canada was modeled and analyzed, but with muchless detail and rigor than the U.S. lower-48. The por-tions of Mexico that are interconnected at borderregions were treated as interconnected net powertransfers.

ERCOT

Source: North American Electric Reliability Council.

WECC MAAC

MAIN ECAR

MAPP

NPCC

SPP

SERC

FRCC

Figure 3-58. NERC Regions


Nuclear and hydroelectric based generation quanti-ties were input into the dispatch models as discreteexogenous values implying the models did not “dis-patch” these units. Additionally, wind power was usedas a proxy for all renewable technologies, but this deci-sion was a simplifying assumption, not an endorse-ment of wind generation technologies over otherrenewable technologies.

The study approach was to model current laws andregulations in environmental emissions, siting, andongoing operations. The power model used for thestudy does not allow discrete generation unit evalua-tion of environmental emissions, but each case, sensi-tivity and scenario output was evaluated to ascertainwhether total calculated emissions met projectedallowance budgets for sulfur dioxide (SO2) and NOx.

Electric Power Demand

Demand for electricity in the United States is cou-pled to the economy. Growth in GDP has consistentlybeen accompanied by growth in electric demand. Therate of electric demand growth as a function of GDP(known as “electric intensity”) has decreased over thepast 50 years, and is likely to continue decreasing overthe period of this study. Through the 1960s, annual

electric energy demand grew at a rate faster than GDP.Coincident with the energy price shocks to the U.S.economy in the 1970s, electric energy demand begangrowing slower than the GDP. The reasons for thischange have been attributed to increased efficiency,energy conservation, the completion of the overallelectrification and air conditioning of U.S. society, anda fundamental shift in the levels of energy-intensiveindustry in the United States. As shown in Figure 3-59,the relationship between GDP growth and powerdemand growth has been fairly constant. No empiricaldata suggest this trend will substantially change duringthe study period. However, higher energy costs and theresultant consumer response, industrial capacity losses,and governmental policies are likely to further acceler-ate the rate of decline in the electric intensity.

The rate of growth in electric energy demand hasbeen modeled in the Reactive Path scenario to varyfrom a starting factor of 0.72 times the forecastedgrowth in GDP in 2003 and decreasing linearly to alevel of 0.62 times the forecasted growth in GDP by2025. Assumptions used for GDP are described inChapter Two of this report. Forecasted GDP growth isone of the most critical variables in assessing potentialgrowth in electric power demands and the resulting

0

2,000

2,500

3,000

3,500

4,000

3 4 5 6 7 8 9 10 11

GDP (TRILLIONS OF 2002 DOLLARS)

DE

MA

ND

(T

ER

AW

AT

T H

OU

RS

)

Figure 3-59. Electric Power Demand vs. GDP, 1982-2002

impact upon natural gas. However, the study partici-pants felt the future relationship between GDP growthand electric power demand growth had a fairly wideband of potential outcomes, particularly in a higherenergy price environment.

Weather conditions are the other major driver ofelectric demand. Hot weather routinely causesdemand to fluctuate by more than 10% within anyregion. Cold weather has slightly lesser impact uponpower demand due to the lower saturation level ofelectric-based residential and commercial heatingequipment compared to air conditioning equipmentparticularly in northern climates. Both modeled sce-narios used normal weather assumptions so there is noinherent weather-driven difference between them.Weather impacts were evaluated using sensitivities.

In the Reactive Path scenario, electric energydemand is estimated to grow from 3,470 terawatthours in 2003 to 5,420 terawatt hours in 2025. Figure3-60 shows historical growth plus the modeling resultsfrom the study efforts. The overall level of electricenergy demand is a primary driver to fuel consump-tion and consequently to demand for natural gas.Relatively small changes in the rate of growth in elec-

tric energy demand can significantly impact theamount of natural gas consumed within the power seg-ment. Each annual terawatt hour of retail sales differ-ential can be equivalent to 5 BCF per year of gasconsumed or conserved. In 2025, the difference inretail power demand between the Reactive Path andBalanced Future scenarios is estimated to avoid con-suming 265 BCF of natural gas for that one year.

The amount of natural gas required by the powersector is a direct function of the overall level of powerdemand at any moment, and natural gas’ competitiveposition within the regional generation capacity supplystack available to meet the power demand. Natural gasis used more heavily during peak electric demand peri-ods. This is due to the fact that while gas-fired genera-tion has one of the lowest capital costs to install, eventhe most efficient of gas generators are more expensiveto operate on a variable cost basis than most nuclearand coal capacity based upon historical fuel prices.Therefore, gas is not dispatched during periods oflower demand when baseload nuclear and coal capaci-ty is available to run. In seasons of higher demand,more gas-fired combined cycle generation is dis-patched, followed by less efficient gas steam units, andfinally during peak hour periods of certain days, gas


0

3,000

4,000

5,000

6,000

TE

RA

WA

TT

HO

UR

S

REACTIVE PATH

BALANCED FUTURE

ENERGY INFORMATION ADMINISTRATION

1990 1995 2000 2005

YEAR

2010 2015 2020 2025

Figure 3-60. Lower-48 Electric Power Retail Sales

combustion turbines are used to meet the highest peakloads. As overall electric energy and peak demandgrows, more of the baseload fleet’s limited remainingavailability is committed, thus requiring other unitsthat have higher variable costs to be more heavily uti-lized. In most regions these units are predominatelythe newer gas-fired combined cycle units. This inex-orably increases the demand for natural gas and repre-sents a significant increase compared to past years.

Generation Capacity

In 2002, lower-48 installed generation capacitytotaled 860 gigawatts, consisting of the fuel typesshown in Figure 3-61. Gas, oil, and dual-fuel unitscombined to be the second largest block of capacity.These units have historically been used to meet inter-mediate and peak demand requirements in most areasof the country. Baseload nuclear, hydro, and coal con-tribute electric energy to the supply demand balance inhigher proportions to installed capacity than gas, oil,and dual-fuel generation units. The relative contribu-

tions of generation by fuel type are also shown inFigure 3-61. Canada’s capacities and fuel usage aredescribed in the electric power segment of the DemandTask Group Report.

The United States has experienced periods of bothrapid generation capacity expansions and limitedexpansions. The 1990s was a period of limited capaci-ty expansion that coincided with strong economicgrowth and related robust power demand growth.This culminated in a period where installed capacitycould not reliably meet peak load in different regions,which resulted in higher power prices and extremevolatility in different regions during the latter years ofthe 1990s. During this time, regulatory driven struc-tural changes were occurring in the wholesale powermarkets. These changes encouraged the introductionof merchant powerplants into the wholesale markets.

The net result of these circumstances is a power plantconstruction boom that will add approximately


WECC

ERCOT/SPP SERC/FRCC

ECAR/MAIN/MAPPNPCC/MAAC

GENERATION CAPACITY(GIGAWATTS)

140 GW

220 GW

160 GW

180 GW

210 GW

920 TWH 500 TWH

900 TWH570 TWH

590 TWH

100,000

200,000

300,000

400,000

0

100,000

200,000

300,000

400,000

GAS OIL OTHER*NUCLEARHYDROCOAL

ELECTRICITY GENERATION BY FUEL (TERAWATT HOURS)

*Includes renewables.

GAS

OIL

OTHER*

NUCLEAR

HYDRO COAL

Figure 3-61. U.S. Electricity Generation and Generation Capacity in 2002, by NERC Region

220 gigawatts of generation capacity between the year1998 and the end of 2005. Almost all of this new capac-ity is gas-fired and less than 20% of the units have alter-nate fuel capability. In this study, capacity additions for2004 and 2005 were limited to projects under construc-tion that did not have any publicly announced delays.

New generation capacity added during the five-yearperiod from 2000 through 2004 rivals the largestamount of new construction experienced by the electricpower industry. Furthermore, the reliance on a singlesource of fuel is unprecedented. As shown in Figure 3-62, the prior construction boom in the 1970s resultedin coal, nuclear, gas, and oil generation capacity beingbuilt. This new reliance on a single fuel is the result ofa confluence of factors: lower emission rates, ease in sit-ing, relative capital costs, short lead time for installa-tion, modular size, and ease in specific site expansion.

Alternate Fuel Capability and Fuel Substitution

Many older steam units were built to be dual fueledwith natural gas and one of the fuel oil products.Energy Information Administration data and otherofficial reports suggest approximately 150,000megawatts of capacity is oil and gas switchable.

However, observed market behavior suggests thiscapacity is becoming more limited in switching capa-bility or can no longer switch to non-gas fuels.Historically, most oil switching occurred in Florida,New England, New York, and the Mid-Atlantic regionbecause they have the most residual fuel oil capacity.Residual fuel oil has historically competed with naturalgas on the margin for generation by steam (boiler)units. These regions also have oil-only units that sub-stitute for natural gas whenever oil is more economic.Additionally, Florida has historically been pipelinecapacity constrained, which leads to oil consumptionto meet demand since gas is not available.

The ease, operational risk, and economics of fuelswitching vary depending upon the technology. Steamunits, combined cycles, and combustion turbines havevery different considerations for the decision to switch.Older steam units can switch “on the fly” through asimple communication with the plant operators forchanges to the fuel burn mix. The process requiressomewhat imprecise adjustments to the oil-gas intakeflows and replacements of various boiler fuel guns.Also, typical fuel-switching requests can be accom-plished under a wide range of plant megawatt outputsand have very limited risk of unit output runback ortripping when being executed. Conversely, combined


0

10

20

30

40

50

60

1970 1975 1980 1985

YEAR

1990 1995 2000 2005

COAL

NUCLEAR

NATURAL GAS

OTHER

Sources: Energy Information Administration; Platt's; American Electric Power Co., Inc.

GIG

AW

AT

TS

Figure 3-62. Comparison of U.S. Annual Installed New Generation Capacity

cycles and combustion turbines have far greater sensi-tivity to the procedures for switching fuels. They mustbe at specific megawatt outputs or in some cases mustbe shut down completely to avoid unit runback or trip-ping off-line from the transmission grid.

Switching economics do not depend solely on thecompeting delivered cost of fuels. Consideration fordifferences in fixed and variable maintenance costs,increased emission costs, unit megawatt derates, andfuel infrastructure capability/costs are distinct parts ofthe calculation. Regulated utilities also consider thefuel recovery and operational and maintenance costrecovery risks as part of the decision to switch. The neteffect of these costs, operational constraints and envi-ronmental issues has led to a significant decline in oilusage for power generation and an increase in the pricedifferential needed to encourage fuel switching. Figure3-63 illustrates this trend.

The modeled population of oil- and gas-fired gener-ation capacity consists of three distinct types of units:those that run exclusively on gas (gas-only), those thatrun exclusively on oil (oil-only), and those that canswitch between gas and oil (dual-fuel). The oil used inthese units ranges from residual oil (Nos. 4, 5, and 6)and distillate oil (No. 2 oil/kerosene). The relative eco-

nomics of dispatching these units depends upon thedelivered fuel price, emissions, and variable operatingand maintenance expenses. Therefore, the term “fuelswitching” applies to two conditions: (1) the shiftbetween the use of gas or oil at dual-fuel units; and (2) when both gas-only and oil-only units are availablewithin a dispatch region, the shift of dispatch betweenthese units by substituting the dispatch of one unit foranother. This results in modeling of both switchingand substitution behavior in determining the capacityand generation output of the regional powerplants.

The Reactive Path scenario assumed a number oflimits on the addition of either oil-only or dual-fuelunits. The NPC assumed that no new oil-capablecapacity would be allowed in the northeastern or westcoast states, and that the amount of switching capabil-ity for the United States as a whole would be limited to25% of the gas and oil capacity total. Additionally, theconstruction of residual oil-only units was limited toregions where it was felt that there would be sufficientexisting infrastructure to accommodate the additionaloil consumption. Due to a number of constraints, suchas permit conditions and the availability of oil, thereare also limitations to the amount of switching possi-ble at dual-fuel units. Therefore, the total capability toswitch from gas to oil is the sum of all the oil-only


0

50

100

150

200

250

300

350

400

450

1975 1980 1985 1990 1995 2000

BIL

LIO

N K

ILO

WA

TT

HO

UR

S

OIL GAS

YEAR

Figure 3-63. U.S. Electricity Generated, by Oil and Gas

capacity plus a fraction of the dual-fuel capacity, basedon the maximum number of hours per year the dual-fuel units can operate on oil.

Since gas and oil capacity primarily serves peak andintermediate load, the overall capacity utilization of allgas- and oil-fired units was found to be relatively lowin the Reactive Path scenario, ranging from 20% in thenear term to nearly 30% by 2025. In the Reactive Pathscenario, the total amount of oil-capable capacityincreased from 81 gigawatts in 2003 to 137 gigawatts by2025. Additionally, the utilization of fuel-switchingcapability during this period was found to increase inthe Reactive Path scenario from a near-term average ofabout 10% to 22% by 2025. These values represent theannual average utilization and are depicted in Figure 3-64 and compared to the Balanced Future scenario.On a seasonal basis, the utilization of the fuel-switch-ing capability were found in the Reactive Path scenarioto be greatest in the winter months when gas priceswere modeled to be at their peaks.

Figure 3-64 shows the amount of fuel switching inelectric power generation for the Balanced Future sce-nario. This scenario assumes actions are taken bypower generators to increase the amount of fuel switch-ing beyond that contemplated by the Reactive Path sce-

nario by retrofitting 25% of existing combined-cycleand combustion turbine facilities for backup fuel. Theoverall lower price environment of the Balanced Futurescenario leads to less overall fuel switching, and lowerannual average usage is forecasted. However, theactions assumed in the Balanced Future scenario allowgreater level of fuel switching at times of peak demand,such as during periods of cold weather. This wasdemonstrated by sensitivity analyses that consideredextreme weather cycles, showing the greater fuel switch-ing capability contributed to lower overall prices. Thenet effect of these changes between the scenarios isBalanced Future has more oil capability that allows it toreduce peak requirements on natural gas, leading tolower annual gas prices and higher annual oil switchingand substitution in power generation fuel choices.

Assumptions and Case Results

Reactive Path

The Reactive Path scenario incorporates assump-tions that range from aggressive to moderately conser-vative on the issues of new build economics, new buildfuel selection, technology advances, re-licensing ofnuclear and hydroelectric units, environmental regula-tion impact, and fuel usage. It is important to note that


0

100

200

300

400

500

600

700

2000 2005 2010YEAR

2015 2020 2025

GE

NE

RA

TIN

G C

AP

AC

ITY

(G

IGA

WA

TT

S) BF O&G CAPACITY BF TOTAL OIL CAPACITY

BF OIL CAPACITY USEDRP O&G CAPACITY RP TOTAL OIL CAPACITY

RP OIL CAPACITY USED

Note: RP = Reactive Path; BF = Balanced Future; O&G = oil and gas.

Figure 3-64. U.S. Gas-to-Oil Switching and Substitution in Reactive Path and Balanced Future Scenarios

many study participants felt the totality of the collect-ed assumptions represented a “success” scenariowherein market-based decisions triumph, regulatoryimpediments are not notably increasing, and environ-mental policies limiting non-gas fuels are not substan-tially expanded. One large driver of natural gasdemand from electric power in the Reactive Path sce-nario is the assumption that approximately 20gigawatts of coal-fired capacity is retired prior to 2010as a result of mercury emission regulations that arescheduled to be proposed in December 2003 by theU.S. Environmental Protection Agency.

Balanced Future

The Balanced Future scenario represents additionalenhancements to electric power assumptions thatreduce gas demand and contribute to a more relaxedbalance between supply and demand.

The complete list of common and differentialassumptions are shown below, but some worth high-lighting are: (1) the assumption on greater efficiency inthe United States’ consumption of electricity as meas-ured by income elasticity in the GDP scalar; (2) a reso-lution of upcoming mercury emission regulations that

avoids substantial coal capacity retirement; and (3)substantial increase in renewable generation capacity.

The two scenarios result in significant differences ingas demand by power generation, as shown in Figure3-65. The Reactive Path scenario consistently has high-er demand starting concurrently with retirement ofold, small coal units due to mercury air quality regula-tions, and continuing with less flexibility to use alter-nate fuels over the study period.

Table 3-13 shows a side-by-side comparison of keymodel outputs including those driven by inputassumptions affecting generating capacity. Table 3-14shows the major power assumptions common to eachscenario or differential to each scenario. A briefdescription of each assumption is bulleted below.

Detailed Assumptions

� Both scenarios assumed an electric demand growthfactor starting at 0.72 times the change in GDP; theReactive Path scenario declined this factor to 0.62while the Balanced Future scenario declined to 0.55over the study period. This reflected greater energyefficiency driven by market forces and enabled bygovernment policies.


0

1

2

2005 2010 2015

YEAR

2020 2025

3

4

5

6

7

8

9

TR

ILLIO

N C

UB

IC F

EE

T BALANCED FUTURE

REACTIVE PATH

Figure 3-65. Natural Gas Consumption in the U.S. Power Sector

� National average coal prices are assumed to be $1.28per MMBtu on a volume-weighted basis, decliningby 1% real dollars annually.

� The forecast price of West Texas Intermediatecrude oil (WTI) was assumed to be constant inreal terms at $20.00 per barrel (2002 dollars). Therefiner acquisition cost of crude oil (RACC) was

assumed to be 90% of WTI. Residual fuel oil anddistillate prices are based on a percentage rela-tionship to RACC on a dollars-per-Btu basis. Theresulting ratios are residual fuel oil at 80% ofRACC and distillate at 140% of RACC. Theresulting prices are: high-sulfur fuel oil No. 6 –$2.48 per MMBtu; No. 2 oil – $4.34 per MMBtu(2002 dollars).


ReactivePath

BalancedFuture

Natural Gas (Trillion Btu per Year Average) 7.13 6.71

Coal (Trillion Btu per Year Average) 23.3 23.5

Oil (Trillion Btu per Year Average) 1.4 0.9

Generation (Terawatt Hours per Year Average) 4,497 4,472

Post-2005 New Gas-Fired Capacity (Gigawatts) 148 128

Post-2005 New Coal-Fired Capacity (Gigawatts) 132 133

Post-2003 Nuclear Capacity Improvement (Gigawatts) 1.9 9.7

Post-2003 New Renewable Capacity (Gigawatts) 73 155

Cumulative Oil/Gas Retirements (Gigawatts) 9 0

Cumulative Coal Retirements (Gigawatts) 20 0

Table 3-13. Scenario Results and Associated Assumptions

Assumption Status Assumption Status

Energy Intensity-GDP based Changed SO2 Emissions Same

Coal Prices Same NOx Emissions Same

Oil Prices Same Mercury Emissions Changed

Nuclear Licenses Same Carbon Emissions Same

Nuclear Capacity Enhancement Changed Coal Plant Siting Same

Hydro Licenses Same Oil Plant Siting Same

Hydro Enhancement Same Renewable Plant Siting Changed

New Build Economics Same Alternate Fuel Capability* Changed

Renewable Capacity Changed Alternate Fuel Capability† Changed

Weather Same Transmission Capability Same

*Retirement quantities of oil/gas steam units.†Percentage of operating hours oil can be used.

Table 3-14. Common and Differing Assumptions Between Scenarios

� Each scenario assumes that all nuclear plants haveone successful license extension.

� The Reactive Path scenario assumes nuclear capaci-ty is enhanced by approximately 6%, but that somecapacity is lost due to regulatory actions for a netincrease of 2%.

� The Balanced Future scenario assumes nuclearcapacity is enhanced by a net amount equal to 10%.

� Each scenario showed hydropower as an exogenousinput of megawatt hours that was derived from his-torical averages for each month by geographicregion. No capacity enhancements or degradationsare included in the input amount.

� New build economics were the same in both scenar-ios. They varied in sensitivity cases.

� Renewable generation in the Reactive Path scenarioincreased by 73 gigawatts between 2003 and 2025,while it increased by 155 gigawatts in the BalancedFuture scenario. The amounts and geographic dis-tribution of renewable capacity in the Reactive Pathscenario were based on meeting state-level renew-able portfolio standards; these amounts were signif-icantly increased in the Balanced Future scenariobased on analyses by the NPC. In both cases, theseamounts and geographic distributions were exoge-nous inputs to the models. (Wind turbines weremodeled as a proxy for all renewables in both sce-narios, which is not an endorsement of it over othertechnologies. These other technologies are dis-cussed in more detail within the electric section ofthe Demand Task Group Report.)

� Weather was same in both scenarios.

� SO2 emissions were calculated post model runs inboth scenarios.

� NOx emissions were calculated post model runs andpowerplant dispatch was altered if the levels of emis-sions exceeded the allowable amount.

� Mercury emissions per se were not modeled, but thepotential impact of the impending mercury regula-tions were modeled by retiring approximately 20gigawatts of coal capacity in the Reactive Path.These generating units were older than 40 years,smaller than 200 megawatts, and not co-located witha larger unit or plant site.

� Carbon emissions were considered in a sensitivity.

� One main assumption on coal was a modelingrestriction that did not allow any coal-fired plants tobe built in EPA-designated non-attainment areas forozone or other major pollutants. Additionally, nocoal-fired units were added in California,Washington, and Oregon and total additions werelimited in Florida to 4 gigawatts.

– Overall coal-fired generation construction waslimited to 14 gigawatts per year.

– New coal generation capacity was added, beyondthe amount needed to meet the model’s 15%reserve margin requirement, whenever the fueland power price relationships provided financialincentives.

– Maximum annual availability of coal-fired gener-ation was assumed to be 80%.

� Residual fuel oil-fired combined cycles were limitedto the south Atlantic, Gulf Coast, and Gulf Coastwaterways to reflect the fuel logistics. Geographicrestrictions similar to those of coal were used for theeast and west coasts.

� Renewable generation was largely placed inCalifornia and Nevada in the Reactive Path scenario;it was more geographically distributed in theBalanced Future scenario to reflect both successfulmarket penetration of technologies and industryresponse to government policies addressing Renew-able Portfolio Standards requirements.

� Gas-fired generation added after 2005 was able todispatch on distillate oil up to 10% of the time in theReactive Path scenario, while the Balanced Futurescenario allowed it to dispatch up to 15% of thehours if fuel economics justify switching.

� In the Reactive Path scenario, approximately20 gigawatts of existing oil/gas steam units retiredpost-2001, while in the Balanced Future scenariothey did not retire.

� The power transmission grid was assumed to be thesame in both scenarios. The underlying assumptionsincreased interregional capacity by 50% over thestudy period.

New Build Economics

A model outside the electric dispatch model deter-mines new generation capacity. When electric powerdemand grows to a level where the system reserve


margin in any region is less than 15%, the model com-pares the generation capacity options and selects themost economic technology and fuel. The expansionplanning process in the EEA models is a heuristicapproach relying on busbar curves to determine theappropriate capacity factor operating ranges and pro-duction simulation to determine if the newly addedunits are operating in those ranges.

“Busbar” refers to the transmission equipment justat the edge of the powerplant’s site. The costs “behindthe busbar” include all fuel costs, construction costs,financing costs, taxes, operations and maintenanceexpenses, and all the other costs of owning and operat-ing a powerplant. These cost inputs were developed in“real” or “constant dollar” terms and converted to“nominal” dollars using escalation rates for the variouscosts.

In the production simulation, the newly added unitswere integrated with the existing fleet and all unitswere dispatched to meet load. The capacity factor testsonly give proper answers if the analysis is approachedfrom one perspective: units operating above theircapacity factor range should be replaced with the nexttype of unit but units operating below their capacityfactor range may still be the economic choice. This isdue to the fact that the new units may be operating atlower capacity factors because there are existing unitsin a similar dispatch price range. For example, a newefficient coal unit may operate at a high capacity factor,but it may not be economic in some regions because itis simply decreasing the operation of slightly less effi-cient coal units.

Table 3-15 shows the technologies considered and aselection of the input criteria. A more detailed list is


Technology Description

LeadTime

(Years)

Capital Cost(2002 Dollarsper Kilowatt)

2010 Heat Rate(Btu per

Kilowatt Hour)

Maximum CapacityUtilization(Percent)

Conventional PulverizedCoal w/ Scrubber

7 1,200 9,300 85

Integrated CoalGasification CombinedCycle Greenfield

6 1,400 9,000 90

Integrated CoalGasification CombinedCycle Brownfield

5 1,400 9,000 90

Super Critical PulverizedCoal w/ AllEnvironmental

7 1,250 8,600 85

Gas Combined Cycle 3 600 7,000 92

Low-Sulfur DieselCombined Cycle

3.5 600 7,200 90

Distillate Combined Cycle 4 670 7,400 88

E-Class Residual OilCombined Cycle w/Environmental

4 800 8,100 70

Gas Combustion Turbine 1.5 350 10,000 15*

Low-Sulfur DieselCombustion Turbine

2.5 400 10,600 15*

Advanced Nuclear 10 1,500 10,500 92

Renewable – Wind 3 1,100 N/A 30

* 30% maximum capacity factor in West for low hydro years and backup for renewables.

Table 3-15. Generation Technologies Model Input Parameters

included in the Appendices to the Demand Task GroupReport.

Environmental Issues

Uncertainty over the extent and timing of environ-mental rules, particularly air quality regulations, are asignificant factor in electric industry decision-makingfor investments in new generation. The Reactive Pathscenario attempted to predict industry actions inresponse to existing laws and regulations, includingthose scheduled to be implemented under current lawsand rulemakings. Our Balanced Future scenario didnot attempt to model any specific legislative initiative,such as “Clear Skies,” but it did envision action thatreduces uncertainty over emissions standards, providesclear and extended timelines for compliance, and pro-motes cap and trade emission systems that are marketdriven and useful.

One example is the mercury emission control rule,which is scheduled to be released in December 2003 indraft form, and in final form by December 2004.Actual implementation is forecasted to be 2008.However, almost certain litigation by all sides in thedebate may either postpone implementation, or creategreater uncertainty over final implementation. The netconsequence of these types of uncertainty is to stiflesome of the needed investments in control technology

for economically marginal plants, until clarity isachieved. This can result in short or intermediate-termdislocations in non-gas-fired generation capacity,which directly impacts the amount of gas built, anddispatched during the period that non-gas-fired capac-ity is unavailable. Table 3-16 lists some of the knownupcoming environmental issues facing the electricpower industry.

Siting and New Source Standards

The U.S. Environmental Protection Agency describesthe new source standards as follows:

Section 111 of the Clean Air Act, “Standards ofPerformance of New Stationary Sources,” requiresEPA to establish federal emission standards forsource categories which cause or contribute signif-icantly to air pollution. These standards areintended to promote use of the best air pollutioncontrol technologies, taking into account the costof such technology and any other non-air quality,health, and environmental impact and energyrequirements. These standards apply to sourcesthat have been constructed or modified since theproposal of the standard. Since December 23,1971, the Administrator has promulgated nearly75 standards. These standards can be found in the


Regulation Issuance Date Implementation Date

NSR Enforcement Ongoing

NSR Rule – Routine Maintenance 2004

NOx (Section 126) State Petitions 2004

NOx SIP Call 2004

Clean Water Act 316(b) 2004

Mercury controls (MACT) 2003 2008

Ozone (8 hour) 2010

Fine particulate standards 2010

Regional Haze (BART) 2012

NSR = New Source Review MACT = Maximum Achievable Control Technology

SIP = State Implementation Plan BART = Best Available Retrofit Technology

Table 3-16. Major Federal Environmental Regulations Affecting Power Industry

Code of Federal Regulations at Title 40 (Protectionof Environment), Part 60 (Standards of Perfor-mance for New Stationary Sources).

Generally, state and local air pollution controlagencies are responsible for implementation, com-pliance assistance, and enforcement of the newsource performance standards (NSPS). EPAretains concurrent enforcement authority and isalso available to provide technical assistance whena state or local agency seeks help. EPA also retainsa few of the NSPS responsibilities – such as theability to approve alternative test methods – tomaintain a minimum level of national consistency.

In areas with acknowledged air quality issues, thesestandards are typically used to drive decisions towardsnatural gas generation technology without the ability toswitch to alternate fuels, normally No. 2 oil. Thisobserved behavior was a primary reason for modelingassumptions that limited new coal and oil fired genera-tion capacity in general, and that did not allow construc-tion of any new capacity in the northeast United States orthe west coast.

A subset of these rules known as New Source Reviewapplies to modifications within existing units. Theserules are in the process of being reviewed and reissuedto clarify the issues. Litigation resulting from the pro-posed clarifications and modifications to the existinginterpretation of the rules is certain. The uncertaintyof the outcome continues to inhibit the electric powerindustry from modifying and upgrading existing coalgeneration that could quickly provide substantialcapacity. Industrial energy users face the exact sameissues within their plants and processes as discussed intheir section of the report.

Sensitivities and Scenarios Summary

The sensitivities that impact power generation themost are:

1. Fuel Flexibility

2. High Electricity Sales to GDP Elasticity

3. Low Electricity Sales to GDP Elasticity

4. High GDP Growth

5. Low GDP Growth

6. Weather Cases

7. Carbon Reduction

The first 5 sensitivities are graphed in Figure 3-66 withReactive Path and Balanced Future scenarios to compare

natural gas demand. There is a wide variance in overalldemand. The results suggest that continued improve-ments in efficiency for power consumers and flexible fuelarrangements are critical keys to minimizing the amountof natural gas consumed within the power sector.

A continuation of current improvements in efficien-cy coupled with a more flexible fuel regulatory outlookwould save over 3 TCF per year by 2025 when com-pared to a future that has no additional electric powerefficiency. Additional fuel flexibility could result insaving 1.7 TCF per year in 2025 when compared to theReactive Path scenario. The weather sensitivities andcarbon reduction scenario are described in theDemand Task Group Report.

Other Considerations

Power Markets and Transmission

The electric power wholesale market is regional innature due to the operation of the transmission gridand the ability to flow power between different areas ofthe grid. Control of the power transmission grid is verydifferent than the control capability of the gas trans-mission system. Interconnect capacity between theregions is expected to improve due to technologyimprovements in the limiting equipment. However, sit-ing power transmission lines remains very difficult.Consequently, little improvement is expected in theability to move large blocks of power, from regions withlower cost generation (usually coal, hydroelectric, andnuclear) to higher cost areas. Approximately 1,000 cir-cuit miles of new transmission of 230 kilovolts or high-er, which represents approximately a one-half percentincrease in circuit miles, are added each year to theNorth American regions in participation with NERC.This, however, does not translate into the same amountof increase in capacity to transfer load within orbetween regions.

Renewable Power Generation Technologies

The two scenarios assumed significant quantities ofrenewable generation technology is installed during theperiod of this study. Reactive Path assumes 73 gigawattsand Balanced Future assumes 155 gigawatts of capacity.In both scenarios, wind was the modeled technology,but this is not an implied endorsement of wind versusother technologies. These quantities are within thewide range of many projections by governmental agen-cies and non-governmental organizations. The driversto renewable generation in the near future will be the


continuation of Renewable Portfolio Standards (RPS)and potentially an extension of tax credits that havespurred development over the past few years. Currently12 states have RPS in place, 3 states have “best efforts”goals, and 7+ states are actively considering some levelof RPS. Canada is also actively promoting some level ofRPS. Over the long term, economic competitivenesswill be the primary driver to the market penetration ofthese technologies. Figure 3-67 shows the relative com-petitiveness of various technologies with and withoutincentives in 2003 and projected for 2013.

Conclusions, Recommendations, andUncertainty Discussion

Study participants in the Power GenerationSubgroup concluded the following:

� The magnitude of new gas-fired capacity added tothe generation fleet creates the potential for largesurges of demand for natural gas under a variety ofcircumstances:

– Sudden reductions in baseload capacity likenuclear, coal, or hydro, whether by regulatoryaction, poor rainfall, environmental rulings,security issues, or other reasons.

– Hot weather events in the summer driving short-term gas peak dispatch will reduce gas availablefor storage injections.

� Energy efficiency reflected by lower electric intensi-ty reduces the need for capacity investments andreduces the forecasted demand for natural gas.

� Lack of alternate fuel capability in the recent naturalgas new builds, retirement of existing oil/gas steam,and local sentiment in opposition to oil infrastruc-ture at plants is leading to less flexibility in fuelchoices and less price elasticity in gas purchases.This creates a higher probability of electric powergas purchases contributing to gas price volatility.

� Economically rational market solutions should bethe primary driver in fuel choices while recognizingenvironmental standards as a relevant factor.

� Environmental regulation uncertainty and long-term concerns on potential carbon legislation createreluctance to invest in coal-based technology. Coalbuilds included in cases reflect a bias towards mak-ing the investment, rather than waiting for increasedcertainty.

– Construction activity for coal has been at a lowpoint for five years and is expected to continuefor several more years.

� The structure of power markets have limited impacton gas demand, although in some regions less-efficient steam units dispatch at levels higher thanexpected even though new combined-cycle efficientunits are available.


0

2

4

6

8

10

12

2000 2005 2010

YEAR

2015 2020 2025

BALANCED FUTURE

LOW ELASTICITY

HIGH ELASTICITY

LOW GDP

HIGH GDP

FUEL FLEXIBILITY REACTIVE PATH

TR

ILLIO

N C

UB

IC F

EE

T

Figure 3-66. Comparison of Natural Gas Consumption in the Power Sector

0

2

4

6

8

10

12

14

0

2

4

6

8

10

12

14

ONSHORE WIN

D

(CLA

SS 4 W

IND)

OFFSHORE WIN

D

(CLA

SS 7 W

IND)

LOW

-IMPACT H

YDRO �

SMALL &

MIN

IBIO

MASS CO-F

IRIN

G

WIT

H COAL

BIOMASS G

ASIFIC

ATION

COMBINED C

YCLE �

GRID P

OWER

LANDFILL

GAS

CONCENTRATING S

OLAR

POWER �

TROUGH

GEOTHERMAL

NEW G

AS GTCC

NEW P

ULVERIZ

ED COAL

LE

VE

LIZ

ED

CO

ST

OF

ELE

CT

RIC

ITY

CE

NT

S P

ER

KIL

OW

AT

T H

OU

R (

2003 D

OLLA

RS

)

WITHOUT INCENTIVES:

WITH INCENTIVES:

Source: Navigant Consulting, Inc., The Changing Face of Renewable Energy, June 19, 2003 (Public Release, October 2003).

2003

20132003 & 2013

CH

APTER 3 - N

ATURA

L GA

S DEM

AN

D 100

Figure 3-67. Life Cycle Cost of Electricity – Renewable Energy Options, With and Without Incentives


Comparison with 1999 Study Results

Figure 3-68 illustrates the demand projections forthe Reactive Path and Balanced Future scenariostogether with the NPC 1999 Reference Case projection.The 2003 demand forecast of the Energy InformationAdministration’s Annual Energy Outlook (AEO) is alsodepicted for reference.

The most visible difference between 1999’s studyand this study is the approach to evaluating the sup-ply/demand balance. The primary focus of the 1999study was to test supply and delivery systems againstsignificantly increased demand. The current studybuilt up demand as part of the two scenarios and sen-sitivities. Like this study, the 1999 NPC study assesseddemand on a regionally disaggregated basis in the fol-lowing sectors: residential, commercial, industrial,electric power generation, and lease/plant/pipelinefuel. Major differences in approach and assumptionsbetween these studies are as follows:

� This study developed and employed a descriptivemodel for industrial demand that estimated demandfor boiler fuel, feedstock, process heat, and cogener-ation/other demand in ten industry groups. The

1999 study did not perform a detailed assessment ofindustrial demand.

� This study was based on in-depth analyses of alter-native power generation technologies and the heatrates and expected efficiency improvements of thesetechnologies, and it gave consideration to potentialfuture transmission enhancements in NorthAmerica. The 1999 study took a more limitedapproach to generation capacity, made globalassumptions for expected generation efficiencyimprovements, and did not consider transmissionenhancements.

� The 1999 study assumed that new gas-fired genera-tion additions would grow by 88 gigawatts from1998 to 2010. The 2003 study reflected the actualnew generation buildup of over 200 gigawatts from1998 through 2005.

� The 1999 study assumed that 15 gigawatts of nucleargeneration capacity would retire by 2015. The 2003study assumes all nuclear generation facilities will berelicensed once, such that overall nuclear capacitywill remain essentially constant through 2025. The1999 study assumed nuclear capacity factors wouldincrease from 75% to 80% annually. The 2003 study

0

10

20

30

40

2000 2005 2010

YEAR

2015 2020 2025

EIA AEO 2003

NPC 2003 REACTIVE PATH

NPC 2003 BALANCED FUTURE

NPC 1999 REFERENCE

TR

ILLIO

N C

UB

IC F

EE

T

Figure 3-68. Comparison of U.S. Demand Projections


assumes more recent performance that approaches90% capacity factors will persist.

� Coal capacity factors were assumed to increase from64% to 75% in 2015 as part of the 1999 study. The2003 study sets the maximum coal availability at 80%annually, which coal generation achieves by 2015.

� The 2003 study assumes much larger capacity instal-lations of renewable generation than the 1999 study.

Recommendations Related to Natural Gas Demand

To achieve our nation’s economic goals and meetour aspirations for the environment, natural gas willplay a vital role in a balanced energy future. Stable andsecure long-term supply, a balanced fuel portfolio, andreasonable costs will be enabled by a comprehensivesolution composed of key actions facilitated by publicpolicy at all levels of government. The foundation ofdemand-related recommendations is to improvedemand flexibility and efficiency. Natural gas is a crit-ical source of energy and raw material, permeating allsectors of the economy. Each sector of the economycan make contributions to using natural gas resourcesmore efficiently.

The changes in demand require involvement of eachconsumer segment and can be broadly characterized as:

� Energy efficiency and conservation

� Fuel switching and fuel diversity.

In the very near term, reducing demand is the pri-mary means to keep the market in balance because ofthe lead times required to bring new supply to market.While current market forces encourage conservationamong all consumers and fuel switching for large cus-tomers who have that capability, proactive governmentpolicy can augment market forces by educating thepublic and assisting low-income households. Key ele-ments of this recommendation are summarized below.

Encourage Increased Efficiency andConservation through Market-OrientedInitiatives and Consumer Education

Energy efficiency is most effectively achieved in themarketplace, and can be accelerated by effective utiliza-tion of power generation capacity, deployment of high-efficiency distributed energy (including cogenerationwhich captures waste heat for energy), updating build-ing codes and equipment standards reflecting current

technology and relevant life-cycle cost analyses, pro-moting high-efficiency consumer products includingbuilding materials and Energy Star appliances, encour-aging energy control technology including “smart”controls, and facilitating consumer responsivenessthrough efficient price signals.

� Educate consumers. All levels of governmentshould collaborate with non-governmental organi-zations to enhance and expand public educationprograms for energy conservation, efficiency, andweatherization.

� Improve conservation programs. DOE shouldidentify best practices utilized by states for the low-income weatherization programs and encourageadoption of such practices nationwide.

� Review and upgrade efficiency standards. DOE,state energy offices, and other responsible state andlocal officials should review the various building andappliance standards that were previously adopted toensure that decisions reached under cost/benefitrelationships are valid under potentially higherenergy prices.

� Provide market price signals to consumers to facili-tate efficient gas use. FERC, Regional TransmissionOrganizations (RTOs), and state utility commissionsshould facilitate adoption of market-based mecha-nisms and/or rate regimes, coupled with metering andinformation technology to provide consumers withgas and power market price signals to allow them tomake efficient decisions for their energy consumption.

� Improve efficiency of gas consumption by resolvingthe North American wholesale power market struc-ture. FERC and the states/provinces, and if necessarycongressional legislation, should improve wholesaleelectricity competition in the United States, Canada,and interconnected areas of Northern Mexico. FERCshould mitigate rate and capacity issues at the seamsbetween adjoining RTOs to maximize efficient ener-gy flows between market areas.

� Remove regulatory and rate-structure incentives toinefficient fuel use. FERC, RTOs, and state regula-tors should ensure central dispatch authority rules,procedures and, where applicable, cost recoverymechanisms, require dispatch of the most efficientgenerating units while meeting system reliabilityrequirements and minimizing cost.

� Provide industrial cogeneration facilities withaccess to markets. Congress, FERC, RTOs, and,


where applicable, state regulators should ensure thatlaws, regulations, and market designs provide indus-trial applications of cogeneration with either accessto competitive markets or market-based pricingconsistent with the regulatory structure where thecogeneration facility is located.

� Remove barriers to energy efficiency from NewSource Review. Remove barriers to investment inenergy efficiency improvements, and investments innew technologies and modernization of power-plants and manufacturing facilities by implementingreforms to New Source Review such as those pro-posed by the U.S. Environmental Protection Agencyin June 2002.

Increase Industrial and Power GenerationCapability to Use Alternate Fuels

Natural gas has become an integral fuel for industri-al consumers and power generators due to a range offactors, including its environmental benefits, and theseconsumers should continue to be allowed to choosenatural gas to derive these benefits. However, thegreatest consumer benefit will be derived from market-based competition among alternatives, while achievingacceptable environmental performance. The ability ofa customer to switch fuels serves to buffer short-termpressures on the supply/demand balance and is aneffective gas demand peak shaving strategy that shouldreduce upward price volatility. Increasing fuel diversi-ty, the installation of new industrial or generationcapacity using a fuel other than natural gas, serves toreduce gas consumption over the life of the new capac-ity. Most facilities that would consider installing non-gas fueled capacity tend to be large and energyintensive. Therefore, increasing fuel diversity will havea large cumulative effect on natural gas consumptionover the period of this study.

� Provide certainty of air regulations to create a clearinvestment setting for industrial consumers andpower generators, while maintaining the nation’scommitment to improvements in air quality.

– Provide certainty of Clean Air Act provisions.Congress should pass legislation providing cer-tainty around Clean Air Act provisions for SOx,NOx, mercury, and other criteria pollutants.These provisions should recognize the overlap-ping benefits of multiple control technologies.The current uncertainty in air quality rules andregulations is the key impediment to investmentin, and continued operation of, industrial applica-

tions and power generation facilities using fuelsother than natural gas. Congress should ensurethat such legislation encourages emission-tradingprograms as a key compliance strategy for anyemissions that are limited by regulation.

– Propose reasonable, flexible mercury regulations.The Environmental Protection Agency’s December2003 proposed mercury regulations should pro-vide adequate flexibility to meet proposed stan-dards. These regulations should acknowledge thereductions that will be achieved by way of otherfuture compliance actions for SOx and NOx emis-sions, and provide phase-in time frames that con-sider demand pressure on natural gas.

– Reduce barriers to alternate fuels by New SourceReview processes. Performance-based regula-tions should meet the emission limits requiredwithout limitations on equipment used or fuelchoices. State and federal regulators shouldensure that New Source Review processes, andNew Source Performance Standards in general, donot preclude technologies and fuels other thannatural gas when the desired environmental effi-ciency can be achieved.

� Expedite hydroelectric and nuclear powerplant reli-censing processes. FERC, the Nuclear RegulatoryCommission, and other relevant federal, state,regional, and local authorities should expedite reli-censing processes for hydroelectric and nuclearpower generation facilities. These authoritiesshould fully consider the increased future require-ments for natural gas-based generation in the affect-ed regions that could arise from “conditions ofapproval” or denial of relicensing. In the case ofdenial, adequate phase-in time specific to the fueltype of replacement resources should be provided tobring alternative generation resources onto the gridto replace non-renewed facilities.

� Take action at the state level to allow fuel flexibility.

– Ensure alternate fuel considerations in Inte-grated Resource Planning. Where IntegratedResource Planning is conducted at the state regu-latory agency level, state commissions shouldrequire adequate cost/benefit analysis of addingalternate fuel capability to gas-only-fired capacity.

– Allow regulatory rate recovery of switching costs.State public utility commissions should providerate treatment to recover fuel costs and increasednon-fuel operating and maintenance costs when

units switch to less expensive alternate fuels asmatter of practice and policy, since the fuelswitching either directly or indirectly benefitsratepayers by reducing gas price and/or volatilitythrough fuel switching.

– Support fuel backup. State executive agenciesshould ensure that policies of state permittingagencies encourage liquid fuel backup for gas-fired power generation, and encourage a balancedportfolio of fuel choices in power generation andindustrial applications.

� Incorporate fuel-switching considerations in powermarket structures. RTOs, Independent SystemOperators, and tight Power Pools should ensure bid-ding processes and cost caps provide appropriateprice signals to generation units capable of fuelswitching. FERC should ensure that wholesalepower markets, containing any capacity compo-nents, should have market rules facilitating pricingof alternate fuel capability.

Additional Demand Considerations

There are additional actions and policy initiativesthat could be undertaken to create a more flexible andefficient consumer environment for natural gas, whileassuring environmental goals are achieved.

� Permit Reviews. State environmental agencies,in consultation with the U.S. EnvironmentalProtection Agency, should review existing alternatefuel permits, and opportunities for peak-load reduc-tion during non-ozone season. All new permitsshould have maximum flexibility to use alternatefuels during all seasons, recognizing the ozone sea-son may require some additional limitations.During ozone season, cap and trade systems shouldgovern the economic choices regarding fuel choiceto the maximum extent possible.

� Forums to Address Siting Obstacles. With respectto coordination among multiple levels of govern-ment, federal agencies should consider facilitatingforums to address obstacles to constructing newpower generation and industrial capacity. Partici-pants would include the relevant federal, state, andlocal siting authorities, as well as plant developersand operators, industrial consumers, environmentalnon-governmental organizations, fuel suppliers, andthe public. The objective of these forums would be toaddress with stakeholders the impact of siting deci-sions on natural gas markets.

� Potential Limits on Carbon Dioxide Emissions.Ongoing policy debates include discussion of car-bon reduction, including potential curbs on CO2

emissions. Many actions would constitute the market’s response to such limitations, including shutdown and/or re-configuration of industrialprocesses, additional emissions controls includingcarbon sequestration, or the shifting of manufactur-ing to other countries.

Natural gas has lower CO2 emissions than other carbon-based fuels. Therefore, natural gas combus-tion technologies are likely to be a substantial aspect ofthe market’s response to limitations on CO2 emissionsin industrial processes and power generation. Themost significant impact of CO2 emission curbs wouldlikely be restrictions in operation of much of the coal-fired power generation, since coal-combustionprocesses tend to emit the highest levels of CO2.Depending on the level of emission restrictions, therequirements for natural gas in power generationalone could increase substantially. Alternatives to nat-ural gas would be additional nuclear power and/orcoal-fired generation employing carbon sequestrationtechnologies that are unproven on a large scale.Renewable electric generation capacity is likely to playa growing role in the future, but has not demonstrat-ed the ability to have a large impact.

This study tested the impacts on natural gas demandand the resulting market prices, by performing sensi-tivity analyses; the impact on gas demand could besignificant, as discussed elsewhere in this study,depending on the degree to which carbon intensitymight be reduced. Natural gas consumption forpower generation would clearly increase under anyCO2 reduction scheme during the time frame of thisstudy, placing enormous demand pressure on naturalgas. This would likely lead to much higher natural gasprices and industrial demand destruction.

� DOE Research. With respect to governmentresearch, the NPC is supportive of DOE researchwhere it complements privately funded researchefforts. DOE and state energy offices should contin-ue to support research and commercialization ofwind, solar, biomass, and other renewable genera-tion technologies. DOE should continue to supportgovernment and industry partnership in fundingimprovements such as advanced turbines, clean coal,carbon sequestration, distributed generation, andrenewable technologies. DOE should also continueto support the efficient use of natural gas.


CHAPTER 4 - NATURAL GAS SUPPLY 105

This chapter of the Integrated Reportdescribes the methods used by the SupplyTask Group to develop an outlook for natu-

ral gas supplies and discusses the results of thestudy effort. These results provide the basis for thesupply elements of the Summary of Findings andRecommendations. What follows is a summary ofthe supply outlook, along with additional detailsfrom each of the functional subgroups. Full docu-mentation is found in the Supply Task GroupReport and its appendices.

Study Approach

In undertaking its analysis of natural gas supply, theSupply Task Group considered the most importantfactors affecting the current supply situation and thelong-range outlook. This analysis included the fol-lowing:

� A comprehensive review of the North American gasresource base using the best publicly available data.This assessment included a thorough review of bothconventional and nonconventional resources(including tight gas, coal bed methane, and shalegas). In order to gain a solid understanding ofpotentially commercial recoverable resources, thereview also included a detailed assessment of drillingand development costs, and the likely number andsize of future discoveries.

� A comprehensive review of the production perform-ance history for the mature basins of NorthAmerica. This was needed in order to gain anunderstanding of the future production decline ratesof existing reserves, the likely response to future

drilling, and the potential for growth in provedreserves from revisions and extensions to existingfields.

� An evaluation of the effect of the permitting processand access restrictions on development of indige-nous resources.

� An assessment of the effect that technology advancesmight have on the cost and availability of gasresources.

� An assessment of the potential contribution frommajor new supply sources, such as imported lique-fied natural gas (LNG) and Arctic gas.

The Supply Task Group had five subgroups. TheResource Subgroup was led by ExxonMobil,Technology by ChevronTexaco, Environmental/Regulatory/Access by Burlington Resources, LNG byShell, and the Arctic Subgroup was led jointly byExxonMobil, ConocoPhillips, and BP. Given thebreadth of the resource work, the Resource Subgroupwas further subdivided into conventional and noncon-ventional resource groups; the latter was led byAnadarko. The members of the Supply Task Group,whose names are listed in Appendix B, oversaw all ofthe subgroups.

Based on advice of participants from prior NPCstudies, high priority was given to timely completionof the resource and cost estimating work. TheResource Subgroup set out to complete the resourcereview before the end of 2002. They also concludedthat the most efficient way to access industry expertsin key North American geologic plays was to hold aseries of workshops across the country, inviting the

NATURAL GAS SUPPLYCHAPTER 4

contribution of as broad a group as possible. Industryworkshops were held in New Orleans, Denver, MenloPark, Houston, Calgary, and Reston. In some cases,follow-up workshops were held to reconfirm or mod-ify assessments in light of subsequent model projec-tions of resource development.

The Resource Subgroup further decided that pub-licly available data from the U.S. Geological Survey, theMinerals Management Service, and the Canadian GasPotential Committee were the best starting points foran industry review of the resource base. Cooperationby each of these organizations was outstanding. Inthese workshops, each agency was asked to describe forthe group their detailed, play-by-play, resource assess-ment. This discussion then generated debate and com-ment from industry experts. In the course of thisdiscussion, consensus emerged regarding any signifi-cant modifications that the group felt appropriate forthe NPC study. The intent of this work was not tojudge an assessment as “right” or “wrong,” but rather todevelop a “best estimate” that industry could supportfor modeling purposes. At the same time, key costdrivers, access issues, and technology factors were dis-cussed. All of this information was carefully docu-mented for future use by the appropriate subgroups.

This process facilitated an excellent exchangebetween industry and government on natural gasresource assessment. Some important lessons werealso learned that will lead to better industry and gov-ernment assessments in the future. The process wasessentially complete by year-end 2002. This process isdescribed further in this chapter, with additionaldetails contained in the Resource section of the SupplyTask Group Report.

The assessment of technically recoverable resourceand cost was an important part of the study, but justas important was the assessment of future produc-tion performance based upon an analysis of produc-tion history. For each significant producing basin inthe United States and Canada, this analysis includedthe initial production rates, decline rates, and expect-ed reserve recoveries from all gas wells drilled in thepast ten years. This information was essential forassessing the production trends of proved reservesand the likely effect of future drilling on the produc-tion outlook.

One reason for this interest in production perform-ance was the much-questioned supply response to sig-

nificantly increased drilling for gas in 2000-2001.These data were used to reconcile the supply responseto the drilling activity undertaken. Results of this workare described later in this chapter.

The ability to access resources and obtain timelypermits is also a critical factor in determining thefuture contribution of indigenous resources. TheEnvironmental/Regulatory/Access Subgroup deter-mined early on that their evaluation of this issue need-ed to go beyond “stipulations” contained in oil and gasleases, to the “conditions of approval” that accompanythe development of those leases. A team of expertsdeveloped a model of how those conditions impact gasdrilling and development. Those results are alsodescribed later in this chapter.

Similarly, the effects of technology on future supplydevelopment can be significant. The TechnologySubgroup chose a workshop process similar to theResource Subgroup to assess how new technologymight help reduce costs and increase recoveries. Manyareas of technology were evaluated, including subsur-face imaging, drilling and development costs, comple-tions, coal bed methane, deepwater developments, andnatural gas hydrates. The projected effects of technol-ogy on future gas recovery are significant anddescribed later in this report.

Finally, it was clear that a good assessment of thepotential contributions to supply of new, large, longlead-time resources was needed. The LNG and ArcticSubgroups undertook this task. Their work included acomprehensive review of worldwide gas resource avail-ability, an examination of resource development andliquefaction capability together with an assessment ofshipping requirements, and regasification needs.Similarly, the potential for major new pipelines tobring Arctic gas to North American markets wasreviewed and assessed. These analyses represent thefirst comprehensive work by the NPC on these newsources. A summary of the results is presented in thisreport.

This chapter will present the supply outlookdeveloped by the NPC and a summary of theresults from each of the subgroups. In addition,the NPC supply outlook will be compared to otherpublic forecasts and the outlook from the NPC1999 study.

CHAPTER 4 - NATURAL GAS SUPPLY106

Summary of Supply Outlook

The primary charge of the Supply Task Group was todevelop an outlook for natural gas supplies that couldmeet U.S. demand through the year 2025. To assist inthe quantification of this outlook, Energy andEnvironmental Analysis, Inc. (EEA) was contractedand their Hydrocarbon Supply Model (HSM)employed for the study. The HSM is a computer modelthat provides a rigorous and consistent framework foranalyzing and forecasting natural gas, crude oil, andnatural gas liquids supply and cost trends in NorthAmerica. EEA and the HSM were used in the 1992 and1999 NPC studies, although the model was enhancedfor this study as described in the Supply Task GroupReport.

The HSM used input from each of the supply sub-groups to calculate the costs of developing new sup-plies for each of the resource regions. The model thendetermined which new supply sources to develop, inorder of lowest delivered cost, until demand was metthrough 2025. The cost of the last increment of supply,together with the respective transportation cost, estab-lished the end-use price for the gas. The supply out-look that follows is the result of this process.

In addition to the EEA modeling, an NPC groupbegan work on a model based on a license from AltosManagement Partners. This effort was designed tosupplement the efforts of EEA and to provide a con-trasting modeling approach. Additional details on themodeling process are provided later in this report.

Overall Supply Outlook

The total supply outlook for the Reactive Path sce-nario is illustrated in Figure 4-1. Overall, total suppliesgrow at an annual rate of 1% through 2025 to keeppace with North American demand for natural gas.During this period, production from the U.S. lower-48and Canada is projected to remain relatively constant,with the overall growth being met from new Arctic andLNG gas supplies.

This outlook formed the basis for the first supply-related finding:


Finding: Traditional North Americanproducing areas will provide 75% of long-term U.S. gas needs, but will be unable tomeet projected demand.

0

5

10

15

20

25

30

35

1990 1995 2000 2005 2010 2015 2020 2025

OTHER LOWER-48

NON-ARCTIC CANADA

ALASKA

LNG


ROCKIES

MACKENZIE

DELTA

GULF OF MEXICO DEEPWATER

TR

ILLIO

N C

UB

IC F

EE

T

YEAR

Figure 4-1. Components of North American Supply

The material to follow will provide details on theanalysis that supports this finding.

The largest component of North American supplyremains traditional supply sources in the UnitedStates and Canada. These sources supply almost100% of U.S. and Canadian demand today, and areexpected to still represent over 75% of demand in2025, as production levels are maintained in therobust price environment contemplated by theReactive Path scenario. In the U.S. lower-48, growthin the Rockies and deepwater Gulf of Mexico is off-setting declines in virtually all other U.S. maturebasins. In Canada, production from the WesternCanada Sedimentary Basin is also expected to declinethroughout the study period, with some increasefrom offshore Eastern Canada.

The Arctic supply results from new pipelines fromMackenzie Delta and Alaska. Mackenzie Delta isassumed to come on stream at 1 billion cubic feet perday (BCF/D) in 2009 and expanded to 1.5 BCF/D in2015. The outlook for Alaska is for production to startin 2013, reaching 4 BCF/D capacity in 2014. In total,the new Arctic supplies will contribute about 8% ofNorth American supply by 2025.

Small volumes of LNG are currently imported intothe United States, providing 1% of North Americansupply. The outlook in the Reactive Path scenario is forLNG imports to grow to 12.5 BCF/D by 2025, provid-ing 12% of North American supply. This assumes thatthe 4 existing U.S. regasification terminals are fully uti-lized by 2007, and that 7 new terminals will be built. Inaddition, 7 of these 11 terminals will be expanded.

Although it is based upon the best available data andexpertise, the overall supply outlook has inherentuncertainty. A number of factors can significantlyaffect production forecasts from North Americanindigenous sources. For example, there is uncertaintyin the estimated size of the undiscovered resource base.In addition, the predicted rate at which technology willgrow and contribute to production is uncertain. Andfinally, the commercial factors that will influencefuture drilling activity and subsequent production arehighly variable.

In order to quantify uncertainty in the factors influ-encing future production, model sensitivities were runto bracket the range of possible outcomes. These aredescribed later in this chapter.

A key issue in this projection of future production isthat North America has never experienced a sustainedprice environment like the one anticipated by theReactive Path scenario. In this new environment, theuse of past experiences to project the future will be lessreliable. Econometric models can help to describeplausible, internally consistent futures; but of them-selves, they do not create more reliable predictionsthan the experts who provide data for the models.

In the Reactive Path supply outlook, the volumes ofArctic gas and LNG were fixed based on capacity andstart-up timing described above. The lower-48 andCanadian production outlook was the result of model-ing supply/demand balance. The overall size of theresource base plays a central role in determining themodel’s output, but there are several other key consid-erations. These include the costs to develop theresource, the production characteristics of theresource, technology improvements that lower costsand improve recovery, assumptions regarding access tothe resource base, drilling activity levels, and reservedevelopment pace. A discussion of these items follows.

Resource Base

The technical resource base identifies that volume ofnatural gas that is technically recoverable withoutregard for costs or price. The technical resource basewas determined through an extensive workshopprocess involving a wide cross-section of industryexperts. These workshops covered 72 regions of theUnited States, Canada, and Mexico, which were consol-idated into 17 “super-regions” as shown in Figure 4-2.

The NPC’s assessment of the most likely technicallyrecoverable resource volume for North America thatresulted from that process are annotated on the map,with additional details shown in Table 4-1.

The resource base is described for proved reservesthat can be produced from existing wells, growth toproved reserves in existing fields, and undiscoveredconventional and nonconventional (tight gas, coal bedmethane, and shale gas) potential.

The assumptions of technology advancements play akey role in this assessment. Improvements in recovery,exploration tools, and other areas will increase theresource base as shown in the Table 4-2. The technicalresource is shown for technology advances through2015 for comparison to the 1999 NPC study, as well astechnology advances through to 2030. Details of the



ALASKA303

ARCTICCANADA

71

CANADAATLANTIC

85

GULF OF MEXICO329

ROCKIES284

MID-CONTINENT

88

EASTERNINTERIOR

110

GULF COAST ONSHORE183

WESTERN CANADA SEDIMENTARY BASIN 224

GREATBASIN

5

MEXICO 121

WESTCOAST

ONSHORE29

U.S.PACIFIC

OFFSHORE22

WESTTEXAS

64

BRITISHCOLUMBIA

11

EASTERN CANADA6

U.S. ATLANTICOFFSHORE33

Figure 4-2. Super-Region Technical Resources (Trillion Cubic Feet)

basis for the technology improvement factors are pro-vided in the Technology Improvements section of thischapter.

The technical resource base is compared to previousNPC studies and the USGS/MMS assessments inFigure 4-3. This comparison is based on advancedtechnology through 2015. For the U.S. lower-48, thetechnical resource of 1,250 trillion cubic feet (TCF) is14% (210 TCF) lower than the 1999 study. Half of thisreduction is in the growth to proved reserves category.

Similar methodologies were employed to estimate thevolume of growth resource, but lower expected recov-eries were used in this study in line with recent experi-ence. The reduction from the USGS/MMS referenceassessment is primarily from lower nonconventionalpotential.

The overall North American assessment is also lowerthan the 1999 and 1992 NPC studies (see Figure 4-4).The Canadian assessment is lower than the 1999 studybased on lower potential assessed for the Arctic region


ProvedReserves

(as ofDec. 2001)

Growth*to

ProvedReserves

UndiscoveredConventional

Potential

UndiscoveredNon-

conventionalPotential

TotalTechnicalResource

Lower-48Onshore

145 148 189 282 764

Lower-48Offshore

30 57 298 0 384

Alaska 9 36 201 57 303

United States 184 241 687 339 1,451

Canada 60 69 219 50 397

Mexico 28 22 70 0 121

North America 272 332 976 389 1,969

*Includes 55 TCF of discovered non-proved.

Table 4-1. North American Technical Resource Base – Current Technology (Trillion Cubic Feet)

CurrentTechnology

2015Technology

2030Technology

Lower-48 Onshore 764 839 1006

Lower-48 Offshore 384 415 486

Alaska 303 331 395

United States 1,451 1,585 1,887

Canada 397 420 475

Mexico 121 130 147

North America 1,969 2,135 2,508

Table 4-2. Impact of Technology Improvement on the Resource Base (Trillion Cubic Feet)

and a lower nonconventional assessment for theWestern Canada Sedimentary Basin. There was noMexican assessment made in the 1999 study, but thecurrent assessment is about half of the 1992 study as aresult of lower assessed undiscovered potential and alower current level of proved reserves.

Given the uncertainty associated with estimating thetechnical resource base, low and high cases were devel-oped to define the range. The low end of the resourcerange is defined as P90, which means there is a 90%probability of that volume actually being found. Thehigh end is defined as P10, which means there is only a10% chance of that volume being found. The P10 andP90 values bracket the mean, which represents the bestestimate of resource volume. The NPC defined the P10as 135% of the mean and the P90 as 70% of the mean.This resource range was used to evaluate the impact onthe production outlook and will be discussed furtherin the sensitivity section.

Production Performance

A key aspect of the NPC resource study was an eval-uation of the production performance of existing U.S.and Canadian basins from 1990 to the present. Thisanalysis looked at the performance of individual gas


0

500

1,000

1,500

2,000

2,500

3,000

1992

NPC

1999

NPC

2003

NPC

LOWER-48 CANADA

ALASKAMEXICO

TR

ILLIO

N C

UB

IC F

EE

T

Figure 4-4. North American Mean Assessment1999 Base, Advanced Technology

0

200

400

600

800

1,000

1,200

1,400

1,600

1992 NPC 1999 NPC USGS/MMS 2003 NPC

NONCONVENTIONALNEW FIELDGROWTHPROVED

TR

ILLI

ON

CU

BIC

FE

ET

Figure 4-3. Lower-48 Mean Assessment – 1999 Base, Advanced Technology

wells drilled over the period, determining initial pro-duction rates, initial decline rates, and expected recov-eries per well. Also evaluated were base productiondecline trends and the production response toincreased drilling activity. The results of this analysiswere used to estimate future well performance param-eters and calibrate the HSM model results.

Some of the key observations from the analysis areas follows:

� While North American production grew 11 BCF/D(1.8% per year) between 1990 and 2002, growthslowed dramatically after 1996, as shown in Figure4-5. Growth in the lower-48 Rocky Mountains,

deepwater Gulf of Mexico, and more recently, EastTexas/North Louisiana has been offset by produc-tion losses in the other regions, particularly the Gulfof Mexico shelf and Midcontinent. Canadian pro-duction growth, which comprised 65% of the totalgrowth since 1990, slowed dramatically and began todecline, even as the number of Canadian gas wellcompletions more than tripled.

� Conventional gas production in the U.S. lower-48has been declining since 1990 and nonconventionalproduction (tight gas, coal bed methane, and shalegas) has doubled from 12% to 25% of production, asshown in Figure 4-6. Aside from the deepwater Gulfof Mexico, the only U.S. basins maintaining sustain-


EASTERN CANADA

0

10

20

30

40

50

60

70

80

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

HISTORICAL PROJ.

Source: Energy and Environmental Analysis, Inc., Gas Supply Review (GSR).

YEAR

OTHER

MIDCONTINENT

PERMIAN BASIN

E.TEXAS/N. LOUISIANA

EASTERN GULF COAST

S. TEXAS GULF COAST

GULF OF MEXICO

SHELF

ROCKIES

WESTERN CANADA

DEEPWATER

GULF OF MEXICO

SEDIMENTARY BASIN

Figure 4-5. U.S. Lower-48 and Canadian Production by Region

able production increases (Rockies, EastTexas/North Louisiana) have being driven byincreased nonconventional production.

� Average estimated ultimate recovery (EUR) per gasconnection in the U.S. lower-48, excluding noncon-ventional and the deepwater Gulf of Mexico, fell15% between 1990 and 1999 as the resource basematured and technology gains and higher pricesmade smaller prospects economic. As drillingramped up in response to the 2000-2001 price spike,average EUR fell a further 18%, as more marginalwells were drilled. Western Canadian average EURhas fallen dramatically as the basin has matured andthe industry concentrated on lower-risk, shallow-depth drilling (see Figure 4-7).

� Initial Production Rates (IPs) increased markedlythrough the early to mid-1990s, helping the industryto maintain production rates, as the industryemployed technology to accelerate production andimprove drilling economics. Increases in IPs flat-tened in the latter part of the 1990s as per-wellreserves fell and fracture technology implementa-tion neared saturation level in most basins.

Declining EURs and increasing IPs have resulted insteepening initial well decline rates.

� As more and more high decline wells have beenadded to base production, base decline rates havesteadily risen. Figure 4-8 shows that the decline rateof lower-48 base production has increased to over25%, from just over 15% in the early 1990s. Just tomaintain production levels requires first year pro-duction from new wells of 12-13 BCF/D, up from 8 BCF/D in 1992. Western Canada has shown a sim-ilar increase in base decline.

� Industry responded aggressively to the 2000-2001price spike, with the gas rig count climbing to an all-time high over 1,050 as shown in Figure 4-9. Theincremental activity yielded a limited productionresponse as: (1) the resource base continued tomature, (2) additional drilling yielded very low mar-ginal results, (3) much of the incremental activityoccurred in low rate regions, (4) gains from comple-tion/stimulation technology slowed, (5) base declinerates continued to increase, (6) higher gas pricesmade it possible to drill lower quality prospects, and(7) rig efficiency declined.


0

10

20

30

40

50

60

70

1990 1992 1994 1996 1998 2000 2002

WE

T G

AS

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

CONVENTIONAL

COAL BED METHANE

SHALE

TIGHT*

* Tight gas estimated 2000-2002.

YEAR

Figure 4-6. Lower-48 Wet Gas Production by Resource Type

Sources: GTI Gas Resource Database; Energy and Environmental Analysis, Inc.


Figure 4-7. Recovery per Gas Connection

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

1990 1994 2000

U.S. LOWER-48 (EXCLUDING NONCONVENTIONAL AND DEEPWATER GULF OF MEXICO)

WESTERN CANADA

BIL

LIO

N C

UB

IC F

EE

T

YEAR199819961992 2002

Figure 4-8. Lower-48 Base Production Decline Rates

-30

-25

-20

-15

-10

-5

0

1992 1994 1996 1998 2000

RA

TE

OF

DE

CLIN

E (

PE

RC

EN

T)

1993 1995 1997 1999 2001

YEAR


Source: IHS Energy Group.

Development Costs and Technology Improvements

The costs to develop the technical resource, includ-ing exploration and development drilling, productionfacilities, and operating and maintenance, were devel-oped and provided as input to the econometric model-ing effort. Public and commercial databases, fromsources such as American Petroleum Institute andEnergy Information Administration, were used as thereference for estimating the costs. The costs were dis-cussed at the regional resource workshops and bench-marked against the experiences of various industryrepresentatives. Costs were developed for each of themain resource regions and for different depth intervalsto match the granularity of the resource description.Overall, the estimated costs were similar to the 1999study, with the primary difference being higher drillingcosts for deeper reservoirs.

In addition to developing detailed costs, the rate ofdrilling rig attrition was evaluated and an estimatemade of future rig availability. Industry input on thisevaluation led to a significantly lower rate of attritionthan assumed in the 1999 study. The view was that ina robust price environment significant efforts would be

expected to keep rigs working. Ultimately, the rig attri-tion is used to determine the number of new rigs, andthe corresponding costs, required to support the pro-jected drilling activity levels.

As was discussed earlier, assumptions of improvingtechnology increase the technical resource base assess-ment by 8 to 27%. In addition, the technologyimprovement factors developed by the TechnologySubgroup also enhance the ability to commercialize thetechnical resource. Factors such as drilling and com-pletion costs and recovery per well are estimated toimprove with time, thus increasing the volume of com-mercial resource. As a result of the technologyimprovements assumed in the Reactive Path scenario,the overall level of gas production in 2025 is 14% high-er than if no improvements were included.

Additional details on cost development and technol-ogy improvement can be found later in this chapter,with detailed study results contained in the SupplyTask Group Report.

Commercial Resource

While the technical resource base represents thepotential available natural gas resource, calculating


Figure 4-9. Monthly Lower-48 Gas Production

0

10

20

30

40

50

60

0

200

400

600

800

1,000

1,200

GAS RATE

RIGS

AV

ER

AG

E D

AIL

Y V

OLU

ME

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

NU

MB

ER

OF

AC

TIV

E G

AS

DR

ILLIN

G R

IGS

YEAR

Sources: Energy and Environmental Analysis, Inc., Gas Supply Review (GSR); and Baker Hughes.

1990 1994 2000199819961992 2002

how much of that resource can be commercial and atwhat price will ultimately determine the volume ofnatural gas that can be developed. Utilizing thedrilling, development, and operating costs estimates,the technology improvement factors, and the produc-tion performance parameters, the model calculates thevolume of natural gas that can be commercially devel-oped at any given price. Figure 4-10 shows the volumeof lower-48 resource that is commercial at three differ-ent price thresholds. This shows that 760 TCF, or 60%of the total technical resource (advanced technology),can be commercial at a $4.00 per million Btu (MMBtu)price (wellhead price, 2002 dollars).

This relationship of commercial to technicalresource can also be illustrated with a cost-of-supplycurve as shown in Figure 4-11. This plot shows the vol-ume of lower-48 natural gas that is commercial todevelop as a function of price and for the range ofassessed technical resource. Proved reserves areexcluded from these curves since they are the same foreach of the resource curves.

The curve for the mean assessment shows that at aprice of $4/MMBtu, 585 TCF, or 55% of the unproventechnical resource base, can ultimately be commer-cially produced. This compares to the 760 TCF shown

in Figure 4-10 that includes proved reserves. Thecurves for the P10 and P90 resource base represent theprobabilistic range of the lower-48 resource assessed bythe NPC. It should be noted that while these commer-cial volumes may appear large in relation to currentU.S. annual production (18 TCF), it will take manydecades for these resources to ultimately be produced.The vast majority of this resource base will be pro-duced from fields that have yet to be found and fromwells that have yet to be drilled.

Similar curves were developed for each of the pro-ducing regions. The commercial potential varies sig-nificantly by region, as illustrated in Figure 4-12.Factors such as resource quality, undiscovered field sizedistributions, and development costs determine theshape of the supply curve. For example, theMidcontinent and Gulf Coast regions tend to havehigher quality reservoirs that allow for a relatively highpercentage of technical resource to be commercial. Onthe other hand, the Rocky Mountain and EasternInterior areas, are dominated by nonconventionalresources, which are poorer quality and thus highercost supplies.

The volumes of technical resource that can be com-mercial will also vary significantly by resource type as


Figure 4-10. Lower-48 Technical Resource vs. Commercial Resource

0

200

400

600

800

1,000

1,200

1,400

2003 NPC 3 4 5

NONCONVENTIONAL

NEW FIELD

GROWTH

PROVED

COMMERCIALTECHNICAL

DOLLARS PER MILLION BTU (2002$)

TR

ILLIO

N C

UB

IC F

EE

T


Figure 4-11. Cost-of-Supply Curves for Lower-48 Resource Range

0

200

400

600

800

1,000

1,200

1,400

1,600

2 4 6 8

ECONOMIC THRESHOLD (DOLLARS PER MILLION BTU)

UN

PR

OV

EN

RE

SO

UR

CE

(T

RIL

LIO

N C

UB

IC F

EE

T)

TOTAL HIGH

RESOURCE

TOTAL MEAN

RESOURCE

TOTAL LOW RESOURCE

HIGH RESOURCE (P10)

MEAN RESOURCE

LOW RESOURCE (P90)

3 5 7

Figure 4-12. Cost-of-Supply Curves by Lower-48 Region

0

20

40

60

80

100

2 3 4 5 6 7 8

ROCKIES

GULF OF MEXICO

EASTERN INTERIOR

TOTAL GAS

GULF COAST

MIDCONTINENT

UN

PR

OV

EN

RE

SO

UR

CE

S (

PE

RC

EN

T)

DOLLARS PER MILLION BTU (2002$)

shown in Figure 4-13. A larger percentage of growthtechnical resource can be commercial than new fieldsor nonconventional technical resources. This is a resultof the growth resource requiring primarily lower riskdrilling in producing fields to develop and thus havinga lower cost.

Following supply cost development, the economet-ric model was used to project future U.S. and Canadianproduction required to meet expected demand. Thatproduction outlook will be discussed next.

North American Production Outlook

Future North American natural gas supply can becharacterized by three components. The first is pro-duction from traditional areas in the U.S. lower-48,Canada, and Mexico. The second comprises the Arcticareas of Alaska and Canada. And the third source ofsupply is imported LNG. Arctic and LNG supplies arediscussed in a subsequent section of this chapter.

Mexico’s current production is less than 2 TCF/year,but is expected to grow significantly. However,demand is projected to outpace supply, so that Mexicowill remain a net natural gas importer through 2025.Future supply and demand were not modeled in detail

for Mexico, although the net supply/demand balancewas incorporated into the econometric model and isdiscussed in more detail in Chapter Six.

This section reviews the production outlook for theU.S. lower-48 and non-Arctic Canada in the context ofthe Reactive Path scenario. For supply, this includesthe Mean resource assessment and incorporates theexpected technology improvement factors. Access toindigenous resources is assumed to be unchanged, withthe existing Outer Continental Shelf (OCS) moratorianot lifted and leasing and permitting restrictions in theRocky Mountain region continued, although the per-mitting process is assumed to improve to allow a con-tinued high level of drilling activity. The resultingproduction outlook for the United States and Canada(excluding new Arctic developments) is shown inFigure 4-14.

North American gas production grew rapidly in theearly-1990s, as the industry was deregulated. Growthrates slowed considerably in the mid-1990s throughthe early-2000s, as excess productive capacity wasgradually eroded. Production peaked in 2001 follow-ing major drilling ramp-ups in the United States andCanada, and have since declined.


Figure 4-13. Cost-of-Supply Curves by Resource Type

0

20

40

60

80

100

GROWTH NEW FIELDS

NONCONVENTIONALTOTAL GAS

UN

PR

OV

EN

RE

SO

UR

CE

S (

PE

RC

EN

T)

2 3 4 5

U.S. DOLLARS

6 7 8

Looking ahead, in the absence of new Arctic sources,the outlook is for generally flat production through2025 assuming a robust price environment.Individually, both the U.S. lower-48 and non-ArcticCanada are expected to have flat production levels, asthe traditional producing areas in the U.S. lower-48and Canada continue to mature.

Production is continuing to shift from decliningconventional gas production to nonconventionalsources (tight gas, coal bed methane, and shale gas) asshown in Figure 4-15. The ability to continue growingnonconventional production will be critical to sustain-ing production levels.

The next two sections contain more details of the U.S.lower-48 and non-Arctic Canada production outlooks.

U.S. Lower-48 Production Outlook

Looking forward, the ability to maintain the pace ofnew drilling and development activity will play a criti-cal role in sustaining gas supplies. Declines from exist-ing reserves have gradually become steeper, withcurrent base decline rates of over 25% in the first year.Production from existing wells will drop by over 50%

from 2000 to 2005. In order to offset these declines,new wells will be required to develop additionalresources in the growth, undiscovered conventional,and undiscovered nonconventional categories, asshown in Figure 4-16.

Historically, through improvements in technologyand effective development programs, industry has suc-cessfully increased recovery from producing fields and“grown” the reserves that are ultimately produced. AsFigure 4-16 illustrates, these expected resources repre-sent more reserves than are currently categorized asproved by the industry. Growth remains a large andimportant low-cost wedge of future reserves and pro-duction.

Production from undiscovered conventional fieldsrepresents the single largest source of new supply in theNPC outlook. In a tight market environment, signifi-cant exploration will occur for resources undiscoveredtoday. Although these future resources will be increas-ingly small, deep, and of poorer quality, many will becommercially viable in a higher price environment.

The growing contribution to supply from noncon-ventional resources is projected to offset the production


Figure 4-14. U.S. Lower-48 and Non-Arctic Canada Production Outlook

0

5

10

15

20

25

30

1990 1995 2000 2005

YEAR

2010 2015 2020 2025

TR

ILLIO

N C

UB

IC F

EE

T NON-ARCTIC CANADA

U.S. LOWER-48


Figure 4-15. U.S. Lower-48 and Non-Arctic Canada Gas Production by Type

0

5

10

15

20

25

30

2000 2005 2010 2015 2020 2025

NONCONVENTIONAL

CONVENTIONAL

ASSOCIATED

TR

ILLIO

N C

UB

IC F

EE

T

YEAR

Figure 4-16. Lower-48 Production by Resource Category

0

5

10

15

20

25

2000 2005 2010 2015 2020 2025

TR

ILLIO

N C

UB

IC F

EE

T

PROVED

GROWTH

UNDISCOVERED NONCONVENTIONAL

FUTURE DRILLING REQUIRED

UNDISCOVERED CONVENTIONAL

YEAR

decline from conventional sources (proved, growth, andundiscovered conventional in Figure 4-16) in a robustprice environment. The increase in production fromthis segment reflects access to and development of largenonconventional resources, particularly in the RockyMountain region.

Lower-48 production was also evaluated by majorsupply regions. Figure 4-17 shows historical produc-tion from 1990 and the projections for the ReactivePath scenario though 2025. The large producingregions, Gulf of Mexico shelf, Midcontinent, PermianBasin, South Texas, and East Texas are all projected toexperience production declines. These regions arebecoming increasingly mature. Future drilling, whileexpected to continue at high levels, will be targetingsmaller reservoirs, with lower initial production ratesand lower reserves per well.

Offsetting this decline will be increasing productionfrom nonconventional resources and deepwater Gulfof Mexico. Nonconventional resources represent over35% of the undiscovered potential and technologyadvances and the robust price outlook in the ReactivePath scenario make more of this resource commercialto develop. The Rocky Mountain region contains themajority of the nonconventional resource and produc-

tion is projected to grow by 50% by 2020. Increasingnonconventional production is also projected for theEastern Interior.

Production from the deepwater Gulf of Mexicomade this area the fastest growing region during the1990s. This increase is expected to continue, althoughat an overall slower pace. It should be noted that theareas of projected production growth are less maturethan the declining regions and therefore have a greateruncertainty in resource base size.

The supply outlook developed is an aggregate of indi-vidual supply regions and individual wells within thoseregions, all of which are required to meet expecteddemand for natural gas. The Reactive Path scenarioestablishes a supply/demand balance and a price for nat-ural gas required to achieve that balance. While thisprice level ultimately determines the volume of gas thatcan be commercially developed, an assessment was alsomade of the supply volumes that would be developed atvarying prices. Figure 4-18 depicts the volume of lower-48 supply that would be developed at three differentprice levels. For example, if the price outlook for natu-ral gas was a constant $3/MMBtu, lower-48 productionis projected to continually decline through 2025. This isa result of insufficient supply being available at that


Figure 4-17. Lower-48 Production by Region

0

5

10

15

20

25

1990 1995 2000 2005

YEAR

2010 2015 2020 2025

OTHER

GULF COAST ONSHORE

MIDCONTINENT/PERMIAN BASIN

EASTERN INTERIOR

ROCKIES

GULF OF MEXICO

DEEPWATERGULF OF MEXICO SHELF

TR

ILLIO

N C

UB

IC F

EE

T

price level to sustain production. At a price level around$5/MMBtu, the model projects additional supplies to becommercially viable and overall production levels wouldremain relatively flat.

The lower-48 production outlook in the ReactivePath scenario is lower than projections from the 1999NPC study and the government’s preliminary EIA 2004Annual Energy Outlook. Figure 4-19 compares thethree production outlooks. It should be noted that thislower production outlook is with a higher price envi-ronment than either of the other projections. A moredetailed reconciliation of the differences in these out-looks is included at the end of this chapter.

Canadian Production Outlook

Production from the Western Canada SedimentaryBasin was one of the key contributors to the growth ofNorth American production in the 1990s, providingover 50% of total North American growth. Like muchof the U.S. lower-48, however, growth rates in WesternCanada have been rapidly flattening as the basin hasmatured, even as activity rates have been increasing.Production from Western Canada is no longer pro-jected to continue to rise.

Production from offshore Eastern Canada began in2000, and the outlook is for moderate growth fromanticipated future discoveries. Figure 4-20 shows theoutlook for Canadian regional production, excludingthe Mackenzie Delta, which is included in the Arcticsupply region.

While the production outlook for Western Canadaoverall is declining, production from coal bed methaneand shale gas is expected to rise and partially offset thefall in conventional output. Figure 4-21 breaks out thenonconventional components of the WesternCanadian outlook. The potential opening of offshoreBritish Columbia has not been incorporated into theCanadian projection.

In summary, the production outlook for theUnited States and Canada is characterized by declin-ing production from maturing supply regions that isoffset by growing nonconventional production andnew offshore developments. This flat outlook isfully dependent on the ability of future drilling pro-grams to find and develop new reserves to offset rap-idly declining base production, and a robust priceenvironment to support the high drilling activitylevel.


Figure 4-18. Lower-48 Production at Different Prices

0

10

15

20T

RIL

LIO

N C

UB

IC F

EE

T

$5 FIXED PRICE

$4 FIXED PRICE

$3 FIXED PRICE

2000 2005 2010 2015 2020 2025

YEAR


Figure 4-19. Lower-48 Production Outlooks

0

15

20

25

TR

ILLIO

N C

UB

IC F

EE

T30

NPC 1999 ($2.50 - $3.50)

EIA PRELIM. 2004 OUTLOOK ($4.00 - $4.50)

NPC 2003 - REACTIVE PATH ($5.00 - $7.00)

2000 2005 2010 2015 2020 2025

YEAR

Figure 4-20. Canadian Regional Production (Excluding Mackenzie Delta)

0

1

2

3

4

5

6

7

8

1990 1995 2000 2005 2010 2015 2020 2025

TR

ILLIO

N C

UB

IC F

EE

T

WESTERN

EASTERN

YEAR

Note: Prices are in dollars per million Btu (Henry Hub, 2002 dollars).

Drilling Activity

The future outlook for production in the UnitedStates and Canada is critically dependent on the main-tenance of a very active drilling program. Historically,even as production growth has flattened in the UnitedStates and Canada, the number of gas wells annuallyrequired to maintain this production has increaseddramatically (Figure 4-22). In the United States, thenumber of gas wells has increased from less than10,000 per year in the early 1990s to nearly 17,000 in2001. In Canada, the increase has been even more dra-matic, increasing from less than 2,000 in the early1990s to over 10,000 in 2001.

This outlook projects that drilling activity levels willremain at near record highs throughout the studyperiod. Industry “reality-check” workshops were heldto confirm that higher activity levels were plausible inthe basins where the model predicted an increase.

Figures 4-23 and 4-24 show a breakdown of new gascompletions by resource category. Nonconventionalcompletions comprise over 50% of the activity. Coalbed methane wells, after declining recently, return tohistorical levels. The number of shale and tight gaswells increase significantly over the period.

The conventional resource to be developed will bemore challenging than in the past, with reservoirstending to be deeper, hotter, tighter, less productive,and more costly to develop. Technology improvementshave played a critical role in allowing the commercialdevelopment of these resources. The outlook assumesthat advancements will continue. Not only do techni-cal advances provide the means to access increasinglychallenging reservoirs, they also drive down overallcosts, allowing additional commercial developmentfrom less traditional areas.

Reserves Development and Resource Replacement

The high drilling activity levels in the outlook arenecessary for new production to meet demand andalso discover new resources for future development.Figure 4-25 shows that with this high drilling level,lower-48 gas production remains relatively stable in2010-2025. However, the figure also shows that theannual reserve additions and the reserves-to-produc-tion (R/P) ratio are expected to go on a gradual butsustained decline.

The R/P ratio (the number of years that a givenresource would last if produced at the current rate) is a


Figure 4-21. Canadian Production by Category

0

1

2

3

4

5

6

7

2000 2005 2010 2015 2020 2025

TR

ILL

ION

CU

BIC

FE

ET

CONVENTIONAL

COAL BED

METHANE

SHALE

YEAR


Figure 4-22. Gas Well Activity in the U.S. Lower-48 and Canada

0

5

10

15

20

25G

AS

WE

LL

S (

TH

OU

SA

ND

S)

U.S. LOWER-48

CANADA

1990 1995 2000 2005 2010 2015 2020 2025

YEAR

Figure 4-23. U.S. Lower-48 Gas Wells by Type

0

5

10

15

20

2000 2005 2010 2015 2020 2025

GA

S W

ELLS

(T

HO

US

AN

DS

)

CONVENTIONAL

TIGHT

COAL BED METHANE

SHALE

YEAR


Figure 4-24. Canadian Gas Wells by Type

0

5

10

15

20

2000 2005 2010 2015 2020 2025

CONVENTIONAL

COAL BED METHANE

SHALE

GA

S W

ELLS

(T

HO

US

AN

DS

)

YEAR

Figure 4-25. Lower-48 Reserve Additions, Production, and Reserves-to-Production Ratio

0

5

10

15

20

25

0

1

2

3

4

5

6

7

8

9

LOWER-48 RESERVE ADDITIONS

GAS PRODUCTION

RESERVES-TO-PRODUCTION RATIO

RE

SE

RV

E A

DD

ITIO

NS

AN

D R

AT

E

(TR

ILLIO

N C

UB

IC F

EE

T P

ER

YE

AR

)

RE

SE

RV

ES

-TO

-PR

OD

UC

TIO

N R

AT

IO

(YE

AR

S)

2000 2005 2010 2015 2020 2025

YEAR

useful index to measure how fast a resource is beingdepleted. This figure is significant because it showsthat in the Reactive Path scenario, insufficient reservesare being discovered and developed in the U.S. lower-48 to fully replace production.

Access

The Environmental/Regulatory/Access Subgroupundertook a major evaluation of the complex regulato-ry/environmental issues faced by the industry and theimpact they have on access to potential natural gasresources. A significant part of this evaluation wasfocused on the Rocky Mountain region and assessedthe impact that permitting “conditions of approval”have on resource access. This work quantified areaswhere timing restrictions made them “effectively” off-limits to development, as well as other areas that hadadded costs or time delays associated with drilling.The corresponding technical resource for these areaswas then determined, with the overall results shown onFigure 4-26. The technical resource associated with theoffshore moratoria areas is also shown on this map.

The Reactive Path scenario assumes that these accessrestrictions (the offshore moratoria and the currentRockies conditions of approval process) remain inplace. In the Balanced Future scenario, the offshore

moratoria are lifted in a phased manner starting in2005 and the permitting process in the Rockies isimproved, resulting in a 50% reduction over five yearsin the resource volume effectively off-limits and in costimpacts and timing delays. This results in an addition-al 114 TCF of technical resource (9% of the lower-48total) that would be available for potential commercialdevelopment. By region, the volumes are 34 TCF fromthe Rockies, 25 TCF from the eastern Gulf of Mexico,33 TCF from the offshore Atlantic, and 21 TCF fromthe offshore Pacific. Overall, in this sensitivity lower-48 production increased by 3 BCF/D, or 4% of U.S.supply, in 2020. This resulted in a reduction in averageHenry Hub price (nominal) of $0.60/MMBtu, whichcorresponds to a potential savings in natural gas coststo consumers of $300 billion for the 2005-2025 period.

The results of the access evaluation led to the secondsupply-related finding:


Figure 4-26. Lower-48 Technical Resource Impacted by Access Restrictions

Note: Volumes are undiscovered,

technically recoverable.

33TCF

25TCF

21TCF

125 TCF

69

OFF-LIMITS

Finding: Increased access to U.S. resources(excluding designated wilderness areasand national parks) could save consumers$300 billion in natural gas costs over thenext 20 years.


Arctic Gas and LNG Outlook

The final supply finding is related to new supplysources:

As noted earlier, the supply contributions from newArctic developments and LNG imports were based onassumptions for timing and the capacity of each newdevelopment. These assumptions were then input tothe model as a fixed annual volume. Those annualprofiles for the Reactive Path scenario are shown inFigures 4-27 and 4-28.

Gas resources in the Arctic are remote from anyexisting pipeline infrastructure and are located in aharsh environment, so significant investment will berequired to bring these resources to market. The keyhurdles associated with commercializing these

resources are costs, permitting, Alaska state fiscalissues, and market risks. There are some favorabledevelopments regarding these resources. Industry ismaturing technology advancements to reduce capitalcosts and the supply/demand picture supports theneed for additional supplies. Also, the governments ofthe United States, Alaska, and Canada recognize thesignificant risks with such large-scale projects and areworking to put frameworks in place to address some ofthe hurdles.

The Mackenzie Gas Project is assumed to startdelivering 1 BCF/D of natural gas in 2009 with a50% capacity expansion in 2015. Most of the neces-sary resource has already been discovered in threemajor fields and the project is being actively pro-gressed.

An Alaskan gas pipeline project is assumed to startdelivering 4 BCF/D of natural gas in 2013.Approximately 35 TCF of discovered resource innorth Alaska will anchor the project, although addi-tional undiscovered resource will be needed tomaintain deliveries at full capacity for a 30-yearproject life.

Finding: New, large-scale resources such asLNG and Arctic gas are available and couldmeet 20-25% of demand, but are higher-cost, have longer lead times, and face majorbarriers to development.

Figure 4-27. Arctic Production Profile

0

2

4

6

2005 2010 2015

YEAR

2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

ALASKA

CANADA

The assumptions over the timing and volumes ofArctic gas are consistent for the Reactive Path and theBalanced Future scenarios, reflecting earliest feasiblestart-up dates. A sensitivity case was developed wherethe Alaska pipeline project is not developed during thestudy period. In this case, the average gas price is pro-jected to increase by 7% through 2025 as a result of the4 BCF/D Alaska production not being available.

From its current 1% share of North American sup-ply, LNG is projected to grow to a 12% share by 2025in the Reactive Path scenario. This will require largecapital investment in the field development of foreignsources, and its subsequent liquefaction, ship-bornetransportation, and eventual regasification in NorthAmerica. Although the needed volumes of natural gasresource are potentially available globally, the pace ofLNG infrastructure development in both producingcountries and in North America will provide the upperlimit for import volumes. In addition, concurrentlygrowing markets in the Far East and Europe will pro-vide strong competition for sales into North America.

Figure 4-28 shows the build-up of LNG supply fromthe four current U.S. regasification terminals that have

been operating at less than capacity since their con-struction. Starting in 2007, seven new terminals andseven expansions are projected in the Reactive Pathscenario.

In the Balanced Future scenario, permitting time isreduced from two years to one year, and two addition-al terminals and two additional expansions areassumed as a result of improved permitting processes.The incremental LNG imports for this scenario areshown in Figure 4-28.

To evaluate the impact of a lower rate of importedLNG growth, a sensitivity case was developed in whichonly two new LNG terminals were constructed due topermitting difficulties. In this case, LNG importcapacity was reduced to 6 BCF/D and the average gasprice increased by 10% from 2005 to 2025.

Resource Base Sensitivity

The range around the resource base volume was thelargest uncertainty tested in the supply outlook. Therange of technical resource was probabilisticallyassessed at 35% above the mean estimate to 30% below


Figure 4-28. North American LNG Imports

0

2

4

6

8

10

12

14

16

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

EXISTING

EXPANSION OF EXISTING

NEW – REACTIVE PATH

NEW – BALANCED FUTURE

YEAR

the mean estimate, each with a 10% probability ofoccurrence. Figure 4-29 shows the lower-48 produc-tion projections for this resource range in the ReactivePath scenario. In these sensitivity runs, the LNG andArctic import assumptions were not changed. The lowresource base sensitivity shows production declining17% in 2020 from the Reactive Path scenario, while thehigh resource base run shows an 11% increase. Clearly,maintaining production levels in the U.S. lower-48would not be possible if a lower resource base were inplace.

In addition to the volumes impact, the average gasprice increased 38% in the low resource base sensitivi-

ty and was 25% lower in the high resource base sensi-tivity. This further demonstrates the effect that theresource base uncertainty has on the future outlook fornatural gas supplies and prices.

Conclusions

From an overall perspective, supply in a robustfuture price environment will grow to meet risingdemand. Growth from the Rocky Mountain and deep-water Gulf of Mexico areas will offset declines in othertraditional basins in the United States and Canada.Additional growth will come from new sources of sup-ply such as LNG imports and Arctic gas.


Figure 4-29. Lower-48 Production Outlook and Resource Base Sensitivities

0

TR

ILLIO

N C

UB

IC F

EE

T

10

15

20

25

REACTIVE PATH MEAN RESOURCE

HIGH RESOURCE

LOW RESOURCE

2000 2005 2010 2015 2020 2025

YEAR

Resource Assessment

This section describes the assessment of natural gasresources in North America. This assessment, togeth-er with cost and production performance data wereused as inputs by an econometric model to determinethe size of the of commercial resource base and toderive an outlook for natural gas production through2025.

Assessment Process

The resource assessment was based on best practiceslearned from prior NPC studies and from other simi-lar studies. It was designed to use publicly availabledata, to be play-based, and to provide a thoroughreview by geoscientists and engineers. The resultingassessment represents an industry consensus.

Many sources of public and commercial data wereused. For the United States, data from the MineralsManagement Service (MMS) and United StatesGeological Survey (USGS) comprised the baseline. ForCanada, the Canadian Gas Potential Committee(CGPC) assessment was primarily used. For Mexico, acombination of IHS Energy Group (IHS) and USGSdata were used. Production-performance and field-size data were derived from the Energy InformationAdministration (EIA), IHS, and NRG Associates(Nehring). Cost data were derived from the AmericanPetroleum Institute (API) in the United States and thePetroleum Services Association of Canada (PSAC) inCanada.

Early on, best practice teams were organized to for-mulate methodologies for reserve growth, new field(undiscovered) assessment, cost, etc. Following that,major workshops were held for the purpose of reach-ing industry consensus on the various assessmentparameters for significant plays and basins.Subsequently, a further series of workshops was held tore-validate, or change, assessment parameters inresponse to information learned from the models usedto develop long-term forecasts. The work process isshown diagrammatically in Figure 4-30.

The smallest unit used for assessment is the “play”(or “Assessment Unit” as it is called in updated USGSterminology). A play has a coherent set of petroleumgeology characteristics. North America containsapproximately 700 plays. For the purposes of the cur-rent NPC study and for use in supply modeling, theseplays have been aggregated into 72 regions. In their

turn, the regions have been aggregated into 17 super-regions as shown in Figure 4-31.

Although comprised of many different plays, eachsuper-region displays its own distinguishing features.For instance, the Rockies super-region contains pre-dominantly nonconventional gas resource and has farmore access restrictions than any other lower-48onshore area. The super-regions are discussed in alater section of this chapter.

Definitions

In most cases, natural gas is a mixture of hydrocar-bons (primarily methane) plus small amounts of non-combustible gases. Natural gas may be produced inassociation with oil, or it may come from non-associ-ated gas fields. Approximately 87% of North Americangas is non-associated.

All volumes of natural gas referenced in this studyare dry gas remaining after liquefiable portions andnon-hydrocarbon gases have been removed as requiredby marketing considerations.

Technical resource is defined as that quantity of gasrecoverable with current technology without regard tothe economics of doing so. Commercial resource esti-mates are derived from econometric models.

In this study, remaining technical resources includeproved reserves, proved growth, and undiscovered, oryet-to-be-found, resources.

Proved reserves are defined as those reserves thathave a high confidence of being produced, and byimplication, they are already economic.

The estimated volume of gas that a field will ulti-mately produce is known as the estimated ultimaterecovery (EUR). At any time during the life of a field,the EUR is equal to the sum of those volumes that havebeen produced (cumulative production) plus theremaining proved reserves. Statistically, it can beshown that with the passage of time successive fieldEUR estimates tend to grow due to improved knowl-edge gained through operational experience during thelife of the field. Growth is the estimated technicalresource remaining in a field above the current esti-mate of proved reserves.

Undiscovered resource is the total volume of natu-ral gas expected to be found in the future that is notdue to growth of existing fields. It assumes current


CH

APTER 4 - N

ATURA

L GA

S SUPPLY

132

WORKSHOPS(VALIDATE RESOURCE, SUPPLY COST, ACCESS, AND TECHNOLOGY ISSUES FOR SIGNIFICANT REGIONS)

GULF OF

MEXICOROCKIES ALASKA CANADA

EAST

INTERIORMEXICO

PERMIAN/

ANADARKO

GULF

COAST

NONCONVENTIONALCONVENTIONAL

CASE

1

CASE

2

CASE

3

ETC.

MODEL RUNS(MULTIPLE CASES

FOR LEARNING &

UNCERTAINTY

ANALYSIS)

GROWTH

PROVED

DEV./

COST

NEW

FIELDS

ACCESS

TECH-

NOLOGY

BEST PRACTICE(PROVIDE

CONSISTENT

METHODOLOGY)

CORE

RESOURCE

TEAM

Figure 4-30. Resource Assessment Process


ALASKAARCTIC

CANADA

CANADA

ATLANTIC

GULF OF MEXICO

ROCKIES

MID-

CONTINENTEASTERN

INTERIOR

GULF COAST

ONSHORE

WESTERN

CANADA

SEDIMENTARY

BASIN

GREAT

BASIN

MEXICO

WEST

COAST

ONSHORE

U.S.

PACIFIC

OFFSHORE

WEST

TEXAS

BRITISH

COLUMBIA

EASTERN

CANADA

U.S. ATLANTIC

OFFSHORE

Figure 4-31. The 17 Super-Regions

technology and is not necessarily economic.Undiscovered resource is sometimes termed new fieldor yet-to-find.

Technology advancement (described later in thischapter) will tend to increase the size of the undiscov-ered resource depending upon the model-based timingof exploration and development. The assessmentsreported in this section are based upon current tech-nology and are independent of modeling assumptions.

Technical Resource Base

Figure 4-32 shows the relative contributions of tech-nical resource in North America. Of the 1,969 TCFNorth American technical resource, 69% is undiscov-ered. The remaining 31% is associated with knownfields in the proved and growth categories. In general,the uncertainty in the undiscovered category is largerthan in the growth category, and the uncertainty in“growth” is larger than in the proved category.

The undiscovered resource is split into two cate-gories: conventional and nonconventional. Althoughthe distinction is not absolute, conventional resourcesare located in discrete accumulations. They tend tohave better production performance characteristicsand they are amenable to traditional exploratory tech-niques. Nonconventional resources, including coal bed

methane, shale gas, and basin-centered gas, are typical-ly continuous accumulations that are much larger inarea than conventional discrete accumulations. Theyalso tend to have poorer production performance.Prior to drilling, traditional exploratory techniques arerelatively inaccurate at predicting productivity in anonconventional accumulation.

Figure 4-33 shows the relative contribution of con-ventional and nonconventional undiscovered resourcein North America.

Methodology

For the purposes of this report, it is more man-ageable to contrast the 17 super-regions (Figure 4-31) than to work with the 72 individual regions,which are the basic supply unit used in the sup-ply/demand modeling. For example, the Rockiessuper-region is comprised of 11 regions, each withits own supply characteristics. Figure 4-34 illus-trates the 72 regions.

In their turn, each region contains many plays,which are the basic units of technical resource assess-ment. The three major sources of baseline assessmentdata use variations of the same methodology to esti-mate the volume and distribution of undiscoveredfield sizes in each play.


NONCONVENTIONAL

28%

CONVENTIONAL

72%

Figure 4-33. North American Undiscovered Resource

UNDISCOVERED

69%

GROWTH*

17%

PROVED

14%

* Growth includes 55 trillion cubic feet in discovered

non-proved fields.

Figure 4-32. North American Technical Resource


Figure 4-34. The 72 Supply Regions

For conventional resources, the assessment processdefines a distribution of all discovered and undiscov-ered fields within a play. This distribution is a rela-tionship between the sizes and numbers of fields.Geoscience experts provide estimates such as the vol-ume of undiscovered resource as well as the maximumand mean size of undiscovered fields, which, whencombined with the data from discovered fields, fullydefines the total distribution of resources within a play.The distribution of undiscovered fields is then easilyderived. An example field size distribution is shown inFigure 4-35.

This process works reasonably well for the distri-bution of field sizes greater than economic thresholdbecause the anchor points from discovered fields arewell quantified. Generally, sub-commercial discover-ies are either underreported or poorly quantified,because there is no economic incentive to be precise.Therefore an adjustment has to be made to theassessment of undiscovered small fields below 6 BCF.This is done using a theoretical distribution and canadd approximately 20% to the technical resource vol-ume. It is important to estimate the distribution ofcurrently uneconomic fields, because the threshold

size for economic fields will decline with advances intechnology.

Nonconventional technical resource assessment isbased on an entirely different methodology. Discretegas fields do not exist in the conventional sense.Resources are generally distributed over large areaswhere nearly every well may find natural gas,although production rates may be highly variableand sometimes non-commercial. In this case, theassessment is based upon the total areal extent of thenonconventional reservoir unit, the area that onewell can drain, and the EUR that can be recovered bya well.

The assessment of growth was done at the regionlevel and was based on an analysis of the historicaldevelopment of fields. The method makes the assump-tion that a field is developed rationally by drilling thebest opportunity at any given time. It further makesthe assumption that, on average, “best” equates tohighest available EUR per well. For all fields discoveredin any particular year within a region, the rate at whichthe EUR per well decreases is analyzed. The sequenceof completions is divided into cohorts, which are more


1

10

0 10 20 30 40

100

1,000

FIELD SIZE RANK

MIL

LIO

N B

AR

RE

LS

OF

OIL

EQ

UIV

ALE

NT DISCOVERED

UNDISCOVERED

Source: OCS Report MMS 2001-087.

Figure 4-35. Example of a Field Size Distribution

fully described in the Supply Task Group Report.Growth is then calculated by extrapolating this EURper well analysis to an economic limit in the future.Figure 4-36 is an example of growth analysis showinghow EUR per well trends are projected.

Assessment estimates are often cited as precise val-ues, but in practice all values are subject to a range ofuncertainty. Generally a play resource assessment isthe statistical mean of its field size distribution (or poolsize distribution in much of Western Canada). Inorder to define a range of uncertainty around themean, this study has chosen to use a P10 value as thehigh side and a P90 value as the low side. P10 meansthere is a 10% chance that the high-side value willactually occur. P90 means there is a 90% chance thatthe low-side value will occur.

To arrive at the correct resource distribution for anaggregation of plays, the Monte Carlo method wasused. After several statistical tests, it was decided to usea high side of 135% of the mean and a low side of 70%of the mean for the complete North American aggre-gation. Although some simplifying assumptions were

made in defining this uncertainty range, the industryconsensus was that it was reasonable.

Technical Resources of the United States,Canada, and Mexico

The proportion of North America’s proved, growth(including discovered non-proved), and undiscoveredtechnical resources in each country is shown in Figure4-37. The United States has 1,451 TCF of technical gasresource, Canada has 397 TCF, and Mexico has 121TCF. In each country, undiscovered is the largest cate-gory of technical resource, ranging from 58% inMexico to 78% in the United States. The remainingresource is split approximately equally between provedand growth in all three countries.

The top three super-regions in terms of volume arethe Gulf of Mexico with 329 TCF, followed by Alaskawith 303 TCF, and the Rockies with 284 TCF (shownin Figure 4-2). Although these three super-regionseach contain a large technical resource base, they arequite distinct in character. In the Gulf of Mexico, agrowing proportion of new production will come


0.1

1.0

10.0B

ILLIO

N C

UB

IC F

EE

T P

ER

WE

LL

PRE 1951

1951-1960

1961-1970

1971-1980

1981-1990

1991-2001


FIELD VINTAGE

0 5 10 15 20

COHORTS

Figure 4-36. Growth-to-Known Methodology

from costlier, deeper water developments. In theRockies, a growing proportion of new production willcome from costlier nonconventional resources. Onthe other hand, in Alaska most of the resource isstranded due to the hostile Arctic environment andlack of a commercially viable export pipeline. Table 4-3 contains details of North America’s technicalresource base.

A short description of the characteristics of each ofthe significant super-regions in the United States,Canada, and Mexico follows.

United States

The United States contains 11 super-regions,described below. Current annual lower-48 productionof around 19 TCF satisfies 85% of demand. In 2025,lower-48 production will satisfy about 70% of demand.

Three of the super-regions provide just over 70% ofcurrent U.S. gas production: the Gulf of Mexico, 27%;the Gulf Coast Onshore, 25%; and the Rockies, 18%.In terms of technical resource, the same three super-regions contain 63% of the remaining 1,451 TCF. Thusthe relative production contribution from the U.S.super-regions will change through the study period.

Alaska. Alaska contains a very large undiscoveredresource (258 TCF), located both onshore and off-shore. North Alaska has a large discovered gas resource(40 TCF), which is currently stranded due to lack ofpipeline infrastructure. Development depends on thecommercial viability of constructing a pipeline to mar-kets in Canada and the U.S. lower-48. The remotenessand harsh environment add significantly to explo-ration and development cost. In addition, access toresource in the Arctic National Wildlife Refuge(ANWR) and the National Petroleum Reserve, Alaska(NPRA) is still a contentious issue. The potentiallylarge nonconventional undiscovered resource has alarge assessment uncertainty mainly because there is alack of data.

U.S. Pacific Offshore. This area has moderateundiscovered resource potential (21 TCF), but it isunder a moratorium for new leases. Some wells weredrilled offshore in northern California and Oregon inthe 1960s with minor gas shows but without commer-cial success. Southern California has minor gas pro-duction associated with oil.


Figure 4-37. Technical Resources ofUnited States, Canada, and Mexico

UNITED STATES: 1,451 TCF

PROVED

13%

GROWTH

17%

UNDISCOVERED

70%

CANADA: 397 TCF

PROVED

15%

GROWTH

17%

UNDISCOVERED

68%

MEXICO: 121 TCF

UNDISCOVERED

58%

GROWTH

19%

PROVED

23%

CH

APTER 4 - N

ATURA

L GA

S SUPPLY

139

Discovered Remaining Undiscovered

Super-RegionProved

Reserves

Growth toProved

Reserves*

TotalDiscoveredRemaining

UndiscoveredConventional

Potential

UndiscoveredNonconven-

tional Potential

TotalUndiscovered

Potential

TotalTechnicalResource

Alaska 9 36 45 201 57 258 303West Coast Onshore 3 3 6 10 13 23 29Great Basin 1 1 2 3 0 3 5Rockies 50 26 75 36 173 209 284West Texas 16 21 38 20 7 27 64Gulf Coast Onshore 38 60 98 77 8 86 183Midcontinent 24 32 56 27 5 32 88Eastern Interior 14 5 18 15 76 92 110Gulf of Mexico 29 55 84 244 0 244 329U.S. Atlantic Offshore 0 0 0 33 0 33 33U.S. Pacific Offshore 1 1 2 21 0 21 22Western Canada Sedimentary Basin 57 28 86 93 46 138 224Arctic Canada 0 25 25 46 0 46 71Eastern Canada Onshore 0 0 1 2 4 6 6Canada Atlantic Offshore 2 15 18 68 0 68 85British Columbia 0 0 0 11 0 11 11Mexico 28 22 51 70 0 70 121North American Total 272 332 604 977 389 1,366 1,969

United States 183 241 425 687 339 1,027 1,451Canada 60 68 128 219 50 269 397Mexico 28 22 51 70 0 70 121

* Growth includes 55 TCF of discovered non-proved in Alaska (14 TCF), Arctic Canada (25 TCF), Canada Atlantic (15 TCF), and Gulf of Mexico (1 TCF).

Table 4-3. North American Technical Resource Base – Current Technology (Trillion Cubic Feet)

West Coast Onshore. Approximately half thetotal undiscovered resource of 23 TCF is noncon-ventional. This occurs in the north and is unlikelyto be commercial during the study period becauseof poor reservoir quality and a thick volcanic over-burden.

Great Basin. This large area has an extremely smallpotential (3 TCF) owing to a combination of geologi-cal factors. Most of this is concentrated in a small areain the east (Paradox Basin). Recent exploration resultsin other areas of the Great Basin have been disap-pointing.

Rockies. The total undiscovered potential here isvery large (209 TCF) and is 80% nonconventional.There are significant access issues and, until 2002,there was a shortage of pipeline export capacity.Nevertheless, production has grown and the Rockiessuper-region is one of the few areas where indigenousproduction is likely to continue growing. Discovery ofthe world class San Juan coal bed methane play in the1980s led to significant, although less prolific, coal bedmethane plays in other parts of the Rockies. Waterdischarge and operational footprint issues are likely tobe future concerns. Advances in well completiontechnology have improved the viability of nonconven-tional tight gas and shale gas plays. Access and tech-nology will determine how much of the technicalresource base becomes commercial.

West Texas/New Mexico. Total undiscovered poten-tial is moderate (27 TCF), because the main producingareas, such as the Permian Basin, are mature. However,downspacing and infill drilling provide some opportu-nities for field growth. The super-region also containsthe large nonconventional Barnett Shale play in theFort Worth basin, where the recent production ramp-up has been driven by improvements in completionand stimulation technology.

Midcontinent. Although it is an important areaof current production, this super-region containsonly moderate undiscovered resource (32 TCF),mostly (85%) conventional. The Anadarko Basinhas potential for further deep conventional discover-ies, and other basins have small nonconventionalpotential.

Gulf Coast. The Gulf Coast is an important area ofcurrent production with a large undiscovered potential(86 TCF), over 90% conventional. Although reason-

ably well explored, the complex geology allows for thepossibility of new trends, particularly deeper. Usingimproved completion and “sweet spot” detection tech-nology, there is also the possibility of finding addition-al moderately large nonconventional tight and coal bedmethane resources.

Gulf of Mexico. This is the most prolific producingsuper-region, even though the mature shelf plays inshallower water are in rapid decline. Total undiscov-ered resource (244 TCF) is mainly in the deeper waterplays where the complex geology due to salt causeshigher exploration risk. Risk and deep drilling makethis the highest cost area for exploration and develop-ment in the U.S. lower-48. The eastern Gulf of Mexicocontains moderate undiscovered potential, but accessto that region is restricted.

U.S. Atlantic Offshore. Although this area has mod-erate undiscovered resource (33 TCF), it is under aleasing moratorium. It was fairly well explored in the1970s with no commercial discoveries, but the deeperwater has not been tested. There was a gas discoveryoffshore New Jersey, but at the time it was not eco-nomic to develop. Recent adjacent Canadian discover-ies are reason for moderate optimism for potential inthe north of this super-region.

Eastern Interior. This area contains a very large,mainly nonconventional, undiscovered resource (92TCF). Almost three-quarters of this potential is locat-ed in the Appalachian Region. However, productionhas barely grown over several decades. The main issuesare low recoveries per well and the disparate mineralownership. Technology improvement and a sustainedhigher price environment will cause moderate produc-tion growth in the Eastern Interior.

Canada

Canada contains five super-regions. The currentannual production of 6 TCF more than satisfies inter-nal demand. The surplus of about 3 TCF is exportedto the United States. The Western Canada SedimentaryBasin contributes 97% of current production, but only56% of Canada’s 397 TCF of technical resource. As inthe United States, the relative production contributionof Canada’s five super-regions will change throughtime.

Western Canada Sedimentary Basin. The WesternCanada Sedimentary Basin is mature and its produc-


tion has plateaued. The remaining undiscovered con-ventional resource (93 TCF out of a total undiscoveredresource of 138 TCF) is located in increasingly smalleraverage pool sizes. Nonconventional resources are notas well assessed as in the United States and have a largeuncertainty range. Coal bed methane development isimmature compared with the Rockies. Unlike theRockies, access is a relatively minor issue.

Arctic Canada. A fairly large volume of strandedresources (25 TCF) has been discovered onshore andoffshore, although much is remote. Approximately30% of the stranded gas will be developed as part ofthe Mackenzie Gas Project. Undiscovered resource is46 TCF. However, much of this will not be developedthrough 2025 because of remoteness and Arctic condi-tions.

Canada Atlantic (offshore). Like the CanadianArctic, stranded resources (15 TCF) have been discov-ered, particularly off Labrador. The undiscoveredresource is also large (68 TCF). High cost and lack ofpipelines will limit development of much of thisresource through 2025.

British Columbia (onshore and offshore).Excluding that part of British Columbia assessed in theWestern Canada Sedimentary Basin, there is moderateundiscovered potential (11 TCF) in the inter-montaneand the offshore/coastal basins. Offshore access isrestricted, although there is potential that restrictionswill be lifted.

Eastern Canada (onshore). This very large area hasonly small undiscovered conventional and nonconven-tional resource (6 TCF). There is some coal bedmethane activity in Nova Scotia.

Mexico

For the purposes of this study, Mexico has beendefined as a single super-region. Current annual gasproduction is 1.8 TCF. The annual shortfall of 8% ofdemand is provided by exports from the UnitedStates. Mexico has started an ambitious program toincrease its exploration and development of gasresources.

Mexico has a moderate technical resource (121TCF), which is mainly non-associated in the north andassociated with the prolific oil production in the south.Compared to adjacent U.S. areas, Mexico has beenmore lightly explored, particularly offshore.

Resource Comparisons

As described earlier, the technical resource availablefor potential future production is composed of threecategories: proved, growth, and undiscovered. The rel-ative distribution of these three components of theresource base in the North American super-regions isdescribed below.

Proved

Figure 4-38 ranks the super-regions by their provedgas reserves. The Western Canada Sedimentary Basin


Figure 4-38. Super-RegionsRanked by Proved Reserves

57

50

38

29

28

24

16

14

9

3

2

1

1

WESTERN CANADA SEDIMENTARY BASIN

ROCKIES

GULF COAST ONSHORE

GULF OF MEXICO

MEXICO

MIDCONTINENT

WEST TEXAS

EASTERN INTERIOR

ALASKA

WEST COAST ONSHORE

CANADA ATLANTIC OFFSHORE

GREAT BASIN

U.S. PACIFIC OFFSHORE

0 10 20 30 40 50 60 70

TRILLION CUBIC FEET

contains the most proved reserves (57 TCF), followedby the Rockies (50 TCF), the Gulf Coast Onshore (38TCF), and the Gulf of Mexico (29 TCF). Provedreserves in Alaska consist of 2 TCF from south Alaskaand 7 TCF of fuel gas for ongoing Prudhoe Bay oildevelopment. Northern Alaska has an additional 33TCF of discovered gas resource, which is not bookedas proved because of the current lack of a pipeline tomarket.

Of the three technical resource components,proved reserves is the least uncertain. In a given area,industry will generally develop proved reserves moreeconomically than growth or undiscovered re-sources.

Growth

As described earlier, estimates of EUR generallyincrease over time. The difference between the cur-rent estimate of proved reserves and what is ultimate-ly produced is known as growth. Figure 4-39 showsthat 53% of the 277 TCF of reserve growth will become from three super-regions. The largest is theGulf Coast Onshore with 60 TCF, followed by theGulf of Mexico with 55 TCF, and the Midcontinentwith 32 TCF.

Growth estimates have intermediate uncertaintybetween proved and undiscovered. In a given area,industry will generally develop growth more economi-cally than undiscovered resources.

Undiscovered

Figure 4-40 shows the super-regions ranked by vol-ume of undiscovered technical resource. This techni-cal resource is further split into conventional andnonconventional. Alaska ranks highest with 258 TCF,with 57 TCF of this nonconventional. The Gulf ofMexico ranks second in terms of undiscoveredresource (244 TCF), although it has the largest conven-tional resource of all super-regions. The Rockies ranksthird with 209 TCF of undiscovered resource, 173 TCFof which is nonconventional. The Mexico super-region ranks 7th with 70 TCF.

Of the three categories of technical resource,undiscovered estimates are the most uncertain. Sincemost of the North American super-regions are rela-tively mature, the average remaining undiscovered

field size is small. Combining this with the higherrisk of exploration failure causes a smaller proportionof undiscovered to be economic compared to provedor growth.

Figure 4-41 shows each super-region’s relative contri-bution of conventional undiscovered resource. The dis-tribution is skewed, with only a few major contributors.


Figure 4-39. Super-Regions Ranked by Reserve Growth

60

55

32

28

26

22

22

21

5

3

1

1

0 10 20 30 40 50 60 70

GULF COAST ONSHORE

GULF OF MEXICO

MIDCONTINENT


ROCKIES

MEXICO

ALASKA

WEST TEXAS

EASTERN INTERIOR

WEST COAST ONSHORE


GREAT BASIN

TRILLION CUBIC FEET

The Gulf of Mexico ranks first with 25%, followed byAlaska with 21%, and the Western Canada SedimentaryBasin with 9%.

In contrast, Figure 4-42 shows the relative contri-bution of nonconventional undiscovered resource.This distribution is even more skewed than for con-ventional undiscovered. Only 9 of the 17 super-regions contain nonconventional resource of anysignificance. In this case, the Rockies ranks first with44%, followed by the Eastern Interior with 20%,Alaska with 15%, and the Western CanadaSedimentary Basin with 12%.

Main Conclusions from Super-Region Comparison

� Of the total 1,969 TCF of North American technicalresource, 69% is undiscovered, 17% is growth, and14% is proved. In any one area, proved will general-ly be developed first, followed by growth and thenundiscovered.

� Four super-regions (Gulf of Mexico, Rockies,Western Canada Sedimentary Basin, and Alaska)contribute a large proportion (62%) of NorthAmerica’s undiscovered resource.

� In terms of nonconventional resource, just foursuper-regions (Rockies, Eastern Interior, Alaska, andWestern Canada Sedimentary Basin) contribute90% of the undiscovered potential.

� The current North American proved reserve base,which now totals 272 TCF, is expected to grow by277 TCF, or by 102%. The Gulf Coast Onshore,the Western Canada Sedimentary Basin, the Gulfof Mexico, and the Rockies will contribute over63% of the proved reserves plus growth total, pro-viding a large proportion of near-term productionvolume.

� Although North American production willincrease slightly by 2025, the relative contribu-tions of the super-regions will change signifi-cantly. Decline will be most severe in the GulfCoast Onshore, West Texas, and the WesternCanada Sedimentary Basin. On the other hand,these declines will be offset by productionincreases in the Rockies, Eastern Interior, Alaska,and Mexico.


Figure 4-40. Super-Regions Ranked byUndiscovered Technical Resource

0 50 100 150 200 250 300

ALASKA

GULF OF MEXICO

ROCKIES


EASTERN INTERIOR

GULF COAST ONSHORE

MEXICO

CANADA ATLANTIC OFFSHORE

ARCTIC CANADA

U.S. ATLANTIC OFFSHORE

MIDCONTINENT

WEST TEXAS

WEST COAST ONSHORE


WESTERN BRITISH COLUMBIA

EASTERN CANADA ONSHORE

GREAT BASIN

TRILLION CUBIC FEET

NONCONVENTIONAL

CONVENTIONAL

258

244

209

138

91

86

70

68

46

33

32

27

23

21

11

6

3


Figure 4-42. Distribution of Nonconventional Undiscovered Resource

ROCKIES

44%

EASTERN

INTERIOR

20%ALASKA

15%

EASTERN CANADA ONSHORE 1%

MIDCONTINENT 1%

WEST TEXAS 2%

GULF COAST ONSHORE 2%

WEST COAST ONSHORE 3%WESTERN

CANADA

SEDIMENTARY

BASIN

12%

Figure 4-41. Distribution of Conventional Undiscovered Resource

GULF OF

MEXICO

25%

ALASKA

21%

WESTERN

CANADA

SEDIMENTARY

BASIN

9%GULF

COAST

ONSHORE

8%MEXICO

7%

CANADA

ATLANTIC

OFFSHORE

7%

EASTERN CANADA ONSHORE <1%

GREAT BASIN <1%

WEST COAST ONSHORE 1%

WESTERN BRITISH COLUMBIA 1%

EASTERN INTERIOR 2%

WEST TEXAS 2%

U.S. PACIFIC OFFSHORE 2%

MIDCONTINENT 3%

U.S. ATLANTIC OFFSHORE 3%

ARCTIC

CANADA

5%

ROCKIES

4%


Cost Methodology

A critical part of the NPC study was estimatingreasonable costs for use in the model to determinecommercial resources. Costs were needed for allaspects of onshore and offshore gas development –exploration and development drilling, productionand lease facilities, and operations and mainte-nance. Where possible, public and commercialdatabases were used to estimate costs. Sourcesincluded, among others, the API Joint AssociationSurvey on Drilling Costs, the PSAC Well CostStudies, and the EIA Oil & Gas Lease Equipmentand Operating Costs report. In areas where ade-quate public and commercial data were not avail-able, costs were based on available information andcirculated for review and comment to industryexperts familiar with costs in that area. Costs werethen revised based on the input received. At each ofthe regional workshops, which were held primarilyto review the resources, costs were also discussed inorder to determine the key factors that might affectcosts in that region (i.e., infrastructure, weather,drilling depths, etc.).

The costs used in the model are average costs forgeneric operations. For example, the well costs are forgeneric wells at an average drill depth. Actual costswill vary with regards to water depth, drill depth, porepressure, rig type, etc., depending on specific loca-tions. The same is true for development costs. Actualcosts will depend on location, infrastructure, meto-cean conditions, well productivity, etc. All costs usedin the modeling exercise were expressed in year 2000dollars.

Gulf of Mexico

The Gulf of Mexico was divided into eight super-play regions and subdivided by six water-depth inter-vals. Well depths were based on the resource weightedaverage reservoir depth, referenced to sea level, for agiven super-play and water depth. Costs were devel-oped for both shallow-water and deepwater scenarios.For the shallow-water scenarios, two water depthswere assumed (100’ and 400’), and costs were devel-oped for both exploration wells and platform develop-ment wells. For the deepwater scenarios, four waterdepths were assumed (1000’, 2000’, 4000’, and 6000’),and costs were developed for exploration wells andboth subsea-completed and platform-completeddevelopment wells.

The Minerals Management Service provided the ini-tial drilling and completion (D&C) costs. These costswere sent to industry experts for review and wereadjusted based on their comments.

Non-drilling costs make up a substantial part ofoffshore development costs. These non-drillingdevelopment costs include production platforms,production equipment, subsea equipment, aban-donment costs, and the costs of gathering pipelines.These costs are dependent on the development con-cept selected for a field. Field size, water depth, loca-tion, and well productivity determine this concept.For this study, the following development conceptswere used for costing purposes in the Gulf ofMexico: a steel pile jacket (SPJ), which is a bottom-founded fabricated steel structure; a tension leg platform (TLP), which is a seabed anchored buoy-ant/compliant substructure constructed in steel orconcrete; a spar, a buoyant concrete caisson that isanchored to the seabed; and subsea production sys-tems/tiebacks.

The initial non-drilling development costs werebased on spreadsheets from EEA and were bench-marked and adjusted based on the WoodMackenzie database. Development costs were cal-culated for 20 different field sizes for all of thewater depths and drill depths associated with eachsuper-play. These costs were plotted and sent toindustry experts for review and were adjustedbased on their comments.

Operating costs for the Gulf of Mexico were basedon the Wood Mackenzie U.S. Gulf of MexicoDeepwater Study (November 2001). Summaries ofthese costs were sent to industry experts for review andwere adjusted based on their comments.

Lower-48 Onshore

Drilling and completion costs for the lower-48onshore wells were based on the API JointAssociation Survey on Drilling Costs. This surveyhas been conducted annually since 1959 and is sentto operators who have conducted drilling operationsduring the year. The survey provides total D&C costsfor oil, gas, and dry wells on a state and regional basisby depth intervals. All cost components such as per-mitting, location construction, mobilization, rentalsand services, tangible items, and stimulations areassumed to be included. The API Joint Association

Survey on Drilling Costs also contains a breakdownof D&C costs for coal bed methane, horizontal, andsidetrack wells.

For this NPC study, the 1999 and 2000 surveys wereused to determine an average base case cost for oil,gas, and dry wells in 26 regions. The regions weredivided into four depth intervals at 5,000-foot incre-ments.

Onshore development costs, also known as leaseequipment costs, consist of everything needed toproduce a well downstream of the wellhead tree(i.e., flowline and connections, separators, dehydra-tors, pumps, and storage tanks). Onshore develop-ment costs were derived from EIA’s Report “Oil andGas Lease Equipment and Operating Costs 1986Through 2000.” This report, which has accompany-ing Excel spreadsheets, presents estimated costs ofall equipment and services that are in effect duringJune of each year. The aggregate costs for typicalleases by region, depth, and production rate areaveraged, and these averages provide a generalmeasure of the changed costs from year to year forlease equipment and operations. The report is inthe public domain and can be viewed on EIA’s web-site.

Alaska – Onshore and Offshore

Cost parameters (water depth and reservoirdepth) were provided for fourteen regions. D&Ccosts for six onshore regions were initially basedon limited data contained in the API JointAssociation Survey. These costs were sent toindustry experts and adjusted based on additionalindustry experience. D&C costs for eight offshoreregions were initially based on previous EEA-generated costs. These costs were also sent toindustry experts and adjusted based on additionalindustry input.

Onshore development costs consist of the samecomponents as onshore U.S. lower-48 estimates, withadditional costs added for access road and utility con-struction due to the remote locations of the newfields. Onshore development and operating costs,expressed on a per-well basis, are based on EEA-generated costs that were reviewed and adjusted byindustry experts.

Offshore development costs consist of the samecomponents as U.S. Gulf of Mexico estimates. The

development concepts considered in this study foroffshore Alaska were: an offshore platform, usuallya steel pile jacket; a gravel island; a gravity basedstructure (GBS), a platform constructed in concretewhich sits on the seabed; and subsea productionsystems/tiebacks.

Initial development costs were based on EEA-generated costs benchmarked and adjusted usingthe Wood Mackenzie database. These costs weresent to industry experts for review and were adjust-ed based on their comments.

Offshore operating costs, expressed as costs perwell, are based on EEA-generated costs that werereviewed and adjusted by industry experts.

Atlantic Offshore

Cost parameters (water depth and reservoir depth)were provided for six regions. Due to lack of recentdrilling and development activity in the AtlanticOffshore region, D&C, development, and operatingcosts were based on Gulf of Mexico costs for similarwater depths and reservoir depths. Adjustment fac-tors based on industry experience were used toaccount for the differences in infrastructure, logis-tics, weather, drilling conditions, etc. D&C costadjustment factors ranged from 1.4 to 1.75, whiledevelopment cost adjustment factors ranged from1.4 to 1.6.

Pacific Offshore

Cost parameters (water depth and reservoir depth)were provided for six regions. Due to minimaldrilling and development activity in the PacificOffshore region, D&C, development, and operatingcosts were based on Gulf of Mexico well costs forsimilar water depths and reservoir depths.Adjustment factors based on industry experiencewere used to account for the differences in infra-structure, logistics, weather, drilling conditions, etc.D&C cost adjustment factors ranged from 1.1 to 2.0while development cost adjustment factors rangedfrom 1.2 to 1.5.

Western Canada Onshore

D&C costs were based on the Petroleum ServicesAssociation of Canada’s well-cost studies. ThePSAC is the national association of Canadian oil-field service, supply, and manufacturing compa-nies; it develops two well-cost studies per year, one


for the summer drilling season and one for thewinter drilling season. These studies contain D&Ccosts in a detailed Authority for Expenditure (AFE)format for the “typical” or “most popular” wellsbeing drilled. The studies generally include 30 to35 wells.

Western Canada was divided into five geograph-ical regions and then subdivided by depth intervals.The summer 2002 and winter 2003 cost studieswere used to determine average D&C costs, whichwere then de-escalated to 2000 dollars for modelinput.

Onshore development costs consist of the samecomponents as the U.S. lower-48 onshore estimateswith additional costs added for access road and util-ity construction due to the remote locations of thenew fields. Onshore development and operatingcosts per well were based on EEA-generated coststhat were reviewed and adjusted by industryexperts.

Canada Offshore and Onshore Other

Cost parameters (water depth and reservoirdepth) were provided for three onshore regions andeleven offshore regions. D&C costs were initiallybased on preliminary work (December 2002) by theCanadian Energy Research Institute (CERI) and pre-vious EEA-generated costs. These costs were sent toindustry experts for review and were adjusted basedon their comments. Offshore wells were assumed tobe drilled with a jack-up or semi-submersibledrilling rig.

Onshore development costs consist of the samecomponents as the U.S. lower-48 onshore estimateswith additional costs added as required for accessroad and utility construction due to remote loca-tions of the new fields. Onshore development andoperating costs per well, are based on EEA generatedcosts that were reviewed and adjusted by industryexperts.

Offshore development costs consist of the samecomponents as U.S. Gulf of Mexico estimates.Development concepts considered in this studyfor offshore Canada were: an offshore platform,usually a steel pile jacket; a gravel island; a gravitybased structure (GBS), a platform constructed inconcrete which sits on the seabed; and subsea pro-duction systems/tiebacks. Initial development

costs were based on EEA-generated costs, prelimi-nary work by the Canadian Energy ResearchInstitute, and the Wood Mackenzie database.These costs were sent to industry experts forreview and were adjusted based on the commentsreceived.

Mexico

Cost parameters (water depth and reservoirdepth) were provided for seven onshore and threeoffshore regions. D&C costs for onshore areaswere based on data from the Pemex website formultiple services contract for the Burgos Basin.The costs for the Burgos Basin were extrapolatedto other areas based on average depth. OffshoreD&C costs were based on Gulf of Mexico well costsfor similar water depths and reservoir depths. Anadjustment factor of 1.5 was used to account forthe differences in infrastructure, logistics, drillingconditions, etc.

Development costs for Mexico came from the IHSMexico study. For onshore developments, the devel-opment plan included wellsites tied back to a dedi-cated production facility incorporating separation,condensate stabilization, gas dew-pointing, gasexport, compression and condensate export. Gasand condensate are exported to main gas and oilexport pipelines that send oil and gas to a processingplant.

Two development concepts were used in the IHSstudy for offshore developments. For small fieldsand shallow water, a lightweight steel jacket sup-porting wellheads was used. For larger fields and indeepwater, the development plan consists of steeljackets supporting wellheads, production, quarters,and compression.

Nonconventional Gas

Costs for nonconventional gas development (coalbed methane and tight gas) were handled in eachgeographical area using essentially the same costmethodology used for conventional developments.For coal bed methane developments, adjustmentsrequired for the unique production style (i.e., dewa-tering of the coal prior to onset of gas production)were made. Dehydration and storage tank costswere removed from the lease equipment cost com-ponent, and costs for water handling equipmentand compression were added. In addition, capital


and operating and maintenance costs for water dis-posal, dependent on the type of disposal (i.e., sur-face discharge or re-injection), were included.Costs for stimulation of tight gas zones wereaccounted for in the D&C costs.

Rig Fleet Availability

An important consideration in the development offuture resources is the future availability of equipment,particularly drilling rigs. Most of the rigs currentlyavailable for use today were built in the late 1970s andearly 1980s. Since this time period, which also had thepeak number of active rigs, both the onshore and off-shore rig fleets have declined. One annual survey thattracks the number of available U.S. rigs is the Reed-Hycalog Rig Census. Conducted since the 1950s, thissurvey tracks the number of rigs available by surveyingdrilling contractors. Rigs are considered available ifthey have worked during a 45-day qualification period.Rigs not considered available include those requiring acapital expenditure (excluding drillpipe) of $100,000for a land rig and $1,000,000 for an offshore rig andthose rigs stacked for more than three years. Also notconsidered are rigs that cannot drill below 3,000 feet.

Figure 4-43 shows the Reed-Hycalog Census datafor the past 15 years. In 1987, the total number ofrigs available, both land and offshore, was 3,331(peak was in 1982 with 5,644 available rigs). In2001, the total available was 1,722.

For this study, major drilling contractors wereconsulted in order to estimate future attrition ratesfor rigs. Different rig fleet attrition scenarios werediscussed and a consensus was reached on a scenariofor both the onshore and offshore rig fleets. Figures4-44 and 4-45 show the estimated rig fleet availabil-ity out to 2025.

For the onshore fleet, a period of slight growth andstabilization was assumed out to 2005. For the next20 years, the following attrition rates to the existingrig fleet were assumed: 2006-2010, 1%; 2011-2015,1.5%; 2016-2020, 2%; 2021-2025, 3%.

For the offshore fleet, a period of slight growth andstabilization was also assumed out to 2005. For thenext 20 years, the following attrition rates to the exist-ing rig fleet were assumed: 2006-2010, 2%; 2011-2015,2.5%; 2016-2020, 3%; 2021-2025, 3.5%.


0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

YEAR

AV

AIL

AB

LE

RIG

S

ALL RIGS LAND RIGS

Note: Includes 48 New Builds during the time period shown.

Source: Reed-Hycalog Census.

Figure 4-43. Rig Fleet Availability


0

50

100

150

200

250

300

2000 2010 2020

YEAR

RIG

S A

VA

ILA

BLE

(BA

SE

D O

N E

XIS

TIN

G F

LE

ET

)

Figure 4-45. Projected Offshore Rig Fleet Attrition

0

800

1,000

1,200

1,400

1,600

2000 2010 2020

YEAR

RIG

S A

VA

ILA

BLE

(BA

SE

D O

N E

XIS

TIN

G F

LE

ET

)

Figure 4-44. Projected Onshore Rig Fleet Attrition


Production Performance Analysis

In order to help estimate future well performanceparameters and to calibrate HSM model results, ananalysis of historical production performance wasundertaken for the U.S. lower-48 and WesternCanada. Analyses were conducted for the periodfrom 1990 to 2002, in order to put current trends intoa long-term context. The analysis then focused on thelast four years of production performance with theaim of understanding the reasons behind the lack ofsignificant, sustained production response followingthe large ramp-up of industry activity in 2000 and2001.

Production performance was analyzed using fourparameters to describe key trends and to understandthe causes of those trends. The four parameters were:

� Gas Well Drilling Activity vs. Production

� Individual Gas Well Performance

– Estimated Ultimate Recovery

– Initial Production Rate

– Initial Decline Rate

� Base Decline of Existing Reserves

� Reserves and R/P Ratios.

Production performance parameters were summa-rized on a regional basis, although most areas wereanalyzed on a much more granular basis looking atindividual formation response, response by depthtranche, and response by resource type (i.e. coal bedmethane vs. conventional performance).

The NPC utilized EEA’s Gas Supply Review (GSR) asits primary data source to analyze production per-formance. The majority of the data for the analysisemanated originally from the IHS ProductionDatabase under license to EEA. Thus, source refer-ences in this chapter to EEA’s Gas Supply Reviewinclude data supplied by Petroleum Information/Dwights LLC; Copyright (2003) Petroleum Informa-tion/Dwights LLC.

To standardize the vast amount of data and performstandard analyses, the IHS production data were con-ditioned by EEA to ensure completeness and standard-ization. The production performance parametersgenerated in the analysis were used either as directinputs to the HSM, or to check HSM outputs.

U.S. and Canadian Production Overview

Annual average production of dry gas in the U.S.lower-48 and Canada has increased 11 BCF/D (0.9BCF/D per year) from 57.8 BCF/D in 1990 to 68.9BCF/D in 2002, an increase of 19% (1.8% per year). Asillustrated in Figure 4-46, peak production rates haveincreased substantially less, as the excess gas deliver-ability present in the early 1990s has been progressive-ly used to satisfy increasing summer power demandand to fill incremental gas storage.

As the North American gas basins have matured,total supply growth has slowed considerably, from2.3% per year in the early 1990s, to 0.6% per year overthe 1996-2002 period. U.S. lower-48 production hasremained essentially flat since 1996. Gas productiondeclined in 2002 from the United States and Canada asa whole and from the Western Canada SedimentaryBasin, North America’s largest producing region.

Production gains were largely concentrated in threeregions, Western Canada, the Rocky Mountains, andthe Deepwater Gulf of Mexico, as technology andadvancing infrastructure allowed the industry toexploit these less mature, highly prospective areas.These three areas now account for approximately 45%of total gas production, up from 27% in 1990. In themore mature Gulf of Mexico shelf and onshore lower-48 basins, while the industry was able to maintain sta-ble production levels through the mid-1990s, as thesebasins continued to mature their production beganfalling, offsetting increases in other basins.

In the U.S. lower-48, the character of the productionhas also been changing. Conventional gas productionactually fell throughout the 1990s, as shown in Figure4-47. Nonconventional production – namely coal bedmethane, shale gas, and tight gas – grew from 12% ofproduction in 1990 to over 25% in 2002.

U.S. Lower-48: Activity vs. Production

Industry activity levels for natural gas exploration,development, and production, as measured by rigcount and gas well connections, have historicallyexhibited a strong correlation to the price of naturalgas, as shown in Figure 4-48. The early 1990s werecharacterized by low gas prices ($1.75 average GulfCoast Spot Price) and a gas rig count of approximately400 rigs. Gas prices rose in the 1997-1999 period,averaging $0.45/MMBtu higher, or $2.20/MMBtu. Asprices rose, the gas rig count rose to an average of 540


rigs. Gas prices continued to climb and Gulf CoastSpot Prices averaged $3.65/MMBtu from 2000 through2002. As prices increased, rig activity also rose, with anaverage gas rig count of 780 rigs over the period. InJune 2003, the rig count stood at approximately 900 gasrigs and a Gulf Coast Spot Price at $5.80.

As measured by gas well completions, a similar pat-tern of increasing activity emerges and is shown inFigure 4-49. In the early part of the decade, the indus-try averaged 400 gas rigs and 9,700 gas completions peryear. As the rig count increased by 35% to 540 gas rigsduring the 1997-1999 period, average gas completionsrose by 25% to 12,100 gas completions per year. Overthe 2000-2002 period, gas completions increased to

19,300 gas completions per year, almost double whatthey were averaging a decade before.

While yearly annual production rates grew in theearly 1990s, this growth was primarily due to a changein production patterns, rather than a true increase inwellhead deliverability. Peak production ratesremained flat at 50 BCF/D year on year. Increasingproduction during summer months for storage and tomeet growing summer power demands allowed annu-al average production to increase.

Since the early 1990s, the rig activity level needed tosustain production has increased substantially asshown in Figure 4-50. While an average gas rig count

EASTERN CANADA

0

10

20

30

40

50

60

70

80

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

YHISTORICAL PROJ.


YEAR

OTHER

MIDCONTINENT

PERMIAN BASIN

E.TEXAS/N. LOUISIANA

EASTERN GULF COAST

S. TEXAS GULF COAST

GULF OF MEXICO

SHELF

ROCKIES

WESTERN CANADA

DEEPWATER

GULF OF MEXICO

SEDIMENTARY BASIN

Figure 4-46. U.S. Lower-48 and Canadian Production by Region



0

10

20

30

40

50

60

70

1990 1992 1994 1996 1998 2000 2002

WE

T G

AS

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

CONVENTIONAL

COAL BED METHANE

SHALE

TIGHT*

* Tight gas estimated 2000-2002.

YEAR

Figure 4-47. Lower-48 Wet Gas Production by Resource Type

0

200

400

600

800

1,000

1,200

1990 1992 1994 1996

YEAR

1998 2000 2002 2004

RIG

CO

UN

T

0

1

2

3

4

5

6

GU

LF

CO

AS

T S

PO

T P

RIC

E (

DO

LLA

RS

)

JUNE 2003

RIG COUNT = 900

PRICE = $5.60

AVERAGE 1990-1996

RIG COUNT = 400

PRICE = $1.75

AVERAGE 1997-1999

RIG COUNT = 540

PRICE = $2.20

AVERAGE 2000-2002

RIG COUNT = 780

PRICE = $3.65

GULF COAST SPOT PRICE

RIG COUNT

Figure 4-48. Lower-48 Gas Rig Count and Gulf Coast Spot Price

Sources: Baker Hughes (rigs) and Platts Gas Daily (prices).

Sources: GTI Gas Resource Database; Energy and Environmental Analysis, Inc.


0

200

400

600

800

1,000

1,200

1990 1991 1992 1993 1994 1995 1996

YEAR

1997 1998 1999 2000 2001 2002

0

4

8

12

16

20

24

AVERAGE 1990-1996

RIG COUNT = 400

COMPLETIONS/YEAR = 9,700

RIG COUNT

ANNUAL COMPLETIONS

Sources: Baker Hughes; and American Petroleum Institute.

GA

S W

ELL C

OM

PLE

TIO

NS

(T

HO

US

AN

DS

)

RIG

CO

UN

T

* AVERAGE 1997-1999

RIG COUNT = 540 (+35%)

COMPLETIONS/YEAR = 12,100 (+25%)

* * AVERAGE 2000-2002

RIG COUNT = 780

COMPLETIONS/YEAR = 19,300

* * *

Figure 4-49. Lower-48 Gas Rig Count and Gas Completions

0

30

35

40

45

50

55

1990 1992 1994 1996

YEAR

1998 2000 2002

PR

OD

UC

TIO

N (

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

0

200

400

600

800

1,000

1,200

GA

S D

RIL

LIN

G R

IGS

/GU

LF

OF

ME

XIC

O R

IGS

GAS DRILLING

RIGS

PRODUCTION

GULF OF MEXICO

DRILLING RIGS

2004

Figure 4-50. Lower-48 Total Dry Gas Production and Rig Count

Source: Energy and Environmental Analysis, Inc., GSR (production) and Baker Hughes (rigs).

Source: Baker Hughes (rigs) and American Petroleum Institute (completions).

of 400 was sufficient to sustain production peak ratesin the early 1990s, the industry has averaged muchhigher rig counts recently. The latter half of the 1990shas seen two complete industry drilling cycles (1997-1999 and 2000-2002). In the 1997-1999 ramp-up, pro-duction increased by 2 BCF/D as the gas rig count rosefrom 400 to 650. Following this peak, productiongradually fell off as activity levels declined. In the2000-2002 cycle, the rig count increased from 400 toover 1,050 rigs, nearly double the peak rates in 1997 toachieve an increase in production of a roughly similaramount. As prices fell and drilling slowed, productionfell dramatically, even with rig activity levels above thepeak rig count in 1996-1998.

Western Canada: Activity vs. Production

Since 1990, 65% of the incremental supply of NorthAmerican gas has come from increasing Canadian pro-duction, primarily from the Western CanadaSedimentary Basin. However, production growth inWestern Canada has slowed dramatically, so much sothat 2002 was the first year that the Western CanadaSedimentary Basin experienced declining production(Figure 4-51). In the early 1990s, as gas export infra-structure grew, Western Canadian production grew by4.5 BCF/D from 1990-1995 from an average of 3,000

gas well completions per year (Figure 4-52). Growthrates slowed through the rest of the 1990s, even as gaswell completions peaked at over 10,500 completions in2001, or over three times the completions in the begin-ning part of the decade. The 2001 and 2002 produc-tion rates were boosted by significant production fromthe Ladyfern Field in British Columbia, a field that isbeginning to decline.

Individual Gas Well Performance

Individual gas well performance was analyzed usingthree basic parameters: (1) Estimated UltimateRecovery, (2) Initial Production Rate, and (3) InitialDecline Rate.

Estimated Ultimate Recovery

Estimated Ultimate Recovery is an estimate of howmuch gas an individual gas connection will produceover its economic life. EURs generally decline as basinsmature, as the industry targets the larger, more eco-nomic prospects first. However, the EUR profile can bevery complicated, with natural EUR decline being mit-igated by a number of factors:

� Technology (e.g., 3D seismic opening deeper, under-explored parts of the basin)


0

5

10

15

20

1990 1992 1994 1996YEAR

1998 2000 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

0

5

10

15

20

GA

S W

ELL

CO

MP

LET

ION

S (

TH

OU

SA

ND

S)

PRODUCTION

ANNUAL COMPLETIONS

Source: TransCanada.

Figure 4-51. Western Canada Sedimentary Basin Production and Gas Well Completions

� Well mix (e.g., targeting shallow, low-risk exploitation wells vs. higher risk exploration wild-cats)

� Economics (e.g., increase in gas prices spurringrapid development of in-fill locations)

� Basin character, reserve type, and other items.

Figure 4-53 shows that during the 1990s the EURper average gas connection in the U.S. lower-48 fellfrom 1.4 BCF in 1990 to 1.2 BCF in 1999, a decline of15%. As drilling and completion activity increased sig-nificantly in 2000 and 2001, average EURs fell a further17% to just under 1 BCF.

In Western Canada, EURs have shown even a moremarked decline, falling from about 1.7 BCF in the early1990s, to 0.3 BCF in 2001, as annual completionsincreased from approximately 3,000 per year to over10,000 gas completions in 2001.

The Gulf of Mexico shelf showed a more rapidreduction in EURs than onshore. On the shelf, EURsfell 34% between 1990 and 1999 from 5.1 BCF to 3.3BCF (Figures 4-54 and 4-55). Judging from the lack of

recent rig response on the shelf, EURs from similarplays likely did not improve after 1999.

Initial Production Rates

In a period of falling EURs, the industry has beenable to partially compensate by accelerating individualwell production. As regulatory constraints on gas wellproduction were eased in the early part of the 1990s,the rapid application of completion and stimulationtechnology, combined with producer’s economic driveto lower R/P ratios, caused IPs to increase rapidly.Average gas well IPs increased from 1.1 million cubicfeet per day (MMCF/D) in 1990 to just under 1.6MMCF/D by 1996, as shown in Figure 4-56. AverageIPs remained near that level in the late 1990s andincreased to a peak in 1999 of 1.6 MMCF/D beforefalling in 2000.

On the Gulf of Mexico shelf, peak rates andplateau times reached their maximum in the 1996-1997 time frame (Figure 4-57). Since then, peak rateshave fallen marginally and plateaus have shortenednoticeably. Onshore, IPs rose rapidly to 1996, butrose only marginally through the latter half of the


PRODUCTION

GROWTH

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1991 1992 1993 1994 1995 1996

YEAR

1997 1998 1999 2000 2001 2002EST.

AN

NU

AL P

RO

DU

CT

ION

GR

OW

TH

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

0

2

4

6

8

10

12

14

GA

S W

ELL C

OM

PLE

TIO

NS

(T

HO

US

AN

DS

)

ANNUAL

COMPLETIONS

Source: TransCanada.

Figure 4-52. Western Canada Sedimentary Basin Production Growth and Gas Well Completions


0

0.4

0.8

1.2

1.6

2.0

1990 1994 2000

BIL

LIO

N C

UB

IC F

EE

T P

ER

GA

S C

ON

NE

CT

ION

0

4

8

12

16

20

GA

S W

ELL C

ON

NE

CT

ION

S P

ER

YE

AR

(TH

OU

SA

ND

S)

YEAR

LOWER-48 RECOVERY

LOWER-48 CONNECTIONS

WESTERN CANADA RECOVERY

CANADIAN CONNECTIONS

(EXCLUDING DEEPWATER GULF OF MEXICO, NONCONVENTIONAL GAS)

1992 1996 1998


Figure 4-53. Estimated Ultimate Recovery per Gas Well Connection

0

200

400

600

800

1,000

1990 1991 1992 1993 1994 1995

YEAR

1996 1997 1998 1999 2000 2001

GA

S W

ELL C

ON

NE

CT

ION

S

1.0

2.0

3.0

4.0

5.0

6.0

BIL

LIO

N C

UB

IC F

EE

T

AVERAGE ESTIMATED ULTIMATE

RECOVERY PER CONNECTION

ANNUAL CONNECTIONS


Figure 4-54. Gulf of Mexico Shelf Recovery per Gas Well Connection (Excludes Norphlet)



0

1

2

3

4

5

6

7

0 1 2 3 4 5

BILLION CUBIC FEET

MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

19901991199219931994199519961997199819992000


Figure 4-55. Gulf of Mexico Shelf Average Daily Gas Well Productionvs. Cumulative Production, by Year of First Production

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

MONTHS

0 10 20 30 40 50 60 70

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000


Figure 4-56. Lower-48 Conventional Average Daily Gas Well Production vs. Time, by Year of First Production


0

1

2

3

4

5

6

7M

ILLIO

N C

UB

IC F

EE

T P

ER

DA

Y

MONTHS

0 10 20 30 40 50 60 70

19901991199219931994199519961997199819992000


Figure 4-57. Gulf of Mexico Shelf Average Daily Gas Well Productionvs. Time, by Year of First Production

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

MONTHS

0 10 20 30 40 50 60 70

199019911992199319941995199619971998199920002001


Figure 4-58. Lower-48 Onshore Conventional Average Daily Gas Well Productionvs. Time, by Year of First Production

decade (Figure 4-58). IPs fell marginally in 2000, andthen more substantially during the 2001 drillingramp-up.

One of the drivers of the increase in IPs has beenthe increasing application of fracture stimulationtechnology. More wells are being fracture stimulatedand often with larger stimulations. Stimulation tech-nology has also advanced, allowing longer fracturelengths to rapidly drain larger areas of a tight reser-voir. In East Texas, for example, while a significantpercentage of wells were fracture stimulated in theearly part of the decade, that percentage increaseduntil almost 100% of completions are now fracturestimulated as shown in Figure 4-59. East Texas ischaracterized by tight reservoirs, but many otherbasins are also reaching a high level of utilization.The trend of increasing IPs from fracture technologythat was enjoyed through the middle of the 1990s willlikely not be continued.

Initial Decline Rates

As higher IPs have increasingly been bringing pro-duction forward, and EURs have been falling, declinerates have been progressively steepening, as shown inTable 4-4. While both the onshore and Gulf of Mexicoshelf have witnessed increasing decline rates, the effecthas been more pronounced on the Gulf of Mexicoshelf, with its rapidly falling EURs.

Base Decline Rates

As the industry has continued to add more andmore high decline wells to base production, overall

1990 1998 2000

1st Year % of 1st Year % of 1st Year % of Decline EUR Decline EUR Decline EUR

(%) Produced (%) Produced (%) Produced

E. Texas/ N. Louisiana 40 22 61 25 64 N.A.

South Texas Gulf Coast 41 27 62 34 67 N.A.

Anadarko Basin 28 12 52 21 58 N.A.

Permian Basin 40 16 37 17 53 N.A.

Gulf of Mexico Shelf (w/ Norphlet) 30 28 53 48 74 N.A.

Rockies (non-Coal Bed Methane) 38 12 44 16 64 N.A.


Table 4-4. Gas Well Decline Rates

0

20

40

60

80

100

120

1990-

1995

1996-

1998

1999

YEAR

2000 2001

CO

MP

LE

TIO

NS

(P

ER

CE

NT

)


Figure 4-59. Percentage of Completions in East Texas that were Fracture Stimulated

Source of data: IHS Energy Group.



0

10

20

30

40

50

60

1992 1994 1996 1998 2000 2002

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y


YEAR

19991998199719961995199419931992

19911990

2002

20012000

Figure 4-60. Lower-48 Daily Wet Gas Production from Gas Wells, by Year of Production Start

0.0

-2.0

-4.0

-6.0

-8.0

-10.0

-12.0

-14.0

0

-5

-10

-15

-20

-25

-30200120001999199819971996

YEAR

1995199419931992

-27%

-26%

-22%

-25%-25%

-21%-20%

-18%-19%

-17%

EQ

UIV

ALE

NT

PR

OD

UC

TIO

N L

OS

S

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)


DE

CLIN

E R

AT

E (

PE

RC

EN

T)

Figure 4-61. Lower-48 Base Gas Production Decline Rate if No New Wells had been Drilled,and Equivalent Production Loss



decline rate has increased. In 1992, the base declinerate was 17%. To simply hold production flat, the 1992gas drilling program needed to replace deliverability of8 BCF/D.

Over the period, the base decline rate has increasedas shown in Figures 4-60 and 4-61, and perhaps moreimportantly, the amount of production from new gaswells required to maintain production levels has dra-matically increased. To keep production flat in 2000and 2001, the gas-drilling program had to replace overa quarter of production, or almost 13 BCF/D.Compared to just 10 years ago, the recent yearlydrilling programs have had to replace an additional 4-5 BCF/D just to maintain production levels more than50% higher than earlier levels.

Proved Reserves

During the 1990s, the overall lower-48 provedreserve base has remained remarkably consistent,beginning the decade at 158 TCF of dry gas and finish-ing the decade at 158 TCF of gas. Over that periodlower-48 gas production totaled 177 TCF. Accordingly,the industry proved 177 TCF, or about 18 TCF/year.

Reserve additions in 2000 and 2001 were largerthan the historical average. In a period in which 39 TCF of gas were produced, gas reserve additionstotaled 57 TCF of gas in 2000-2001, an average of28 TCF/year, or about 10 TCF/year higher than theindustry averaged in the 1990s. While the industrywas able to increase total proved reserves, provedproducing reserves actually fell, as shown in Figure4-62. At year-end 2001, non-producing reservestotaled just under 53 TCF of gas, up 24% from just ayear earlier. Total R/P increased from 8.50 in 2000 toan estimated 9.35 in 2002, but producing R/P hasremained steady.

The two regions that have exhibited the strongestreserve growth, the Rocky Mountains and EastTexas/North Louisiana have a significant percentage ofnonconventional reserves. Basins with only conven-tional reserves either showed declines or only modestincreases.

In Western Canada, the industry has not been able toreplace production. In the 1991-2002 period, provedreserves and R/P have steadily fallen from 70 TCF ofreserves with an 18 R/P in 1991 to an estimated 57 TCF


0

20

40

60

80

100

120

140

160

180

200

1996 1997 1998 1999

YEAR

2000 2001 2002

TR

ILLIO

N C

UB

IC F

EE

T

0.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

RE

SE

RV

ES

-TO

-PR

OD

UC

TIO

N R

AT

IO (

R/P

)

RESERVES PRODUCING RESERVES R/P PRODUCING R/P

Sources: Energy Information Administration; and Energy and Environment Analysis, Inc., Gas Supply Review (GSR).

ESTIMATED

Figure 4-62. Lower-48 Wet Gas Proved Reserves


of reserves with an R/P of approximately 9 in 2002.This is illustrated in Figure 4-63.

Regional Production Summaries

In an overall environment of slowing productiongrowth, individual producing regions have followedtheir own unique production profiles, as basins havebeen explored and developed, matured, and new tech-nologies have been applied to the resource.

As Figure 4-64 demonstrates, in the more matureregions of the Gulf of Mexico shelf and the matureonshore areas of the U.S. lower-48, overall flat produc-tion levels were maintained in the early part of thedecade. After 1996, as these areas continued to mature,they began a sustained decline in production, whichwas slowed by the big drilling ramp-up in the 2000-2001 time frame.

In the frontier areas of the U.S. lower-48 andCanada, technological advances and infrastructureconnections have opened up less mature opportuni-ties. Production from the Rockies and deepwater Gulfof Mexico has grown from 5.0 BCF/D in 1990 to 14.2BCF/D in 2002. While increases in these regions con-

tributed to an overall increase in production in theearly part of the 1990s, declining production in themature regions has more recently more than offsetthese increases.

Declining Regions

Gulf of Mexico Shelf. The Gulf of Mexico shelfbegan the 1990s as the largest producing region inNorth America, with peak production rates of 14BCF/D. While shelf production held fairly steady inthe mid-1990s, after 1996 the shelf began declining ata rate of 0.8-1 BCF/D per year as shown in Figure 4-65. The reason was that new drilling could notkeep up with the rapidly declining EURs and steepdecline rates as the shallow, 3D seismic-driven“bright spot” play rapidly matured. While thedrilling ramp-up of 2000-2001 flattened the declineon the shelf, the shelf appears to have resumed itsrapid decline as drilling rates have once again begunto fall. By the end of 2002, the shelf had become thethird largest producing region in North America,behind the Western Canada Sedimentary Basin andthe Rocky Mountains. However, industry has recent-ly begun to explore and develop deeper prospects on


0

10

20

30

40

50

60

70

80

1991 1992 1993 1994 1995 1996

YEAR

1997 1998 1999 2000 2001 2002

TR

ILLIO

N C

UB

IC F

EE

T

0

2

4

6

8

10

12

14

16

18

20

RE

SE

RV

ES

-TO

-PR

OD

UC

TIO

N R

AT

IO

Source: Canadian Association of Petroleum Producers.

EST.

Figure 4-63. Western Canada Sedimentary Basin Proved Reserves

the shelf, which if successful could help reduce futureproduction decline.

Eastern Gulf Coast. After rising marginally in theearly 1990s, the Eastern Gulf Coast has been decliningsince about 1996, with peak production falling over 1BCF/D from a peak of 4.7 BCF/D in 1996 to 3.5 BCF/Dat the end of 2002.

Midcontinent. The Midcontinent region started thedecade as the third largest producing region in NorthAmerica (second in U.S. lower-48) at peak rates of

10 BCF/D. Production has gradually fallen to less than6.5 BCF/D currently as EURs have steadily declinedthroughout the period.

Permian Basin. Peak gas production in the PermianBasin has slowly dropped from 4.3 BCF/D in 1990 to thecurrent production rate of approximately 3.8 BCF/D.

Holding Steady/Slightly Increasing Regions

South Texas Gulf Coast. In the early-mid part ofthe 1990s, production increased from a peak rate of


0

2

4

6

8

10

12

14

16

18

20

1990 1991 1992 1993 1994 1995 1996

YEAR

1997 1998 1999 2000 2001 2002

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

MIDCONTINENT

S. TEXAS GULF COAST


DEEPWATER GULF OF MEXICO

ROCKIES

WESTERN CANADA

SEDIMENTARY BASIN

EASTERN CANADA

OTHERE. TEXAS/N. LOUISIANA

PERMIAN BASINE. GULF COAST

HISTORICAL PROJ.

2003

Figure 4-64. North American Gas Production by Region



0

2

4

6

8

10

1990 1991 1992 1993 1994 1995

YEAR

1996 1997 1998 1999 2000

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

0 - 12,000 FT

12 - 15,000 FT

> 15,000 FT


Figure 4-65. Gulf of Mexico Shelf Production by Depth

0.0

2.0

4.0

6.0

8.0

1990 1991 1992 1993 1994 1995 1996

YEAR

1997 1998 1999 2000 2001 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

GA

S W

ELL C

ON

NE

CT

ION

S (

TH

OU

SA

ND

S)

0

1

2

3

4

GAS PRODUCTION

ASSOCIATED GAS PRODUCTION

ANNUAL CONNECTIONS


Figure 4-66. South Texas Gulf Coast Production and Gas Well Connections

5.8 BCF/D in 1990 to 7.1 BCF/D in 2001 (Figure 4-66), an increase of 1.3 BCF/D as regional 3D seismiccoverage allowed deeper prospects and smaller, shal-low prospects to be more accurately imaged andexploited. As the basin matured, growth slowed in thelater part of the 1990s. Production declined to 6.3BCF/D by the end of 2002, down almost 1 BCF/Dfrom its maximum.

East Texas/North Louisiana. After holding steadythroughout much of the 1990s, production has grownfrom 3.5 BCF/D in 1999 to 4.6 BCF/D by the end of2002 as nonconventional tight sands of the CottonValley Formation (Figure 4-67) and shales of theBarnett Shale (Figure 4-68) were exploited usingapplied fracture stimulation technology. While drillinghas ramped up substantially, average well productivityhas also held generally flat so that recently producershave been able to generate sustained increases in pro-duction.

Increasing Production Regions

Western Canada Sedimentary Basin. As gas exportinfrastructure was expanded in the early 1990s, explo-

ration and development interest picked up rapidly inthis region. The Western Canada Sedimentary Basinhas grown to be the largest producing region in NorthAmerica. While the basin grew strongly in the early1990s, growth has slowed considerably as the basinrapidly matured. Average well productivity has fallendramatically (Figure 4-69) and basin decline rates havesteepened (Figure 4-70). In recent history, 2002 wasthe first year of flat to declining production in WesternCanada.

Rocky Mountains (including the San Juan Basin).Production from the Rocky Mountains has grownsteadily throughout the decade (Figure 4-71), evenwith periods of low regional prices, and the Rockiesare currently the second largest producing region inNorth America (first in U.S. lower-48). While muchof the Rockies growth has come from nonconven-tional resources (Figure 4-72), both conventionaland nonconventional production rates have beenincreasing.

Deepwater Gulf of Mexico. 3D seismic technolo-gy and advancing development/production technol-ogy have opened this frontier area to drilling. While


0.0

0.5

1.0

1.5

2.0

1997 1998 1999 2000

YEAR

2001 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

0

25

50

75

100

MO

NT

HLY

GA

S W

ELL C

ON

NE

CT

ION

S


PRODUCTION

CONNECTIONS

Figure 4-67. Cotton Valley Production and Gas Well Connections


0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

CUMULATIVE PRODUCTION (BILLION CUBIC FEET)

PR

OD

UC

TIO

N R

AT

E

(MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

1990-95

1996-98

1999

2000

2001


Figure 4-69. Western Canada Sedimentary Basin Production Rate vs. Cumulative Production

0.0

0.25

0.50

0.75

1.00

0

20

40

60

80

1997 1998 1999 2000

YEAR

2001 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

MO

NT

HLY

GA

S W

ELL C

ON

NE

CT

ION

S


PRODUCTION

CONNECTIONS

Figure 4-68. Barnett Shale Production and Gas Well Connections


0

2

4

6

1990 1992 1994 1996

YEAR

1998 2000 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

GA

S W

ELL C

ON

NE

CT

ION

S (

TH

OU

SA

ND

S)

0

2

4

6

PRODUCTION

ANNUAL CONNECTIONS


Figure 4-71. Rockies Production and Gas Well Connections, Excluding Coal Bed Methane

0

0.2

0.4

0.6

0.8

1.0

1.2

1990 1992 1994 1996 1998 2000 2002

VINTAGE YEAR

INIT

IAL

RA

TE

(M

ILLI

ON

CU

BIC

FE

ET

PE

R D

AY

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

ES

TIM

AT

ED

ULT

IMA

TE

RE

CO

VE

RY

(BIL

LIO

N C

UB

IC F

EE

T)

INITIAL RATE

INITIAL DECLINE RATE

EUR


0

10

20

30

40

50

60

INIT

IAL

DE

CLI

NE

RA

TE

(P

ER

CE

NT

PE

R Y

EA

R)

Figure 4-70. Western Canada Sedimentary Basin Production Performance Trends


0

2

4

6

8

10

12

14

16

1990 1992 1994 1996

YEAR

1998 2000 2002

GA

S P

RO

DU

CT

ION

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

GA

S W

ELL

CO

NN

EC

TIO

NS

0

200

400

600

800

1,000

1,200

1,400

1,600

ANNUAL CONNECTIONS

HISTORICAL PROJ.TOTAL GULF OF MEXICO

SHELF


DEEPWATER

Figure 4-73. Gulf of Mexico Shelf and Deepwater Production and Connections

1990 1992 1994 1996

YEAR

1998 2000 2002

0

2

4

6

0

2

4

6

ANNUAL CONNECTIONS

PRODUCTIONCO

AL B

ED

PR

OD

UC

TIO

N

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

CO

AL B

ED

CO

NN

EC

TIO

NS

(TH

OU

SA

ND

S)


Figure 4-72. Rockies Coal Bed Methane Production and Connections

primarily an oil play, gas production has grownstrongly as new developments and infrastructure arebeing built out (Figure 4-73). While shelf produc-tion has been falling rapidly, increased deepwaterproduction has been able, until recently, to sustaintotal Gulf of Mexico production at approximately13-14 BCF/D.

Future production growth will depend largely onwhether the industry can sustain the recent pace ofadding proved reserves and developing large fields,which typically require high-cost, long lead-time proj-ects. The pace of development will also be moredependent on oil-price driven economics than naturalgas economics.

2000-2001 Drilling Campaign

The 2000-2001 drilling campaign saw gas rigactivity increase from a low of 400 rigs in 1999 toover 1,050 rigs in 2001, which was essentially 100%of rig capacity. This was almost double the peak rigrate in 1997-1998. However, the productionresponse was similar, up approximately 2 BCF/D.When drilling slowed to average 700 rigs in 2002,still above the peak drilling rates in 1997-1998, pro-

duction fell dramatically, rather than rising. Whatwas the difference?

� The resource base has continued to mature. Averagewell EUR has been on a long-term decline, and thedrilling campaign of 2000 and 2001 only acceleratedthis trend.

� Marginal drilling was dominated by low productivi-ty wells. In terms of first year buildup, the onshorebasins, with the exception of East Texas, showedaverage first year buildup falling 15% to 25% (Figure4-74).

� The majority of gas completions occurred in low rate,high R/P areas such as the Powder River Basin coalbed methane wells (Figures 4-75, 4-76, and 4-77).

� While technology had allowed industry to sustainproduction at lower activity levels in the mid-1990s by accelerating well production, these tech-nologies had largely matured and reachedsaturation.

� As individual well production was accelerated, basedeclines steepened over the period. It took 8 BCF/D


0 0.1 0.2 0.3 0.4 0.5 0.6

SOUTH TEXAS

ROCKIES

MIDCONTINENT

PERMIAN BASIN

EAST TEXAS

FIRST YEAR DELIVERABILITY (BILLION CUBIC FEET)

1999

2001

Figure 4-74. First Year Production – 1999 and 2001


of new production to replace base declines in theearly part of the 1990s. That had increased by over50% to almost 13 BCF/D.

� Higher gas prices made it possible to drill lowerquality prospects.

� As rig activity increased, rig efficiency fell as meas-ured in completions per rig-year and footage perrig-year.

Model Calibration

Production performance parameters generated inthis analysis were either used as inputs (either directlyor as directional trends) to the HSM model, or wereused to check HSM outputs. For example, the pro-duction profile of the HSM model’s Proved Reservebase was compared with the actual decline of pre-1998completions. As historical individual gas well per-formance parameters were generated for each region,depth interval, and production type, they werechecked against conventional decline analysis andused to provide a check of future well performanceparameters.


0 200 400 600

FIRST YEAR CALCULATED BUILDUP

(MILLION CUBIC FEET PER DAY)

GULF OF MEXICO

SOUTH TEXAS

EAST TEXAS AND NORTH LOUISIANA

PERMIAN BASIN

ROCKIES (EXCL. POWDER RIVER BASIN)

MIDCONTINENT

POWDER RIVER

Figure 4-77. Incremental Drilling Buildupby Region – 2001 vs. 1999

Source: Energy and Environmental Analysis, Inc., GSR.

0 1 2 3 4 5

AVERAGE FIRST YEAR RATE

(MILLION CUBIC FEET PER DAY)

GULF OF MEXICO

SOUTH TEXAS

PERMIAN BASIN



MIDCONTINENT

POWDER RIVER

Figure 4-76. First Year Well Production by Region


0 1,000 2,000 3,000

GULF OF MEXICO

SOUTH TEXAS

PERMIAN BASIN



MIDCONTINENT

POWDER RIVER

INCREMENTAL CONNECTIONS

Figure 4-75. Incremental Drilling by Region – 2001 vs. 1999



Technology Improvements

The Technology Subgroup under the Supply TaskGroup was formed with representation from varioussegments of the oil and gas industry to assess the roleand impact that technology will have on natural gassupply in North America. Several workshops andmeetings were organized to provide a forum for indus-try experts to discuss the role that current and futuretechnology will play in sustaining the supply of naturalgas. From this process, a forecast of technologyimprovement parameters was developed for input intothe natural gas supply model used for the study. Also,sensitivity cases were run to assess the effects of a rangeof high and low rate of advancement of technologydevelopment and application. For the purpose of thestudy, technology was defined broadly as any new orimproved product, process or technique that enhancesthe overall result compared with the current resultsobserved today.

Technology Subgroup Process for the Study

Scope

The Technology Subgroup was established to pro-vide insights into the role and impact of upstreamtechnology in delivering natural gas supply during thestudy period. Composed of thirteen members from across-section of industry organizations, the Techno-logy Subgroup determined its scope to be:

� To design a methodology for measuring the impactof future technologies in the econometric model

� To estimate the technology improvement parame-ters for the scenarios developed and a range of sen-sitivity cases

� To compose an upstream technology commentaryfor the final report that provides a current-stateindustry view of research and development, itsimpact on the outlook, and the role of technology inthe future deliverability of North America naturalgas through the year 2025

� To recommend actions that will facilitate the use ofnew technologies to improve the economics andincrease the future deliverability of natural gas.

Workshops and Special Technology Sessions

To achieve these goals, the Technology Subgroupscheduled a series of workshops providing a forum tounderstand previous studies, provide input into the

supply model, and prepare the report. In addition tothe workshops, six special technology sessions wereheld to discuss with industry experts specific issuesrelated to core, high-impact technology areas. Theselected technology areas were Coal Bed Methane,Drilling, Completions, Subsurface Imaging/Seismic,Deepwater Development, and Natural Gas Hydrates.Held in January and February 2003, these special ses-sions enabled the Technology Subgroup to hear theviews and foresights of a large cross-industry expertcommunity, a total of 128 in all. These sessions werehelpful in assessing the effect of technology on futuresupplies. They also helped to improve the quality oftechnology input parameters for the econometricmodel.

Methodology for Developing TechnologyImprovement Parameters for the Model

The Technology Subgroup reviewed the economet-ric model to understand how technology improvementis factored in the supply model. Some of the membersattended the various regional resource assessmentsworkshops to gain an understanding of the technolo-gies applied and challenges ahead. The subgroup thenreached consensus on the technology improvementparameters for the Reactive Path scenario for eachassessment region and in some cases, by type of reser-voir. The technology improvements parameters devel-oped for input into the supply model are as follows:

� Exploration success – annual percent improvementin the ratio of completed versus non-completedexploration wells

� Development success – annual percent improvementin the ratio of completed versus non-completeddevelopment wells

� Estimated ultimate recovery per well – annual per-cent improvement in the estimated ultimate recov-ery (EUR) of natural gas per well

� Drilling cost – annual percent improvement indrilling costs per well including site preparation, rigmobilization, drilling, and installing casing

� Completion cost – annual percent improvement inthe completion cost per well, including perforating,stimulating, and installing down-hole productionequipment

� Initial production rate per well – annual percentimprovement in the initial production rate estimat-ed in the model for each well completed


� Infrastructure costs – annual percent improvementin the major surface infrastructure costs associatedwith the development of new fields, such as offshoreplatforms, sub-sea production and gathering sys-tems, field processing plants, and field gathering lines

� Fixed operating expenses – annual percent improve-ment in the operating expenses associated with theproduction of natural gas.

Three time periods were used to forecast the tech-nology improvement parameters where technologyapplication would likely change through the studyperiod. They are as follows:

� First five-year period – 2003-2008

� Second seven-year period – 2009-2015

� Third ten-year period – 2016-2025.

Furthermore, two cases of improvement parameters,beyond the Reactive Path scenario, were generated tocreate a range of possible outcomes for technologyimpact in the supply model. The two cases were:

� High pace of technology advancement and applica-tion

� Low pace of technology advancement and applica-tion.

After reviewing the model runs, the input parame-ters were then checked for reasonableness and consis-tency with the expectations described during thediscussions at the workshops and special sessions.Some modifications were made for the final modelruns.

Historical Perspective ofTechnology Contributions

The Technology Subgroup reached the consensusthat technology has historically contributed signifi-cantly to the ability of the petroleum industry to find,develop, and produce natural gas resources. If theindustry relied on the same tools and methodologiesused thirty years ago, it would only be able to producea small fraction of what is currently being producedtoday. How much of an impact technology has had isdifficult to precisely determine, because the industrydoes not measure the impact of technology directly.However, one can find indirect evidence of technolo-gy’s impact by looking at cost trends or productionperformance trends in any given area or field. Also,there is indirect evidence that identifies improvement

in the ability of the industry to explore for natural gas.Most evidence, however, is anecdotal.

Projected Technology Improvements

Even with the noted technology advancements, overthe last ten years, investments in upstream researchand development have declined and the industry hasbeen cautious in using high-cost, high-risk technolo-gies regardless of their potential. This reluctance isparticularly evident if the technology is perceived tohave a longer-term impact. With this observation andthe maturity of the exploration and production envi-ronment, the subgroup postulated that technologywill play a somewhat lesser role in gas resourceenhancement in the near future. Technology will gainslight momentum beyond five years as the industryinvests more in technology developments, motivatedby the challenges of the resources and higher gasprices. This is not intended to imply that there willnot be continued improvements. Indeed, there will becontinued improvements in both tools and tech-niques, but there are no foreseeable major break-throughs on the horizon.

With this back-drop, the Technology Subgroupdeveloped a series of technology improvement param-eters for the Reactive Path scenario in the supply modelthat reflect the anticipated rate of improvement in eachmajor core technical area of application.

Different improvement parameters were determinedfor each major region and in some instances, for eachtype of reservoir, as for example coal bed methane ordeep, high-temperature, high-pressure reservoirs.Also, to reflect the anticipated behavior of the industry,different improvement parameters were adopted foreach of the different time periods, 2003-2008, 2009-2015, and 2016-2025+. The consensus of the membersof the Technology Subgroup was that for most of thetechnical areas and geologic regions, the later timeperiods would probably see a faster pace of improve-ment than the early time period.

The actual technology improvement parametersused in the Reactive Path supply model are provided inthe full report appendices. However, in order to get asense of the magnitude of these improvement parame-ters, Table 4-5 summarizes the improvement parame-ters by averaging them for each parameter.

The values shown in Table 4-5 were not calculatedfrom any theory or formula. Instead, the values were

determined by the Technology Subgroup, using allavailable information and insights generated duringthe study. The parameters were based more on collec-tive experience and intuition, than on theory.However, the Technology Subgroup agreed that theparameters seemed reasonable given all of the discus-sions developed at the workshops and special technol-ogy sessions.

It was appropriate to also look at a range of param-eters that reflect a high and low pace of technologyadvancement and application. The TechnologySubgroup developed parameters for these two addi-tional cases assuming various factors that influence thepace of developing and applying new technologies.These parameters are described in the full report.

Issues and Challenges

Several issues and challenges will face the NorthAmerican petroleum industry and governments asthey pursue research, development, and applicationof new technologies to enhance the supply of natu-ral gas.

Although many of the North American producingbasins are maturing, significant technically recoverableresources still remain. Higher prices will support tech-

nology development; however, declining reserves maymake it difficult to justify major investments in newtechnology. Majors, independent companies, and theservice industry will all play a role in support of therequired technology development.

Industry must also improve processes for the accept-ance and utilization of new technology. The shifttoward more collaborative research increases the diffi-culty of testing and deploying new technologies.Professional societies, trade associations, and academicand government research institutions, along with theindustry will need to increase efforts to communicateand work together to effectively deploy new applica-tions.

Another challenge will be to effectively transfer theknowledge and experience of the existing professionalworkforce near retirement to the new generation enter-ing the industry and research institutions. Otherwise,the risk of “reinventing the wheel” will loom over theindustry.

With the expected tight supplies of natural gas,potentially higher prices, and ever-increasing technicalchallenges, the petroleum industry, research institu-tions, and governments need to adopt innovative tech-nology strategies to respond to these challenges.


Table 4-5. Technology Improvement Parameters for the Reactive Path Scenario Supply Model

% Improvement % Annual Extrapolated

Technology Area Improvement* for 25 Years

Improvement in Exploration Well Success Rate 0.53 14

Improvement in Development Well Success Rate 0.46 11

Improvement in Estimated Ultimate Recovery per Well 0.87 24

Drilling Cost Reduction 1.81 37

Completion Cost Reduction 1.37 29

Improvement in Initial Production Rate 0.74 20

Infrastructure Cost Reduction 1.18 26

Fixed Operating Cost Reduction 1.00 22

* These numbers reflect the average of the parameters, not the actual parameters used in the supply model.

Key Findings

The key technology findings are as follows:

� Technology improvements play an important rolein increasing natural gas supply. During the lastdecade, 3D seismic, horizontal drilling, and improvedfracture stimulation have had significant effects onnatural gas production in many basins in NorthAmerica. Also, due to advanced designs in deepwaterdevelopments, additional production from the Gulfof Mexico slope regions has been realized.

In addition to these step-change technologies, con-tinued improvements in core technical areas havebeen implemented as a result of industry’s continu-ing efforts to search for more cost-effective ways tofind, develop and operate oil and gas fields. Thistrend is especially evident in the production of non-conventional gas reservoirs such as coal bedmethane, shale gas and tight sand formations. Newdesigns in drilling bits, improved well planning andmodern drilling rigs have also lowered drilling costsin many regions. Advances in remote sensing, infor-mation technologies and data integration tools haveserved to keep operating expenses in check.

As modeled in the Reactive Path scenario and illus-trated in Figure 4-78, by the year 2025, advancedtechnologies contribute 4.0 TCF/year of the 27.8TCF/year produced in the United States andCanada. This amounts to 14% of the natural gasproduced during that year.

� Adding new North American natural gas supplieswill require finding, developing and producingmore technologically challenging resources thanever before. Overall, when assessing the natural gasresources that will be found and developed over thenext 25 years, they can be generally described asdeeper, hotter, tighter, more remote, in deeper waterand smaller, harder-to-find prospects. The combina-tion of more difficult natural gas resources and high-er prices of natural gas should catalyze increasedefforts in research, development and application ofnew technologies by the industry and governments.

� Investments in research, development, and appli-cation of new technology have declined over thelast 10 years. Although it is difficult to obtain infor-mation concerning how much the total oil and gasindustry spends on technology improvementsfocused on North American natural gas assets, overthe last decade the trend in upstream research anddevelopment spending has been downward, as

reported by the U.S. major energy producersthrough the EIA (see Figure 4-79).

Forecasting future technology investment is diffi-cult. As a result, the implication of technologyimprovements on production and prices are cast interms of a range of outcomes as shown in Figures 4-80 and 4-81. The low advancement sensitivitycase reflects a slower pace of technology develop-ment and application caused by reduced investmentin research. The high advancement case reflects afaster pace of technology development and applica-tion. The improvement parameters developed forthese sensitivity cases are described in theTechnology section of the Supply Task GroupReport.

Service industries and joint-sponsored research pro-grams are playing an increasing role in research anddevelopment. This can be viewed as a cost-effectiveand less redundant method for research. It may alsohave the effect of slowing down the application ofthe new technologies, and diminishing focus onlong-term or high-risk research. In many cases,long-term or high-risk research has been undertak-en through joint industry and/or government-sponsored programs.


0

22

24

26

28

30

32

2000 2005 2010

YEAR

2015 2020 2025

TR

ILL

ION

CU

BIC

FE

ET

WITH NO TECHNOLOGY

ADVANCEMENT

WITH TECHNOLOGY

ADVANCEMENT

Figure 4-78. Impact of Technology on U.S. and Canadian Natural Gas Production

� The gas exploration and production industryshould collaborate more effectively with theDepartment of Energy on the planning and exe-cution of complementary, not competitive,research and development programs. TheDepartment of Energy plays an important role infacilitating and sponsoring joint research anddevelopment programs within the gas supplyindustry.

During fiscal year 2003, the Department of Energyplans to fund $47.3 million towards jointly spon-sored natural gas technology research and develop-ment programs. This represents 53% of the fundingallocated by DOE to sponsor oil and gas R&D pro-grams but only 9% of the total $529.3 million fundsdirected at fossil energy programs. As a comparison,coal research attracts $349 million in DOE funding.With the new insights developed from this study, theDepartment of Energy should address the obviousquestion of whether the current funding leveltowards natural gas research is appropriate in rela-tionship to other R&D programs and the increasingchallenges facing the new natural gas resources with-in the United States. Figure 4-82 shows R&D pro-gram funding in 2002.

In addition to the question concerning the level ofR&D funding by the DOE, another important issueis whether the funds are focused on the right naturalgas technologies. The DOE’s role is to support thepublic interest in technology pursuits that the indus-try is not adequately addressing. It is thereforeessential that effective communication and collabo-ration exist between the DOE and the industry’stechnology developers to accomplish this goal andprevent duplication. This is not an easy task sincethe developers are split among many entities, such asnational labs, sponsored research organizations, gasproducers, service companies, consultants, and uni-versities. In addition, it is critical to have effectivecollaboration and communication by technologyusers to ensure mutual understanding of the prob-lems to be solved and how effective application canbe achieved.

Service companies have been hesitant to participatein jointly funded DOE-industry projects for thedevelopment and demonstration of advanced tech-nology, assuming incorrectly that their proprietarytechnologies would be made public. The Depart-ment and the service industry need to increasetheir discussions regarding future technology direc-tions to ensure that the two do not duplicate effortsand to increase the opportunities for service compa-nies to participate in government-supported tech-nology development.

� Environmental and safety considerations are sig-nificant drivers in the development and applica-tion of new technologies. The oil and gas industrycontinues to focus a significant amount of technol-ogy development effort to address environmentalconcerns and reduce potential safety issues in thefield. In some cases, these new technologies andapproaches also contribute to improved operationalperformance. As an example, new smaller modularrig designs to reduce the environmental footprintalso reduce downtime for rig moves improving eco-nomics. Drilling and completing multi-lateral andlong-reach horizontal wells reduce the number ofwell locations for equivalent reservoir drainage andsimultaneously increase the recovery per well.Environmentally compatible drilling and comple-tions fluids may reduce the cost associated with zerodischarge requirements in certain sensitive areas.Designing rigs and equipment to reduce safety haz-ards such as manual pipe handling can also improvethe drilling efficiencies and shorten downtime while


0

300

400

500

600

700

800

900

1990 1995 2000

MIL

LIO

NS

OF

DO

LLA

RS

Source: Energy Information Administration,

Form 28 Companies.

YEAR

Figure 4-79. Upstream R&D Expenditure History


0

22

24

26

28

30

32

2000 2005 2010 2015 2020 2025

YEAR

TR

ILLIO

N C

UB

IC F

EE

T

NO

ADVANCEMENT

LOW

ADVANCEMENT

HIGH

ADVANCEMENT

REACTIVE PATH

Figure 4-80. Impact of Technology Change on U.S. and Canadian Natural Gas Production

-2.00

-1.50

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

2000 2005 2010 2015 2020 2025

DO

LLA

RS

PE

R M

ILLIO

N B

TU

HIGH ADVANCEMENT

LOW ADVANCEMENT

YEAR

REACTIVE PATH

Figure 4-81. Impact of Technology Change on Gas Price at Henry Hub (2002 Dollars)

tripping in and out of the hole. As the industry andgovernment regulatory agencies search for accept-able methods to access new areas and reduce costs ofcompliance with environmental and safety regula-tions, these advances in technologies may enablebalanced solutions.

Natural Gas Hydrates

The Technology Subgroup had the charge to investi-gate the technologies associated with natural gashydrates and determine the feasibility of their contri-bution in the Hydrocarbon Supply Model. To gatherdata on the subject of natural gas hydrates, a workshopwas held on January 28, 2003, in Houston, Texas. Theobjectives of the workshop were to (1) determinewhether the production of natural gas from natural gashydrate deposits would be feasible between now andthe year 2025, and (2) identify the technologies thatwould be required to produce natural gas from naturalgas hydrate deposits in North America.

The following conclusions were derived from thespecial session.

� Natural gas production from naturally occurring gashydrate deposits should not be included as a majorsource of gas production in the NPC gas supply

forecast before the year 2025. Their contribution asa significant supply of natural gas is anticipatedbeyond 2025.

� Production from natural gas hydrate deposits in thedeepwater Gulf of Mexico and other deepwater areasaround North America will depend on both thedevelopment of significant technology advance-ments and infrastructure availability. Technologydevelopment will depend on the level of both gov-ernment and private industry funding.

� Production from natural gas hydrate deposits in theArctic areas of Alaska and Canada will depend pri-marily upon available pipeline capacity. If commer-cial production of gas hydrates is determined to befeasible, it is more likely to be a source of fuel usedin the Arctic oil and gas field operations.

Synthetic Gas/Coal Gasification

Synthetic gas or syngas, a mixture of hydrogen andcarbon monoxide, was previously known as “town gas”and was used in many domestic and commercial appli-cations. It was largely displaced after the explorationand production industry developed cheap natural gassupplies last century. Current equipment, systems andinfrastructure are now designed for natural gas, withwhich syngas cannot be blended in existing infrastruc-ture. However, it can displace natural gas where anentire system is converted or designed to use syngas.

Historically, syngas was typically generated fromcoal, but current technology allows almost any hydro-carbon to be gasified. Gasification produces clean syn-gas and leaves contaminants concentrated in an easilyhandled slag. It is therefore often used to convert low-value, impure hydrocarbons such as refinery bottoms,coal, and petroleum coke into a useful product in anenvironmentally sound way.

The major syngas uses are as fuel, for instance inboilers or gas turbines, to generate hydrogen and asfeedstock to make chemicals. As a fuel, it can normal-ly be blended with natural gas.

At high gas prices, coal gasification begins to lookeconomically attractive and may have some applicationas an alternative fuel source in future power plantsbuilt in North America. At lower gas prices, it is almostcertainly uneconomic although it may suit niche appli-cations (e.g., where environmental issues are impor-tant) and geographic areas where suitable feedstock ischeap (e.g., in coal mining areas or near refineries pro-ducing coke or bottoms) and/or natural gas is costly.


RESEARCH ORGANIZATIONS – 12%

SERVICE COMPANIES – 5%

STATE, FEDERAL, & LOCAL TRIBES – 6%

Source: Department of Energy – Office of Fossil Energy.

UNIVERSITIES

19%

CONSULTANTS

& OTHERS

20%PRODUCERS

21%

NATIONAL

LABS

17%

Figure 4-82. Upstream R&D Funding by Performer (FY 2002)


Access Issues

The natural gas study represents an opportunity tobroaden understanding and to offer recommendationsfor easing restrictions that adversely affect access tonatural gas resources. The objectives of the analysis ofaccess issues were:

� To identify the regulatory and environmental issuesaffecting access to gas resources. These issues derivefrom varied interests (mineral extraction, wildliferesource management, environmental protection,historic preservation, local, state, and federaleconomies, recreation) and diverse organizations(administrative agencies, industry, special-interestgroups, general public). The attempt to suit thesemany interests has resulted in regulations that arecomplex and often difficult to understand andadminister.

� To quantify the volumes of natural gas affected byaccess restrictions and the impact on gas markets ofalternative access policies.

� To recommend actions that might reasonably betaken to support environmentally sound gasresource development.

Focus

The Environmental/Regulatory/Access Subgroupstudied a limited number of the lower-48 basins thatwere reviewed by the Resource Subgroup. The accessanalysis focused on those basins with large remainingpotential and with significant access constraints. Thesebasins are located in the Rocky Mountain region of theUnited States (Green River, Uinta/Piceance, PowderRiver, San Juan, Wyoming Thrust Belt) and offshoreAtlantic, Pacific, and Eastern Gulf of Mexico.

The study also examined the issue of access inCanada, and, referencing the recently published studyon Potential Canadian Gas Supply (conducted by theCanadian Energy Research Institute) estimated thepercentages of the various producing basins that arecurrently off-limits to leasing. These percentages wereused in the long-range modeling process. Althoughthere are other access and environmental issues inCanada, generally speaking, access issues in Canada areless significant than in the United States. Therefore,alternative policy cases related to Canadian accessrestrictions were not conducted and likely would nothave had a material impact on the 2003 NPC study.

Expert Teams for Analysis

Two teams of industry experts – one for the RockyMountains and one for the U.S. offshore – were assem-bled to perform the detailed data gathering, compila-tion, and evaluation work. These teams brought withthem extensive, practical industry experience in deal-ing with environmental and regulatory issues thataffect petroleum resource development. These teamsalso had access to specialized contractors to provideadded assistance and expertise. Equally important, theteams sought input from associations and regulatoryagencies, such as the Bureau of Land Management(BLM), Forest Service, and Minerals ManagementService, in order to ensure that their work looked atthese issues from the broadest possible perspective.

Rocky Mountain Area

Prior Reports – Federal Lease Stipulations

Participants in the 2003 NPC study had three previ-ously published Rocky Mountain access-relatedreports to use as points of reference:

� The 1999 NPC Study

� The 2001 Greater Green River Basin Study

� The 2003 Energy Policy and Conservation Act Study– Scientific Inventory of Onshore Federal Lands’ Oiland Gas Resources and Reserves.

These reports are primarily or exclusively concernedwith the effects of federal lease stipulations; they gen-erally regard lands as falling into two categories, eitheroff-limits (areas not leased for various considerations)or available for lease. The participants in the 2003study quickly determined that the three prior reportshad done an excellent job quantifying the impacts oflease stipulations, and that the direction of the 2003study should be to examine the impacts arising frompost-leasing requirements.

Conditions of Approval – Post Leasing

The term “conditions of approval” (COA) refers todevelopment requirements that arise during the per-mitting process that takes place after leases areobtained. These COAs are governed by several con-trolling authorities, but the most significant and wide-ranging tend to be those based on federal legislationconcerning environmental policy, species protection,and historic preservation.


In many instances where leasing is permitted, theseCOAs are actually more of an impediment to explo-ration and development than the lease stipulations.This is not to say that COAs are inappropriate, butrather to provide a baseline for improvement in COAprocesses.

Methodology – Green River, Uinta/Piceance,Powder River, San Juan

The teams developed extensive maps showing thesurface areas subject to COA issues. To perform thismapping, NPC contracted with Hayden-Wing andAssociates, an environmental consulting firm locatedin Laramie, Wyoming. Hayden-Wing is widely recog-nized for its expertise in wildlife surveys, environmen-tal impact statements, wetland evaluations, anddevelopmental permitting.

In addition to the preparation of these maps,Hayden-Wing quantified the percentage of the landareas in these basins that are covered by each habitatand migratory range. They also estimated the frequen-cy of occurrences requiring specific survey or mitiga-tion actions on the part of oil and gas operators, suchas active raptor nests, active Sage Grouse leks, big gamebirthing habitats, and other similar circumstances.

Thus, for example, a given area of a studied basin hasthe potential for a particular protected species habitat.Surveys within that habitat will find the species a cer-tain percent of the time. If the species is found, COAswill be placed upon resource access.

The Rocky Mountain expert team estimated the costand time delay caused by the COAs related to thespecies occurrences described above, as well as forarchaeological activities governed by the NationalHistoric Preservation Act and for environmental analy-ses and environmental impact statements required bythe National Environmental Policy Act (NEPA). Thesedata were developed by analyzing costs and delaysincurred on actual projects conducted in these basins.The team also determined whether each COA alsoapplied to state and fee lands, in addition to federallands. The team drew from its own extensive experi-ence as well as that of Hayden-Wing, industry associa-tions, and regulatory agencies for cost, time, andapplicability information.

Once all of the data had been compiled, they wereanalyzed with the help of a probability analysis pro-

gram created by EEA specifically for this project. Thedata consisted of:

� A list of circumstances that might potentially resultin COAs

� An estimate of the percentage of each basin areawhere those circumstances might be found and ini-tial surveys would have to be done

� An estimate of the conditional probability that theinitial survey would find that the circumstance is, infact, present for a given well in that area and furthermitigation steps would have to be taken

� Cost, time delays, and surface occupancy bans forsurveys and mitigation steps

� Activity sequencing and grouping to ensure logicalscenarios and to eliminate redundancies or inconsis-tencies.

A simplified data input matrix is shown in Table 4-6.

The program analyzed the cumulative effect of theCOAs in each basin for 1,000 hypothetical wells.Separate runs were made to determine the impacts onfederal lands, state lands, and fee lands, and forexploratory versus development wells. The quantifica-tion process from map areas to cost-and-delay outputinformation is shown schematically in Figure 4-83.

By calculating a weighted average based on thecumulative acreage of each of the three land types, theteam was able to estimate the average cost and timingdelay per well in each basin associated with COAs.Some well locations, although available for leasing,were rendered effectively off-limits due to the cumula-tive effects of COAs. For purposes of this report, anyacreage that was rendered unavailable for surface occu-pancy for 9 months or more per year due to the cumu-lative effect of COAs is considered to be “effectivelyoff-limits to development.”

Efforts were next made to normalize the areas ofacreage effectively off-limits due to COAs to the playareas within each basin, and to subtract the percentageof lands in each basin already determined by the 2003EPCA study to be off-limits due to leasing restrictions.This allowed the team to determine the net basin-widepercentage of natural gas resource that is effectivelyoff-limits due to COAs for each of the major basins.

These findings are summarized in Table 4-7.

CH

APTER 4 - N

ATURA

L GA

S SUPPLY

180

Surface

Authority

Activity Sequence, Probability,

and Grouping

Wildcat

Wells

Development

Wells

Ite

m N

um

be

r

Ite

m

Fe

de

ral

Ac

tio

n

Sta

te L

an

ds

Pri

va

te L

an

ds

Co

nti

ng

en

t

on

Ite

m X

Ha

pp

en

ing

Fir

st

Pro

ba

bil

ity

or

Co

nti

ng

en

t

Pro

ba

bil

ity

%

Co

rre

lati

on

wit

h I

tem

X

Co

rre

lati

on

Fa

cto

r

Ite

m

Nu

mb

ers

of

Co

mm

on

Gro

up

Ad

de

d C

os

t

(Do

lla

rs)

Tim

e D

ela

y

(Mo

nth

s)

Ad

de

d C

os

ts

(Do

lla

rs)

Tim

e D

ela

y

(Mo

nth

s)

No

Ac

ce

ss

1 Raptor Survey General 1 1 0 99.4 700 1

2 Raptor Nest Survey 1 1 0 28.1 1 3,500 1 350 1

3 Active Raptor NestsFound: No Access

1 1 1 2 5.7 X

4 Active Raptor Nests

Found: Mitigation

1 1 1 2 30.0 3 -1 3 106,000 3 106,000 3

5 Big Game Survey 1 1 0 40.0 1,200 360

6 Big Game Found:

Relocate/Directional

1 1 0 5 10.0 150,000 3 150,000 3

7 Big Game Found:Mitigation

1 1 0 5 90.0 6 -1 6 5,280 3 5,280 3

8 Blackfooted FerretSurvey

1 1 0 54.5 7,500 6 7,500 6

9 Blackfooted FerretFound: No Access

1 1 0 8 1.0 X

10 Blackfooted FerretFound: Mitigation, etc.

1 1 0 8 10.0 9 -1 9 106,000 3 106,000 3

Note: Approximately 50 key items per basin.

Table 4-6. Rockies Access Analysis Matrix


ETC

(APPROX.

60 MAPS)

CALCULATE PERCENT

OF STUDY AREA

• PER-WELL RANGE OF

COSTS AND AVERAGE

• PER-WELL RANGE OF

DELAYS AND AVERAGE

• LEGISLATIVE/ADMINISTRATIVE

WITHDRAWN AREA

• EFFECTIVE AREA INACCESSIBLE

DUE TO REGULATION

DEVELOP CURVES FOR

COST AND FOR DELAY

$ K

WELL 1 2 3 . . . 1,000

RAPTOR HABITAT WINTER RANGE T & E SPECIES

QUANTIFY

REQUIREMENTS

IF SPECIES/ACTIONS ARE

PRESENT: COST, DELAYS

AND PROBABILITY OF

• SURVEYS

• MITIGATION

• STUDIES, EA, EIS

• NO SURFACE ACCESS

CONDUCT

ANALYSIS TO

DETERMINE

COST AND

TIME DELAYS

FOR 1,000

POTENTIAL

WELLS

WITHIN STUDY BOUNDARIES,

MAP THE AREAS HAVING SPECIES

OR ACTIONS OF INTEREST

CALCULATE PERCENT

OF STUDY AREA

CALCULATE PERCENT

OF STUDY AREA

DETERMINE ACCESS

INPUTS TO MODEL

QUANTIFY

REQUIREMENTS



AND PROBABILITY OF

• SURVEYS

• MITIGATION



QUANTIFY

REQUIREMENTS



AND PROBABILITY OF

• SURVEYS

• MITIGATION



Figure 4-83. Access Quantification Process

Analysis Extension – Wyoming Thrust Belt

In addition to the four major producing basinsdiscussed above, the Rocky Mountain area expertteam examined access restrictions in the area definedby this study as the Wyoming Thrust Belt. Thisexamination was conducted by comparing land useplanning maps of this region to the play areas devel-

oped by the NPC Resource Subgroup. The expertteam also applied its knowledge of administrativeleasing policies currently being employed in the areaby the governing agencies. Using this information,the expert team determined that roughly 80% of theresource underlying the Wyoming Thrust Belt area iscurrently withdrawn from leasing due to administra-tive decisions.


CategoryGreenRiver

Uinta/Piceance

PowderRiver San Juan

Resources Off-Limits

Federal Statutory/Administrative No Leasing

8.7% 4.1% 4.6% 2.3%

Prohibitive Conditions ofApproval

12 Months Off-Limits 24.5% 15.2% 5.7% 6.2%9-11 Months Off-Limits 6.9% 1.8% 20.3% 0.1%

Total Restricted Percentage 40.1% 21.1% 30.6% 8.6%

Average Added Costs perWell Due to Conditions ofApproval (Thousands ofDollars)

Federal/State Exploratory 240-250 146-152 103-108 63-68

Federal/State Development 90-95 104-108 57-61 53-57

Weighted Average 103 107 62 55

Fee Exploratory 48-52 54-56 15-20 30-34

Fee Development 54-58 68-70 17-21 30-35

Weighted Average 56 69 19 33

Average Time Delay (Months)

Federal/State Exploratory 12-14 9-11 2-4 5-7

Federal/State Development 20-22 7-9 13-15 6-8

Fee Exploratory 2 2 6 1

Fee Development 2 2 2 1

Note: Percentages refer to new field and nonconventional resources only. Proved reserves andgrowth of old fields are not included.

Table 4-7. Rockies Access Restrictions

Permitting Pace

The process of issuing permits requires resourcesthat are not always immediately available. This is par-ticularly true whenever a new play develops in a givenarea and there is a sudden increase in demand for per-mits. Therefore, a limitation on the yearly increase inthe number of permits issued by the governing agen-cies was developed and included in the modelingprocess. For periods of very heavy permit loads (over300 wells drilled per year), the increase is limited to15% of the prior year’s level.

Offshore United States

The single largest access restriction to natural gasresources in the offshore United States is thePresidential Order issued by former President Bush.President Clinton extended the reach of that restrictionby withdrawing additional acreage in the Eastern Gulfof Mexico, thus creating a moratorium on explorationand development activities that covered not only theentire Atlantic seaboard, but also extended to most ofthe Pacific Coast and most of the Eastern Gulf ofMexico as well. This moratorium is to last throughJune 30, 2012. In addition, the remainder of theEastern Gulf of Mexico remains off-limits due to oppo-sition from the state of Florida and the subsequentdecision by the Department of the Interior to greatlyreduce the Lease Sale 181 area.

The OCS expert team analyzed the effects of existingenvironmental and access-related restrictions in theoffshore U.S. in terms of time delays and increasedcosts per well, and ensured these data items were accu-rately reflected in the EEA model. The team also iden-tified the key issues affecting access to development inthe OCS and compiled detailed analyses of them,including the public policy recommendations con-tained elsewhere in this report, and developed themodeling assumptions for the modeling cases con-ducted throughout the course of this study.

For the Reactive Path scenario, the following model-ing assumptions were made:

� All Presidential Order moratoria remain in placethrough 2025.

� All waters placed off-limits due to administrativedecisions of the Department of the Interior remainoff-limits during the time period covered by thisstudy.

� No additional acreage beyond that covered in theMinerals Management Service’s 2002-2007 Five-Year Leasing Program will be offered during the timeperiod covered by this study.

Access Analysis

Sensitivity Case Summaries

To estimate the potential effects on price andrecoverable natural gas resource of future implemen-tation of the Onshore and Offshore recommenda-tions contained in this report, the NPC specifiedseveral modeling sensitivities to be run by EEA. Sincethe Reactive Path scenario described elsewhere in thisreport assumes that current restrictions to access willremain constant throughout the scope of this study,the NPC also specified decreased supply access casesdesigned to estimate the effects that might result froma continuation of the steady increase in restrictionsthat have occurred in the United States over the past30 years.

The Balanced Future scenario included increasedaccess assumptions that are the combination of twosensitivities discussed below – the Gradual IncreaseRockies Access case and the Increased Offshore Accesscase.

Increased Rockies Supply Access Cases. The NPCdetermined it would be useful to perform twoenhanced access cases as a part of its modeling work:(1) an analysis designed to estimate the impact of thecurrent regulatory regime (the “Full Effect Case”) onnatural gas prices and available resource; and (2) ananalysis designed to estimate the potential positiveimpacts from the implementation of the public policyrecommendations contained in this report (referred toherein as the “Gradual Increase Case”) on prices andavailable resources.

In running these two cases, the following assump-tions were made:

� Full Effect Case – The cost and timing restrictionsarising from the cumulative effects of post-leasingCOA are immediately lifted. The percentage ofthe resource that is off-limits to leasing by statuteremains unchanged. It is important to note thatthis case is not intended to advocate the repeal ofthese COAs. Rather, it is simply intended as ameans of estimating the impacts in terms of high-er prices and foregone resource of the current


regulatory regime. As noted elsewhere in thisreport, the NPC fully recognizes and supportsefforts by the various governing agencies to pro-tect endangered species, wilderness areas, andarchaeological artifacts.

� Gradual Increase Case – The cost and timing restric-tions arising from the cumulative effects of post-leasing COA are decreased by 50% over a five-yearperiod beginning in 2004. As in the other cases, thepercentage of resource off-limits to leasing by statuteremains unchanged.

Increased Offshore Supply Access Case. In theOCS, the NPC wanted to test the potential effects onthe price and available resource from a lifting of thepresidential moratoria that are currently in place onthe Atlantic and Pacific Coasts of the U.S. lower-48, aswell as the Eastern Gulf of Mexico. In this case, the fol-lowing are assumed:

� The moratorium ends in July 2005.

� Leasing of offshore tracts does not commence untilJuly 2007. This two-year period is to allow time forfederal, state, and local jurisdictions to develop theadministrative infrastructure needed to manage thisactivity, and for appropriate areas for leasing to beselected.

� Full-scale production does not commence until2012 to allow time for seismic analysis, exploratory,and development programs.

Decreased Supply Access Cases. For the decreasedaccess cases, the NPC made the following assump-tions:

� The costs and timing delays arising from post-leas-ing COAs in the Rocky Mountain area double overa period of 10 years beginning in 2004. Thisincludes the average added cost per well, the aver-age initial time delay, and the percentage ofresource that becomes effectively off-limits todevelopment. In the judgment of the RockyMountain expert team, this gradual increaseapproximates the trend that has taken place duringthe prior decade.

� The percentage of resource statutorily off-limits toleasing as quantified in the 2003 EPCA Reportremains unchanged.

� The total “No Access” percentage of resource (statu-tory + COA) in each basin was capped at no morethan 50%.

� In the OCS, the model assumed a one-year halt indrilling takes place in 2005, and the environmentalcost of compliance doubles over a ten-year period.

Sensitivity Modeling Results

Onshore. The Southern California market obtainsmost of its natural gas supply from the Rockies/SanJuan Basin areas that were the subject of the NPC’swork related to post-leasing COAs. Figure 4-84 showsthe impact on Southern California natural gas pricesthat result from different levels of access to resources inthe Rockies.

As expected, the Gradual Increased Rockies Accesscase begins to show an easing of prices as the impactsof the COAs are reduced beginning in 2004. Theresult shows a consistent price differential of 30-50cents per MMBtu in the later years of this study whencompared to the Reactive Path scenario. However, asmentioned earlier, the Reactive Path scenario assumesthat the current level of COA-related effects willremain static throughout the duration of this study,which represents an improvement over the steadyincrease in such restrictions that has been observed inrecent years. Thus, the NPC believes it would be use-ful to compare the Gradual Increased Rockies Accesscase to the Decreased Rockies Access case, whichassumes that this trend towards more COA-relatedrestrictions will continue if there are no significantchanges in public policy.

In terms of the available natural gas resource, theGradual Increased Rockies Access case shows animmediate ramp-up in available volumes from theOnshore lower-48 compared to the Reactive Path sce-nario in 2004, which continues throughout the dura-tion of the study. This differential peaks in the outyears at more than 1 TCF per year. When one com-pares the Gradual Increased Rockies Access case to theDecreased Rockies Access case, the differential in vol-umes reaches 1 TCF as early as 2010, and increasessteadily, reaching 2 TCF per year during the latter partof the study.

Offshore. The Increased Offshore Access case showsa consistent lowering of prices for the South Floridamarket compared to the Reactive Path scenario begin-


ning in 2011, as new volumes from the Atlantic OCSand the Eastern Gulf begin to come on stream. Thiscase showed a similar price impact for the New YorkCity market as well. In addition, the Increased OffshoreAccess case shows a steady ramp-up of new volumesfrom the Offshore Atlantic beginning in 2011, growingto almost 700 BCF per year by 2025, as well as similarproduction volumes from the Eastern Gulf of Mexico.

In the Decreased Offshore Access case, there is anear-term price increase for the South Florida marketresulting from the one-year drilling moratoriumassumed in 2005.

Sensitivity Conclusion

Figure 4-85 shows the change in Henry Hub pricefrom the Reactive Path scenario for the IncreasedAccess (Combined) case (used in Balanced Future sce-nario) and the Decreased Access (Combined) case,which assumes a continuation of growing accessrestrictions.

Overall, the sensitivity cases related to the issue ofaccess support the policy recommendations that follow

below. None of the recommendations are by them-selves a panacea for alleviating the tight supply anddemand outlook that is forecast by the Reactive Pathscenario in this study. However, when taken as awhole, the NPC believes that the prompt implementa-tion of these recommendations would effectivelyincrease available natural gas in the lower-48 areas ofthe United States, which in turn would have a signifi-cant effect on prices paid by consumers.

Recommendations

The following recommendations will reduce permit-ting response time by streamlining processes, institut-ing performance metrics, clarifying statutoryauthority, and ensuring adequate agency resourcing.

Onshore – Increase access (excluding desig-nated wilderness areas and national parks)and reduce permitting costs/delays 50%over five years.

Onshore Advisory Task Force. The Secretary ofthe Interior, in consultation with the Secretaries ofEnergy and Commerce, should charter an advisory


-1.50

-1.00

-0.50

0.00

0.50

1.00

2000 2005 2010 2015 2020

FULL EFFECT

GRADUAL INCREASE

DECREASED ACCESS

YEAR

DO

LLA

RS

PE

R M

ILLIO

N B

TU

Figure 4-84. Rockies Access Impact on Southern California Prices (2002 Dollars)

committee under the Federal Advisory CommitteeAct. The Committee’s charge would be to reviewthe various statutory and regulatory regimes thatgovern use of public lands in Western states, andmake recommendations designed to reduce redun-dancy, streamline processes, and reconcile conflict-ing policies.

Endangered Species Act. Because there are noqualification requirements to nominate a species forlisting, species are frequently proposed by groups andindividuals to hinder land management planning andproject permitting. This has created such a backlogthat nominated species are given the same level ofprotection as listed threatened and endangeredspecies without supporting scientific data.Qualification requirements and technical review pro-cedures should be established to prevent unwarranteddelays.

Land Use Planning. Provision of reasonable accessto energy resources while protecting environmentalvalues is a key challenge for land managers and pro-ducers. Federal land management agencies arerequired to prepare land-use plans that allocate public

land uses, protect economic and environmentalresources and values, and establish future managementdirection. These plans should use ReasonablyForeseeable Development scenarios and maximizeland-use planning and cost efficiencies. Land-useplanning and project monitoring need to be estab-lished as high priorities and agencies need to ensureadequate personnel to meet land use-plan, leasing, andproject permitting expectations.

Roadless Areas. Modify the Forest Service’s road-less rule to exempt oil and gas exploration and devel-opment activities because such activities aretemporary in nature, subject to extensive environmen-tal regulation, and are fully reclaimed after productionceases.

BLM Wilderness Study Areas. Identify areas ofhigh natural gas potential within the BLM wildernessstudy areas and open them for multiple use.

Staffing. Congress should ensure that staffing forland management agencies is fully funded to enabletimely and adequate planning, leasing, permitting, andproject monitoring.


-1.00

-0.80

-0.60

-0.40

-0.20

0.00

0.20

0.40

0.60

2000 2005 2010 2015 2020 2025

DO

LLA

RS

PE

R M

ILLIO

N B

TU

INCREASED ACCESS

DECREASED ACCESS

YEAR

REACTIVE PATH

Figure 4-85. Impact of Access Restrictions on Henry Hub Price (2002 Dollars)

Permitting. Streamline the permitting process bysuch steps as use of categorical exclusions for wells andright-of-ways, encouragement of joint drilling andright-of-way applications, and use of sundry noticesinstead of full permit applications for successive wellson multi-well drill pads. Set performance goals; suchas 90% of all drilling applications must be completedwithin 35 days. Consider use of a permitting focusteam to provide assistance to field offices with highpeak workloads.

Cultural Resources. Due to liberal interpretationof current regulations, operators are frequentlyrequired to perform exhaustive cultural resource stud-ies far beyond the scope of their projects. Improvedmethods for determining site significance are critical-ly needed. In-depth consultation and review shouldnot be mandated if a site is not unique or lacks signif-icance. BLM should ensure that its national historictrail and visual resource management guidelines areused objectively and consistently to avoid unintendedeffects to private landowners, lessees, and state andfederal revenues.

NEPA Process. Considerable frustration existsaround the inability of agencies to meet NEPA require-ments in a timely and efficient manner. It is imperativethat agency-specific accountability and performancemetrics be developed and implemented to measureand report results to the public and Congress. TheCouncil on Environmental Quality’s (CEQ) 1997report “The National Environmental Policy Act, AStudy of Its Effectiveness After 25 Years” offers manypositive and effective recommendations that should beimplemented by agencies. Further recommendationsinclude the following:

� Directing federal agency compliance with CEQ reg-ulations at 40 CFR 1500 to 1508 (e.g., scope of envi-ronmental analysis, public participation anddocumentation) and relevant executive orders (e.g.,requiring permit streamlining and energy impactassessments)

� Setting performance goals and targets, along withperformance enhancement measure, for action onleasing and permitting for each BLM office, andreporting results to the public

� Developing internal programs aimed at improvinginformation exchange and technology transfer withother agencies, and the manner by which relevant or

new information from inventory, monitoring,research, and planning activities is incorporated inland-use plans.

Offshore – Lift moratoria on selected areasof the federal OCS starting in 2005.

Removal of OCS Moratoria. The President,Congress, and state governors should review therationale for continued moratoria on leasing anddevelopment of prospective natural gas resources. Areview process should be structured to identify cur-rent moratoria areas containing high resource poten-tial, with a view towards the lifting of these moratoriain a phased approach beginning in 2005. ThePresident and Congress should consider all currentlyexisting factors when conducting this review, includ-ing but not limited to the outlook for domestic sup-ply, the significant natural gas resources that underliethe waters currently subject to presidential ordermoratoria, the environmental advantages of produc-ing and transporting OCS natural gas, and the outstanding safety and environmental record demon-strated by the oil and gas industry in other OCS areasover the last 30 years.

OCS Leasing of Available Lands. The Departmentof the Interior should provide continued access tothose OCS areas identified in the current 2002-2007 5-Year Leasing Program.

OCS Education and Outreach. The Secretary ofthe Interior, in consultation with the Secretaries ofEnergy and Commerce, should launch a process thatwill lead to an energy education and outreach pro-gram encouraging a national dialogue about the exist-ing and potential role of OCS-derived natural gas inmeeting our nation’s energy needs, with the goal ofincreasing public/stakeholder awareness of OCS natu-ral gas activities and the key role it plays in thisnation’s economy.

Consideration of Existing Studies. Numerousother studies and recommendations have been devel-oped to address energy availability. Some examplesare “Energy for a New Century: Increasing DomesticEnergy Supplies,” 1998 OCS Policy CommitteeSubcommittee on Environmental Information forSelect OCS Areas; “2001 Report from theSubcommittee on Natural Gas OCS Policy,” OCSPolicy Committee Subcommittee on EnvironmentalInformation for Select OCS Areas; “2003 Report of the



Subcommittee on Education and Outreach” of theOCS Policy Committee and the U.S. Commission onOcean Policy principles and recommendations.Additional information on these is found in an appen-dix to the Supply Task Group Report. TheDepartment of Interior in consultation with key stake-holders including states, industry, and non-govern-mental organizations, should review therecommendations from these various studies foraction by Congress.

OCS Energy Permit Approvals. Congress shouldmake statutorily permanent the requirement thatall decisions regarding access to the OCS must (1)consider impact on the nation’s energy supply, dis-tribution, and use, and (2) decision-makers must beheld accountable for the impact their decisions willhave on energy supply. Congress should providefull funding to the federal agencies so they conductall the necessarily research, analysis, and approvalsof OCS-related natural gas activities in a timelyfashion.

The OCS and the Role of States. Congress shouldsupport mitigation of any negative impacts OCSdevelopment may have on infrastructure and coastalcommunities by directing a portion of the bonus bidsand royalty revenue stream from existing royalties toaffected coastal states. Additionally, the OCS royaltystream should be reviewed as a possible fundingsource of MMS activities that support the OCS oil andgas program.

Congress and the Administration should considerlegislative proposals that would definitively establishthe roles and responsibilities a coastal state has withregard to reviewing and taking a role in OCS leasingand development activities based on distance fromthe shorelines to OCS activities, and distribute OCSrevenues to the coastal states according to these leas-ing “zones.” At some distance seaward from theshoreline, the federal government should have thesole discretion to lease OCS resources. Congress, theAdministration, and coastal states should considersuch leasing guidelines.

OCS Inventory. Congress should provide MMSfunding and authority to obtain a more accurateassessment of the OCS resources. This would

include working with affected stakeholders, includ-ing coastal states.

Coastal Zone Management (CZM). If a state allegesthat a proposed activity is inconsistent with its CZMPlan, it should be required to specifically detail theexpected effects, demonstrate why mitigation is notpossible and identify the best available scientific infor-mation and models which show that each of the effectsare “reasonably foreseeable.” State CZM Plans shouldnot be approved by the Secretary of Commerce if suchimplementation would effectively ban or unreasonablyconstrain an entire class of federally authorized andregulated activity, e.g. gas drilling, production, andtransmission.

Endangered Species Act/Marine Mammals Pro-tection Act. Regulatory changes designed to protectmarine species should be based on best available scien-tific information and data to avoid inappropriate orunnecessary actions being taken with no benefit to theintended species. Reasonable lease stipulations andoperational measures designed to protect listed speciesshould be practical and cost effective and aimed toachieve minimal delays in ongoing operations.Congress should provide funding to NOAA and MMSto study the relationship between oil and gas activitiesand marine mammals in the Gulf of Mexico, with theinitial focus on sperm whales.

U.S. Commission on Ocean Policy. The Commis-sion’s guiding principles and recommendations, cur-rently being drafted, must be implemented in amanner consistent with energy-oriented principles toeffectively support improved OCS resource access anddevelopment.

Summation

The NPC believes it is possible to meet the nation’senvironmental/endangered species goals while at thesame time encouraging fuller development of criticalnatural gas resources. The NPC urges the governmentto give serious consideration to the implementation ofthe policy recommendations contained in this report,recommendations that would enable the industry,government, and other interested parties to worktogether to develop and implement innovativeapproaches and solutions to these difficult and com-plex issues.


LNG Imports

A LNG subgroup was formed as part of the NPCSupply Task Group to develop a short- and long-term(2025) outlook of potential LNG imports into NorthAmerica. In addition to forming an LNG import out-look, the LNG Subgroup developed a “primer” onLNG, covering a description of the LNG value chain,LNG history, global LNG supply and demand, compet-itive supply cost, and U.S. terminal permitting/devel-opment. A summary of issues and recommendationsfacing U.S. terminal development, which may affect thelevel of LNG imports, was also developed.

The team made use of publicly available data toidentify potential North American LNG import termi-nal locations and to estimate the timing of LNGimports. The approach used was to:

� Research and develop estimates of LNG supply,transportation, and regasification costs

� Utilize announcements of potential new U.S. LNGimport terminals and global LNG supply

� Evaluate the competitive global LNG market

� Establish “standard” model assumptions for timingof terminal permitting and construction, terminalsize, and buildup of imports

� Identify the timing of potential supply and LNGimport terminal additions

� Identify “controlling” assumptions that might affectthe pace of new LNG imports

� Develop three scenarios for use in modeling input

� Identify issues that might affect the pace of LNGimports

� Compile and use research in support of the LNGdiscussion

� Propose recommendations to address the identifiedissues.

LNG Overview

LNG, or liquefied natural gas, is the liquid form ofnatural gas that has been cooled to a temperature of–256°F or (–161°C) and maintained at atmosphericpressure. It is an odorless, colorless, non-corrosive,and non-toxic liquid. The process for liquefying natu-ral gas reduces the volume of the gas to approximately

1/600th of its original volume. This process enables itto be transported economically in specially designedocean vessels throughout the world.

LNG development projects are located where thereis minimal local demand for natural gas or in areasremote far from traditional economic pipeline trans-portation systems. Once the gas is transformed into aliquid it is then transported in special LNG vessels andimported into countries where local demand exceedsthe availability of supplies from domestic sources.

The LNG industry is often described by the expres-sion the “LNG chain,” as illustrated in Figure 4-86.This is a reference to the fact that LNG projects arelarge and required critical mass and alignmentthroughout the many phases of production, trans-portation, and distribution of the product if they are tobe successful. All links of this “chain” must worktogether for natural gas to be produced, liquefied andexported, transported, imported and regasified, andsold as natural gas into an end-user market. LNG proj-ects require massive reserves (7-10 TCF), produce sig-nificant volumes (0.5-1.0 BCF/D), and requireinvestments as large as $4-$10 billion. Due to the largescale of these projects, and the considerable financialrisk involved in undertaking them, a secure market forthe natural gas is usually a necessary condition for theirdevelopment. That is the reason why most of theworld’s LNG is sold under long-term contracts (20-25years), although short-term and spot markets sales arebeing introduced as the market matures.

LNG Safety

The LNG industry has demonstrated an excellentsafety record in its almost 40-year history. This is theresult of the attention to detail in engineering, con-struction, and operations in all aspects of the LNGchain. The industry also has to meet stringent safetystandards set by countries such as the United States,Japan, Australia, and European nations.

The LNG facilities (liquefaction and regasificationand storage) are industrial sites and must meet allcodes, rules, regulations, and environmental standardsenforced by local jurisdictions.

LNG ships are specially designed, double-hulledships. The LNG in these ships is stored in special con-tainment systems at atmospheric pressure and at


–256°F. These vessels are designed to protect the cargotanks and to prevent leakage or rupture in an accident.The International Maritime Organization (IMO) hasdeveloped internationally ratified standards for theconstruction and operations of all ships, includingLNG ships.

In 2003, there are 17 global LNG liquefactionexport facilities (one in the United States), 40 regasifi-cation import facilities (four in the United States andone in Puerto Rico), and 136 LNG ships, handlingabout 120 million tons annually. The industry hassuccessfully completed more than 33,000 LNG voy-ages, covering more than 60 million miles withoutmajor accidents or safety issues in port or on the highseas.

LNG Global Supply

LNG is currently produced in 12 countries, includ-ing the United States, from 17 operating liquefactionfacilities. New supply projects are under constructionand additional projects have been proposed, as shownin Figure 4-87.

The global LNG industry began in 1964 with thestartup of the Algeria LNG export facilities, followedby the United States in 1969, and Libya in 1970.Significant Asian supplies were developed throughoutthe 1970s with new export facilities in Brunei,Indonesia, Malaysia, and Australia. Middle East sup-plies were first introduced in 1977 from the UnitedArab Emirates, with significant additions in the late1990s and 2000 from Qatar and Oman. Additional

Atlantic Basin supplies from Nigeria and Trinidad/Tobago where added in the late 1990s.

New supply projects are under construction orunder evaluation. Norway is currently developingtheir first LNG export project and expansions are beingbuilt in Trinidad/Tobago, Nigeria, Qatar, Australia, andOman. Proposed projects include potential produc-tion from additional expansions in Trinidad/Tobago,Nigeria, Algeria, and Qatar plus new developments inNigeria, Australia, Indonesia, Angola, Egypt, Sakhalin,Venezuela, Bolivia, and Peru.

LNG Global Demand

LNG demand has grown continuously from the timethe first deliveries of Algerian LNG entered Europe in1964, as shown in Figure 4-88. Since the 1970s, LNGdemand has grown by about 8% per year, primarilydue to the fast-growing Asian markets (largest areJapan and Korea). While continued growth is forecastfor Asia, developing markets in the United States andEurope are expected to support demand growth in therange of 6-10% per year. Growth at this rate will resultin a doubling of the existing market by 2010.

The growing LNG demand outlook in Asia, Europe,and North America will also mean increasing competi-tion for new supplies.

Model Assumptions

The inputs for the LNG cases were exogenous to themodel, meaning the volume profile was hard coded andnot determined by the model. This treatment is based

FIELD PRODUCTION LIQUEFACTION REGASIFICATION & STORAGE

SHIPPING

Figure 4-86. The LNG Value Chain


EXISTING

UNDER

CONSTRUCTION

PROPOSED

Figure 4-87. Global LNG Supply

0

20

40

60

80

100

MIL

LIO

N T

ON

S

TURKEYTAIWANS. KOREAJAPANW. GERMANYBELGIUMITALYU.S.SPAINFRANCEU.K.

BELGIUM

KOREA

TAIWAN

TURKEY

PUERTO RICO

DOMINICAN REPUBLIC

ITALY

SPAINU.S.

JAPAN

FRANCE

Source: Cedigaz.

U.K.

1965 1970 1975 1980 1985 1990 1995 2000

YEAR

120

Figure 4-88. Historical LNG Demand – LNG Import Countries and their Start Dates

on the assumption that most of the projected LNGimports will be long-term base-load volumes. Once thedevelopment decision is made for these capital-intensiveprojects, these volumes should not be affected to anygreat extent by daily or monthly fluctuations in prices.The exogenous inputs include terminal locations/nodes,volumes, and timing of imports.

The following is a summary of the key modelassumptions:

� Long-term prices support increased LNG imports

� New terminals sized for 750 MMCF/D base load

� New terminal expansions sized for 750 MMCF/Dbase load

� New terminal permitting time of 2 years

� New terminal construction time of 3 years

� Ramp-up rate of 3 years upon commencement ofimports

� Existing U.S. LNG terminals supplied first, followedby their expansions, followed by new build terminals

� Location of new terminals driven by available down-stream pipeline access and ease of permitting

� Timing of imports driven by supply availability,shipping, and new LNG import terminal develop-ment

� Limited shipping and LNG supplies available in thenear term.

The outputs from the model runs indicate that U.S.long-term natural gas prices will support an increase inLNG imports. Dependent on supply developmentcost, location, and transportation cost, LNG can beimported into North America in a range of $2.00 to$4.00. Since the supply cost was determined not to bethe critical assumption affecting LNG imports, theteam focused on the assumptions with respect to tim-ing and potential quantity of LNG imports.

The existing U.S. LNG import terminals have a base-load (continuous, steady) capacity of 400-750MMCF/D. Many of the recently announced LNG ter-minals are in the 700 MMCF/D to 1.5 BCF/D range.Although the capacity of new terminals will very likelyvary, the team elected to use a generic size of 750MMCF/D, with expansions of 750 MMCF/D. The onlyexception is a terminal located in Baja California.Because the recently announced proposals for termi-nals there are for 1 BCF/D, the model assumed

1 BCF/D for this terminal. The model inputs assumethese volumes (750 MMCF/D or 1 BCF/D) are base-load volumes, not peak-load volumes.

The rate of entry of additional LNG imports will beprimarily driven by the time required to secure permitsand construct new LNG import terminals. Uponapplication, the permitting process for an onshore U.S.LNG import terminal can take two to three years. Thetiming for an offshore terminal is approximately oneyear. Construction of an onshore terminal would takeabout three years; offshore terminals may take slightlylonger. Combining these factors, the team assumedthat each new terminal development would take fiveyears to complete (two years for permitting and threeyears for construction).

A buildup of three years was assumed for each newterminal before full utilization was to be achieved. Thisassumption is not caused by market demand restraintsbut is due to the combination of supply developmentand new ship construction. LNG competes in a globalmarketplace with significant growth potential, not onlyin North America, but also in Asia and Europe. Thiscompetition and anticipated growth means thatgrowth will be constrained by limitations of keyresources needed by upstream supply projects and bythe availability of suitable shipyards for building newLNG carriers. LNG supply liquefaction facilities aretypically constructed in series of processing unitsreferred to as trains. Depending on size, multipletrains are typically constructed one to two years fol-lowing the initial train. This construction profileimpacts the LNG supply availability, resulting in abuildup profile.

The model assumes the four existing U.S. terminalswill first be fully utilized, then expanded (three expan-sions have been announced to date). Once the existingterminals and their expansions are fully utilized, addi-tional volumes will come from new terminal develop-ment. The locations of the new terminals are drivenprimarily by three factors: the availability of existing orpotential expansion of downstream pipelines, the per-ceived ease of permitting, and other physical con-straints. The bulk of the new U.S.-based terminalsmodeled are located in the Gulf of Mexico due todeclining Gulf of Mexico shelf production, sparecapacity of existing infrastructure, the availability ofdeepwater ports (onshore) or existing offshore pipelinesystems (offshore), and the perceived ease of permit-ting (as compared to other U.S. locations). Due to the


growing gas demand in the northeast, two new LNGterminals are assumed to be built along the northeastcoast. Two terminals are modeled for Mexico, one onthe east coast and one on the west coast. These termi-nals are needed to meet the growing demand for natu-ral gas in Mexico and California.

The timing and number of new LNG terminals wasguided by many considerations. These include globalLNG competition, the complexity involved with thedevelopment of the LNG Value Chain (field produc-tion, liquefaction, shipping, regasification import ter-minals, connection to downstream markets), and theavailability of limited locations for new import termi-nals. Except for the Low LNG Imports case, it isassumed North America will experience robust LNGgrowth. North America will be competing with Asiaand European markets for LNG supply. The NorthAmerica market is very attractive for nearby AtlanticBasin supplies such as Trinidad, Nigeria, and potentialfuture supplies from other countries such as Venezuela.However, because of transportation costs, Asian andEuropean markets may remain more attractive forNorth African and Middle Eastern supplies. This com-petition, combined with political uncertainty, will havean effect on potential imports into North America.

The timing of new LNG supply will also depend onthe origination of the supply source. Most of the iden-tified LNG supply potential is located in developingcountries where challenges such as government insta-bility, boycotts, and civil unrest are found. Many ofthese supplies will also require project financing. Whilerecent U.S. gas prices have shown significant increasesfrom historically low levels, prices are neverthelesshighly volatile and the long-term outlook is still uncer-tain. Future financing may only be available if lendersare convinced that U.S. natural gas prices have indeedachieved a “step change” compared to recent historicaltrends. The other constraining factor is availability ofNorth America locations for terminal development.Onshore import terminals require access to sizeableacreage (50-100 acres); they must also be located on adeepwater port, and they must have cost-effective accessto downstream natural gas markets. There are only afew onshore locations in the United States and Mexicothat meet these criteria. Offshore terminals requireappropriate oceanographic conditions (for unloadingLNG ships), appropriate soil conditions (to support theoffshore structures), and access to offshore pipelines.These criteria also limit the number of locations whereoffshore terminals can be built.

Potential imports of LNG are limited for the nearterm by a lack of available LNG supply and ships. Allfour U.S. LNG terminals will be fully operational bythe end of 2003 (Elba Island was reactivated inDecember 2001; Cove Point was reactivated in August2003). However, most of the existing LNG supply andshipping is already dedicated to other markets. Whilesome spot cargoes will allow for increased imports, theexisting terminals are not expected to be fully utilizeduntil the recently announced LNG supply projects(Norway-Snohvit, Trinidad, Nigeria, and Egypt) areconstructed. These new supplies are scheduled tocommence production in the 2005-2007 time frame.

Case Descriptions

Three LNG scenarios were developed. These wereReactive Path, Balanced Future, and a Low Sensitivitycase. Each of the three cases indicate significantincreases in LNG imports, growing from a base ofabout 600 MMCF/D beginning in 2003. Table 4-8 is asummary of the three input cases.

Each of the scenarios shows a growing demand forLNG in North America in every year of the forecast.This is due to increasing U.S. natural gas prices, com-bined with new LNG supply and reduced LNG supply


Table 4-8. NPC LNG ScenariosNorth America LNG Imports(Billion Cubic Feet Per Day)

Reactive Balanced LowYear Path Future Sensitivity

2000 0.6 0.6 0.6

2005 2.3 2.3 2.3

2010 7.3 7.5 5.5

2015 8.8 10.7 5.8

2020 11.6 13.3 6.5

2025 12.5 15.0 6.5

Total # of Terminals

Existing Terminals 4 4 4

New Terminals 7 9 2

Expansions* 7 9 4

* Includes three expansions of existing terminals.

cost. There are four existing LNG terminals in theUnited States with the last one built in 1981. Since theearly 1980s, two of the terminals (Cove Point and Elba)have been mothballed. The other two terminals(Everett and Lake Charles) had minimal imports until1999 when the first Trinidad LNG supply project wasdeveloped. The outlook for potential LNG imports haschanged over the past three years with both Elba(2001) and Cove Point (2003) reactivated, Everettexpanded (2003), and two of the terminals (LakeCharles and Elba) announcing expansions.

All scenarios assume the four existing terminals willbe fully utilized by late 2007 (when supplies becomeavailable) and that all expansions will be constructedand fully utilized by the end of the decade. This base-line will result in an increase in LNG imports of 600MMCF/D, up to over 3.9 BCF/D by 2010.


In addition to the baseline, the Reactive Path sce-nario includes a total of seven new import terminalsand the expansion of four of the new terminals. This

scenario includes the development of five new importterminals by 2010. Two additional terminals andexpansions of four new terminals are added between2010 and 2020. Figure 4-89 shows the potential loca-tions for these terminals. The import volumes gradu-ally increase from 600 MMCF/D in 2003 to a peak of12.5 BCF/D by 2025, as shown in Figure 4-90.

Reactive Path Assumptions

� Existing terminals are fully utilized by 2007

� Existing Terminal Expansions

+ 2005 - Lake Charles and Elba+ 2007 - Cove Point

� New Terminals

+ 2007-2010: 5 Total– Gulf of Mexico #1 & #2 (2007, 2009)– Northeast #1 (2009)– East Coast (Altamira) Mexico (2007)– West Coast (Baja) Mexico (2008)

+ 2010-2020: 2 Total + 4 Expansions– Gulf of Mexico #3 (2012)– Northeast #2 (2020)


LAKE

CHARLES

ELBA ISLAND

COVE PT

EVERETT

EXISTING

NEW THRU 2020

GOM

OFFSHORE

EAST COAST

BAJA

ALTAMIRA

GULF COAST

ONSHORE

LOCATIONS OF

LNG TERMINALS

Figure 4-89. LNG Terminal Locations – Reactive Path Scenario

– Gulf of Mexico #1, #2, & #3 Expansions(2016, 2018, 2020)

– Northeast #1 Expansion (2016)


The Balanced Future scenario builds from theReactive Path scenario and assumes increased LNGsupply and shipping availability, along with less delayin import terminal permitting. This scenario incorpo-rates two additional terminals, one in Bahamas (serv-ing the Florida market) and one on the U.S. WestCoast. The Balanced Future scenario also includesexpansions of the second Northeast terminal and anexpansion of the Florida terminal, and it acceleratesstart-up of the new terminals by one year. Figure 4-91shows the locations for these potential terminals.

In addition to the baseline, the Balanced Future sce-nario includes a total of nine new import terminalsand the expansion of six of the new terminals. TheBahamas terminal is developed in 2010, with an expan-sion by 2012. The second Northeast terminal is accel-erated to 2011, with an expansion in 2023. The WestCoast terminal will be developed in 2021. The importvolumes gradually increase from 600 MMCF/D in2003 to a peak of 15.0 BCF/D by 2025, as shown inFigure 4-92.

Balanced Future Assumptions (additions to ReactivePath highlighted in bold)



+ 2005 - Lake Charles and Elba

+ 2007 - Cove Point

� New Terminals

+ 2007-2010: 6 Total

– Gulf of Mexico #1 & #2 (2007, 2009)

– Northeast #1 (2009)

– East Coast (Altamira) Mexico (2007)

– West Coast (Baja) Mexico (2008)– Florida (Bahamas) (2010)

+ 2010-2025: 3 Total + 6 Expansions

– Gulf of Mexico #3 (2011)

– Northeast #2 (2011)

– Florida Expansion (Bahamas) (2012)

– Gulf of Mexico #1, #2, & #3 Expansions(2015, 2017, 2019)


– West Coast (2021)



0

2

4

6

8

10

12

14

16

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

EXISTING

EXPANSION

NEW

YEAR

Figure 4-90. North American LNG Imports – Reactive Path Scenario


LAKE

CHARLES

ELBA ISLAND

COVE PT

EVERETT

EXISTING

NEW THRU 2015

GOM

OFFSHORE

EAST COAST

BAJA

ALTAMIRA

GULF COAST

ONSHORE

BAHAMAS

WEST

COAST

LOCATIONS OF

LNG TERMINALS

ADDITIONS TO

REACTIVE PATH

Figure 4-91. LNG Terminal Locations – Balanced Future Scenario

0

2

4

6

8

10

12

14

16

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

YEAR

EXISTING

EXPANSION

NEW

Figure 4-92. North American LNG Imports – Balanced Future Scenario

Low Sensitivity Case

The Low Sensitivity case assumes a combination ofregulatory delay and successful public opposition tonew terminal development. This scenario assumes thatthe baseline is developed with a total of only two newimport terminals (Gulf of Mexico (2007), and Baja(2008)) and one Gulf of Mexico expansion (2016).The import volumes gradually increase from 600MMCF/D in 2003 to a peak of 6.5 BCF/D by 2025 (seeFigure 4-93).

Low Sensitivity Assumptions (changes from ReactivePath highlighted in bold)



+ 2005 - Lake Charles and Elba+ 2007 - Cove Point

� New Terminals

+ 2007-2010: 2 Total– Gulf of Mexico #1 (2007) – West Coast (Baja) Mexico (2008)

+ 2010-2020: 1 Expansion– Gulf of Mexico #1 (2016)

Controlling Inputs

The controlling inputs concerning additional LNGimports will be the availability of new LNG supply andthe ability for new LNG terminals to be permitted andconstructed.

LNG is a global market and the United States will becompeting for LNG supply resources. The Reactive Pathscenario assumes North America LNG imports willgrow to 12.5 BCF/D or about 95 MTA (million tons perannum) of LNG over a timeframe of about 20 years. Toplace that in perspective, the global LNG market, whichbegan about 30 years ago and has spread to eleven coun-tries, is some 13.5 BCF/D in 2003, or approximately 100MTA. Each of the main market areas, Asia, Europe, andNorth America are forecast to grow by an average of6.5% annually. This growth will require over 30 BCF/Dof new supply. As each of the three demand areas hassignificant LNG demand growth potential, there will besignificant competition to secure supplies.

Case Results

As stated earlier, the volume of imported LNG washard coded in the model. Therefore, the resulting vol-ume output of the model was equal to the input. Each


0

2

4

6

8

10

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

YEAR

EXISTING

EXPANSION

NEW

Figure 4-93. North American LNG Imports – Low Sensitivity Case

of the three cases assume that volumes of importedLNG grow, peaking by 2025 as shown in Table 4-9.

Figure 4-94 illustrates the model results of the vol-ume of LNG imports for the three cases.

The increased number of new LNG import termi-nals in the Reactive Path and Balanced Future scenar-ios has a significant impact on increased imports. In2002, U.S. LNG imports make up less than one percentof total U.S. natural gas demand. This percentage willincrease significantly with the Reactive Path andBalance Future scenarios, resulting in LNG providing14% and 17%, respectively, of the U.S. supply of natu-ral gas by 2025. The Low Sensitivity case, while not asrobust, will still result in LNG making up about 8% ofU.S. natural gas supply by 2025.

As the amount of LNG imports is increased in allthree scenarios, each has an effect on price. Figure 4-95 shows the price variance of the Balanced Future(higher LNG imports) and Low Sensitivity (importsreduced to half of the Reactive Path) cases in relation tothe Reactive Path. The LNG imports are the same in allthree cases through 2007 because the new terminalsand associated new LNG supplies are not on line untilafter 2007. The impact of the different volume of LNGimports is illustrative through the pricing output of themodel. The Balanced Future, with additional LNGimports starting in 2010, has a moderate pricing bene-fit of about 5% over the 2010-2025 timeframe. It isimportant to note that the Low Sensitivity Case (com-bination of regulatory delay and successful publicopposition) has a much more significant impact onlong-term price, with price increases of 10-12%. Thesecases illustrate the significance LNG imports will havein meeting the growing North America demand andthe importance of getting new terminals permitted andbuilt in a timely manner.

Recommendations

This aggressive outlook for LNG import terminalconstruction will require streamlined permitting andconstruction to achieve the projected buildup.Expediting the approval process throughout all agen-cies (federal, state, and local) is critical to overcome themany obstacles that may surface, including local oppo-sition. Leveraging off the recent positive shifts by theFederal Energy Regulatory Commission (FERC) (posi-tive changes on regulatory process, active leadershiprole in recent reactivation of Cove Point and ElbaIsland, and implementation of Memorandums ofUnderstanding among federal agencies workingtogether) and changes made to regulatory policies inlate 2002 governing both onshore and offshore LNGimport terminals, will provide a springboard forimpacting positive changes down through the locallevel. The goal of the following recommendations is toreduce the time required for LNG facility permitting toone year.

� Agencies must coordinate and streamline permit-ting activities and clarify positions on new terminalconstruction and operation. Project sponsors facemultiple, often-competing state and local reviewsthat lead to permitting delays. A coordinated effortamong federal, state, and local agencies led by FERCwould reduce permitting lead time. Similarly,streamlining the permitting process by sharing dataand findings, holding concurrent reviews, and setting


0

2

4

6

8

10

12

14

16

2000 2005 2010

YEAR

2015 2020 2025

REACTIVE PATH

BALANCED FUTURE

LOW SENSITIVITY

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

Figure 4-94. North American LNG Import Cases

2005 2010 2025

Reactive Path 2.3 7.3 12.5

Balanced Future 2.3 7.5 15.0

Low Sensitivity 2.3 5.5 6.5

Table 4-9. LNG Import Assumptions(Billion Cubic Feet Per Day)

review deadlines would provide greater certainty tothe overall permitting process. FERC should furtherclarify its policy statement on new terminals so as tobe consistent with corresponding regulations underthe Deep Water Port Act, including timing for theNEPA review process and commercial terms andconditions related to capacity rights.

� Fund and staff regulatory agencies at levels neces-sary to meet permitting and regulatory needs in atimely manner. The expected increase in the num-ber of terminal applications will require higher lev-els of government support (federal, state, and local)to process and avoid delays. Additional agencyfunding/staffing will also be required once these newterminals become operational, particularly to sup-port the large increase in LNG tanker traffic.

� Undertake public education surrounding LNG.The public knowledge of LNG is poor, as demon-strated by perceptions of safety and security risks.These perceptions are contributing to the publicopposition to new terminal construction and jeop-ardizing the ability to grow this required supplysource. Industry advocacy has begun, but a moreaggressive/coordinated effort involving the DOEand non-industry third parties is required.

Emphasis should focus on understandings, safety,historical performance, and the critical role thatLNG can play in the future energy supply.

� Update natural gas interchangeability standards.Standards for natural gas interchangeability in com-bustion equipment were established in the 1950s.The introduction of large volumes of regasified LNGinto the U.S. supply mix requires a re-evaluation ofthese standards. FERC and DOE should championthe new standards effort to allow a broader range ofLNG imports. This should be conducted with par-ticipation from local distribution companies, LNGpurchasers, process gas users, and original equip-ment manufacturers. DOE should fund researchwith these parties in support of this initiative.

� Review and revise LNG industry standards if nec-essary. In order to promote the highest safety andsecurity standards and maintain the LNG industry’ssafety record established over the past 40 years ofoperations, FERC, the U.S. Coast Guard, and theU.S. Department of Transportation should under-take the continuous review and adoption of industrystandards for the design and construction of LNGfacilities, using internationally proven technologiesand best practices.


-0.80

-0.40

0.00

0.40

0.80

1.20

2000 2005 2010

YEAR

2015 2020 2025

BALANCED

FUTURE

LOW

SENSITIVITY

DO

LLA

RS

PE

R M

ILLIO

N B

TU

REACTIVE PATH

Figure 4-95. Impact of LNG Import Cases on Henry Hub Price (2002 Dollars)


Arctic Developments

The North American Arctic regions in NorthwesternCanada and Alaska contain significant gas resourcesthat can help meet future North American gasdemand. Discovered resources include about 35 TCFon the North Slope of Alaska and 9 TCF in theMackenzie-Beaufort basin.

These gas resources are remote from any existingpipeline infrastructure and are located in an Arcticenvironment, so significant investment will berequired to bring these resources to market. The keyhurdles associated with commercializing theseresources are costs, permitting, Alaska state fiscalissues, and market risks. Even though these resourceswere discovered over 30 years ago, these hurdles haveprevented the development of commercially viableprojects to date.

Industry is maturing technology advancements toreduce capital costs and the supply/demand picturesupports the need for additional supplies. Also, thegovernments of the United States, Alaska, and Canadarecognize the significant risks of such large-scale proj-ects and are working to put frameworks in place toaddress some of the hurdles.

This NPC study assumes that appropriate govern-ment frameworks are achieved in a timely manner andthat conditions support the commercial viability ofArctic gas projects. Consequently, it is assumed thatthese projects will come on line at what is consideredthe earliest feasible dates – 2009 for a Mackenzie GasProject and 2013 for an Alaska gas pipeline. The vol-umes assumed to be transported by these projects areshown in Figure 4-96.

If these projects are delayed (due to delays in estab-lishing government frameworks, delays in permitting,or for other reasons) there could be adverse conse-quences for consumers in the form of reduced gas sup-plies and/or higher energy prices. The NPC alsorecognizes that these projects may not be commerciallyviable due to the large investments needed, as well as thepotential for additional government requirements orburdens that could increase project costs and impedethe projects’ ability to compete with alternatives.

Canadian Arctic Gas Background

Resource

Table 4-10 summarizes the NPC study teamassumptions on the Discovered and Undiscovered

Potential resource available from the Canadian Arcticand is based on information from the Canadian GasPotential Committee.

Active drilling in the Canadian Arctic began in thelate 1960s with interest being sparked by the huge oiland gas discovery made in a similar geological play atPrudhoe Bay in 1967. A number of onshore gas dis-

0

2

4

6

2005 2010 2015

YEAR

2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

ALASKA

CANADA

Figure 4-96. Total Arctic Gas Transported toMarket

RegionDiscov-

ered

Undis-coveredPotential

Mackenzie Corridor 0.7 4.6

Mackenzie/ Beaufort Sea 8.8 21.2

Arctic Islands 16.4 9.4

Total 25.9 35.2

Source: Canadian Gas Potential Committee, 2001.

Table 4-10. Canadian Arctic Gas Resource(Trillion Cubic Feet)


coveries were made in the early 1970s in theMackenzie-Beaufort region, as well as some gas discov-eries in the more remote Arctic Islands region. Thatregion is located approximately 1,000 miles northeastof the Mackenzie Delta and is in a very harsh Arcticenvironment.

Current Status of Project Development

The Mackenzie Delta Producers Group (ImperialOil, ConocoPhillips Canada, Shell Canada Limited,and ExxonMobil Canada) and the Mackenzie ValleyAboriginal Pipeline Corporation are currently workingto develop a Mackenzie Gas Project, including aMackenzie Valley Pipeline. The pipeline would trans-

port onshore natural gas resources from the Taglu,Parsons Lake, and Niglintgak gas fields, and would beaccessible to other natural gas discoveries in theMackenzie Delta and Mackenzie Valley regions. Thegas would be transported through the MackenzieValley Pipeline to existing gas pipelines in northwest-ern Alberta for further transportation to market.

Figure 4-97 shows a schematic of the proposedMackenzie Gas Project. This project is currently in theproject definition phase. This phase includes technical,environmental, consultation, and commercial workrequired to prepare, file, and support regulatory appli-cations for field, gas-gathering, and pipeline facilities.It is currently estimated that regulatory applications

Source: Imperial Oil Resources.

Y U K O N

B R I T I S HC O L U M B I A A L B E R T A

N O R T H W E S T

T E R R I T O R I E S

WHITEHORSE

YELLOWKNIFE

BEAUFORT SEA

ANCHOR FIELDS:

TAGLU

NIGLINTGAK

PARSONS LAKE

GREAT BEAR

LAKE

GREAT SLAVE

LAKE

ALBERTA

TERMINUS

NORMANWELLS

INUVIK

FORT

SIMPSON

N U N A V U T

INUVIK AREA

FACILITY

COMPRESSION FACILITY NATURAL GAS PIPELINE

NATURAL GAS AND N.G. LIQUIDS PIPELINESGATHERING SYSTEM

MA

CK

EN

ZIE

RIV

ER

ARCTICCIRCLE

HAY RIVER

A L A S K A

Figure 4-97. Proposed Mackenzie Gas Project

will be filed in 2004, supporting start-up of theMackenzie Gas Project in 2009. A study commissionedby the Government of the Northwest Territories(GNWT) and TransCanada PipeLines Limited indi-cates that direct investments may total $7.6 billionCanadian (2002 dollars). This estimate consists of $4.3billion for field development costs and $3.3 billion forpipeline construction.1

Risks and Hurdles

There are significant risks and hurdles associatedwith commercializing Canadian Arctic gas as evidencedby the fact that the gas has yet to be commercialized inspite of being discovered over 30 years ago. Major risksand hurdles include permitting, costs, and market.

Alaska Arctic Gas Background

Resource

Oil and gas have been produced on the Alaska NorthSlope since the late 1970s. In the absence of a market,most of the gas has been reinjected to enhance therecovery of oil. The size of the discovered gas resourceis well understood given the extensive developmentand long production history in the Prudhoe Bay Field.

Table 4-11 summarizes the discovered resource avail-able from the Alaska North Slope. All the discoveredresource data except for Point Thomson is from theJanuary 2001 MMS report entitled Prospects forDevelopment of Alaska Natural Gas: A Review. The PointThomson data are from ExxonMobil, as reported in theJune 15, 2002 issue of the Alaska Oil & Gas Reporter.The ExxonMobil data for Point Thomson (8 TCF) ishigher than that reflected in the MMS report (5 TCF).

Figure 4-98 contains a map showing the majorNorth Slope fields. Most of the discovered resource iscontained in the massive Prudhoe Bay field, where thegas is being reinjected to enhance liquids recovery. Thesecond largest resource is the Point Thomson field,which is currently undeveloped.

The undiscovered potential for the Alaska Arctic inthis NPC study, based on USGS and MMS data, totals213 TCF, including 44 TCF of nonconventional coal

bed gas and 97 TCF offshore. Access to these newresources will be an important factor in the successfulcommercialization of Alaska gas. Some of thisprospective acreage is currently available to industry,while other areas are currently not. Government poli-cies to access gas-prone acreage in Alaska will play akey role in ensuring the necessary gas resources con-tinue to be produced from Alaska well into the future.

Attempts to Commercialize

Alaska gas development projects have been pro-posed, planned, and studied since oil and gas was firstdiscovered on the North Slope in 1967. The optionshave included various pipeline, LNG, and gas-to-liq-uids concepts. To date, none of these options havebeen commercially viable.

Current Status of Project Development

The major North Slope gas producers ExxonMobil,BP, and ConocoPhillips (the Producers) completed acomprehensive study during 2001-2002 to assess thefeasibility of delivering Alaskan gas to lower-48 mar-kets. This study assessed the cost, technology, regulato-ry, and environmental issues associated with theproject. The Producers spent $125 million on this studythat involved 110 owner company representatives andover 1 million staff-hours (including contractors).

The Alaska gas pipeline system under considerationwould transport approximately 4.5 BCF/D with thepossibility of an expansion to increase capability to 5.6 BCF/D through intermediate compression.Approximately 0.5 BCF/D would be extracted for fueluse and for natural gas liquid (NGL) extraction,resulting in approximately 4 BCF/D being delivered tomarket. The major system components include a Gas


1 An Evaluation of the Economic Impacts Associated with theMackenzie Valley Gas Pipeline and Mackenzie Delta GasDevelopment, published May 13, 2002, by Wright MansellResearch Ltd.,and available at the GNWT website:www.gov.nt.ca.

Resource TCF

North Slope

Prudhoe Bay 23

Point Thomson 8

Other North Slope 4

Total 35

Table 4-11. Alaska North Slope DiscoveredResource (Trillion Cubic Feet)

Treatment Plant, Alaska to Alberta pipeline system,NGL extraction plant, and an Alberta to lower-48pipeline system. Figure 4-99 shows an overallschematic of the system studied by the Producers.

The Producers concluded that both northern andsouthern routes were within current technical capabil-ity. The Producers also concluded that the macroeco-nomic development from an Alaska gas pipeline issignificant. Total government direct revenues could beover $100 billion. In addition to these direct tax androyalty revenues, there would be significant economicstimulus through creation of thousands of jobs.However, the Producers concluded that an Alaska gaspipeline project via any route was currently not com-mercially viable. They determined that the projectrisks outweighed rewards, that additional engineeringwork was not justified at that time, and that futureactivity must match progress with governments andcommercial viability.

The Producers also concluded that governments willplay a key role in reducing project cost and schedulerisk. Mitigation of these risks could be achievedthrough enactment of federal regulatory enabling leg-islation to provide efficiency and clarify the regulatory

process for the U.S. portion of the pipeline, clarity withthe NEB/First Nations regulatory process, and fiscalcertainty for the project with the state of Alaska.

The three major North Slope Producers continue towork on potential cost reduction concepts and withgovernments to establish appropriate frameworks toaddress these risks.

Risks and Hurdles

Four key risk areas must be addressed before anAlaska gas pipeline will attract investment capital fromthe private sector. The four risk areas are cost, permit-ting, state fiscal, and market risk. In addition, the U.S.government is debating a fiscal package related to theAlaska gas pipeline project.

� Cost. An Alaska gas pipeline project will be thelargest-ever privately funded development project.Both the large investment required for the projectand the prospect of cost overruns represent signifi-cant project risks.

� Permitting. Numerous permits or approvals will berequired from the U.S., state, local, Canadian, terri-torial, and provincial governments. In addition,


MILES

0 10 20

NORTHSTAR

ENDICOTT

PRUDHOEBAYKUPARUK

MILNEPOINT

POINT

THOMSON

DEVELOPMENTS

MAJOR DISCOVERIES

ALPINE

LIBERTY

BADAMI

TRANS-ALASKA

PIPELINE SYSTEM

Figure 4-98. Major Alaska North Slope Fields


NORTH SLOPE GAS TREATMENT~4.5 BILLION CUBIC FEET PER DAYREMOVE CO2, OTHER IMPURITIESINITIAL COMPRESSION

�NORTHERN ROUTE� OPTIONALASKA TO ALBERTA PIPELINE52-INCH, BURIED PIPELINE1,800 MILES

�SOUTHERN ROUTE� OPTIONALASKA TO ALBERTA PIPELINE52-INCH, BURIED PIPELINE2,100 MILES

NATURAL GAS LIQUID EXTRACTIONPROCESS GAS TO DELIVERY SPECIFICATION(ALBERTA OR CHICAGO)

ALBERTA TO LOWER-48 PIPELINE~4 BILLION CUBIC FEET PER DAY52-INCH BURIED PIPELINE1,468 MILES

Figure 4-99. Alaska Gas Pipeline System

agreements with First Nations will be required for anAlaska gas pipeline project. The permitting processand potential legal challenges could cause significantdelays in an overall project schedule. In addition,permit stipulations could add significant costs to aproject that could make it more difficult to becomecommercially viable.

� State Fiscal. The state of Alaska and the Producersrecognize the need to establish fiscal certainty forthis high-risk project. The absence of clear and pre-dictable methods to calculate royalty and tax pay-ments to the state of Alaska over the life of a pipelineproject represents a significant uncertainty.

� Federal Fiscal Activity. In addition to the state ofAlaska’s activity to address state fiscal risk, the U.S.federal government is currently debating the needfor a federal fiscal package. There are differing viewswithin industry on the likely cost of and need for afederal fiscal package.

� Market. There is also significant market uncertain-ty in terms of the demand for gas and the price cus-tomers will be willing to pay for natural gas over a30+ year project life. For example, in the late 1970sit was expected that there would be sufficientdemand for natural gas in the U.S. lower-48 and thatprices would be sufficient to warrant construction ofthe Alaska Natural Gas Transportation System(ANGTS). However, by the early 1980s it was clearthat the high-cost ANGTS project was not econom-ic and could not be financed. For an Alaska gaspipeline project to be commercially viable, the mar-ket outlook over a 30+ year life must be sufficientlyencouraging to justify the large investment required.

Arctic Supply Assumptions for NPC Study

Canada

For purposes of this NPC study, it was assumed thatthe permitting, cost, and market hurdles identified ear-lier in this chapter are overcome and that a MackenzieGas Project starts up in 2009 and transports the vol-umes shown in Figure 4-100.

While the initial volumes that might be transportedby a Mackenzie Gas Project could range from 800MMCF/D to 1,200 MMCF/D, for the purposes of thisstudy it was assumed that the project would initiallytransport 1 BCF/D. This would consist of 800MMCF/D of gas from three anchor fields (Taglu,Parson’s Lake, and Niglintgak) as well as 200 MMCF/Dfrom other fields. It was further assumed that addi-

tional economic discoveries of gas are made to allowexpansion to 1.5 BCF/D in the year 2015 and to keepthe line full through the end of the study period(2025). Between 2009 and 2025, a total of 8 TCF wouldbe transported to market.

Alaska

For purposes of this NPC study, it was assumed thatthe permitting, state fiscal, cost, and market hurdlesidentified earlier are overcome and that an Alaska gaspipeline project starts up in 2013 and transports Alaskagas to Alberta. From Alberta it is assumed that the gasis transported through a combination of existingpipeline capacity or newly installed capacity to marketsin the U.S. lower-48. Figure 4-101 shows the volumestransported to Alberta. During the initial year (2013),it was assumed that only 2.5 BCF/D is transportedbecause not all the compressor stations would be com-missioned that first year. During the second and sub-sequent years, a full 4.0 BCF/D would be transported.Between 2013 and 2025, a total of 18 TCF would betransported to market.

In addition to gas from known discoveries, an addi-tional 16 TCF of “yet-to-find” gas would be required to


0

0.5

1.0

1.5

2.0

2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

YEAR

Figure 4-100. Canadian Arctic Gas Volumes

keep the pipeline full for a 30-year project life. In addi-tion, the pipeline could be expanded if additional eco-nomic discoveries were made.

Sensitivities

Given the commercial, regulatory, and cost-relateduncertainties associated with the Alaska pipeline proj-ect, sensitivity cases were developed that looks at threealternate outcomes:

� No Alaska pipeline is built during the study period

� The pipeline is delayed 5 years and production startsin 2018

� The pipeline is expanded by 1 BCF/D to 5 BCF/D in2020.

The results from the no Alaska pipeline and delayedpipeline cases were a projection of average natural gasprices 15% higher from 2013 through 2017. In the nopipeline case, the average natural gas price was also 7%higher from 2020 through 2025.

Recommendations

The projections in this study are generally favorablefor development of Arctic resources. Based on theseprojections, the NPC has assumed that both the

Mackenzie Delta pipeline and the Alaska pipeline areconstructed in a “success case” time period, withMackenzie gas initiating production in 2009 andAlaska in 2013. The timetable for Alaska gas is veryaggressive, and can only be met with prompt govern-ment action. Currently pending enabling legislation,which at a minimum would provide regulatory cer-tainty, creates an opportunity to take action and toensure the legislative requirements of such a massiveinfrastructure project are met.

Infrastructure projects of this magnitude require thefollowing:

� Congress should enact enabling legislation in 2003for an Alaska gas pipeline. Passage of this legisla-tion in 2003 is required to support deliveries of thisgas to the market in 2013. Council members andPrudhoe Bay producers agree that Congress shouldimmediately enact legislation that provides regula-tory certainty to such a project.

� Canadian agencies should develop and implementa timely regulatory process. The various govern-ments in Canada (federal, territorial, provincial)and the First Nations should continue to workcooperatively to develop and implement a timelyregulatory process. An efficient process must be inplace in early 2004 to support a 2009 Mackenzie GasProject start-up and a 2013 Alaska gas pipeline proj-ect start-up.

� Alaska needs to provide fiscal certainty for theproject. The state of Alaska should provide fiscalcertainty to project sponsors in a manner that issimple, clear, not subject to change, and that canimprove project competitiveness. Such action by theAlaska legislature in 2004 is required to support a2013 project start-up.

� Governments should refrain from potentiallyproject-threatening actions. Governments shouldavoid imposing mandates or additional restrictionsthat could increase costs and make it more difficultfor a project to become commercially viable.

� Infrastructure improvements incidental to Alaskagas pipeline construction must be planned in a time-ly and coordinated manner. The U.S. and Canadiangovernments – federal, state, provincial, and territori-al – should study and/or consult with one another andindustry participants and affected communities toassess contemplated infrastructure improvements insupport of Arctic gas development in advance of thetime when these improvements are needed.


0

1

2

3

4

5

2005 2010 2015

YEAR

2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

Figure 4-101. Alaskan Gas Volumes


Comparison to Other Supply Outlooks

The production outlook for the NPC 2003 ReactivePath scenario is lower than the projections from the1999 NPC study and the government’s preliminary EIA2004 Annual Energy Outlook. Figure 4-102 comparesthe three outlooks for production from the U.S. lower-48. While the Reactive Path outlook is for lower pro-duction, it is occurring in a more robust priceenvironment than either of the other two outlooks. Ina similar price environment, the 2003 NPC studywould project even lower production.

The production outlook for the NPC 2003 ReactivePath for Canada (excluding Arctic) is for relatively flatproduction through 2015, as shown in Figure 4-103.The National Energy Board of Canada (NEB) and theCanadian Energy Research Institute (CERI) have simi-lar near-term forecasts, with the NPC 2003 projectingless decline in the out years. There are some defini-tional differences between NPC and NEB/CERIaccounting for operational fuel use, so the NPC 2003projection has been adjusted to exclude lease and plantfuel for comparison to the marketable gas basis used byNEB/CERI.

In order to better understand the key differences inthese outlooks, a detailed review evaluated the maincomponents of each production outlook. The resultsof that review are summarized below.

Energy Information AdministrationPreliminary Annual Energy Outlook 2004

Results of the NPC 2003 Reactive Path scenario(“NPC 2003”) were compared to a preliminary releaseof the Energy Information Administration’s AnnualEnergy Outlook 2004 (“prelim. AEO 2004”). Detaileddiscussions were held between NPC study participantsand DOE and EIA staff to better understand the factorsinfluencing the differences in the outlooks. It is hopedthat this type of comparative analysis will lead to bet-ter overall projections by all parties concerned. Theresults of that analysis are presented below.

Lower-48 Total Production

Figure 4-104 shows lower-48 total production out-looks for NPC 2003 and prelim. AEO 2004. NPC 2003projects flat to modestly increasing total lower-48 pro-duction, rising from 50 BCF/D in 2002 to a peak of 54BCF/D by 2015. The prelim. AEO 2004 outlook is for

0

15

20

25

TR

ILLIO

N C

UB

IC F

EE

T

30

NPC 1999 ($2.50 - $3.50)

EIA PRELIM. 2004 OUTLOOK ($4.00 - $4.50)

NPC 2003 – REACTIVE PATH ($5.00 - $7.00)

2000 2005 2010 2015 2020 2025

YEAR

Figure 4-102. Lower-48 Production Outlooks

Note: Prices are in dollars per million Btu (Henry Hub, 2002 dollars).


NPC 2003 – REACTIVE PATH

NEB – SUPPLY PUSH

CERI (EST.)

NEB – TECHNO-VERT

0

4

5

6

7

8

2000 2005 2010 2015 2020 2025

TR

ILLIO

N C

UB

IC F

EE

T

YEAR

Note: NEB = National Energy Board of Canada; CERI = Canadian Energy Research Institute.

Figure 4-103. Canadian Production Outlooks for Marketable Gas (Excluding Arctic)

0

50

60

70

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

PRELIM. AEO 2004

NPC 2003

YEAR

Figure 4-104. Lower-48 Production

gradually increasing production, from 51 BCF/D in2002 to 59 BCF/D in 2020.


The technical resource bases for both NPC 2003 andprelim. AEO 2004 start from the same general datasources, the USGS for the onshore basins and the MMSfor the offshore. However, the NPC has made severaladjustments to the USGS/MMS data to:

� Change the overall field size distribution for certainplays

� Add more small fields (especially offshore)

� Adjust the USGS/MMS resource quantities perexpert industry opinion

� Use a different methodology to estimate Growth toKnown.

In turn, the EIA adjusted the USGS nonconven-tional resource base.

Table 4-12 compares the prelim. AEO 2004 accessibletechnical resource base (as of 1/1/2002) with the NPC2003 accessible technical resource base (as of 1/1/99).In order to put the studies on a similar basis, 56 TCF ofgas have been subtracted from the NPC 2003 technicalresource base estimate to account for three years ofproduction. Overall, the prelim. AEO 2004 unproventechnical resource base of 981 TCF of lower-48 gas is19% (211 TCF) higher than the NPC 2003 technicalresource base.

Undiscovered. Undiscovered resources are 33% (96TCF) higher in NPC 2003 than in prelim. AEO 2004.Onshore undiscovered resources are essentially equiva-lent, while offshore resources are 57% higher in theNPC study. While both studies started with the MMSassessment, the NPC added more small fields to thetechnical resource base and included the state waters.

Growth (EIA – Inferred Resources). Growth toproved reserves are 44% (132 TCF) lower in NPC 2003than in prelim. AEO 2004. Growth was modeled dif-ferently, with the NPC using the “cohort” methodologyas described in the Supply Task Group Report, and pre-lim. EIA 2004 using USGS growth curves. Overall,conventional resources are 6% (36 TCF) lower in NPC2003 than in prelim. AEO 2004.

Nonconventional. Nonconventional resources are45% lower (175 TCF) in NPC 2003 than in prelim.

AEO 2004. As detailed in Table 4-13, a significant dif-ference is in the Rocky Mountains, where the EIA tech-nical resource of 220 TCF (as of 1/1/2002) is over 90TCF higher than the 124 TCF (as of 1/1/99) of gas inthe NPC study. In addition, the prelim. AEO 2004 hasincluded higher nonconventional resources in theMidcontinent, the Louisiana-Mississippi Salt Basins,the Permian Basin (including the Barnett Shale), andthe Texas Gulf Coast.

It should be noted that there is possibly a definition-al difference between the two studies as regards non-conventional, tight gas resources. The NPC 2003definition of nonconventional is: “Large accumula-tions having regional spatial dimensions with diffuseboundaries which cannot be represented in terms ofdiscrete, countable reservoirs delineated by down-diphydrocarbon-water contacts. Common featuresinclude gas down-dip from water, lack of obvious trapsand seals, close proximity to source rock, and abnormalpressure (high or low).”

Reserve Additions per Gas Well

NPC 2003 projects lower average reserve additionsper gas well, as illustrated in Figure 4-105. The NPC2003 projected that lower-48 onshore per well reserveadditions will slowly decline from 0.8 BCF/well in 2000to 0.65 BCF/well by 2020. The prelim. AEO 2004 proj-ects a similar declining trend, however, it starts at over1 BCF/well in 2002 and declines to 0.8 BCF/well by2020.

The NPC 2003 outlook has higher conventionalreserve additions per gas well (green lines) while pre-lim. AEO 2004 has forecast much more robust non-conventional reserve additions per well (red lines).NPC 2003 projects that nonconventional reserves perwell will average approximately 0.6 BCF/well through2020. In contrast, prelim. AEO 2004 forecasts non-conventional per well reserve additions climbing to 1.4 BCF/well through 2007 and then remaining at thatlevel through 2020.

Drilling Activity

The number of annual gas wells projected in theNPC 2003 and prelim. AEO 2004 are depicted inFigure 4-106. The NPC is projecting annual gas wellsof just under 15,000 per year in the near term, rising to17,500 by 2013. Prelim. AEO 2004 forecasts the num-ber of gas wells will average approximately 16,000 over


CH

APTER 4 - N

ATURA

L GA

S SUPPLY

210

Prelim.

AEO 2004 NPC 2003 Difference

As of 1/1/02 As of 1/1/99

Less 3 years

of production As of 1/1/02 Change Percent

Undiscovered 289 388 3 385 96 33%

Onshore 149 168 1 167 18 12%

Offshore 139 220 2 218 79 57%

Deep (> 200 meters) 106 129

Shallow (< 200 meters) 33 91

Growth to Known (Inferred Resources) 301 204 35 169 (132) -44%

Onshore 243 148 26 122 (121) -50%

Offshore 58 56 9 47 (11) -19%

Deep (> 200 meters) 10 8

Shallow (< 200 meters) 47 48

Conventional Total 590 592 38 554 (36) -6%

Nonconventional 391 234 18 216 (175) -45%

Tight Gas 260 141 10 131 (129) -50%

Shale 54 35 6 29 (25) -46%

Coal Bed Methane 78 49 3 46 (32) -41%

Other – Low Btu 0 10 0 10 10

Total Lower-48 Unproved 981 826 56 770 (211) -22%

Note: Resources do not include areas where drilling is officially prohibited.

Table 4-12. Comparison of EIA Preliminary AEO 2004 and NPC 2003 Lower-48 Gas Resources(Trillion Cubic Feet of Gas Technical Recovery; Current Technology; Accessible Resource)

CH

APTER 4 - N

ATURA

L GA

S SUPPLY

211

NPC (as of 1/1/99) USGS Prelim. AEO 2004 (as of 1/1/02)

Tight ShaleCoal BedMethane Total Tight Shale

Coal BedMethane Total Tight Shale

Coal BedMethane Total

Appalachia + Warrior + Midwest

35 27 14 76 45 34 17 97 18 37 13 67

Midcontinent (Anadarko, Arkoma)

- - 5 5 - - 5 5 13 4 17

Louisiana – Miss Salt

6 - - 6 6 - - 6 34 34

Rocky Mountains 86 - 28 124 172 - 45 216 158 2 60 220

Green River 39 - 1 40 81 - 2 82 65 2 67

San Juan 19 - 7 26 26 - 24 50 26 17 42

Uinta/Piceance 18 - 5 23 19 - 2 21 22 9 31

Wind River - - 0 0 - - 0 0 24 24

Northern Plains 7 - - 7 43 - - 43 16 2 17

Powder River 1 - 13 14 1 - 14 15 28 28

Other 2 - 2 4 2 - 2 4 6 5 10

Low Btu 10

Permian Basin (Including Barnett Shale)

7 - 7 - 3 - 3 5 15 21

Texas Gulf Coast 3 - - 3 - - - - 24 24

Other 12 0 1 13 12 - 1 13 7 7

Total 141 35 49 234 235 38 67 340 260 54 78 391

Table 4-13. Nonconventional Technical Resource Base Comparison

the next ten years, before rising slowly to over 17,000per year by 2015.

Onshore average reserve additions per well, onshoredrilling activity, and production for 2015 are depictedin Figures 4-107, 4-108, and 4-109. While convention-al reserve additions per well are lower, the prelim. AEO2004 has forecast significantly higher activity levels,leading to higher onshore conventional production.While prelim. AEO 2004 is forecasting somewhat high-er nonconventional production, big differences in non-conventional recoveries are somewhat offset by lowerprelim. AEO 2004 activity levels.

Technology

Both the EIA and NPC outlooks incorporate technol-ogy improvement assumptions in their supply models.The model algorithms address these improvementparameters in slightly different ways, but in some casescomparisons can be made between the prelim. AEO 2004and the NPC 2003 cases. For conventional resources,EIA and NPC use technology improvement parametersthat annually improve either cost or success rates in var-ious areas such as drilling, infrastructure, and operating


0.0

0.4

0.8

1.2

1.6

2.0

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

WE

LL

NPC 2003 TOTAL ONSHORE

PRELIM. AEO 2004 CONVENTIONAL

NPC 2003 NONCONVENTIONAL

PRELIM. AEO 2004 TOTAL ONSHORE

PRELIM. AEO 2004 NONCONVENTIONAL

NPC 2003 CONVENTIONAL

YEAR

Figure 4-105. Lower-48 Onshore Reserve Additions per Gas Well

5

0

10

15

20

25

2000 2005 2010

YEAR

2015 2020 2025

GA

S W

ELLS

(T

HO

US

AN

DS

)

NPC 2003

PRELIM. AEO 2004

Figure 4-106. Lower-48 Onshore Gas Wells

cost or exploration and development success rates.Overall these assumptions compare closely between pre-lim. EIA 2004 and NPC 2003 (see Figure 4-110). Also,NPC includes an EUR/well improvement parameter forconventional resources where EIA does not.

For nonconventional resources, the most criticaltechnology parameter to compare is EUR/wellimprovement. Again, the algorithm between the EIAand NPC models vary in handling this parameter.However, from discussions and analysis, it appears thatNPC used higher annual improvements in EUR/wellthan did EIA in their prelim. AEO 2004 for noncon-ventional resources as shown in Figure 4-111.

Lower-48 Regional Production Response

As depicted in Figure 4-112, NPC 2003 and prelim.AEO 2004 production outlooks for the offshore Gulfof Mexico are quite similar. However, onshore pro-duction in prelim. AEO 2004 is forecast to rise from37 BCF/D in 2000 to 45 BCF/D by 2020, an increaseof 8 BCF/D over the 20-year period. The NPC 2003,by contrast, projects relatively flat onshore produc-tion. Figure 4-113 shows production profiles for


0 0.5 1.0 1.5

BILLION CUBIC FEET PER WELL

CONVENTIONAL

NONCONVENTIONAL

PRELIM. AEO 2004

NPC 2003

Figure 4-107. Onshore Average Reserve Additionsin 2015

0 5 10 15

WELLS PER YEAR (THOUSANDS)

PRELIM. AEO 2004

NPC 2003

CONVENTIONAL

NONCONVENTIONAL

Figure 4-108. Onshore Drilling Activity in 2015

0 5 10 15 20 25

PRELIM. AEO 2004

NPC 2003

TRILLION CUBIC FEET

CONVENTIONAL

NONCONVENTIONAL

OFFSHORE

Figure 4-109. Production in 2015


0.0 0.5 1.0 1.5

EXPLORATION SUCCESS

DEVELOPMENT SUCCESS

DRILLING COST

INFRASTRUCTURE COST

OPERATING COST

IMPROVEMENT FACTOR AT YEAR 2025 (RELATIVE TO 2002)

NPC 2003

PRELIM. AEO 2004

Figure 4-110. Technology Improvement for Conventional Resources

0.0 0.5 1.0 1.5 2.0

COAL BED METHANE

TIGHT SANDS

SHALE

EUR/WELL IMPROVEMENT FACTOR AT YEAR 2025 (RELATIVE TO 2002)

NPC 2003

PRELIM.

AEO 2004

Figure 4-111. Technology Improvement for Nonconventional Resources


0

10

20

30

40

50

60

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003 – OFFSHORE

PRELIM. AEO 2004 – OFFSHORE

NPC 2003 – ONSHORE

PRELIM. AEO 2004 – ONSHORE

YEAR

Figure 4-112. Lower-48 Gas Production – Onshore vs. Offshore

0

5

10

15

20

25

2000 2005 2010 2015 2020 2025

YEAR

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003:

ROCKIES

GULF COAST

MIDCONTINENT

WEST TEXAS

PRELIM. AEO 2004:

ROCKIES

GULF COAST

MIDCONTINENT

SOUTHWEST

Figure 4-113. Lower-48 Onshore Production for NPC 2003 and Prelim. AEO 2004

NPC 2003 and prelim. AEO 2004 for the followingregions:

� Gulf Coast Onshore

� Rocky Mountains

� Midcontinent

� West Texas (NPC 2003) and Southwest (prelim.AEO 2004).

The following observations can be made on theonshore regional comparisons:

� Rocky Mountains – The Rocky Mountains forecastsare similar, with increasing gas production.Preliminary AEO 2004 forecasts are somewhat moreaggressive in the near term and then rise at similarrates to the NPC.

� Permian Basin and Midcontinent – The forecasts ofthe more mature basins are generally similar. ThePermian Basin production plots are almost coinci-dent. In the Midcontinent, while similar, the prelim.AEO 2004 forecasts generally flat production, whilethe NPC outlook is for gradually declining production

� Gulf Coast – There is a large difference betweenNPC 2003 and prelim. AEO 2004, as NPC 2003 out-

look is for generally declining production and pre-lim. AEO 2004 forecast constant Gulf Coast produc-tion rates.

In terms of resource type, nonconventional produc-tion is expected to grow in both the NPC 2003 and pre-lim. AEO 2004 outlooks at roughly similar rates. Interms of conventional gas production, both studiesproject similar declines through 2010. Post 2010, theNPC projects that conventional production will con-tinue to decline, while prelim. AEO 2004 forecasts thatconventional production will flatten, as shown inFigure 4-114.

LNG, Arctic Production

The outlooks for LNG imports into the U.S. lower-48 are similar for the prelim. AEO 2004 and NPC 2003Balanced Future cases, as shown in Figure 4-115. TheNPC Balanced Future scenario assumes two additionalLNG receiving terminals than the Reactive Path sce-nario and quicker terminal permitting.

In terms of Arctic gas, Figure 4-116 shows the out-look for Alaska production, where the prelim. AEO2004 assumes a 2018 start-up of the Alaska pipelinewhile the NPC assumes 2013. For the Mackenzie


0

5

10

15

20

25

30

2000 2005 2010 2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

PRELIM. AEO 2004 – CONVENTIONAL

NPC 2003 – CONVENTIONAL

PRELIM. AEO 2004 – NONCONVENTIONAL

NPC 2003 – NONCONVENTIONAL

YEAR

Figure 4-114. Onshore Natural Gas Production

Delta pipeline, both outlooks assume a 2009 start-upas shown in Figure 4-117. However, prelim. AEO2004 has forecast initially higher volumes (1.85BCF/D vs. 1.1 BCF/D) with the NPC 2003 projectingan expansion in 2015 to increase the production ratesto 1.6 BCF/D.

Summary

The process of comparing and reconciling the EIAand NPC outlooks has been aimed at learning andimproving the quality of future outlooks. Some of thekey parameters that are contributing to the differentoutlooks are nonconventional well recovery expecta-tions and the overall level of projected conventionaland nonconventional drilling. In addition, althoughboth outlooks use the USGS/MMS resource assess-ments as the reference, each apply adjustments thatresult in a 19% difference in assessed technical resourcebase. Areas of similar outlooks and assumptions werealso highlighted.

The identification of the key drivers to the differingoutlooks will enable both organizations to review theirassumptions and methodologies, with an objective ofproducing improved outlooks in the future.

1999 NPC Study

The production outlooks for NPC 2003 and theNPC 1999 Reference Case (hereafter called “NPC1999”) were compared for key components, includingtotal production, technical resource base, recovery perwell, drilling activity level, technology improvements,and the resultant regional production outlooks.

Total Production

Figure 4-118 shows the production outlooks overallfor the U.S. lower-48 and for Canada for NPC 2003(solid lines) and NPC 1999 (dashed lines). In NPC2003, total production is forecast to remain near flat atapproximately 70 BCF/D through 2015. In contrast, inNPC 1999 total production is forecast to grow, reach-ing an annual production level of 92 BCF/D by 2015,21 BCF/D higher than the outlook in NPC 2003.

The biggest difference in the production outlookbetween the two studies, in both absolute and percent-age terms, is the forecast for lower-48 production. Inthe near term, as is evidenced by Figure 4-118, actual2002 production levels were approximately 6 BCF/Dlower than projected in the NPC 1999 outlook.


2

0

4

6

8

10

12

2000 2005 2010YEAR

2015 2020 2025

NPC 2003

BALANCED FUTURE

NPC 2003

REACTIVE PATH

14

PRELIM.

AEO 2004

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

Figure 4-115. LNG Net Imports to U.S. Lower-48


0.0

0.5

1.0

1.5

2.0

2000 2005 2010

YEAR

2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003PRELIM. AEO 2004

Figure 4-117. Mackenzie Natural Gas Production

0

2

4

6

8

2000 2005 2010YEAR

2015 2020 2025

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003

PRELIM. AEO 2004

Figure 4-116. Alaskan Natural Gas Production

Looking forward from 2002, the NPC 2003 outlookwas also less robust. NPC 2003 projects flat to mod-estly increasing lower-48 production, rising from50 BCF/D in 2002 to a peak of 54 BCF/D by 2015. Incontrast, the NPC 1999 outlook was for more stronglyrising production, increasing from 56 BCF/D in 2002to almost 72 BCF/D by 2015.

While a number of factors contributed to the differ-ent outlooks, three critical factors helped influence themore recent study:

� A lower assessment of the technical resource base

� The marginal productive response to theprice/drilling run-up in 2000 and 2001

� The documentation in NPC 2003 of the rapidlymaturing asset base, significant deterioration ofindividual gas well recoveries, and continuingincrease in base decline.

Canadian production forecasts in NPC 2003 andNPC 1999 are more similar. Production levels for 2002and 2003 are almost coincident. Post-2004, a 10-15%difference between the two outlooks graduallyemerges. NPC 2003 forecasts essentially flat produc-tion levels of 18 BCF/D through 2015 as western

Canadian production remains flat through 2007 andthen gradually declines. Increasing eastern Canadianproduction offsets that decline. In contrast, in NPC1999 Canadian production is forecast to grow from 18BCF/D in 2003 to 20 BCF/D by 2005 driven by pro-duction increases in western Canada, and then toremain at that level through 2015.

NPC 2003 documented the maturing nature of theWestern Canada Sedimentary Basin. Productionincreases, which were very strong in the early to mid-1990s have been slowing, and 2002 was the first year inrecent history of declining production. Recoveries perwell have been falling even more dramatically inWestern Canada than in the U.S. lower-48.


Figures 4-119 and 4-120 compare the technicalresource base assessed in all three NPC studies. In theU.S. lower-48, the technical resource base of 1,250 TCFis a reduction of 210 TCF (14%) from the 1999 study.About half of this occurred in the growth category,where the production performance analysis performedin the 2003 study projected lower future well recoveriesand hence less reserve growth in existing fields. Given


10

0

20

30

40

50

60

70

80

90

100

2000 2005YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003 – TOTAL PRODUCTION

NPC 1999 – TOTAL PRODUCTION

NPC 2003 – U.S. LOWER-48


NPC 2003 – CANADA*


* Excluding Arctic.

Figure 4-118. U.S. Lower-48 and Canadian Production


0

200

400

600

800

1,000

1,200

1,400

1,600

1992 NPC 1999 NPC USGS/MMS 2003 NPC

NONCONVENTIONAL

NEW FIELD

GROWTH

PROVED

TR

ILLIO

N C

UB

IC F

EE

T

Figure 4-119. Lower-48 Mean Assessment – 1999 Base, Advanced Technology

0

500

1,000

1,500

2,000

2,500

3,000

1992 NPC 1999 NPC 2003 NPC

CANADA LOWER-48

ALASKAMEXICO

TR

ILLIO

N C

UB

IC F

EE

T

Figure 4-120. North American Mean Assessment – 1999 Base, Advanced Technology

that the growth resource has the lowest supply cost,this lower assessment in the 2003 study would result inlower near term production.

The Canadian assessment is lower in the 2003 studyas well, primarily as a result of lower potential assessedin the Atlantic and Arctic frontier regions and a lowernonconventional resource assessment in the WesternCanada Sedimentary Basin. Since little of the frontierresource is developed in either study, only the lowernonconventional assessment had an impact on thelower production outlook.

Recovery per Well

One of the most significant differences betweenNPC 2003 and NPC 1999 is in terms of forecast recov-eries per gas well. In the two main producing regions,the U.S. lower-48 and Western Canada, NPC 1999forecast significantly higher recovery per well thanNPC 2003, as is depicted in Figure 4-121. In the U.S.lower-48, NPC 2003 well recoveries averaged approxi-mately 1.2 BCF/well in 2000 and were forecast toremain essentially flat through 2007, before fallinggradually to 1.0 BCF/well by 2015. In contrast, NPC1999 projected average well recoveries to rise from

1.7 BCF/well in 2000 to 1.8 BCF/well by 2006, andthen begin to decline, ending at 1.2 BCF/well in 2015.Through 2010, lower-48 well recoveries averagedapproximately 30% to 40% higher in NPC 1999 thanin NPC 2003.

In Western Canada, NPC 2003 forecast a gradualdecline in per-well recoveries, from 0.8 BCF/well in2000 to 0.6 BCF/well in 2010. NPC 1999, in contrast,forecast a significant increase in per-well recoveries,rising from 1.1 BCF/well in 2000 to 1.65 BCF/well by2010. By 2010, NPC 1999 per well recovery rates InWestern Canada were almost the times the average per-well recovery in NPC 2003.

The per-well performance analysis completed forNPC 2003 documented that average per-well recover-ies have been falling in Western Canada, as shown inFigures 4-122 and 4-123. While well mix has exacer-bated that trend, as the industry has concentrated ondrilling lower risk, shallow development opportunities,declining recoveries are evident across most drillingdepths and play types. Only the Devonian, favorablyimpacted by the rapid exploitation of the large wellrecoveries from the LadyFern Field, showed any devia-tion from this trend.


0.2

0.0

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2000 2005 2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

WE

LL





YEAR* Excluding Arctic.

Figure 4-121. Lower-48 and Canadian Recovery per Gas Well


0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

CUMULATIVE PRODUCTION (BILLION CUBIC FEET)

PR

OD

UC

TIO

N R

AT

E

(MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

1990-95

1996-98

1999

2000

2001


Figure 4-122. Western Canada Sedimentary Basin Production Rate vs. Cumulative Production

0

0.2

0.4

0.6

0.8

1.0

1.2

1990 1992 1994 1996 1998 2000 2002

VINTAGE YEAR

INIT

IAL R

AT

E

(MIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

ES

TIM

AT

ED

ULT

IMA

TE

RE

CO

VE

RY

(BIL

LIO

N C

UB

IC F

EE

T)

INITIAL RATE

INITIAL DECLINE RATE

EUR


0

10

20

30

40

50

60

INIT

IAL D

EC

LIN

E R

AT

E (

PE

RC

EN

T P

ER

YE

AR

)

Figure 4-123. Western Canada Sedimentary Basin Production Performance Trends

Within the U.S. lower-48, NPC 1999 forecasts werehigher both onshore and offshore. In the Gulf ofMexico, Figure 4-124 shows the NPC 2003 average wellrecoveries of 15 BCF/well in 2000, gradually falling to10 BCF/well by 2013, before rising marginally through2015. By contrast, NPC 1999 projected average Gulf ofMexico recoveries rising from 20 BCF/well in 2000 to 28BCF/well in 2005, before declining to 18 BCF/well by2015. NPC 2003 documented the large decrease in perwell recoveries in the Gulf of Mexico shelf, from over 5BCF/connection in 1990 to 3 BCF/connection in 2000.

As shown in Figure 4-125, onshore lower-48 wellrecoveries were also lower in NPC 2003. NPC 2003projects that average well recoveries would stay nearflat through 2015, at 0.8 BCF/D. This contrasts withthe NPC 1999 forecast of average well recoveriesincreasing from 1.1 BCF/well to 1.2 BCF/well through2011, before falling. As shown in Figure 4-126, in allcritical onshore basins, NPC 1999 forecast higher aver-age well recoveries than NPC 2003.

Drilling Activity

The outlook for drilling activity was higher in NPC2003 for both the U.S. lower-48 and Canada (see

Figure 4-127). In the U.S. lower-48, the NPC 2003outlook is generally 5-10% higher than NPC 1999,although differences in specific years could be higher,for example, nearing 25% in 2004 and 2005, spurredby higher expected gas prices. Differences in Canadawere larger, where NPC 2003 activity levels were 50%higher than those projected in NPC 1999.

In Canada, these higher activity levels were able topartially mitigate the much lower per-well recoveries.In the U.S. lower-48, marginally higher activity levelswere unable to offset the large differences in forecastper-well recoveries, and accordingly, the productionoutlook is markedly lower.

Technology

The 1999 and 2003 studies each handled technolo-gy improvement in a similar fashion. Both factored inannual improvement in costs, exploration success, andwell recoveries. The 2003 study added two parametersnot included in the 1999 study: operating expenseimprovement and development well success rateimprovement. A comparison of the improvement fac-tors between the 1999 study and the 2003 study isillustrated in the following figure. Overall, the 1999


5

0

10

15

20

25

30

2000 2005

YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

WE

LL

NPC 2003

NPC 1999

Figure 4-124. Gulf of Mexico Recovery per Well

0.2

0.0

0.4

0.6

0.8

1.0

1.2

1.4

2000 2005YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

WE

LL

NPC 2003 – ONSHORE*


* Plus Offshore West Coast.

Figure 4-125. Lower-48 Onshore Recovery per Well


0.0

2000 2005

YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

WE

LL

0.5

1.0

1.5

2.0

NPC 1999 – GULF COAST NPC 2003 – GULF COAST

NPC 1999 – MIDCONTINENT/PERMIAN

NPC 1999 – ROCKIES NPC 2003 – ROCKIES


Figure 4-126. Lower-48 Onshore Recovery per Well in Critical Basins

0

5

10

15

20

25

2000 2005

YEAR

2010 2015

WE

LLS

(T

HO

US

AN

DS

)





* Excluding Arctic.

Figure 4-127. U.S. Lower-48 and Canadian Gas Wells

study assumed technology improvements to have ahigher impact than the 2003 study. The largest vari-ance is in the exploration success rate improvement.The 1999 study, following a decade of significantimprovement due to 3D seismic technologies, project-ed exploration success rate to improve at approxi-mately 2% per year. With the recent slowdown in 3Dtechnology application, noted by the 2003 TechnologySubgroup, the improvement in exploration successwas estimated to be 0.8% per year in the 2003 study(see Figure 4-128).

Lower-48 Regional Production Response

As detailed in Figure 4-129, lower-48 forecast pro-duction differed both offshore and onshore. In theGulf of Mexico, NPC 2003 forecast generally flat pro-duction of 14-15 BCF/D. In contrast, NPC 1999 pro-jected near-term Gulf of Mexico production to climbstrongly to 20 BCF/D by 2005, and then to continue toincrease gradually to 22 BCF/D by 2010.

Onshore, NPC 2003 forecast flat to marginal pro-duction growth. In contrast, NPC 1999 forecastonshore production increasing from 38 BCF/D in 2000to over 50 BCF/D by 2015.

Figure 4-130 shows the regional, per basin lower-48production outlooks.

� Rocky Mountains – NPC 2003 forecast productionfrom the Rockies is similar to NPC 1999. Both out-looks have overall gas production rising steadilyfrom 8-9 BCF/D currently to 12-13 BCF/D by 2015.

� Permian Basin and Midcontinent – In the maturePermian/Midcontinent basins, while NPC 2003projects production to fall to 10 BCF/D, NPC1999forecast production growing to 12-13 BCF/D.

� Gulf Coast – The biggest difference between the twostudies onshore is in the Gulf Coast forecasts, partic-ularly in East and South Texas. NPC 2003 forecastthat Gulf Coast production will fall at just under 2%per year from approximately 14 BCF/D currently tojust under 11 BCF/D by 2015. In contrast, NPC1999 projected production from the Gulf Coast togrow from 14 BCF/D to over 19 BCF/D by 2015.

Overall, most of the factors impacting productionoutlooks (resource base, technology, and productionperformance) were more favorable in the 1999 study.Production performance, and specifically recovery perwell, is the biggest difference in the two outlooks and isthe driver to the lower outlook in the 2003 study.


0.0 0.5 1.0 1.5 2.0

EXPLORATION SUCCESS

DRILLING COST

INFRASTRUCTURE COST

RECOVERY FACTOR – CONVENTIONAL

RECOVERY FACTOR – NONCONVENTIONAL

IMPROVEMENT FACTOR AT YEAR 2025 (RELATIVE TO 2002)

NPC 2003

NPC 1999

Figure 4-128. Technology Improvement Comparison – NPC 1999 vs. NPC 2003


10

0

20

30

40

50

60

2000 2005YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 2003 – GULF OF MEXICO

NPC 1999 – GULF OF MEXICO



* Plus Offshore West Coast.

Figure 4-129. Lower-48 Gulf of Mexico and Onshore Production

5

0

10

15

20

2000 2005

YEAR

2010 2015

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NPC 1999 – GULF COAST NPC 2003 – GULF COAST


NPC 1999 – ROCKIES NPC 2003 – ROCKIES


Figure 4-130. Lower-48 Production Outlooks for Onshore Basins


Supply Recommendations

The principal recommendations of the Supply TaskGroup are related to increasing the diversity of supplythrough new supplies entering the market. To facilitatethis, government policies must remove impedimentsthat inhibit delivery of the additional supplies.

The new supply sources are broadly characterized as:

� Lower-48 resources that are currently restricted orface permitting impediments

� North American Arctic gas

� Increased LNG imports.

Support for all new supply sources is required tomeet the expected growth in natural gas demand. Therecommended actions to facilitate development ofthese new supply sources are discussed in the Access,Arctic, and LNG sections of this chapter, and are sum-marized below.

Access

Onshore – Increase Access (ExcludingWilderness Areas and National Parks) andReduce Permitting Costs/Delays 50% overFive Years

� Improve government land-use planning.Governing agencies should use ReasonableForeseeable Development scenarios as planningtools rather than to establish surface disturbancelimitations. Land use planning and project moni-toring should be a priority in order to facilitate time-ly plan revisions and project permitting.

� Expedite leasing of nominated and expired tracts.The federal government should expedite the leasingof nominated tracts and expired leases. This can befacilitated by use of existing planning documentsand reducing requirements for extraneous environ-mental analysis where appropriate.

� Expand use of categorical exclusions or sundrynotices as alternatives to processes imposed by theNational Environmental Policy Act. Every surfacedisturbance activity requires environmental analysisprior to permitting. NEPA costs and delays can bereduced through the use of categorical exclusions orsundry notices instead of environmental assess-ments for minimal disturbance activities andthrough improvement of data sharing and coordina-tion by state and federal land management agencies.

� Streamline and expedite permitting processes. Thepermitting process should be streamlined by estab-lishing performance goals for each office, reducingon-site inspections, increasing use of sundry noticesin lieu of Application for Permit to Drill (APD), andusing dedicated teams to support high workloadfield offices. This should be continuously monitoredand refined by efficient and comprehensive report-ing, benchmarking, and best practices programswithin the Bureau of Land Management and ForestService, etc.

� Establish cultural resource report standards andeliminate duplicate survey requirements. This isthe most frequent cause of delays and expense forAPD and right-of-way approvals. Significant costreductions and time savings can be realized by elim-inating duplicate surveys, developing clear standardsfor determining site significance, and establishingclear cultural report review requirements amonggoverning agencies.

� Establish qualification requirements and technicalreview procedures for nomination of endangeredspecies. There currently exists no qualificationrequirements to nominate a species for listing, andonce nominated, these species are given the sameprotection as listed endangered species. This resultsin delays to land management planning and projectpermitting until a ruling on the nominated species.It is recommended that this process be changed toestablish qualification requirements and technicalreview procedures to prevent such unwarranteddelays.

� Fund and staff federal agencies at levels, and inmanners, appropriate for timely performance ofresponsibilities. Federal land management agenciesneed to ensure adequate resources to efficiently han-dle responsibilities for updating land-use plans,administering the NEPA process, processing leaseand permit applications, and resolving appeals andprotests in a timely manner. The Bureau of LandManagement should consider the formation of ded-icated teams to assist field offices with high permit-ting workloads.

Offshore – Lift Moratoria on Selected Areasof the Federal OCS by 2005

� Lift, in a phased manner, moratoria on selectedOCS areas having high resource-bearing potential.Federal and coastal state governments, working withindustry and other stakeholders, should develop a


plan to identify current moratoria areas of theEastern Gulf of Mexico and Atlantic and PacificCoasts containing a high resource potential, with aview toward lifting the moratoria in a phasedapproach beginning in 2005.

� Update resource estimates for MMS-administeredareas. The federal government (Minerals Manage-ment Service) should coordinate the development ofupdated estimates of natural gas resources underly-ing the OCS submerged lands and identify the datagathering activities that could be undertaken toimprove the technical support for this estimate.

� Ensure continued access to those OCS areas identi-fied in the 2002-2007 5-Year Leasing Program.

� Ensure that Marine Protected Areas are meetingtheir intended purposes. Regulatory requirementsfor protection of marine species should be based onthe best available scientific analysis to avoid inap-propriate or unnecessary action having uncertainbenefit to the intended species. Lease stipulationsand operational measures should be practical, costeffective, and aimed to achieve minimal delays inongoing operations.

� Require federal and state joint development ofCoastal Zone Management (CZM) Plans. Ensurethat federal and state authorities improve coordi-nated development and review of CZM Plans tounderstand the impact on federally authorized andregulated OCS activities. If a state alleges that aproposed activity is inconsistent with its CZMPlan, it should be required to specifically detail theexpected effects, demonstrate why mitigation isnot possible, and identify the best available scien-tific information and models which show that eachof the effects are “reasonably foreseeable”. TheSecretary of Commerce should not approve stateCZM Plans if such implementation would effec-tively ban or unreasonably constrain an entireclass of federally authorized and regulated activi-ties, such as gas drilling, production, and trans-portation.

Arctic� Congress should enact enabling legislation in 2003

for an Alaska gas pipeline. Passage of this legisla-tion in 2003 is required to support deliveries of thisgas to the market in 2013. The NPC and PrudhoeBay producers agree that Congress should immedi-ately enact legislation that provides regulatory cer-tainty to such a project.

� Canadian agencies should develop and implementa timely regulatory process. The various govern-ments in Canada (federal, territorial, provincial)and the First Nations should continue to workcooperatively to develop and implement a timelyregulatory process. An efficient process must be inplace in early 2004 to support a 2009 Mackenzie gasproject start-up and a 2013 Alaska gas pipeline pro-ject start-up.

� Alaska needs to provide fiscal certainty for theproject. The state of Alaska should provide fiscalcertainty to project sponsors in a manner that issimple, clear, not subject to change, and that canimprove project competitiveness. Such action by theAlaska legislature in 2004 is required to support a2013 project start-up.

� Governments should refrain from potentiallyproject-threatening actions. Governments shouldavoid imposing mandates or additional restrictionsthat could increase costs and make it more difficultfor a project to become commercially viable.

� Infrastructure improvements incidental to Alaskagas pipeline construction must be planned in atimely and coordinated manner. The U.S. andCanadian governments – federal, state, provincial,and territorial – should study and/or consult withone another and industry participants and affectedcommunities to assess contemplated infrastructureimprovements in support of Arctic gas developmentin advance of the time when these improvements areneeded.

LNG

The goal of the following recommendations is toreduce the time required for LNG facility permitting toone year.

� Agencies must coordinate and streamline theirpermitting activities and clarify positions on newterminal construction and operation. Projectsponsors currently face multiple, often-competingstate and local reviews that lead to permitting delays.A coordinated effort among federal, state, and localagencies led by FERC would reduce permitting leadtime. Similarly, streamlining the permitting processby sharing data and findings, holding concurrentreviews, and setting review deadlines would providegreater certainty to the overall permitting process.FERC should further clarify its policy statement onnew terminals so as to be consistent with correspon-ding regulations under the Deep Water Port Act,

including timing for the NEPA review process andcommercial terms and conditions related to capaci-ty rights.

� Fund and staff regulatory agencies at levels neces-sary to meet permitting and regulatory needs in atimely manner. The expected increase in the num-ber of terminal applications will require higher lev-els of government support (federal, state, and local)to process and avoid delays. Additional agencyfunding/staffing will also be required once these newterminals become operational, particularly to sup-port the large increase in LNG tanker traffic.

� Update natural gas interchangeability standards.Standards for natural gas interchangeability in com-bustion equipment were established in the 1950s.The introduction of large volumes of regasified LNGinto the U.S. supply mix requires a re-evaluation ofthese standards. FERC and DOE should championthe new standards effort to allow a broader range ofLNG imports. This should be conducted with par-ticipation from LDCs, LNG purchasers, process gasusers, and original equipment manufacturers. DOEshould fund research with these parties in support ofthis initiative.

� Undertake public education surrounding LNG.The public knowledge of LNG is poor, as demon-strated by perceptions of safety and security risks.These perceptions are contributing to the publicopposition to new terminal construction and jeop-ardizing the ability to grow this required supplysource. Industry advocacy has begun, but a moreaggressive/coordinated effort involving the DOEand non-industry third parties is required.Emphasis should focus on understandings, safety,historical performance, and the critical role thatLNG can play in the future energy supply.

� LNG industry standards should be reviewed andrevised if necessary. In order to promote the high-est safety and security standards and maintain theLNG industry’s safety record established over thepast 40 years of operations, FERC, the Coast Guard,and the U.S. Department of Transportation should

undertake the continuous review and adoption ofindustry standards for the design and constructionof LNG facilities, using internationally proven tech-nologies and best practices.

Additional Supply Considerations

There are additional actions and policy initiativesthat could be undertaken to potentially enhance sup-ply sources. Among those are the role played by taxand other fiscal incentives or packages, and the desir-ability of additional government-sponsored researchspending.

Two strongly held views of fiscal incentives emergedduring the study discussions. Supporters of suchincentives believe additional production would resultfrom pursuit of marginal opportunities and/or highcost supply alternatives, helping to ease the tight sup-ply/demand balance. Others believe market forces areand will be sufficient to stimulate additional invest-ment without the need for tax-relates incentives orsubsidies. Potential fiscal incentives such as tax creditsfor nonconventional resource development, low-BTUgas, stripper oil well and deep gas drilling incentives,and an Alaska pipeline fiscal package were discussed,but the NPC makes no recommendation in this regard.

With respect to government research, the NPC issupportive of a role for DOE in upstream research,particularly where it complements privately fundedresearch efforts. DOE’s natural gas research programhas a significant role in technical studies and relatedwork that support public policy decision-makingregarding natural gas supply. DOE currently spendsabout $50 million per year on jointly sponsored natu-ral gas technology research. This represents 53% of thefunding for oil and gas research, but only 9% of thefunds directed at fossil energy programs in total. TheNPC believes DOE should evaluate whether this levelof funding is appropriate in relation to other DOE pro-grams in light of the increasing challenges facing natu-ral gas. Further discussion of this issue is included inthe Technology section of the Supply Task GroupReport.


CHAPTER 5 - TRANSMISSION, DISTRIBUTION, AND STORAGE INFRASTRUCTURE 231

For purposes of organization, and in acknowl-edgement of the differing issues of the majorpipeline transmission and distribution market

segments, the Transmission & Distribution Task Group(T&D Task Group) has chosen to separately report onthe areas of pipeline transmission, distribution, andstorage. In aggregate, the subsections form a coherentanalysis, just as the separate but conjoined efforts ofthe study’s Task Groups (Demand, Supply, andTransmission & Distribution) have been combinedinto an integrated document.

Study Approach

In order to incorporate a wide range of industryexpertise, the T&D Task Group was comprised of 26U.S. and Canadian representatives from the followingnatural gas industry sectors: pipeline transmission; dis-tribution; storage; marketers; and producers. Whenissues arose outside of the specific participant knowl-edge areas, experts within the represented companies,as well as firms not directly represented on the panel,were contacted for their views. Care was taken to co-ordinate with the other Task Groups (Supply andDemand) through liaison members. This liaisonapproach was also followed with the important ad hocgroups, such as Arctic Gas and LNG Imports.Government representatives included participation byrepresentatives from the U.S. Department of Energy(DOE), the Federal Energy Regulatory Commission(FERC), and the Energy Information Administration.

This analysis relied upon supply and demand dataprovided by the other Task Groups as well as data fromthe Energy Information Administration, the AmericanGas Association (AGA), the Interstate Natural Gas

Association of America (INGAA), and other industryassociations. NPC member companies also provideddata. Early in the study, the T&D Task Group deter-mined and set the major exogenous variables requiredfor the analysis. Examples of these determinationsincluded: selecting pipeline capacity expansions andnewbuilds within the first five years; setting the “lag” ordelay between a price signal and the construction of arequired pipeline developed subsequent to the first fiveyears; determining the cost differentials for construc-tion (pipeline, storage, and distribution) by region; andestimating the amount of storage required for humanneeds (residential/small commercial) services.

With regard to the issues facing the T&D TaskGroup, EEA’s Gas Market Data and Forecasting Systemmodel makes economically justified decisions to routenatural gas, expand pipeline capacities, and constructnew storage facilities. The modeling software consistsof a complex nodal (physical flow) structure, which isfundamentally based on unit pricing concepts.Decisions to flow gas through existing facilities and/ordecisions to build pipelines between nodes, add incre-mental storage facilities, build additional facilities atthe citygate, etc., are “calculated” in the model on ayear-by-year basis. The network used in the modelincorporates 115 supply/demand nodes and 317 trans-portation corridors (see Figure 5-1). The model willalways attempt to use existing facilities to their maxi-mum, while at the same time looking for pricing sig-nals that would support facilities expansion either toexisting facilities or with greenfield projects.

Model output was then carefully reviewed by theT&D Task Group to search for and correct any anom-alies. Once the results of the major scenarios (ReactivePath and Balanced Future) were approved, sensitivities

TRANSMISSION, DISTRIBUTION,AND STORAGE INFRASTRUCTURE

CHAPTER 5

of the Supply and Demand Task Groups (which resultin differing data inputs to the T&D model) were alsoreviewed for their impact on T&D results. In addition,the T&D Task Group chose to evaluate its own sensi-tivities to validate certain stresses upon NorthAmerican infrastructure.

Summary of Results

The study shows that continued expansion of gastransmission, storage, and distribution facilities will berequired to meet the future needs of gas consumersand suppliers, but there remains a critical dependencyon the existing natural gas infrastructure. Neededexpansions or enhancements include increasing thecapacity of existing infrastructure, developing pipelinelaterals connecting new supply, storage, and generationfacilities, expanding of distribution networks, andbuilding multi-billion dollar pipelines that link Arcticsupply regions to the North American grid.

Two scenarios and multiple sensitivities were ana-lyzed with respect to the timing and location of new

major supply sources as well as cases related todemand reduction. A status quo approach to natu-ral gas policy yields undesirable outcomes because itdiscourages economic fuel choice, new suppliesfrom traditional basins and Alaska, and new lique-fied natural gas (LNG) terminal capacity. The NPCdeveloped two scenarios of future supply anddemand that move beyond the status quo. The twoscenarios were the Reactive Path and the BalancedFuture. The Reactive Path scenario assumes contin-ued conflict between natural gas supply and demandpolicies that support natural gas use, but tend to dis-courage supply development. This scenario resultsin continued tightness in supply and demand, lead-ing to higher natural gas prices and price volatilityover the study period. The Balanced Future scenariobuilds in the effects of supportive policies for supplydevelopment and allows greater flexibility in fuel-switching and fuel choice. This results in a morefavorable balance between supply and demand, priceprojections more in line with alternate fuels, andlower prices for consumers.


ELBAISLANDLNG

SOUTHATLANTICOFFSHORE

LAKE CHARLES LNG

COVE PT.LNG

EVERETT

EASTCANADA NODES

NACO ETC.

JUAREZ

REYNOSANORTHERNNODES

MID-ATLANTICOFFSHORE


WEST COASTOFFSHORE

Figure 5-1. Supply/Demand Nodes and Transportation Corridors

The major results for the Balanced Future are sum-marized below. These results will be compared to theReactive Path in the Scenarios and Sensitivities sectionof this chapter.

� Transmission. Estimated expenditures for newNorth American transmission pipelines, includingsustaining capital, are $2.7 billion/year (2002 dol-lars) over the study period, from 2004 to 2025. Thiscompares to $3.5 billion/year expended between1996 and 1999.

While capital for new infrastructure declines in theprojection, especially in the later years, sustainingcapital increases and becomes a greater percentage oftotal capital requirements. This is a result of invest-ments for continuing compliance with the PipelineSafety Improvement Act and the fact that increasinginvestments are required for an aging infrastructureto assure its safe and reliable operations.

� Distribution. Estimated expenditures for newNorth American distribution pipelines, includingsustaining capital, are $5.3 billion/year (2002 dol-lars) over the study period, from 2004 to 2025. Thisapproximates to amounts expended between 1996and 1999. The successful development of this distri-bution system infrastructure will depend on severalkey factors, including:

– Obtaining inter-agency coordination and regula-tory certainty in all permitting processes

– Obtaining access to expansion capital

– Maintaining the historical levels of reliability andflexibility of natural gas services as gas demandgrows and load patterns change

– Developing mechanisms to foster research anddevelopment.

� Storage. Estimated expenditures for new NorthAmerican storage facilities, including sustainingcapital are $0.4 billion/year (2002 dollars) over thestudy period, from 2004 to 2025. This is slightlyhigher than that expended between 1996 and 1999.It is important to note that these estimates do not

include the cost of base gas, which is projected to beone of the largest components of future storageexpenditures. Other observations related to storageinfrastructure are:

– Projected growth in weather-sensitive demandwill require up to 700 billion cubic feet (BCF) ofadditional working gas capacity by 2025.

– Given that the geologic base for potential storagecapacity is highly exploited, new storage facilitiesmay be located further from the markets theyserve and may be increasingly expensive todevelop.

– A return to normal weather (30-year average)would require overall storage utilization ratesabove those experienced in the 4 years prior toDecember 2002.

– Demand for gas storage can be as much as 25%higher than normal in a year in which winterweather is significantly colder than normal.North American storage capacity has not beentested by such a winter for many years and, assuch, it is likely that current storage capacity willbe severely challenged to meet such demands.

Figures 5-2 and 5-3 show capital expenditures forNorth America. As can be seen, there is significantvolatility in the amount spent on transmission facili-ties, but expenditures generally decline in the outeryears. In addition, as the established infrastructureages, a significant portion of the ongoing transmissionexpenditures are used to sustain existing capacity.From 2000 to 2002, sustaining capital is estimated as21% of total transmission expenditures. By 2020 to2022, sustaining capital will increase to almost 75%.Sustaining capital for transmission, distribution, andstorage is estimated as 21% of total expenditures for2000-2002. By 2020, sustaining capital for the threesegments is projected to be 45% of total expenditures.

Sustaining capital for transmission was calculatedon the basis of replacing 700 miles of pipe and 77,000horsepower of compression each year. This is viewedas a conservative estimate because it is a small fractionof the existing 290,000 miles of pipe and 16,000,000horsepower of compression, much of which is over 40years old. For instance, if we assumed a 50-year life forpipelines, then the appropriate replacement rate forpipe would be over 5,800 miles per year. The basis forusing the lower number is that it better matches thehistorical level of replacement. Because of the impacts


Pipeline and distribution investments willaverage $8 billion per year, with anincreasing share required to sustain thereliability of existing infrastructure.

of the Pipeline Safety Improvement Act of 2000, how-ever, we doubled the historical levels for the purposesof the study. At some point in the future, the progres-sive aging of pipelines and compressors will result in afurther significant increases in the miles of pipe andhorsepower replaced per year.

Pipeline and storage infrastructure developmentshave generally been financially supported by contractswith a term of ten to twenty years. In a free market,shippers make long-term contract commitments whenthey see the need for the service that will be provided.Recently, the average transportation contract term fornew/proposed and existing pipeline and storage infra-structure has trended shorter. Much of the trend is theresult of market choices, while some is caused by theimpact of regulatory policies which may create barriersto choice. When such barriers exist to shippers makinglong-term commitments, investment in pipeline andstorage infrastructure is impacted, as the related rev-


0

2

4

6

8

10

12

14

2000 2005 2010 2015 2020 2025

BIL

LIO

NS

OF

2002 D

OLLA

RS

YEAR

16

GAS DISTRIBUTION - NEW

STORAGE - NEW

TRANSMISSION - NEW

GAS DISTRIBUTION - SUSTAINING

STORAGE - SUSTAINING

TRANSMISSION - SUSTAINING

ARCTIC PIPELINES

Figure 5-2. Detailed North America Capital Expenditures for Transmission, Distribution, and Storagein Balanced Future Scenario

Regulatory barriers to long-term contractsfor transportation and storage impairinfrastructure investment.

enue stream is viewed as more short-term in natureand less likely to support long-term infrastructureinvestment.

Transmission

The United States’ pipeline transmission infrastruc-ture has been developed over a period of eight decadesand has provided the nation with reliable access toNorth American natural gas supply. The infrastructuregrew rapidly in World War II to meet the needs of theburgeoning wartime economy and continued itsgrowth during the industrial economic expansion ofthe 1950s and 1960s. In the 1970s, the pipeline trans-mission system grew from 255,000 miles to 266,000miles and expenditures averaged $2.7 billion per year.Despite the negative impacts of a faltering economyand price deregulation, the transmission system grewfurther in the early 1980s to 271,000 miles.

U.S. natural gas consumption has grown signifi-cantly from its low point in 1986, rising from 16.2 tril-lion cubic feet (TCF) (44.4 BCF/D) to an estimated22.6 TCF (61.4 BCF/D) in 2001.1 During this period,the dominant growth sector was electric generation,including industrial combined heat and power, and the

gas transmission grid in the U.S. grew from 281,000miles to 285,000 miles.2 The U.S. grid is a significantpart of the North American grid of large-diameterpipelines, which is shown in Figure 5-4.

Despite the large amount of pipeline transmissiongrowth, there have still been periods in which thedemand for capacity has exceeded its supply. Theseconstraints have resulted in increased price differen-tials between upstream supply regions and down-stream markets. For example, Western Canadianprices were significantly below those of the down-stream markets during the 1990s, with price differen-tials sometimes rising above $1.25 per million Btu(MMBtu). As a result, capacity was added.

The California supply/demand imbalance during2000 and 2001 also led to multiple pipeline construc-tion projects including expansions on the Trans-western, El Paso, and Kern River pipelines and the con-version of Southern Trails Pipeline from oil to gas


1 Energy Information Administration, Natural Gas Annual,Table 6.5, Natural Gas Consumption by Sector, 1949-2001.

2 Department of Transportation RSPA 7100.2-1.

0

2

4

6

8

10

12

14

2000 2005 2010

YEAR

2015 2020 2025

NEW INFRASTRUCTURE

SUSTAINING CAPITAL

ARCTIC PIPELINES

BIL

LIO

NS

OF

2002 D

OLLA

RS

NEW INFRASTRUCTURE

SUSTAINING CAPITAL

ARCTIC PIPELINES

Figure 5-3. Total North American Capital Expenditures for Transmission, Distribution, and Storagein Balanced Future Scenario


Figure 5-4. North American Pipeline Grid (24" Diameter and Greater)

service. In aggregate, these projects brought over 1.3 BCF/D of new capacity to California.

The one U.S. region that has experienced an ongoingcapacity shortfall is the Rocky Mountain supply area.In response, a number of new export projects haverecently been proposed for the region, includingAdvantage, Western Frontier, Front Range, CheyennePlains, Bison, Southern Trails, TransColorado/SilverCanyon, Powder River Basin North, Northwest PipelineRockies Expansion, and Ruby. Periodic constraintsappear to be the result of a rapid growth in supply thatsurged ahead of potential shippers’ commitments to thelong-term pipeline contracts required to facilitate newpipeline construction. Market participants will decidewhich of the projects will move forward and when.

Results from the Study

In the United States, pipeline capacity utilizationfactors in the Reactive Path scenario are projected toundergo significant changes during the 22-year fore-cast period:

� The Midcontinent production region (Oklahoma/Kansas) has some of the largest changes in capacityfactors, with usage factors on pipelines runningfrom the Midcontinent to the Midwest marketregion dropping from 94% in 2000 to 54% in 2025.

� The Texas intrastate market sees major flow realign-ment, with capacity factors on pipelines runningfrom the Permian Basin to East Texas, droppingfrom 81% at the start of the period to 7%. IfMexican production fails to grow at the rate forecastby SENER, then the steady growth in demand pro-jected over the period may cause U.S. exports toMexico to increase rather than decrease.

� Capacity factors from Northern Louisiana to theMidwest market areas drop from 75% to 57% asArctic supply and/or Canadian supply replaces GulfCoast gas in the Midwestern energy markets in thelatter part of the study period. There is also somepotential reduction in utilization factors in pipelinesmoving gas from the Gulf Coast to the Mid-Atlantic,assuming LNG landed in East Coast market centershelps to serve demand growth in that region as wellas create additional upstream delivery capabilitythrough existing pipeline resources.

� The one supply region showing little excess capacityis the Rocky Mountains. This region shows signifi-cant production growth over the study period,growing from 4.4 BCF/D in 2000 to 9.2 BCF/D in

2018 before experiencing a slow decline to 8.7 BCF/D in 2025. As a result of the increase intransmission capacity prior to 2018 and a subse-quent decline in production, capacity factors onpipelines leading east of the region have a capacityutilization rate lower in 2025 than in 2000. Thecapacity factors on pipelines leading to California,however, are above 93% for the entire period.

� In Canada, Western Canada Sedimentary Basin pro-duction peaks at 17.9 BCF/D in 2005. Capacity uti-lization to eastern Canada drops from approxi-mately 94% in 2000 to 81% in 2025. Production inthe Maritimes area of eastern Canada rises to 1.3 BCF/D in 2011, undergoes a gentle decline toapproximately 1.0 BCF/D in 2019, and then risesonce again to 2.2 BCF/D in 2025.

� The Balanced Future scenario also features increasedsupply access to the Rocky Mountain and OuterContinental Shelf regions. As a result, flow patternschange from those in the Reactive Path. For exam-ple, the Midcontinent to Midwest capacity factor is74% in 2025 in the Balanced Future versus 54% inthe Reactive Path. Other notable changes in theBalanced Future include over 1.5 BCF/D of produc-tion from the Atlantic offshore that flows into EastCoast markets, a drop in capacity factors fromCanada to the Pacific Northwest from 70-80% to 50-60%, and a drop in west-to-east Canadian long-haulutilization of 81% to 73%.

Pipeline capacity must also be constructed to trans-port gas from storage fields to high consumption cen-ters. This is particularly true for storage developed toserve the Mid-Atlantic and New England markets. Asnoted in the Storage section of this chapter, these tworegions will require an additional 135 BCF of workinggas storage by 2025. Because the nearest suitable andundeveloped reservoirs exist in the western portions ofPennsylvania and New York, eastern Ohio, andOntario, incremental pipeline capacity of approxi-mately 2.0 BCF/D will have to be constructed to linknew storage capacity to the coastal market centers,which include New York City, Boston, andPhiladelphia. The incremental pipeline capacityrequired by 2025 is shown in Figure 5-5.

Future Environment

In describing throughput trends, it is illustrative toexamine the balance of flows into major marketregions. For this purpose, a major market region is



4000 1500

28

100

144

223

1400

320

11422000383

2540

9001500

467

80

1098

1448

346

50

200

75

146

146

150

450

360

400

250

570

1500750

1000

1500

1500

1500

750

1000

1500

1460

3000

500-2000

2000

300

PIPELINE CAPACITY

LNG IMPORTS

Figure 5-5. New Pipeline and LNG Capacity Change from 2003 to 2025 in Balanced Future Scenario(Million Cubic Feet Per Day)

defined as one in which consumption exceeds produc-tion (New England, Northeast, Mid-Atlantic, SouthAtlantic, Florida, East South Central, Midwest, UpperMidwest, West North Central, Pacific Northwest, andCalifornia).

Between 2000 and 2010, there is an aggregate netconsumption growth (consumption minus intra-regional production) of 4.5 BCF/D in the primarymarket regions. Incremental LNG deliveries into thesemarket regions are projected to account for 3.3 BCF/Dof this increased demand. As such, only 1.2 BCF/D ofadditional long-haul deliveries are needed from netsupply to net consumption regions.

Between 2010 and 2020, lower-48 consumption inthe major market regions has a further increase of3.6 BCF/D. In this period, LNG imports into net mar-ket areas is projected to increase by 1.5 BCF/D, result-ing in a need to increase long-haul transport from tra-ditional supply regions such as the Gulf of Mexico.From 2020 to 2025, net demand in major marketregions is projected to remain stable. During thisperiod, the net market area increase in consumption isexceeded by projected increases in LNG deliveries.Thus, no additional long-haul capacity development isrequired.

In Canada, net consumption growth in the majormarket regions (defined as regions where demandexceeds supply, namely Ontario, Quebec, andManitoba) is 0.36 BCF/D from 2000 to 2010, or 1.0%per year. Between 2010 and 2020, growth is again pro-jected to be 1.0% per year or 0.40 BCF/D. From 2020to 2025, the net consumption is projected to declineslightly. Over the study period, there will be no growthin long-haul capacity to eastern Canada as demandgrowth will be met through enhancement and utiliza-tion of existing pipelines.

The projected changes in flows across the majorNorth American pipeline corridors are displayed inFigure 5-6 (2004 to 2010) and Figure 5-7 (2010 to2020), which are both taken from the BalancedFuture scenario. As a result of the decreasing supplyin the mature regions of the United States, pipelinesconnected to these areas will see a gradual decline inthroughput. This should be particularly true for thesouthern sections of pipelines serving the WestTexas/Permian Basin to Midwest corridor. The mid-dle/northern sections of these systems (i.e., Kansas,Nebraska, etc.) will be re-supplied, however, by

growing Rocky Mountain production fed eastwardvia new pipelines, such as the completed Trailblazerexpansion, the Cheyenne Plains project, theAdvantage proposal, and the Western Frontier pro-posal.

A significant source of new supply is LNG imports,which rise from less than 0.6 BCF/D in 2000 to almost6 BCF/D in 2010 and then to 12-15 BCF/D by 2025.When located on the Gulf Coast, these supplies help tomaintain throughput in pipelines originating from theGulf Coast. When located directly in market regions,these facilities will access demand typically with onlyshort-haul infrastructure expansion required. LNGreceived in the market regions also has the effect ofincreasing upstream pipeline delivery capability, as gasthat previously used the long-haul path will be dis-placed to potential upstream markets by the LNGreceived downstream.

As mentioned above, production from the WesternCanada Sedimentary Basin peaks in 2005 and thenundergoes a long-term decline to 2025, when produc-tion drops to 14.3 BCF/D. Part of the productiondecline is replaced by Arctic gas from Mackenzie Deltaand Alaska. The first flow from Mackenzie Delta intoAlberta is expected in 2009 at 1.0 BCF/D, increasing to1.5 BCF/D in 2016. The Alaska production is projectedto begin in 2013 at 2.5 BCF/D and then increase in2014 to 4.0 BCF/D for the remainder of the forecastperiod. The combined Arctic flow more than offsetsthe projected decline in western Canadian productionin the early part of the study. To accommodate thesechanges in supply, however, major new pipeline sys-tems will need to be constructed from the frontierregions to interface with existing pipeline infrastruc-ture in northern Alberta.

Additional pipeline capacity will also be required toexport Alaskan gas from Alberta to U.S. and Canadianmarkets. Options for transporting this gas includeusing existing capacity spared by a decline in WesternCanada Sedimentary Basin production, expandingexisting pipelines, and constructing new pipelines.The NPC analysis suggests that an additional 0.5 to2.0 BCF/D of new or expansion capacity may beneeded to move the gas from Alberta to downstreammarkets. The amount of export capacity is very sensitive to changes in the western Canadian sup-ply/demand balances and could change significantlyby the time investment decisions are made regardingAlaskan gas.


Future development costs for long-haul pipelineinfrastructure and for connection to new storage andpowerplant facilities are forecast to be slightly belowhistorical levels. The cost to construct the new NorthAmerican pipeline facilities is expected to average $2.0 billion/year (2002 dollars) over the period to 2025.The projected investments are somewhat front-loaded,with the average for the years 2003 to 2010 expected tobe almost $2.3 billion/year. These capital expenditurelevels compare to an investment rate of $3.5 billion/year, which occurred between 1996 and 1999. Theexpected decline in the rate of capacity developmentresults from several factors, including a substantialincrease in LNG imports delivered to major marketcenters and the flow of new supplies into existingpipelines that currently have or are forecast to havespare capacity. Both of these actions promote effi-

ciency by maximizing utilization of existing infrastruc-ture while minimizing the need for new construction.

Scenarios and Sensitivities

The two scenarios generated results that were veryclose in terms of total North American transmissionpipeline miles constructed and expenditures. TheReactive Path projection had 41,200 miles of interstatepipeline constructed over the 2003 to 2025 period at acost of $1.98 billion/year. In the Balanced Future fore-cast, the analogous numbers were 43,500 miles and$2.02 billion/year. These cost projections in both sce-narios are below actual expenditures for the lastdecade, which indicates interstate pipeline develop-ment should not be a limiting factor in achieving thenecessary supply/ demand balance.


DECREASED FLOW

INCREASED FLOW

LNG IMPORTS

657

(363)

(388)

(189)

(12)

(134)

(291)

259

(187)

(135) (377)

181

39

(102)

159

(195)

(13)

185

18

(299)

1

(2)

1047

498688

15

50

8480

489

(3)

(55)

126

(407)

40

71

8

82

(15)

(33)

(44)

(95)

(144)

(434)

(150)

(27)

900

490

(34)

(117)

259

(566)

(108)

(41)

(193)

900

500

56

(339)

Figure 5-6. Flow Change from 2004 to 2010 in Balanced Future Scenario (Million Cubic Feet per Day)

The location of the infrastructure costs variedbetween the two scenarios, however, with more of theBalanced Future’s expenditures occurring in theUnited States versus Canada. The Balanced Futureshows a decline in infrastructure requirements for botheastern and western Canada due to increased produc-tion from U.S. areas currently limited by access restric-tions. Since the Balanced Future postulates improvedaccess to U.S. domestic resources, more infrastructureis required in the United States. Spending in theUnited States is thus $77 million per year higher in theBalanced Future while Canadian expenditures declineby $37 million per year.

An important sensitivity is the one in which newLNG import facilities are not approved for construc-tion in the Mid-Atlantic and Northeast regions, caus-

ing that LNG to be landed at sites within the Gulf ofMexico. Although no new transmission capacity isrequired for the Reactive Path, incremental pipelinecapacity of approximately 0.3 BCF/D must be builtfrom the Gulf Coast to Florida markets in the BalancedFuture to accommodate the incremental LNG pro-posed in that scenario. Although little incrementalinfrastructure is required, this sensitivity results inhigher prices in the Mid-Atlantic and Northeast mar-kets due to a tighter supply/demand balance andpipeline capacity constraints. According to the resultsof the sensitivity analysis, the delivered costs to NewYork City are about $0.07/MMBtu higher by 2010. Thevariance between the two scenarios widens to$0.30/MMBtu in 2015 and to $0.44 in 2025. Theanalysis quantifies the higher gas prices associated with


DECREASED FLOW

INCREASED FLOW

LNG IMPORTS

804

(467)

(231)

8

(442)

45

565

(32)

1128

1871055

266

(1)

(2)

88

411

632

(46)

(92)

(196)

(3)

174

48

129

(29)

265

(93)

79

(43)72

(113)(13)

(100)

100

(226)

600

(463)

441(182)

230

(221)

402

(274)

(447)

835

96

(627)(849)

107

100

(292)

900

63

(99)

237

(7)

(212)

(730)

13

(16)

(30)

115

165

100

10

173

47

456

(406)

70

750

1000

1000

750

1250

800

0

0

0

Figure 5-7. Flow Change from 2010 to 2020 in Balanced Future Scenario (Million Cubic Feet per Day)

not allowing facilities to be built in the region that con-sumes approximately 8.6 BCF/D or 14% of the currentU.S. total. For instance, for a consumption of 8.6BCF/D, the difference in delivered prices of$0.30/MMBtu in 2015 results in an increased energycost of $942 million for that year alone.

Another significant impact to gas transmissionrequirements occurs in the Cold Weather sensitivity.In this forecast, one of the coldest 23-year sequences ofweather over the last 70 years was used to determinewinter demand. The years used in the forecast were1956 to 1978, with the temperature patterns in 1956shifted to 2003, 1957 to 2004, etc. The average priceover the full 23-year projection was little changed asthe temperature average for the period was only 3%lower than the temperature pattern used in theReactive Path and Balanced Future scenarios. Thestandard deviation of the price, however, was muchhigher, as the 23-year forecast had episodes of weatherthat were much colder than normal. Thus, the stan-dard deviation of the average price for the ReactivePath was $0.69/MMBtu whereas the standard devia-tion for the Cold Weather sensitivity was$0.98/MMBtu. The $0.29/MMBtu variation is suffi-cient to support the development of additional trans-

mission or storage infrastructure. The effect of colderthan normal or warmer than normal weather onannual prices is shown in Figure 5-8.

Challenges to Building and Maintaining theRequired Transmission Infrastructure

Contracting Challenges

During the first seven decades of its history, the nat-ural gas transmission industry’s development wasunderpinned by long-term contracts held by local dis-tribution companies (LDCs). The LDCs ensured thefinancial integrity of pipeline infrastructure by signing20-year contracts under which pipelines were respon-sible for the bundled purchase and delivery of the gasto the LDC citygate.

This integral relationship between the transmissionand LDC industries began to change in 1983 withFERC’s issuance of Order No. 380, which allowedLDCs to modify their existing gas purchase obliga-tions with pipelines. Further changes occurred in1986 when FERC, in Order No. 436, adopted open-access policies on interstate pipelines, which allowed“shippers” to use a pipeline’s capacity to schedule thedelivery and receipt of gas. In combination, these


-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2000 2005 2010 2015 2020 2025

COLD WEATHER

WARM WEATHER

RANGE OF

WEATHER-INDUCED VOLATILITY

YEAR

AV

ER

AG

E A

NN

UA

L D

OLLA

RS

PE

R M

MB

TU

Figure 5-8. Weather Sensitivity Minus Balanced Future (at Henry Hub)

Orders gave other parties the ability to begin to com-pete directly with the pipelines for the gas merchantfunction.

FERC Order No. 636, adopted in 1992, furtherchanged the competitive marketplace by essentiallyeliminating the historical pipeline gas sales function.As a result of this paradigm shift in future regulation,pipelines were limited to providing transportation andstorage services only and could no longer buy or sellnatural gas, except for limited operational reasons.This “unbundling” of the transportation and storagefunctions required each upstream supplier and down-stream consumer to inherit the responsibility toarrange for the purchase or sale of gas on their ownbehalf. New tariffs were written and contracts wereentered into for unbundled transportation and storageservices. In addition, FERC required that a secondarymarket in transportation and storage services beallowed to develop, wherein shippers could “release” aportion of their contracted capacity to a creditworthythird party for their use, either on a short-term orlong-term basis.

Soon thereafter, a new business segment of gas mar-keters evolved. By the end of the 1990s, marketers hadsignificantly expanded their role to include a broadportfolio (through the capacity release process or oth-erwise) of pipeline transmission and storage capacitycontracts as well as acting as a managing agent of suchresources for others. In their role as managing agent,marketers’ goals were to optimize the use of pipelineand storage assets held by their counter parties, such asLDCs and industrial users, generally reducing theseparties’ daily participation in the evolving market. Insuch business structures, LDCs and end-users“swapped” use and optimization of these assets forongoing gas management and reduced risk.Correspondingly, marketers saw such arrangements asopportunity and potential upside, as they could usethem in a variety of ways that the LDCs and end-usersmight not.

When LDCs and other major consumers began pur-chasing gas supplies from marketers, their contractswere generally chosen to be of short duration, i.e., 1-3years. In such a scenario, marketers often mirroredtheir risk, becoming short-term holders of pipelinecapacity as a means of matching their overall contrac-tual exposures. Some marketers did, however, sub-scribe to longer-term contracts to facilitate the con-struction of new infrastructure.

In this unbundled interstate pipeline world, the nextmarket evolution was the unbundling in the 1990s ofthe sales and transportation functions of many LDCs.This unbundling of LDC services was mandated bysome state public utility commissions (PUCs) with theexpectation that it would increase competition andlower prices to consumers behind the citygate. By theend of the 1990s, unbundling was complete in manystates for the industrial gas and electric generation cus-tomers and was underway in some states in the resi-dential and commercial sectors. One belief at manyPUCs during this time was that unbundling LDCs,with the advent of competition, should no longer enterinto long-term pipeline capacity contracts since theirshare of the future gas sales behind the citygate wasuncertain. In fact, many LDCs were prohibited or dis-couraged from maintaining these contractual commit-ments.

During this period, producers became increasinglyimportant as subscribers of new supply area pipelinecapacity, especially capacity associated with greenfielddevelopments (often referred to as a supply-push sce-nario). Where it made sense to commit to proposedinfrastructure projects to assure their product wasavailable to market, many producers have done such.The producer’s goal was to ensure that they could reli-ably transport and sell their gas at a liquid, i.e., highvolume, sales point where it could receive a marketprice that was not reduced by a capacity constraint.

Another subscriber to capacity during the 1990s wasthe marketing affiliate of interstate pipelines.Although the pipelines could no longer buy and sell gasthemselves, they were allowed to have an affiliatedcompany that did so. By the end of the 1990s, marketaffiliates of pipelines were subscribing to largeamounts of capacity in new transmission projects, par-ticularly where third parties weren’t willing to do so.For newly constructed capacity, the FERC requiredsuch contracting with their affiliate to be under an “at-risk” condition to the pipeline when it chose to buildon this somewhat speculative basis, i.e., withoutdemonstrating long-term contracts from third partiesfor the proposed capacity.

Today, the recent turmoil in the gas-marketing sec-tor has dramatically reduced the number of independ-ent and affiliated marketers as prospective subscribersto existing and/or proposed pipeline transmissioncapacity. Even where such firms might want to con-tract for capacity, their current creditworthiness may


make them too great a risk for pipelines to consider.With some LDCs still being discouraged or prohibitedfrom entering into longer-term contracts by theirPUCs, considerable uncertainty exists regarding theidentity of the parties that will contract for unutilizedcapacity on existing pipelines or who will sign long-term capacity contracts for future pipeline projects.

Contracting New Capacity

As stated previously, a key concern for the pipelinetransmission industry is the entity that will contractfor new and existing pipeline capacity. To give a per-spective, the Power Generation, Marketing,Production, and LDC sectors contracted for 91% ofthe firm transmission capacity subscribed in theUnited States as of December 2002. The percentageholdings of these sectors have, however, undergone amarked transformation over the last five years. TheMarketing sector increased its share of total firmcapacity from 13% to 24% over the period. With thisbusiness segment in turmoil over the last two years,this has exacerbated the uncertainty surrounding theidentity of companies that will contract for firm trans-mission capacity in the future.

The Power Generation and Production sectors’pipeline capacity holdings grew at a smaller rate of5 BCF/D and 2 BCF/D, respectively. The LDC andIndustrial sectors, the most important segments ofindustry growth as recently as ten years ago, wereessentially unchanged over the interval.

The LDC sector holds the largest amount of firmcapacity of all the sectors discussed. Between 1998 and2002, LDC firm capacity remained constant at50 BCF/D. The key issue faced by the distribution andtransmission industries is the recontracting of existingLDC contracts for firm pipeline capacity. During thenext five years, 71% of all LDC firm capacity expires.As a result of the large amount of contract expirations,LDCs are viewed as unlikely to contract significantamounts of new firm transportation capacity, espe-cially given the current reluctance of some PUCs toallow them to enter into long-term contracts.

Another marked change within the industry relatesto the expiration profile of firm transportation con-tracts. At year-end 2002, 77 BCF/D or 64% of the totalfirm transportation contracts were set to expire withinthe following five years. In 1998, the comparableamount was 51%. The 13% increase in expirations

between the two five-year periods again indicates acontinuing movement to shorter-term commitments.The result is that regulatory practices (prudencereviews and ratemaking) may be inhibiting efficientmarkets and discouraging the financial incentives todevelop and maintain pipeline infrastructure. Thisinformation is graphically displayed in Figure 5-9.

Given the importance of the Power sector to thegrowth projections in this study, it is worthwhile tofocus on that sector in more detail. The gas-firedpower generation capacity increased approximately128,000 megawatts from 1998 to 2002. This generationconsisted of combined-cycle gas turbine (CCGT)installations that generally are intermediate dispatchand tend to operate more than their gas turbine coun-terparts, which are generally used for hourly electricpeaks. If this generation capacity were to have beencompletely utilized, a significant amount of daily gastransmission capacity would have been required forsupply to the plants. In a survey of nationwide con-tracts, however, firm gas transmission capacity forpower generators increased by 13 BCF/D, indicatingthat participants in this sector chose to contract for lessthan 100% firm transportation capacity, determiningthat was within a manageable level of need andrisk/exposure. From the survey, the expiration profileof the Power sector’s gas transmission capacity is dis-tributed across the next 20 years. However, the con-tracts are skewed in 2003 as the Marketers’ tended tosource numerous of these facilities, i.e., their short-term contracting orientation/strategy, as can be seen inFigure 5-10.


19981998

2002

2002

0

10

20

30

40

50

60

70

80

<5 YEARS >5 YEARST

OT

AL E

XP

IRA

TIO

NS

OF

FIR

M C

ON

TR

AC

TS

(P

ER

CE

NT

)

Figure 5-9. Firm Contract Expirations

It is important to note that approximately 190,000megawatts (57%) of gas power generation capacity atyear-end 2002 relies on non-firm gas transmissioncapacity. These were market choices, as operators ofthese facilities have assumed the risk of service inter-ruption by not securing firm contracts. Possible impli-cations are as follows:

� As the utilization rate for these generation plantsincreases and surplus pipeline capacity declines, gasaccessibility using interruptible pipeline capacitywill become increasingly problematic.

� During the summer season, increasing power uti-lization will often conflict with traditional gas stor-age injections and will strain the pipeline and stor-age system resources. For example, gas injectionsmay be pushed into only the evening hours and/ormore injections may be required earlier or later inthe summer season, i.e., in the shoulder months ofApril, May, and October.

� The current fleet of gas-fired generation – many ofwhich do not have fuel flexibility to consider alter-nate fuels – and future power development facilitiesmay not be able to depend on immediately availablesurplus pipeline capacity.

Fortunately, the natural gas industry has time torespond to any increased pipeline transmissionrequirements. The recent, rapid buildup of gas-firedgeneration has increased generation reserve marginsabove the required levels in most regions such that lit-tle new generation construction is likely to occur overthe next few years. Under projected power demandgrowth in this study, the levels of throughput on long-haul pipelines should increase over time as generatorsare increasingly utilized, but the full potential of theirdemand and the need for new supporting pipelineinfrastructure should not be felt until after 2008.

For a major pipeline expansion or a new project, themaximum pipeline tariff or transportation rate is nor-mally, but not necessarily, calculated using an annual-ized cost component and contract volumes. Thus, theapplicable tariff rate is frequently the same in a low-demand month (April) as a high-demand month(January). This non-varying cost for firm transport,when combined with large swings in seasonal marketdemand, can result in large variations in capacity uti-lization, citygate prices, and the realized market valueof pipeline capacity.

This seasonal variability in the realized market val-ues of pipeline capacity may increase its worth. The


0

1

2

3

4

5

6

7

2005 2010 2015 2020

YEAR CONTRACTS EXPIRE

VO

LU

ME

EX

PIR

ING

(BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y)

MARKETERS

UTILITIES

Figure 5-10. Firm Contract Expiration by Year and by Shipper Class

increased worth results from having the downside riskof holding capacity capped at the rate paid for thecapacity while the upside value is not limited in today’smarket. Since the firm shipper has bought the right tocall on the capacity at any time, the combination(capped costs, assured access, uncapped sales prices,and observed price volatility) creates a potentially valu-able option. Because gas prices are volatile, the samerelationship holds true for monthly and daily timeintervals. In all three cases, the holder of transporta-tion capacity has asymmetric risk with a fixed down-side exposure and an uncapped but highly uncertainupside potential. Some market participants would liketo “hold” this option; others would not. This optionalso has value in the secondary market for transporta-tion and storage capacity that has developed. However,this opportunity is troublesome for some LDCs whereregulatory barriers exist that impede them from con-tracting for capacity to serve their customers.

The response time or “lag” between the occurrenceof a price signal, i.e., an increased price differentialbetween two points, and the time at which a proposedproject can gain sufficient commitments to go forwardcan vary significantly between one project and another.In cases where a number of companies are in agree-ment that the basis is significant and lasting, the periodbetween a project proposal and construction can befairly short. In the case of the most recent Kern RiverExpansion, the developer held an open season inAugust 2000, filed for a FERC certificate in November2000, made a final investment decision in March 2001,and was in commercial operations by May 2003.

One of the challenging problems in new pipelineproject development is the fact that non-contractingparties on both ends of the pipeline system may ulti-mately benefit from new capacity construction becauseof the new infrastructure’s impact on price and basisvalue. For example, all western Canadian producersbenefited from the price increase that followed thedevelopment of the Alliance Pipeline, not just the pro-ducers who actually contracted for the capacity toChicago. As is typical in a free market, there may beconsiderable jockeying among potential project ship-pers to contract for only the minimum (or no) amountof capacity while still having a pipeline project pro-ceed. Pipelines, of course, must seek fairly large proj-ects so that the benefits of scale can keep proposedcosts and tariffs down. Since a perceived ideal positionis to allow others to commit but to still be able to reapall or a portion of the benefits from the removal of a

capacity constraint, gaining a critical mass of long-term commitments can be problematic for a pipelinedeveloper. This is why some projects have multipleopen-seasons, why competitive projects surface whenpreviously announced projects appear to falter, andwhy some projects just don’t proceed. This is typical ofa market at work, but can be very frustrating forpipeline developers and parties who desire to see suchprojects implemented.

Similarly in market regions, new capacity projectsare also problematic when important consuming sec-tors are either inhibited or not motivated to sign long-term firm contracts. Merchant power generators, forexample, may choose to not subscribe to firm contractsfor all or a portion of their supply, as these importantgas consumers may not believe a 24-hour, 365-daypipeline service is required, or the insurance valueassociated with such capacity certainty is not costeffective.

Another customer sector that may be disinclined tosubscribe to long-term pipeline contracts is the LDC.Since LDCs have been the anchor tenants for most ofthe pipeline capacity constructed over the last sevendecades, continuing market evolution and resultantregulatory policies may have created barriers to long-term capacity contracts that have impeded infrastruc-ture investment. Historically, with an obligation toserve human needs customers, LDCs have maintaineda level of pipeline capacity to do such. In the unbun-dled environment today, certain service requirementsare still mandated. Where applicable, regulatory bod-ies must ensure that providers of last resort (POLRs) orother entities providing service to human needs cus-tomers – whether gas or electricity – are allowed tomake pipeline capacity commitments necessary forlong-term service reliability.

Contractual commitments by various parties arecritical to the expansion of the pipeline network.However, as different approaches to pipeline contract-ing are evolving in a changing gas marketplace, thereappears to be a new paradigm evolving in contractingpractices. First of all, contracts appear to be of shorterterm. Second, it is becoming increasingly difficult forsome pipelines to contract the middle portion of atransportation path. A producer may elect to contractfor pipeline capacity only as far downstream as the firstunconstrained point, while some LDCs, on the otherhand, may chose, or must chose only, to contract forcapacity from the citygate to the nearest upstream liq-


uid market point. These points are usually locatedwithin a market area, which may be located hundredsof miles from a supply region.

This trend creates a bifurcation in the pipelinecapacity market. This “gap in the middle” is an anom-aly of the current natural gas marketplace; thisdilemma will affect the decisions of pipeline operatorsconcerning the creation of new capacity and sustainingthe existing capacity levels between the supply andmarket regions. Left to itself, the natural gas industrywill find equilibrium. Clearly, however, governmentalpolicies should not inhibit the ability of LDCs andPOLRs to extend their contracts into the supplyregions.

Construction Challenges

Regulatory approval of new pipeline proposalsinvolves agency reviews at the federal, state and locallevels. Review levels and procedures by agencies varysignificantly from state-to-state with the only commonreview level and approach occurring at the federalagency level. Examples of project review and approvaldurations range from 6 months to 42 months, depend-ing upon the number of agency approvals and com-plexity of the project. FERC has been making greatstrides in improving the time for approval, but manytimes, the project is held up by some other agency evenafter FERC has issued a certificate. These delays inproject approvals can be a significant driver of projectcost increases. Also, as projects are increasinglydelayed, prospective customers may begin to look foralternatives and ultimately terminate their agreements,with such withdrawals sometimes causing entire proj-ects to collapse.

Previous discussions between the industry and thefederal government on the difficulties in coordinatinga pipeline project among the various federal agenciesled to a memorandum of understanding (MOU) in2002. This MOU established a framework for earlycooperation and participation among “ParticipatingAgencies” to enhance the coordination of the regula-tory processes through which their environmental andhistoric preservation activities could occur. Reviewresponsibilities under the National EnvironmentalPolicy Act of 1969 are met in connection with theFERC authorizations that are required to construct andoperate interstate natural gas pipelines. Among theparticipating agencies are the U.S. Army Corps ofEngineers, U.S. Forest Service, National Fisheries, Landand Minerals Management, U.S. Department of the

Interior, U.S. Department of Transportation, AdvisoryCouncil on Historic Preservation, FERC, Council onEnvironmental Quality, and U.S. EnvironmentalProtection Agency.

The National Environmental Policy Act requires fed-eral agencies to evaluate the environmental impact ofmajor federal actions significantly affecting the qualityof the human environment. The MOU encouragesearly involvement with the public and relevant govern-ment agencies in project development to foster aprocess to facilitate the timely development of needednatural gas pipeline projects.

The chair of the White House Council onEnvironmental Quality has stated that the new proce-dure will improve coordination and speed up naturalgas pipelines that currently encounter years of envi-ronmental reviews by various federal agencies. Theextension and full integration of this type of coordina-tion to the state level will also be required, however,before genuine progress can be made.

Further progress could be made by developing aJoint Agency Review Process that would coordinateactivities between federal, state, and local agencies. Alead agency (perhaps FERC) could be assigned theauthority to complete the review/approval in a timelymanner, while meeting the concerns of all agencies andstakeholders. In order to be effective, this processshould be the “governing” process, i.e., not to be fur-ther limited or delayed when approvals have beenreceived to proceed from other responsible agencies.The areas of greatest concern in this regard arerequirements of the U.S. Army Corps of Engineers,Coastal Zone Management Act, and Section 401 of theClean Water Act, all of which could hinder the orderlyimplementation of FERC certificates. One example ofthis concern is the escalating use of the Coastal ZoneManagement Act to delay pipeline progress as exempli-fied by the serious delays currently experienced by theMillennium and Islander East Pipelines.

The recent FERC emphasis in the United States is toidentify key stakeholders early and involve them in theprocess at the outset of a proposed project. An effec-tive approval process allows third parties to becomeinvolved during designated comment periods. Inthese designated comment periods, external stake-holders, such as landowners or other special interestgroups are given the opportunity to voice any con-cerns with the pipeline route. FERC is required to


accept and reasonably address all stakeholder com-ments, and thus can ask the pipeline company toresearch and possibly resurvey each proposed routechange, involving both civil and environmental sur-veys, which can result in significant project delays andunanticipated cost overruns. The Joint Agency ReviewProcess would minimize these inefficiencies, as theprocess should, via significant upfront participation,agree upon a route or options thereto which canuniquely be investigated.

In addition to interventions in the approval process,delays can arise from stipulations in the approval withregard to construction issues, such as short time win-dows for laying pipe, work space limitations in certainareas, or mandated construction methods. The limitedtime periods or “construction windows” are frequentlyrequired by various state and federal agencies and canadd significant costs and delays during construction ofa pipeline project. These restrictions require carefulplanning of construction timing and implementation,and even then weather conditions or other unantici-pated delays (labor, materials, etc.) during the con-struction window can make it difficult to complete thework during the allotted time period. If a project isdelayed past the end of the construction window, thenthe operator may have to wait until the opening of thenext window (and this could be up to a year following)to complete the project, often at substantial additionalcost to the project.

Environmental agencies can also require pipelinecompanies to limit the width of pipeline constructionrights-of-way to reduce tree clearing or other earth dis-turbances. In some of these cases, the pipeline mustthen be installed by stove-piping the pipeline at thelocation (welding one or two pipe joints at a time andthen burying them as you go – a very tedious process)or by welding a portion of the pipe at a more accessibleoffsite location and hauling it along the right-of-waywith large equipment called “side booms.” These con-struction requirements due to work space limits willincrease project costs substantially. These are, ofcourse, further complicated and magnified if construc-tion windows are involved.

Mandated construction techniques often occurwhen pipelines have to cross water bodies, wetlandareas, or major roadways. Environmental agencies,either state or federal, can order the use of special tech-niques, which can include horizontal directionaldrilling, special top-soil separation, and use of wood

mats in wetland soils. Horizontal directional drillingcan add in the range of $200 to $1,000 per foot in addi-tional costs to the length of pipe. Use of mats at wet-land locations can add an additional $50 to $100 perfoot to the pipeline costs in areas where they are used.

Environmental agencies can also require offsite“mitigation” in wetlands construction. The purchaseof property for offsite mitigation can add substantialdelays and costs to the project. In many cases, theagency will not sign off on construction approvals untilthe corresponding property identified for mitigationhas been purchased. Delays occur since the pipelinecompany has to search for suitable acreage for mitiga-tion, obtain necessary clearances for the mitigationsite, and then complete the purchase of the land. Withthe high level of mitigation ratios (two to one is com-mon and five to one occurs), as well as having to estab-lish the mitigation site for long-term, pristine land usequality requirements, mitigation lands can be veryexpensive to purchase.

For some projects, development responsibilities canextend beyond the actual construction period withincreasing requirements for ongoing monitoring andrepairing the pipeline corridor. Environmental agen-cies are now requiring pipeline companies to developand implement a long-term monitoring program tomonitor, document, and correct/repair pipeline corri-dor restoration. This ongoing monitoring and repairprogram can add significant costs to the projectdepending upon environmental sensitivity of thelands, streams, and rivers crossed.

The typical project timeline for a major interstatepipeline project with an Environmental Assessmentthat is filed under a FERC 7 (c) certificate is normally12 to 20 months from project initiation to the recep-tion of the FERC authorization to construct. The typ-ical project timeline for a FERC 7 (c) filing for a majorproject requiring an Environmental Impact Statementfrom project initiation to the FERC authorization toconstruct is 18 to 24 months.

Once permits are obtained and land is acquired,most U.S. pipeline construction projects are typicallyconstructed in one calendar year or construction sea-son. Projects that transverse through areas with issuessuch as endangered species, high population densityareas, historic artifacts, noise mitigation, and safetyconcerns require 6 to 18 months beyond a more typi-cal timeline. Proposed pipelines that impact areas with


these types of concerns may require numerous reroutesin the planning process to obtain final public and reg-ulatory support for permit approvals.

One clear trend in pipeline construction in both theUnited States and Canada is for the continuing escala-tion of costs. Costs have been increasing about 3 to 4%per year, above the projected 1.5% annual rate used inthe study. The rising costs associated with new con-struction are a barrier to infrastructure development,but the modest nature of the cost increase is notexpected to necessarily make required pipeline facili-ties projects uneconomic.

Operations and Maintenance ofExisting Infrastructure

Operational Challenges

If all the flows entering and exiting a pipeline wereconstant in nature, then it would be a relatively easysystem to operate. Operators could set the compres-sors along the system to calculated levels and thepipeline would be “balanced” thereafter. This is calleda “static” system in engineering and unfortunately it isnot reflective of events in the natural gas industry.

Instead, natural gas transmission pipelines aredynamic systems with conditions constantly varying atlarge numbers of receipt and delivery points. Existingnatural gas wells experience mechanical problems,freeze offs, and production declines that change deliv-eries into the system. At the same time, new gas wellsare added and consumers vary their demand accordingto temperatures, industrial processes, and electric gen-eration needs. The throughput capacity of a systemthus varies with the amounts of gas entering and exit-ing the system, the pressures at each inlet and exitpoint, and the locations of these supply and demandpoints, particularly with regard to compressor stations.

Within the dynamic system described above, thereare three major consumption cycles that affect thetransmission industry. The first is a seasonal variationof demand, from winter to summer. The second cycleis a demand variation within a season or a month. Thelast is the change in hourly consumption during a dailycycle.

The seasonal variation exists largely due to con-sumption within the residential and commercialdemand segments. A large component of annual nat-ural gas demand in the United States, approximately

36%, is for residential and commercial consumers.These consumers rely on natural gas for space heating,water heating, cooking, and other purposes. The firstcomponent, space heating, comprises approximately70% of the residential and commercial load, or 25% oftotal U.S. annual consumption of natural gas.Consumption for space heating, however, is closely tiedto the winter heating season. Thus, approximately50% of natural gas consumption occurs during the fivewinter heating months, November through March.Figure 5-11 shows the strong seasonal cycle of naturalgas consumption in the United States.

Given the strong variation in seasonal demand, theindustry has found it economic to use storage fields tomanage the large differences between winter and sum-mer consumption. Storage is discussed in more detaillater in this chapter, but traditionally, and in large partstill today, gas is injected into the storage reservoirs insummer and withdrawn in winter. This allowspipelines and wellhead production to operate at a moreconsistent and more efficient annual level.

As part of the industry’s drive for economic effi-ciency, transmission lines connected to market areastorage fields (in California, the Midwest, and westernMid-Atlantic) have often been constructed for differ-ent capacity levels from the supply areas to the storagefields than from the storage fields to the markets. Thesegment from supply to storage is typically designedbased on average-day levels while that from storage tothe market is based on a peak-day requirement. Thisdesign recognizes that storage withdrawals must beincremental to flowing supply and could potentiallyinhibit long-haul transport from the supply regionsunless capacity downstream (on the market side) ofstorage was increased.

This dual capacity system on the upstream (produc-tion) and downstream (market) sides of storage hasworked well for many decades. The growing utilizationof natural gas fired turbines in the electric generationmarket is raising concerns about the effect on the sum-mer pipeline and storage capacity usage, however. Assummer cooling demands continue to rise over the next25 years, the increased call on pipeline transmissioncapacity during the summer by electric generation willreduce the industry’s ability to inject proper seasonalvolumes into storage. The resulting competition forcapacity on pipeline segments designed for average-dayuse should therefore raise overall utilization factors(actual flow/designed capacity) and associated trans-


mission revenues. However, there may be additionalexpense, e.g. compressor fuel, as the pipeline and stor-age infrastructure must be more dynamic and moretime-of-day responsive.

One of the issues of increasing importance in thedynamics of the pipeline transmission system evolvesfrom the intra-day market. The market demand dur-ing the course of the day can vary considerably due toresidential, industrial, and electric generation con-sumption. Of these, the latter is of growing impor-tance.

Demand from gas turbine electric generators is asignificant and growing portion of the swing in thepipeline intra-day demand. Gas turbine and CCGTplants are significantly more efficient than the naturalgas or oil fired steam generators they were designed toreplace, with CCGT units having a heat rate of about7,000 MMBtu per kilowatt hour versus 11,000 MMBtuper kilowatt hour for steam facilities. The lower heatrate therefore provides a more efficient electricity out-put. Based on this superior efficiency, CCGT plantswill generally be chosen to produce (or dispatch) elec-tricity before their steam-fired plant counterparts andwill also stay online longer.

Though regions differ, this use of gas-fired facilitiesfor electricity peaking can cause dramatic changes innatural gas consumption. A single 500-megawattCCGT plant can burn 90,000 MMBtu per day or 3,800MMBtu per hour. If a market area pipeline has a totaldaily delivery capacity of 1 to 2 BCF/D, then a singlegeneration plant turned on to meet afternoon demandcan raise consumption on a market area pipeline by 4-9%. The afternoon electric generation demands arethus not easily balanced due to the operating charac-teristics of a pipeline. The electric market has a profiledriven by its electricity consumers and requires aninstantaneous response while a pipeline operates beston a steady, ratable 24-hour flow. Pipeline operators,then, must deal with this growing mismatch betweenelectric load characteristics and gas pipeline facilitydesign using the infrastructure they have. A more flex-ible infrastructure would allow a more effective andmore efficient response to these needs; unfortunately,capital expense would be required to accommodatesuch, as well as necessary filings for tariff service mod-ifications.

It is noteworthy that the electric market and naturalgas transportation markets have differing cost struc-tures. The electric generation market is priced on base-


0

50

60

70

80

90B

ILLIO

N C

UB

IC F

EE

T P

ER

DA

Y1997 1998 1999 2000 2001 2002

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

MONTHSource: Energy Information Administration.

Figure 5-11. Monthly Consumption Data

load, intermediate, peaking, thirty-minute and five-minute intervals. Most pipeline tariffs, on the otherhand, are based on an expected, even, 24-hour offtake.Thus there is a price opportunity variance betweenwhat an electric generator is earning and what apipeline operator receives for the hourly swing serviceit is providing; this is often referred to as the “sparkspread.” Such a price opportunity difference may serveto exacerbate the swing as generators attempt to cap-ture as much of the “opportunity” as possible.Unfortunately, these types of actions may degradeservice to other customers, so pipelines may have tonotify generators to reduce their offtake.

One of the services provided by interstate pipelinesystems is the provision of pressure. In order to effi-ciently move gas, most pipelines in the interstate trans-mission grid were designed to operate at a maximumof 800 to 900 psi (pounds per square inch), as com-pared to normal atmospheric pressure of 14.7 psi.Newer pipelines have been designed to operate at pres-sures of 1,600 to 2,100 psi using thicker walled pipe towithstand these higher pressures. Customers fre-quently benefit from these high pressures. LDCs, forinstance, use 100 to 400 psi for their distribution sys-tem mainlines. They can thus avoid the expense ofcompression for the portions of their system directlyconnected to interstate transmission facilities.

Electric generators also receive significant benefitsfrom the provision of high-pressure gas. The new gasturbines, which comprise over 90% of electric genera-tion plants constructed over the last four years, requirepressure at 450 to 650 psi to operate efficiently. If theseplants are not connected to high-pressure interstate,intrastate, or LDC transmission facilities, they mayhave to install local compression to raise the pressureof their natural gas receipts to the required level at asubstantial incremental operating and capital cost.

Besides pressure, another common factor affectingpipeline transmission customers is gas quality, some-times called gas interchangeability. Natural gas fromdifferent supply sources can be composed of differentpercentages of gases that are produced in conjunctionwith methane. Gases without heating value, such ascarbon dioxide and nitrogen, are subject to strict limitsin receipt areas, and gas volumes exceeding these toler-ance levels can be restricted from pipeline access. Thevariance in gas quality, with Btu levels either higher orlower than the level for which the burner is set, cancause poor, inefficient combustion, which increases the

production of pollutants such as nitrogen oxide, car-bon dioxide, and carbon monoxide. There is legitimateconcern, therefore, about allowing gas with improperlimits to enter the pipeline system.

The main concern for pipelines is not strictly a vary-ing level of heating content but the potential for liquidfallout within the pipeline. Some of the higher levelhydrocarbon gases, pentanes and higher, will becomeliquids at lower pressure and temperature levels. Arapid pressure drawdown, perhaps due to a demandswing or a major pressure reduction at a valve, can coolthe gas and cause this liquid fallout to occur. The pres-ence of liquids can cause problems during compres-sion and can also lead to corrosion if left to settle in lowspots within the pipeline system. Pipeline corrosioncan lead to increased maintenance costs (related toattempts to remove the liquids), removal of capacityfrom service, and, in severe cases, loss of systemintegrity. For this reason, pipelines need to specifywithin their tariffs the standards for monitoring thequality of gas volumes.

LNG imports will be an increasing source of supplyin the study. Much of the LNG produced globally hasa high ethane level. The inclusion of ethane may resultin an imported gas stream with Btu content per cubicfoot above 1,100, the typical market area limit.

Without treatment, such as nitrogen injection, pro-cessing, or blending with low-Btu domestic produc-tion, the ethane-rich LNG could be barred from thedistribution and transmission systems in marketregions. However, recent work done under the aus-pices of the Gas Technology Institute indicates thatLNG with high ethane content does not appear tocause problems at the burner tip. This study is called“Gas Interchangeability Tests” and a draft of the firstpart of the study has been recently released. The initialresults suggest that the Btu limits in practice through-out the industry are too narrow and that alternateindices, such as the Wobbe Index, are much more pre-scriptive of safe combustion. It is hoped that addi-tional studies will help the industry determine not onlywhat is operationally appropriate but that they will alsolead to true interchangeability standards, to be incor-porated in the pipeline and LDC tariffs.

Maintenance Challenges

Besides operational challenges, transmission opera-tors will have to focus significant capital and attention


to maintenance of their systems over the next 25 years.In 2002, Congress passed the Pipeline SafetyImprovement Act, which has major ramifications forthe transmission industry. The Act will causeenhanced maintenance programs and actual continu-ing inspections of all pipelines located in populationcenters. According to the Act’s requirements, over 50%of the riskiest pipeline segments in these regions mustbe “physically” inspected in the next five years. Theremaining facilities must be inspected during the fol-lowing five years and all pipelines must be subse-quently re-inspected at less than seven-year intervals.Though currently unaddressed, recovery of these costswill be of substantial concern to pipeline operators andlevel of costs is of concern to ratepayers.

The inspection requirements of the Act will impactthe industry in several different ways. First, the Act willlead to a marked increase in expenditures for pipelinetesting. There are three major methods that can beused in integrity testing: In-line inspection using“smart pigs;” hydrostatic testing; and external inspec-tion. Each method will have its own set of cost factorsand these will vary per pipeline and region.

The cost of performing these tests is still being evalu-ated. The industry consensus, however, is that the testswill be costly. It is assumed, but not yet certain, thatFERC and other regulatory bodies will allow the cost ofthese tests to be included in pipeline tariffs. During peri-ods of testing, it is clear that besides the direct cost of per-forming the inspection, an additional cost, or revenueloss, may occur from the reduction in throughput capac-ity as a result of inspections. The insertion of a smart pigor the excavations of a pipeline for external surveillanceboth reduce pipeline capacity due to pressure reductionsduring the inspection period. According to the IntegrityRule report from the Interstate Natural Gas Associationof America (INGAA), a smart pig run in a pipelinedesigned for internal inspections will result in a 30%decline in throughput capacity for about three days. Ahydrostatic test requires removal of 100% of the capacityand the process takes an average of 25 days due to theneed to carefully purge the pipeline of natural gas, fill thepipeline with water, test the facility, and then properlydispose of the water.

One effect of the increased inspections, therefore,will be temporary reduction in capacity on the linesbeing tested. The reduced capacity will result in anincreased utilization factor for unaffected capacity andcould result in a short-term increase in effective trans-

portation rates. The result may thus be an increasedshort-term cost to consumers, even without the inclu-sion of expenses to physically perform the tests.

INGAA found that integrity inspections will add anadditional $6.8 billion to interstate transmission costsunder the assumption of a ten-year test cycle. By far thelargest of these costs will be due to short-term capacityreductions on the interstate grid, which is predicted tocost $5.7 billion. Capital expenditures on infrastructureimprovements are estimated as $0.6 billion whileinspection costs are forecast to be $0.4 billion.

Another result of the increased integrity activitycould be a pro-active decision to change historical reg-ulatory policy to allow operators to build capacitiesslightly higher than current contractual commitments.The increased capacity could then be used to maintainnormal throughput during periods when supplies arediverted from an alternate system due to maintenance.

Distribution

In the natural gas industry, the distribution system isdefined as that portion of the gas delivery infrastruc-ture that delivers gas from an interconnection pointwith the interstate pipeline system (the “citygate”) tothe ultimate, end-use customer.4 Exceptions to thisgeneral definition are common, including the increas-ing number of electric generation plants that receivegas directly from a pipeline. However, virtually all res-idential and commercial customers and most indus-trial customers receive their gas from a distributionsystem that is owned, operated and maintained by alocal distribution company (LDC). LDC does not referto the type of ownership (investor owned or munici-pality). Rather, LDCs in this study means the entitythat distributes gas to end-use customers.

As a general rule, LDCs broadly categorize theirservices into firm and interruptible deliveries.Distribution systems are designed to meet all firm cus-tomer demands for gas even under design (colder thannormal) weather conditions. The demands of cus-


4 This definition roughly follows the definition used todetermine those segments of pipe that are regulated bythe Federal Energy Regulatory Commission, i.e., trans-mission, and those regulated by others, i.e., distribution.Distribution regulation is typically provided by states ormunicipalities. This type of regulation covers pricing(rates) and terms of service. It should be noted that theU.S. Department of Transportation, which regulates theoperation and safety of pipes, uses a different definition.

tomers who are served with interruptible service mayor may not be met under certain conditions as definedin the LDC’s delivery tariffs, potentially during designweather conditions.

Because LDCs must design their distribution sys-tems to deliver gas even under design weather condi-tions, the overall capacity utilization is much lowerthan that of interstate pipelines. For example, a resi-dential customer who uses gas for heating, can have apeak wintertime monthly gas consumption that is 10or more times what the same customer’s monthly gasconsumption will be in the summer. The difference ingas usage is even more pronounced if peak day to min-imum use days are compared.

Thus customers with fuel oil backup, such as indus-trial consumers or electric generators, who can inter-rupt their gas usage by switching to alternate fuels,allow the LDC to use its system efficiently and reducecosts to customers. The greatest demands on a distri-bution system can arise when an electric generatingunit uses natural gas at the same time the residentialand commercial customers experience peak usage.Meeting these demands may require the LDC toexpand its facilities, exacerbating its seasonal variancein capacity utilization and potentially increasing thetotal overall cost to serve customers.

Distribution Infrastructure Investment

Construction of new facilities to meet customerdemands requires the extension of gas mains and theconstruction of services to bring the gas into an indi-vidual home or business. The costs of both mains andservices vary depending upon many factors. The coststo install a new service average $460 in undevelopedareas, $1,400 in developed areas, and almost $5,600 inurban areas. The costs used in the NPC analysis arebased on distribution system costs from a Gas ResearchInstitute study of LDC cost trends and are refinedbased on the American Gas Association’s “BestPractices” review. The allocation of indirect invest-ment costs was calibrated to reflect total national LDCinvestment. It should be noted that these costs reflectsmaller average size industrial and electric utility con-nections.

In addition to expansion activities, distribution sys-tems are in a state of constant maintenance andupgrade to maintain safety, ensure system reliabilityand to minimize future maintenance costs. Based onAGA benchmarking information, replacement of

mains ranged from 0.4 to 0.7% per year of existinginstalled mains among surveyed LDCs. For this study,main replacements were assumed at 0.5% per year andservice replacement at 0.75% per year.

These rates imply service lives beyond 25 years. Thismatches the current projections for the lives of materi-als used to build new distribution. As a result, in thisstudy, main and service replacements occur only fordistribution facilities installed before 2002. The facili-ties built in this study are not replaced during thestudy.

Despite the use of cost-saving techniques, main andservice replacements often are significantly more costlythan the construction of new facilities. Frequently,replacements occur in congested public rights-of-waywhere numerous other underground facilities arelocated. Also, replacements often occur in developedurban or suburban areas where pavement restorationand landscaping or lawn restoration is required. Thus,based on AGA benchmarking studies, main replace-ment costs were assumed to cost 50% more than con-struction of new mains. Similarly, replacement of serv-ices was assumed to cost 25% more than the cost ofnew construction. Finally, meter replacement wasassumed to cost 15% more than new construction.

The total annual facility investment requirementsfor distribution companies are similar in the ReactivePath and Balanced Future scenarios. To accommodatethe demand projected in the Balanced Future scenario,the results from the distribution analysis show thattotal annual facility investment requirements for distri-bution companies will average $5.3 billion per year(2002 dollars), with a cumulative investment from2004 through 2025 of $135 billion. This compares toannual expenditures during the 1990s, which averagedslightly more than $4.8 billion. However, funding forthis level of expansion may be more difficult than inthe 1990s because more of an LDC’s cash flow will beneeded in the future for other purposes, including buy-ing higher priced gas and placing it in storage. Thismay result in a greater need to finance expansion of thedistribution systems with external funds than was thecase in the 1990s. LDC access to capital markets will,therefore, be important but, given appropriate regula-tory policy, should not be a constraint.

In determining the costs to expand the distributionsystem, a 1% per year increase in productivity wasassumed. This significantly lowers the projected costs.


Given appropriate funding for research and develop-ment (R&D), achieving increased productivity seemsreasonable. Thus, it is not expected that adequacy ofthe distribution infrastructure will be a constraint inthe future.

The improvement in overall efficiency in the resi-dential sector in the Balanced Future reduces systemthroughput slightly, resulting in a modest decline inrequired mains reinforcement and delayed replace-ments. The decline in power generation demand alsoreduces the required investment to serve new load.This decline in investment is, however, offset by a smallincrease in investment to serve growth in the commer-cial and industrial sector load, as the lower natural gasprices in the Balanced Future scenario result in someadditional growth in commercial and industrialdemand.

As discussed in the Transmission section of thischapter, the United States Congress passed legislationintended to enhance the safety of “transmission” typegas pipelines5 through stricter inspection require-ments. The U.S. Department of Transportation (DOT)is currently developing the rules to implement the leg-islation. Companies are required to perform a baselineinspection within the first ten years of all “transmis-sion” like pipeline located within a high consequence

area. Re-inspection will be required every seven yearsafter the initial inspection.

The AGA estimates that the LDCs operate almost22,000 miles of pipeline that is subject to this newpipeline integrity program. While the exact require-ments mandated by DOT is not known, AGA has esti-mated the cost of compliance for LDCs at $2.7 billionto $4.7 billion over the next 20 years. DOT has inde-pendently estimated industry (including LDC andInterstate Pipelines) costs at $4.7 billion over the next20 years. DOT’s analysis for just LDC pipelines wouldbe somewhere between 35% and 50% of this compos-ite figure, or $1.6 billion to $2.4 billion, somewhat lessthan the AGA figures. It is important to note that thecost of remediation, or addressing anomalies found, isnot included in the above cost calculations for eitherindustry or Office of Pipeline Safety estimates.Historical annual capital expenditures in 1998 dollarscan be seen in Figure 5-12.

For purposes of this study, a cost of $16,000/mile or$3.5 billion for the ten-year period was assumed. Thus,


0

2

4

6

BIL

LIO

NS

OF

199

8 D

OLLA

RS

1975 1980 1985 1990 1995 2000

YEAR

Figure 5-12. Historical Investment in Distribution Infrastructure

5 The DOT definition of “transmission” differs from thedefinition used by the rate setting regulators like FERC.As a result, LDCs operate a significant amount of “trans-mission” pipelines from a DOT perspective.

it was assumed that 60% of the total cost will beincurred in the initial ten-year period, when baselineinspections occur.

For the remainder of the study period, costs to com-ply with the pipeline integrity program will continue.However, since facilities needed to complete theinspections will have already been built, integrity man-agement plans will have been written, and high conse-quence areas will have been identified and mapped, itis anticipated that costs to comply with the pipelineintegrity program will decline. Offsetting this declinewill be the increased amount of pipe included in thepipeline integrity program as LDCs expand their sys-tems as well as inflationary pressures. Thus, annualcosts for LDCs to comply with pipeline integrity stan-dards in years after 2012 were assumed to be 40% ofthe annual costs of the initial period. In summary,from 2004 through 2013, an annual cost of $250 mil-lion was assumed to meet the pipeline integrity stan-dards. From 2014 through 2025, an annual cost of$100 million was assumed.

In addition to pipeline integrity costs, LDCs faceincreased costs to protect against security threats byterrorists. These costs cannot be readily quantified. Asa part of outreach, a limited number of LDCs indicatedthat LDCs expect some increase in costs compared tohistorical trends, but overall increases that are less thanthe costs to comply with new pipeline integrity stan-dards. If these costs represent a 1% increase in thecosts to maintain and expand the gas distribution sys-tems, LDCs would incur new expense of $48 millionper year. These costs are included here only for refer-ence and were not included in the figures shown else-where in this chapter.

Challenges to Building and Maintaining theRequired Distribution Infrastructure

As the LDC marketplace has evolved, the require-ments for serving customers have continued but roleshave changed. All states that have residential and com-mercial choice programs have addressed the providerof last resort (POLR) or supplier of last resort (SOLR)issue to some extent. The POLR/SOLR responsibilityhas been defined in varying ways, but generally is theresponsibility to assure that small gas consumers willnot experience an interruption in the supply of naturalgas to meet their needs. Thus, POLR/SOLR caninclude the responsibility to provide essential needscustomers with gas if the customer’s supplier goesbankrupt or fails to deliver gas for other reasons.

POLR/SOLR responsibility always includes small vol-ume residential gas customers (residential or commer-cial) and seldom, if ever, includes very large customerslike electric generators.

There is debate about what entity should be aPOLR/SOLR. Some states have required that the LDCassume this role, while other states have prohibited theLDC from holding the role. While these policy debateswill continue, it is important to recognize that thedemand for natural gas to serve residential and com-mercial markets will likely continue to grow. In fact,this study projects that the number of residential cus-tomers served by the natural gas industry will growfrom 61 million in 2003 to 81 million in 2025. Thislevel of growth will necessitate that state and federalpolicy makers work with the various industry partici-pants to assure that interstate pipeline and storagecapacity is available to serve future customers. Cleardefinition of the responsibilities of the POLR/SOLRand appropriate commitments from policy makers toallow critical expansions are required to assure reliableservice to customers.

The permitting and construction of new or replace-ment facilities is becoming more expensive as a conse-quence of various growth management, building code,and environmental requirements. Many of these issueshave been discussed at some length in theTransmission section of this chapter. It is worth notinghere, however, that access to public rights-of-waywithin metropolitan regions is becoming more diffi-cult to obtain and more expensive. Increased costsfrom such items are not included in this study.However, governmental bodies need to consider theimpacts (financial as well as safety and reliability) ofadded restrictions on the installation and maintenanceof distribution facilities.

To address these and other concerns, states shouldalso develop a mechanism to coordinate siting issuesamong affected state and local governmental entities,wherever multiple governmental entities have animpact on the siting of LDC facilities. Using theIOGCC/NARUC Pipeline Siting Work Group Report6

as a framework, each state should consider, as needed,


6 Philip N. Asprodites, “Roadmap to Implementation ofthe Final Report of the Interstate Oil and Gas CompactCommission/National Association of Regulatory UtilityCommissioners Pipeline Work Group in Louisiana”(April 2003).

programs that might include the following types of ini-tiatives:

� The governor establishing within the office of thegovernor a coordinating effort to organize and expe-dite the activities of all state and local natural gaspermitting entities.

� States naming a lead agency that would have theauthority to monitor processing schedules withinexisting regulatory requirements.

� The state economic development office (CommerceDepartment) being involved with the coordinationeffort and recommending actions to streamline theprocess.

Coordination and certainty in completing a permit-ting process are keys to meeting the growing need fornatural gas while balancing many other key issues.Consistent government policy and rapid, predictableregulatory decisions are needed to enable timely andcost-effective system expansions.

The business environment in which LDCs operatehas changed dramatically since the 1999 NPC study.Traditionally LDCs provided gas to all customersserved by the distribution system. Based on programsthat are currently operational or announced, however,96% of all industrial customers will have customerchoice. Additionally, at least 72% of all commercialcustomers and 57% of all residential customers willalso have the option to choose their gas supplier.7 Thiscontinuing transition has changed the decisionprocesses related to their distribution system expan-sions.

Another topic of concern to LDCs arises because thereduced gas usage resulting from customer-achievedefficiency gains will lead to less gas flowing in an LDC’ssystem and its current asset base to serve existing cus-tomers. Most LDCs have experienced this phenome-non throughout the 1990s. This normally means thatexpansion capital will be required to attach new cus-tomers just to maintain system throughput and theassociated revenue levels. Actual growth in throughputand revenues will require additional capital invest-ment, beyond the level described, just to keep evenwith customers’ conservation efforts. Previous expan-

sions have largely been financed through internal cashgenerated from the business; however, forecasts suggestthat capital markets will need to provide more of thecapital required to maintain and grow the throughputand associated revenues. Accessing capital at the low-est cost in the competitive markets requires a com-pelling story. To achieve favorable access to capital, tra-ditional rate designs may need to be modified oraugmented to reflect the adverse impact to the finan-cial health of LDC’s caused by customers achieving thedesirable goal of greater efficiency. One such exampleis the state of Oregon recently implementing a “conser-vation” tariff that encourages greater conservation bycustomers while mitigating the potentially adverseimpacts of reductions in LDC revenue.8

LDC working capital needs will expand significantlyat the gas prices suggested by the NPC analysis.Currently, the United States is considered to have ade-quate gas in storage if more than 2.5 TCF has beenstored by the beginning of the heating season. The car-rying cost to store this gas for several months at$6.00/MMBtu is significantly more than the compara-ble costs in the $2 to $3 gas price environment oftenseen in the 1990s.

These types of changes, as well as changes in thebroader energy market, are impacting the businessrisks faced by LDCs. Constant attention to the finan-cial health of the distribution industry will allow ade-quate access to capital markets for all future expansionsneeded to serve customers.

Reliable Gas Service in a Changing Market

As the natural gas marketplace changes, newdemands are placed on the interstate pipeline, storage,and distribution system infrastructure. In particular,electric generating customers can dramatically changethe demands for gas as they follow electric load. Thechanges in gas requirements can occur very quickly.Because electricity cannot be stored, it must be pro-duced at the moment it is used. The customer demandfor electricity reaches a peak in the summer in manyareas of the country as it provides essential coolingservices. Thus, power plants are increasingly consum-ing larger amounts of gas at the same time distribution


7 American Gas Association, “Policy Analysis Issues,” IssueBrief 2002-02, May 7, 2002.

8 Oregon PUC Order No. 02-634, Stipulation AdoptingNorthwest Natural Gas Company Application for PublicPurposes Funding and Distribution MarginNormalization (PUC of Oregon, September 12, 2003).

companies and others are attempting to fill seasonalstorage.

These concerns have been the subject of consider-able discussion and debate among industry partici-pants. Recently, INGAA,9 AGA,10 and the APGA11

developed a framework to discuss these issues.12 Thegroups’ stated goal is to “ensure the continuation of thehistoric reliability of the natural gas industry as gasdemand grows particularly from the power generationsector.” These evolving discussions among industryand government will need to continue to assure ade-quate, reliable, cost-effective natural gas service to allcustomers in the evolving natural gas marketplace.

Productivity Improvements Require R&D Investments

A measure of productivity in LDC operations is gasdelivered per LDC employee. With an average drop instaffing levels of 4% per year since 1990, Figure 5-13demonstrates the increased amount of gas deliveredper distribution company employee, primarily as aresult of implementation of new technologies. Muchof this technology came from research and its corre-lated product and skill-set developments. However,expenditures for gas research have declined in the lastfive years, driven in large part by the reduction in fundscollected through the FERC-mandated gas distributionsurcharge. The collections of these funds will be com-pletely eliminated by the end of 2004.

Research and development have provided and mustcontinue to provide the new techniques and technolo-gies to reduce costs and increase both the safety andreliability of distribution systems. Many LDCs believegovernment funding of research remains a criticalneed. In addition to state- and federal-sponsoredR&D, many LDCs participate in and fund R&D.However, some distribution companies may operateunder regulatory frameworks that discourage R&D. Insuch situations, LDC shareholders, finding themselves

at risk to benefit, may be reluctant to support invest-ment in the research and development needed to con-tinue these productivity enhancements in the future.Given the inability of LDC shareholders to benefitfrom R&D investments in operations, the interventionof government may be required. State regulatory com-missions should consider removing any barriers toLDC’s participation in collaborative research.Similarly, DOE funding of gas utilization technologyresearch must continue and, if possible, expand.

Storage

The ability to effectively store and retrieve largequantities of natural gas has been a key factor in thegrowth and development of the natural gas industry.At its most basic level, the storage function allows forthe generally asynchronous supply and demand func-tions to be efficiently matched. Perhaps the most obvi-ous example of this functionality involves satisfying thehighly seasonal demand for natural gas for space-heat-ing purposes in the residential and commercial sectorsduring the wintertime. Indeed, without the ability tobuild gas inventories in storage prior to the high-demand winter period, it is unlikely that natural gaswould have become such a dominant space-heatingfuel in these sectors. Without storage, the wintertimesurge in demand would require that production beaccelerated greatly for the winter season, then throttledback as temperature-driven demand waned. Huge


9 INGAA (Interstate Natural Gas Association of America)represents interstate pipelines.

10 AGA (American Gas Association) represents investor-owned local distribution companies.

11 APGA (American Public Gas Association) representspublicly owned natural gas local distribution companies.

12 March 7, 2003, Letter to The Honorable Patrick H. Wood,III, Chairman of FERC from David N. Parker, Presidentand CEO of the AGA.

0

80

100

120

140

160

1990 1995 2000

YEAR

MIL

LIO

N C

UB

IC F

EE

T P

ER

WO

RK

ER

Figure 5-13. Gas Delivered per Worker

amounts of pipeline capacity would have to be avail-able to transport the gas to market areas, much ofwhich would then be vastly underutilized at othertimes of the year. Thus, a major function of storage isto augment supply to satisfy seasonal demandincreases.

A second major function of storage is the opera-tional function of load balancing, usually associatedwith pipeline operations. In essence, the function ofload balancing is operating the system in such a waythat receipts of gas into the system roughly equal deliv-eries of gas from the system, within certain operatingtolerances. Thus, interconnections to storage give thepipeline operator a place to inject excess gas whenmore is being received by the pipeline than delivered,as well as an incremental source for withdrawal of gaswhen more is being delivered to customers than isreceived by the pipeline.

A third major function for storage, which has grad-ually grown in prominence, is the rapid cycling orturnover of working gas storage inventory. This ismost often associated with salt cavern storage facilitiesbecause of the ability to inject gas into and withdrawgas from these facilities at very fast rates relative totheir storage capacities. This function also supports a

wide variety of market-based uses, where the purposeof its use is primarily to obtain a profit as opposed tooperational uses. Essentially, this function enables par-ticipants to profit from changes in gas prices over shorttime intervals, taking advantage of periods of highvolatility in gas markets.

Overview

Natural gas may be stored in a number of differentways. It is most commonly held in inventory under-ground under pressure in three types of facilities.These are depleted oil and/or gas reservoirs, aquifers,and caverns developed in salt formations (see Figure 5-14).

Each type has its own physical characteristics(porosity, permeability, retention capability) and eco-nomics (site preparation costs, deliverability rates,cycling capability), which govern its suitability to par-ticular applications. Two of the most important char-acteristics of an underground storage reservoir are itscapability to hold natural gas for future use and therate at which gas inventory can be injected and with-drawn – its deliverability rate. The distribution of stor-age facilities varies regionally by type within the U.S.lower-48, as can be seen in Table 5-1. It is important tonote that while these data indicate total working gas


SALT

CAVERNS

MINES

Source: PB - KBB, Inc.

HARD-ROCK CAVERNS

AQUIFERS

DEPLETED

RESERVOIRS

Figure 5-14. Types of Underground Natural Gas Storage Facilities

capacity of over 4.5 TCF, the largest amount of inven-tory actually cycled in any year has been 2.9 TCF, andevidence suggests that storage capacity may be inca-pable, for a variety of reasons, of cycling more thanthat volume without extreme seasonal price variability.

In addition to the three primary storage types,industry participants also have a number of otheroptions to satisfy the temporary spikes in demand gen-erated by end-users – such as a surge in demand forspace heating during an unusually cold period, or asudden requirement for an electric utility to bringonline a natural gas-fired generator – that can exceedthe ability of traditional storage to handle. These stor-age options usually involve storing LNG, compressednatural gas, or liquefied petroleum gas (usuallypropane) in above-ground storage tanks, and have thecapability to deliver natural gas or a propane-air mixinto the local distribution system when required.These facilities are generally capable of relatively highdeliverability but for short durations.

Following are brief descriptions of the characteris-tics of each of the major storage types. Depletedoil/gas reservoir storage facilities are the most widelydistributed geographically. Aquifer facilities arefound primarily in the Midwest, while most salt cav-ern storage has been developed in the salt formationsalong or near the Gulf of Mexico in Texas, Louisiana,and Mississippi.

Depleted Oil and Gas Reservoirs

Most existing gas storage in the United States is heldin depleted natural gas or oil fields. Cycling in this typeof facility (number of times a year the total working gasvolume may be injected/withdrawn per year) is rela-tively low, and daily deliverability rates are dependenton the degree of rock porosity and permeability. Thesefacilities are usually designed for relatively few injec-tion and withdrawal cycles per year, and often for onlyone cycle.


DepletedGas/Oil Fields

AquiferStorage

Salt CavernStorage Total

Region/State

Numberof

Sites

WorkingGas

Capacity(BCF)

Numberof

Sites

WorkingGas

Capacity(BCF)

Numberof

Sites

WorkingGas

Capacity(BCF)

Numberof

Sites

WorkingGas

Capacity(BCF)

Percentof

WorkingGas

Capacity

Consuming East 242 1,722 34 354 4 5 280 2,081 46

Consuming West 31 606 6 38 0 0 37 644 14

Producing 74 1,087 * * 24 138 98 1,226 27

Total U.S. Lower-48 346 3,414 41 393 28 143 415 3,951 87

Canada 11 598 0 0 1 4 12 602 13

Total North America 357 4,012 41 393 29 147 427 4,553 100

*Any aquifer facilities in this region have been counted as depleted gas/oil fields to preserve data confidentiality.

Notes: Regions are those used by the EIA in its Weekly Underground Storage Survey. BCF = billion cubic feet.

Table 5-1. Regional Distribution of Storage Facilities and Working Gas Capacity

Aquifers

Aquifers, which originally contained water, may besuitable for gas storage purposes if certain geologiccriteria are met. In the United States, aquifers that areused for gas storage are found primarily in theMidwest. There are several reasons why an aquifer isthe least desirable type of underground storage, manyof which contribute to making aquifer storage moreexpensive to develop and maintain than depletedreservoir storage. First, it takes about twice as long todevelop an aquifer storage site compared with anaverage depleted gas or oil field. Second, all new facil-ities must be installed, including wells, pipelines,dehydration facilities, and compressor operations.Third, no native gas is present in an aquifer forma-tion. Thus, base or cushion gas must be acquired andinjected into the reservoir to build and maintaindeliverability pressure.

Once in operation, aquifer reservoirs have onepotential advantage over depleted field storage.Because of the additional support of an aquifer’s water(pressure) drive, in most instances, higher sustaineddeliverability rates than gas or oil reservoirs can bedesigned and attained. Injection and withdrawal activ-ities generally are required to conform to a disciplinedschedule, however, to avoid damage to the reservoirs orloss of gas. Therefore, aquifers only cycle once per year.

Salt Caverns

There are two basic types of geologic formations inwhich cavern structures used to store natural gas aredeveloped: salt domes and bedded salts. Both are cre-ated by injecting water (leaching) into a salt formationand shaping a cavern. Caverns created in salt domesare large caverns as they are constructed within salt for-mations that can extend for thousands of feet.

A bedded salt storage cavern, on the other hand, isgenerally developed from a much thinner salt forma-tion (hundreds of feet or less). As a result, theheight-to-width ratio of the leached cavern in a bed-ded deposit is much less than for a cavern in a saltdome. Typically, bedded salt storage developmentand operation is more expensive than that of saltdome storage.

Because salt cavern storage facilities are essentiallyhigh-pressure storage vessels akin to undergroundtanks, their injection and withdrawal rates are veryhigh and base gas requirements low. Their resulting

ability to cycle working gas inventory numerous timesduring a year makes them ideal for meeting largedemand swings.

LNG Storage

Liquefied natural gas is natural gas that has beencooled to approximately minus 260 degrees Fahrenheitfor storage as a liquid. LNG storage accounts for a verysmall portion of the overall natural gas storage capa-bility in the United States because LNG working gasstorage capacity is just over 2% of the overall capacity.However, LNG storage facilities have relatively highdeliverability rates that allow operators to deliver anamount equal to up to 14% of all underground storage.LNG storage can be grouped in two general categories:peak-shaving storage and marine terminals. Each ofthese categories has specific characteristics and utiliza-tion benefits.

Peak-shaving LNG storage has two main positiveattributes: its high deliverability capability as com-pared to more traditional storage, and its flexibilitywith respect to where the storage can be located.However, peak-shaving LNG storage is more costly onthe basis of dollars per million cubic feet of storagecapacity, when compared to traditional storage.

Marine import terminals receive LNG shipmentsand have on-site storage. The LNG is stored in above-ground storage tanks until it can be regasified andinjected into the pipeline grid. Additionally, the LNGcan be stored until it is trucked, in liquid state,directly to customers. The principal operation of animport terminal is not for gas storage, but rather forreceiving the waterborne LNG imports and thenregasifying LNG for shipment via pipelines to cus-tomers.13 Marine terminal storage can also functionas peak-shaving storage; however, that is not its prin-cipal function.

Propane-Air Storage

Propane-air storage is another method by which gasutilities and industrial customers meet demand duringthe coldest days of the year. Propane is stored in above-ground tanks or underground caverns (usually gran-ite) until needed. Because it vaporizes relatively easily,propane can enter the gas pipeline distribution systemswith little difficulty. Generally, a propane-air mixture


13 Energy Information Administration, US LNG Marketsand Uses, January 2003.

containing 1,400 Btu per cubic foot has burning char-acteristics similar to natural gas. Although propane-airsystems are common as a cheap alternative to pipelinecapacity, there have been concerns over several failuresfor the propane to properly vaporize on especially colddays in the Midwest.

Historical Background and Statistics

Most of the nation’s storage sites were developedbetween 1955 and the early 1980s. During this period,U.S. storage capacity increased over fourfold, fromabout 2.1 TCF in 1955 to 8 TCF in 1985.14 The needfor underground storage grew as consumption of nat-ural gas increased significantly. The mix and require-ments of consumers also changed as demand shiftedtoward the more weather-sensitive residential andcommercial markets. Furthermore, in the mid- andlate-1970s, the interstate market encountered supplyand demand imbalance situations during severalexceptionally cold winters, and as a result service cur-tailments were imposed. Since the mid-1980s, totalstorage capacity has remained at approximately 8 TCF,even with the recent surge in new storage development.However, the daily deliverability from storage hasincreased.

The volatile gas market during the late 1980s set inmotion certain events that heightened interest in newstorage facility development. Interest in new storageresurged as regulatory changes under FERC Orders 436and 636 forced more competition into the market-place. Storage became increasingly important as allpipeline services were unbundled and customers hadto make their own storage arrangements. Between1992 and 2002, deliverability from storage increased by29%, from approximately 65 BCF/D to 83 BCF/D.

Results from the Study

The reference case analysis projects an increaseddemand for North American seasonal storage capacityof close to 1 TCF over the 22-year study period, relativeto the demands on storage in the 1999-2002 period,which averaged 2.3 TCF per year. Recognizing that thebase period (1999-2002) was characterized by rela-tively light demand on storage due to generally warmwinters, it is estimated that the current storage infra-

structure is sufficient to satisfy an increased averageannual demand of approximately 300 BCF, leaving 700BCF of demand that will need to be met by develop-ment of new capacity. As much as 150 BCF of this newcapacity could be required in the very near term ifthere were a return to winter weather patterns closer tohistorical normal levels. As only 109 BCF of storageadditions are projected to occur by 2005 (based onprojects announced), most of any such near-termdemand increase will need to be met through moreefficient use of existing capacity, and measures toincrease capacity and deliverability at existing facilities.

By 2015, total cumulative storage capacity additionswill need to have approached 400 BCF, and by 2025,700 BCF, to accommodate growth in the total gas mar-ket. While many of the best resources for gas storage(based on location and geology) have been developed,this rate of growth in the infrastructure is consideredachievable provided that favorable market conditionsexist to finance the additions. Conventional storage isexpected to account for over 80% of the projectedadditions and high deliverability peak-shaving theremainder.

When discussing the adequacy of storage infrastruc-ture, it must be kept in mind that demands on storagevary greatly, from year to year, depending on weather,and that even if the projected growth in capacity isachieved there will likely be winters when the system isunable to fully supply gas withdrawal requirementswithout some significant short-term reduction in gasdemand, whether price induced or otherwise. A win-ter of significantly colder than normal winter weathercan increase demand for storage capacity by as much as25% relative to a normal year. It has been many yearssince North American storage capacity has been testedby such a winter and it is very likely that current stor-age capacity would be severely challenged to meet suchdemand, with potential for even greater price spikesand demand destruction than what was experienced in2001 and 2003.

Storage additions for the U.S. lower-48 were evalu-ated on the basis of nodes within the nine censusregions, while additions for Canada were split betweennodes in eastern and western Canada.

In the U.S. lower-48, the need for near-term storageadditions is greatest in the Pacific, East South Central,South Atlantic, West South Central, and Mid-Atlanticregions. Near-term storage additions for the Mountain


14 Refers to total storage capacity. For depleted field andaquifer storage facilities, base gas typically occupies one-half or more of total storage capacity, with the remainingcapacity available to store working gas.

region are projected to grow modestly. No near-termstorage additions are projected for the West NorthCentral or New England regions.

Projected additions to peak-shaving and conven-tional North American storage over the 2005-2025period are 550 BCF. Nearly 80 BCF of the projectedadditions are for high deliverability peak-shaving stor-age facilities, with lower-48 additions accounting for90% of this requirement. The need for this type ofstorage will be greatest in the South and Mid-Atlanticregions, collectively accounting for over 30% of theprojected growth in peaking storage. Peak-shavinggrowth in the West South Central and Pacific regionsare projected to grow at 9 BCF each. Eastern Canadawill experience the need for peak-shaving additions aswell; additions in this region are projected to be over 8 BCF.

Projected additions to conventional storage during2005-2025 are largely concentrated in the lower-48market area. Three regions in particular, East NorthCentral, Mid-Atlantic, and South Atlantic, are pro-jected to experience significant storage growthamounting to about two-thirds of the projected overallstorage additions. Combined storage growth in thesethree regions is projected to be about 320 BCF, with thegreatest additions to the East North Central at approx-imately 111 BCF (10% increase over current), followedby nearly 109 BCF (44% increase over current) and99 BCF (23% increase over current) to the Mid-Atlantic and South Atlantic, respectively.

This increase in storage will require additionalpipeline capacity to reach the market centers, particu-larly for storage developed to serve the Mid-Atlanticand Northeast markets, which lack suitable reservoirsfor storage development within the region. Instead,the new storage capacity will have to be developed inthe western portions of Pennsylvania and New Yorkand eastern Ohio. This will result in the constructionof incremental pipeline capacity of approximately 2 BCF/D from these storage sites to the coastal marketcenters, which include New York City, Boston, andPhiladelphia.

The Mountain region is projected to require nearly55 BCF of additional storage (17% growth), and theWest North Central approximately 37 BCF (22%growth) of new storage capacity. Eastern Canada isprojected to see growth of about 40 BCF, or about a20% increase relative to current storage capacity.

Annual average North American daily loadsadjusted for storage are projected to grow 19 BCF/D,from 71 BCF/D to 90 BCF/D, from 2005 to 2025. Thisgrowth will impact storage injection and withdrawalpatterns in certain regions more than others, though ingeneral, seasonal withdrawals will increase in responseto growth in the residential and commercial sectors,and to some extent growth in power generation. Incontrast, growth in the Industrial sector during thissame period is projected to be virtually flat with likelyno impact on storage usage patterns. Injection pat-terns will be impacted more due to growth in powergeneration than anything else.

Daily loads during the 10 highest demand days ofthe year are projected to increase from approximately101 BCF/D to over 126 BCF/D during the studyperiod, while loads during the 60 highest demanddays are projected to grow from 92 BCF/D to 116 BCF/D. Storage plays a critical role in satisfyingincremental load during peak use periods. The high-est load periods occur during the heating season andstorage withdrawals typically satisfy over 50% of thedaily North American load during the highestdemand days of the heating season. Two regions inparticular stand out in this regard. In the East NorthCentral region, the reference case projects demandduring the 60 highest demand days of the year will beover 2.4 times the average daily load. A similar pro-jection is evident for the West North Central wheredemand during the 60 highest demand days of theyear is nearly twice that of the average daily demand.Under such circumstances, storage is ideally suited tosatisfy these incremental seasonal loads, which arepredominantly driven by space heating requirementsin the residential and commercial sector.

Natural gas demand has always been seasonal, but arecent phenomenon is that, due to increased gas-firedgeneration implemented around the continent, a newsummer season peak is also developing. Other thanthe industrial load, which has traditionally been steadyon a daily and seasonal basis, the other major demandsectors (residential, commercial, and electric genera-tion) are weather sensitive and have a high degree ofvariability. Demand in North America is projected togrow by 19% between 2003 and 2015, whereas indus-trial demand is projected to grow by only 3%. Thiswould mean that the stable industrial demand sector isbecoming a smaller percentage of total demand. Thiseffect is more pronounced in the United States, where


industrial demand is projected to decline by 6% from2005 to 2015.

Demand for power generation, which will make upthe majority of projected demand growth, is highlyvariable on an hourly, daily, and monthly basis. As canbe seen in Figure 5-15 (historical 1997) and Figure 5-16 (projected 2025), power generation not onlyincreases the number and magnitude of winterdemand peaks, but it also creates a secondary demandpeak in the summer. It also creates an hourly demandprofile that is even more pronounced than that of atraditional residential/commercial load profile. Thegrowing summer peak shortens the summer seasongas storage injection period, primarily allowing forinjections only in the off-peak electric demand hoursof the day and thus requiring more volume to beinjected into the shoulder (historically lower demand)months of April through June and September throughOctober.

Projected near-term (2003-2005) demand for sea-sonal storage could grow by as much as 450 BCF, rela-tive to the requirements of 1999-2002, with most ofthat increased demand being due to an assumption ofa return to more normal weather. As much as 300BCF of that demand growth may be accommodatedby the existing infrastructure. Based on announcedprojects, it is expected that storage capacity will growby only 109 BCF by 2005. Additions to working stor-age capacity in the U.S. lower-48 amount to 69 BCF,and consist of projects previously announced to themarket.

Any remaining incremental near-term demand forstorage will need to be met by more efficient utilizationof existing capacity and short-term enhancements tothe capacity and deliverability of existing facilities.There is a significant risk that any near-term return tomore normal weather patterns could not be met by theexisting infrastructure without some increase in sea-sonal gas price variability and volatility. The Pacificregion will experience the largest near-term growth atover 29 BCF. The announced capacity expansion proj-ects have an estimated cost of over $1 billion. Over 34 BCF of the near-term storage additions will be highdeliverability salt cavern facilities located in the Mid-Atlantic, South Atlantic, West South Central, andMountain regions, with a total estimated developmentcost of $211 million.

A mix of new salt cavern storage capacity anddepleted reservoir storage projects in the Mountainand East North Central regions make up the remaining4 BCF in the U.S. lower-48. The total cost associatedwith these additions is $38 million. Near-term addi-tions to Canadian storage amount to 40 BCF, all ofwhich are located in western Canada and involve newdevelopment in depleted reservoirs. The estimatedcost associated with these additions is $100 million.

Projected North American storage infrastructureadditions over the 2005-2025 period are approximately550 BCF, 80 BCF of which will consist of high deliver-ability salt cavern facilities. In total, future NorthAmerican storage infrastructure additions over thestudy period carry an estimated cost of nearly $5 bil-lion.

On a regional basis, the development of 111 BCF ofadditional depleted reservoir and aquifer storagecapacity is projected in the East North Central at anestimated cost of $905 million. Storage additions tothe Mid-Atlantic region are forecast at 141 BCF, all ofwhich will likely entail the conversion of depletedreservoirs, at an estimated cost of $1.3 billion.

Growth of storage in the South Atlantic is projectedat about 99 BCF, with an estimated development costof $804 million. The Mountain region is projected toneed almost 55 BCF of additional conventional stor-age capacity with attendant development costs of$468 million.

The remaining additions to lower-48 storage capac-ity are projected at almost 87 BCF, at a total estimatedcost of $1.07 billion. Projected additions to Canadianstorage capacity are 56 BCF, including over 8 BCF ofhigh deliverability storage. All but about 2 BCF ofthese additions are projected for eastern Canada. Thetotal estimated cost of storage additions in Canada is$260 million. Table 5-2 shows the capacity additionsand estimated costs in more detail.

It should be noted that these regional capacity addi-tion estimates are based on model results that may notadequately reflect geological and other factors thatfavor construction of capacity in some regions relativeto others. In particular, it is likely that more of thisrequired capacity will be built in the major producingregions of the United States and Canada, and less in themarket regions than is indicated in the discussionabove.



120

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

100

80

60

40

20

MONTH

POWERINDUSTRIAL

COMMERCIALRESIDENTIAL

140

0



Figure 5-16. 2025 Daily Loads for the United States and Canada in Balanced Future Scenario


MONTH

0

120

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

100

80

60

40

20

140

POWERINDUSTRIAL

COMMERCIALRESIDENTIAL


Figure 5-15. 1997 Daily Loads for the United States and Canada

The Outlook for Storage

In recent years there have been several occasionswhen winter/summer seasonal price spreads havedeclined to such low levels that seasonal storage of gasfor price arbitrage purposes has been uneconomic.However, the operational need to store gas to balancewinter and summer demand with relatively constantproduction of gas supplies has remained. In addition,the gas market’s fluctuation on a day-to-day, week-to-week, and month-to-month basis demands quick turn-around of storage inventories. With credit concerns,i.e., cash flow, becoming an issue for many gas tradersand marketers, storing gas without access to it for 6 to8 months can be a risky proposition.

Electric generation demand in summer will competewith storage injection requirements during the sum-mer cooling season. This issue has been in place for anumber of years. Most multiple cycle storage opera-tors are familiar with the double dipping of inventoriesdue to cooling load in the summer and heating load inthe winter. This is the effect of summer season gener-ation demand on the pipeline infrastructure, and this

effect continues to strengthen. If the pipeline infra-structure is strained due to peak summer loads, sched-uled storage refills can be interrupted. As this phe-nomenon increases, there will be more and more ofthese refill interruptions.

Challenges to Building and Maintaining theRequired Storage Infrastructure

The difficulty of siting storage facilities can beattributed to the need to find a site with the appropri-ate combination of geological features, pipeline prox-imity, and the ability to obtain land, rights, and per-mitting. Once a geologically suitable site is found at anacceptable location with respect to the natural gaspipeline infrastructure, the ability to obtain permit-ting, land, and development rights becomes critical.The primary access limitations on developing storagecapacity are the difficulties in dealing with multiplegovernmental entities, limitations on emissions, andlimitations on storage reservoir operating pressures.

The inconsistency in requirements for FERC andstate facility certifications increases the time and cost


Table 5-2. Projected Storage Capacity Additions and Costs by Region

RegionNumber Region Name

Announced2003-2005

2003-2005 Est.Cost (MM$)

Additions (BCF)2005-2025

2005-2025 Est.Cost (MM$)

1 New England - - 32.1 382

2 Middle Atlantic 5.0 47 108.9 914

3 East North Central 0.9 8 111.4 905

4 West North Central - - 36.9 332

5 South Atlantic 7.6 72 98.5 804

6 East South Central 16.0 149 6.0 227

7 West South Central 6.6 62 9.2 282

8 Mountain 3.2 30 54.3 468

9 Pacific 29.3 274 35.1 225

Total U.S. Lower-48 68.6 642 492.4 4,539

10 Canada East - - 54.3 250

11 Canada West 40.0 100 1.8 10

Total Canada 40.0 100 56.1 260

Total NorthAmerica

108.6 742 548.5 4,799

Notes: MM$ = millions of dollars. BCF = billion cubic feet.

to develop storage facilities. Proving necessity, i.e.,market need, and tailoring implementation plans forminimization of environmental impact are two areasthat can have widely varying meanings depending onthe approval entity involved. When two or more gov-ernmental entities must be satisfied, the complexityinvolved in satisfying all parties increases exponen-tially.

The capital investments that would be required toadd 700 BCF of additional working gas capacity by2025 are significant, yet small relative to the potentialcapital requirements in other sectors of the natural gasindustry. A more significant issue with regard tofinancing storage capacity growth is whether there willbe adequate market signals to encourage such invest-ment.

Storage development costs vary significantly fromregion to region and by facility type. Expansions ofexisting facilities have the potential to add approxi-mately 200 BCF of incremental capacity at an averagecost in the range of $5 million per BCF of working gas,while new projects will require $5 million to $10 mil-lion per BCF. Total financial requirements of adding700 BCF of working gas capacity by 2025 are likely tobe in the range of $4 billion to $6 billion.

Similar to pipeline transmission capacity, contract-ing practices for natural gas storage capacity are cur-rently undergoing significant change, and it is not yetapparent how the market requirement for increasedcapacity will be translated into contractual arrange-ments to underpin investments. Until recently, thefastest growing segment of storage customers was theenergy marketing companies who were primarilyfocused on price arbitrage opportunities. Over the last18 months, a number of storage operators report anoticeable retreat from gas storage contracting on thepart of energy marketing companies, due in no smallpart to the financial difficulties of this segment.

Also, market and regulatory trends of recent yearshave caused local distribution companies to becomeless active in contracting for long-term gas storagecapacity. The introduction of customer choice pro-grams and the uncertainties regarding the LDC’s roleas “supplier of last resort” (as discussed in theDistribution section of this chapter) have presenteddifficulties for LDCs in forecasting their future con-tractual requirements for gas supply, pipeline capacity,and storage capacity. At the same time, in the recent

past there was a strong movement towards LDCreliance on energy marketing companies to managecontracted LDC storage capacity, often through short-term asset optimization arrangements between LDCsand marketers.

Coupled with market conditions that were charac-terized by relatively small summer/winter spreadsthroughout the first five months of 2003, these trendshave resulted in what can be described by storagedevelopers as a very soft market for the development ofnew gas storage capacity, notwithstanding the positivelonger-term fundamentals as echoed herein. In orderfor gas storage capacity development to meet antici-pated future demands, storage developers report theymust see a revitalization of demand for multi-year gasstorage contracts through some combination of cus-tomers such as LDCs and/or others with firm obliga-tions to serve seasonal and peak-day market require-ments for critical needs customers. Another possibilitywould be the emergence, or re-emergence, of a busi-ness sector capable of performing this service rolewhile also in pursuit of price arbitrage opportunities.

Any access restrictions can have an even greaterimpact when the limited number of sites is considered.Depleted reservoirs, aquifers suitable for storage, andsalt formations are all of limited extent in NorthAmerica. Any target storage formation must first bereasonably close to a major pipeline before practicalstorage development can be considered. The merepresence of a candidate geologic formation for storagecavern development may not be sufficient to warrantpractical storage development. For example, eventhough there are extensive bedded salt deposits in theNortheast United States, storage cavern developmenthas been difficult to justify economically in theNortheast because there are few options for disposing(or otherwise utilizing) the massive salt brine volumesthat naturally result from salt cavern development.

Comparison to Other TransportationOther Transportation Outlooks

An assessment of recent pipeline projects indicatesthat North American inter-regional pipeline capacitygrew by 11.4 BCF/D from 1999 to 2002. This capacitygrowth exceeded the prediction of 8.7 BCF/D made inthe 1999 NPC study. Most of the difference occurredin the Southeast and the Rocky Mountains. Thegrowth in the Southeast was related to greater than


expected market expansion (market pull), while theRocky Mountain growth was in response to increasedsupply deliverability (supply push). The estimatedaverage annual cost of this expansion was approxi-mately $6.1 billion.15 This compares to a 1990s aver-age expenditure for the United States and Canada of$2.5 billion.16

The cost of pipeline construction per mile in theearly to mid-1990s increased at an annual rate of 1.5%per year. Costs grew more rapidly from 1998 to 2000,averaging over 11% per year. Costs declined somewhatafter the construction peak in 2000 because a smallernumber of active projects led to lower prices for pipe,materials, and construction crews. However, despitethe recent decline in construction activity, the growthrate in cost per mile increased by 3.1% per year from1993 through 2002, which is twice the rate projected inthe 1999 NPC study. The primary factors leading tolarger than projected cost increases were higherexpenses for right-of-way and labor.

Peak construction years for transmission pipelinesin this study occur when Arctic pipelines are underconstruction (2008-2013). The overall constructionestimates are lower than those that were projected inthe 1999 NPC study, principally because of:

� Lower natural gas demand

� Lower production estimates from mature produc-tion regions

� Significantly higher imports of LNG directly intoEast and West Coast markets

� Utilization of existing pipeline infrastructure totransport gas from growing production regions.

Transmission, Distribution, and StorageRecommendations

Sustain and Enhance Natural GasNatural GasNatural Gas Infrastructure

Although the United States and Canada have anextensive pipeline, storage, and distribution network,additional infrastructure and increased maintenance

will be required to meet the future needs of the naturalgas market. The recommended actions listed below arerequired to ensure efficient pipeline, storage, and dis-tribution systems:

� Federal and state regulators should provide regu-latory certainty by maintaining a consistent costrecovery and contracting environment whereinthe roles and rules are clearly identified and notchanging. Regulators must recognize that aginginfrastructure will need to be continuously main-tained and upgraded to meet increasing throughputdemand over the study period. They must also rec-ognize that large investments will be required for theconstructions of new infrastructure. To make thekinds of investments that will be required, operatorsand customers need a stable investment climate anddistinguishable risk/reward opportunities. Changesto underlying regulatory policy, after long-terminvestments are made, increase regulatory andinvestment risk for both the investor and customers.

� Complete permit reviews of major infrastructureprojects within a one year period utilizing a “JointAgency Review Process.” Projects that connectincremental supply and eliminate market imbal-ances should be the highest priority and expedited.Where available supply is constrained, FERC shouldexpedite timely infrastructure project approvals thatwill help mitigate the current supply demand imbal-ance. Longer term, new project reviews should beexpedited via continuing enhancement andincreased participation in a Joint Agency ReviewProcess, similar to that which FERC has utilizedrecently. A Joint Agency Review Process wouldrequire up-front involvement by all interested/con-cerned parties including appropriate jurisdictionalagencies, allowing the decision process to proceed toapproval and implementation more accurately, moretimely, and at lower overall cost. The final FERCrecord should resolve all conflicts. The areas of great-est concern in this regard are requirements of theU.S. Army Corps of Engineers, Coastal ZoneManagement Act, and Section 401 of the Clean WaterAct, all of which could hinder the orderly implemen-tation of FERC certificates. This process must alsoassure that a project, which has used and successfullyexited this process, may proceed per the directionreceived and will not be delayed by non-participatingparties or other external regulatory standards orprocesses. This suggestion is a more-specific render-ing of the 1999 NPC study’s fifth recommendation:


15 The INGAA Foundation, Inc., Pipeline and Storage Infra-structure for a 30 Tcf Market – An Updated Assessment, 2002.

16 Ibid.

“Streamline processes that impact gas development.”The NPC supports legislation that accomplishes theJoint Agency Review Process as described above.Regulators at federal, state, and local levels, withcooperation of all participating parties, should estab-lish processes and timelines that would complete theregulatory review and approval process within 12months of filing.

� Regulatory policies should address the barriers tolong-term, firm contracts for entities providingservice to human needs customers. Many LDCs willnot enter into long-term contracts in today’s marketout of fear that regulators may subsequently deemthem imprudent in the future. Similarly, power pro-ducers, especially those that provide peaking service,are reluctant to contract for firm pipeline servicebecause charges for firm service cannot be economi-cally justified in power sales. As discussed in Finding9 in the Summary volume of this report, this practiceis impairing the investment in infrastructure. Theresult is that regulatory practices that limit long-termcontracts (prudence reviews and ratemaking) inhibitefficient markets and discourage the developmentand enhancement of pipeline infrastructure. The reg-ulatory process must allow markets to transmit thecorrect price signals and enable market participantsto respond appropriately. Regulators should encour-age, at all levels of regulation, policies that endorse theprinciples of reliability and availability of the naturalgas commodity. All regulatory bodies should recog-nize the importance of long-term, firm capacity con-tracts for entities providing service to human needscustomers and remove impediments for parties toenter into such contracts.

� FERC should allow operators to configuretransportation and storage infrastructure andrelated tariff services to meet changing marketdemand profiles. At the interstate level, FERCshould continue to allow and expand flexibility intariff rate and service offerings and continue toallow market-based rates for storage servicewhere markets are shown to be competitive sothat all parties can more accurately value servicesand make prudent contracting decisions. Toensure that existing and future transmission, dis-tribution, and storage facilities can be adapted tomeet the significantly varying load profiles ofincreased gas-fired generation, FERC and stateregulators need to allow and encourage operatorsto optimize existing and proposed pipeline andstorage facilities. In some cases, this will requiresignificantly more flexible facility design basedupon peak hourly flow requirements, and/or amodification to existing facilities to provide foroptimizing storage injections in off-peak hoursor in shoulder months.

� Regulators should encourage collaborativeresearch into more efficient and less expensiveinfrastructure options. Funding for collaborativeindustry research and development is in the processof switching from a national tariff surcharge-fundedbasis to voluntary funding. Because of the benefitsof reduced costs, system reliability, integrity, safety,and performance, DOE should continue funding forcollaborative research. Regulators need to encour-age and remove impediments regarding cost recov-ery of prudently incurred R&D by the operators tofund necessary collaborative research.


CHAPTER 6 - MEXICO SUPPLY/DEMAND OUTLOOK 269

This chapter outlines the basic assumptions andoutlook for Mexico’s demand, supply, and asso-ciated infrastructure for natural gas.

Mexico consumes approximately seven quads(quadrillion Btu) of energy per year. Oil is the princi-pal source of energy for Mexico, and is used primarilyto provide heat for industrial applications, raw materi-als for chemical manufacturing, and fuel for powergeneration. Natural gas represents about 23% ofMexico’s total energy demand, as shown in Figure 6-1.It is used primarily in chemical and manufacturingprocesses and has recently become the fuel of choicefor new power generation facilities.

Mexico’s gas demand of approximately 4.5 billioncubic feet per day (BCF/D) is met primarily byindigenous gas production augmented by pipelineimports from the United States. There are two gastransmission systems with about 6,000 miles ofpipeline operated by Pemex that traverse the countrynorth and south, as shown in Figure 6-2.Additionally, there are approximately 5,000 miles ofdistribution pipelines. The gas transmission and dis-tribution system primarily serves industrial cus-tomers and the expanding power generation industry.

Background

Mexico continues to develop its natural gasindustry to meet its burgeoning gas and powerdemand growth. Major natural gas policy objec-tives by the Fox Administration and Mexico’sEnergy Regulatory Commission (CRE) are as fol-lows:

� Drive economic expansion by using natural gas forpower generation and industrial growth

� Use natural gas to meet environmental standardsin critical areas of the country, thereby decreasingconsumption of fuel oil.

The country’s natural gas sector is the most liber-alized part of Mexico’s energy system and portionsof its pipeline and distribution infrastructure areopen to private investment. However, energyinvestment spending by the state has traditionallyfocused on finding and developing new crude oilreserves for internal use and export. Several factors

MEXICO SUPPLY/DEMAND OUTLOOKCHAPTER 6

COAL 4%

NUCLEAR/

OTHER

10%

NATURAL GAS

23%

OIL

63%

Source: CRE (Mexico's Energy Regulatory Commission).

Figure 6-1. Primary Energy Consumption in Mexico

present significant challenges to the expansion ofthe gas market in Mexico:

� Limited capital is available to fund major newexploration and production projects, and foreigninterests have been effectively excluded from petro-leum exploration and production.

� There is a lack of gas pipeline infrastructure tomove natural gas from the main producing regionsin the south to the developing consuming regions inthe north.

� The focus of reforms in Mexico’s energy system hasbeen primarily in the area of electricity generation.

Mexico’s petroleum and natural gas resourceshave been held as property of the state since nation-alization of the petroleum industry in 1938.Further, labor unions generally do not support for-eign participation in the exploration and develop-ment of petroleum resources. Pemex retains exclu-sive rights to exploration and production ofMexico’s natural gas according to Article 27 of theConstitution. In order to achieve the expansion ofthe gas supply required to meet future demand, thecurrent administration has recently proposed a

Multiple Services Contract (MSC) concept to allowlimited foreign participation in exploration and pro-duction of oil and natural gas. Under the MSC con-cept, private companies bid to develop and operateproduction under a fee for services arrangement. Itis as yet unclear whether this concept will attractsufficient interest and capital to achieve the desiredgrowth objectives.

Economic Growth

Future demand for natural gas in Mexico will bedetermined by a number of factors. Economicgrowth, characterized by the change in gross domesticproduct (GDP), is a key factor influencing energydemand, including the demand for natural gas.Mexico’s GDP growth averaged 3.7% per year from1970 to 2002, generally declining over that period.Figure 6-3 shows annual GDP growth for Mexico,together with the GDP growth assumptions used inthe NPC study to project natural gas demand.Mexico’s projected GDP growth was developed in amanner consistent with the assumptions used in thisstudy for the United States (3% per year) and Canada(2.6% per year). These assumptions reflect an eco-nomic rebound after the recent economic recession.


MEXICALI

AGUASCALIENTES

MATAMOROS

MERIDA

CANANEAHERMOSILLO

GUADALAJARA

VERACRUZ

NACONOGALES

PEMEX

TRANSMISSION

SYSTEM

PRIVATE

OPEN-ACCESS

PIPELINE

LÁZARO

CÁRDENAS

CIUDAD

JUAREZ

SAN LUIS

POTOSI

IMPORTSIMPORTS

PIEDRAS

NEGRAS

IMPORTS

IMPORTS

NUEVO LAREDO


Figure 6-2. Mexican Gas Transmission Systems

However, given the uneven nature of the historicalgrowth, the range of average future GDP growth inMexico could be 2% to 4% per year.

The economic growth outlook for Mexico used inthe NPC projections are summarized as follows:

� 2003 – GDP growth of 3.0%

� 2004 – GDP growth of 3.5%

� 2005 to 2010 – GDP growth of 3.5 to 3.0% per year

� 2011 to 2020 – GDP growth of 3.0% per year.

Natural Gas Demand

Mexico’s natural gas demand reached 4.8 BCF/Din 2002 and has been growing at a rate of approxi-mately 5% per year since 1992. Figure 6-4 shows his-torical demand for the major sectors of natural gasconsumption. Growth rates for the 10-year period of1993-2002 are also shown for each area. The largestsector of demand is petroleum-related activity asso-ciated with Pemex. Power generation is the highestgrowth segment, as gas-fired power technologies

have been increasingly deployed in Mexico to satisfygrowing power demand. Industrial demand for nat-ural gas has shrunk since 1998 reflecting rising gasprices in North America. The significant drop inindustrial natural gas demand in 2002-2003 illus-trates the inter-relationships of Mexican industrieswith other industrial concerns in North America.Residential and commercial consumption is verysmall due to low space-heating requirements and thewidespread use of liquefied petroleum gas (LPG) forhome cooking.

Gas-fired power generation is likely to be the pri-mary source of increasing gas demand in Mexico, asevidenced by recent growth in combined-cycle gener-ation capacity from 5,700 megawatts in 2001 to 7,900megawatts in 2002. Mexico remains short of electric-ity generation capacity, and virtually all new capacityadditions are anticipated to be gas-fired.

The NPC study assumes natural gas demand willincrease in Mexico to accommodate the general eco-nomic growth as well as the specific requirements ofnew gas-fired generation capacity. Historically, gasdemand has been limited by the amount of indige-nous supply made available by Pemex plus imports.


-8

-6

-4

-2

0

2

4

6

8

10

1970 1975 1980 1985 1990

YEAR

1995 2000 2005 2010

PE

RC

EN

THISTORICAL PROJECTED

2011-2020

Figure 6-3. Mexican GDP Growth

It is unclear whether Pemex will be able to increasethe rate of internal supply growth. Therefore, thelevel of pipeline and liquefied natural gas (LNG)imports will be a critical determinant in futuredemand.

Three cases have been developed to consider poten-tial future demand for natural gas in Mexico:

� “High Demand Case,” with overall natural gasdemand growing at 6.0% per year, assuming naturalgas-fired power generation grows at a rate of 14%per year, consistent with the outlook of SENER(Mexico’s Energy Ministry).

� “Continued Trends,” wherein demand grows at ahistorical rate of 4.7% per year

� “Low Demand Case,” with overall natural gasdemand growing at 2.9% per year, reflecting importsupply limitations that could result from higherfuture natural gas prices ($4 to $6 per million Btu atHenry Hub).

Figure 6-5 shows the demand projections for thesethree cases. The outlook of the U.S. Energy

Information Administration is also included for refer-ence.

Natural Gas Supply

Current Production

Mexico currently produces approximately 4.5BCF/D, primarily from the regions shown in Figure 6-6. About 70% of Mexico’s natural gas is producedin association with oil, or “associated gas,” with theremaining non-associated gas produced from gasfields.

Figure 6-7 shows Mexico’s natural gas productionsince 1980. Production has been relatively flat since1997 with declines in associated gas fields of thesouthern regions offset by increases in non-associatedgas from northern fields. As productive capacity inthe south continues to decline, most of the new sup-ply will likely come from the northern areas ofMexico. The Pemex national pipeline system willneed to adapt to these changes.

The NPC study assumes that production growthrates will be 3% per year for each of the three


0

1

2

3

4

5

6

1993 1994 1995 1996 1997

YEAR

1998 1999 2000 2001 2002

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

OTHER

POWER

INDUSTRIAL

PETROLEUM

5%

14%

-1%

7%

GROWTH RATE

(1993-2002)


Figure 6-4. Mexican Gas Demand


0

2

4

6

8

10

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

12

14

1990 1995 2000 2005

YEAR

2010 2015 2020

DOE/EIA

HIGH CASE

CONTINUED TRENDS

LOW CASE


Source: Historical data from CRE (Mexico's Energy Regulatory Commission).

Figure 6-5. Mexican Gas Demand Outlook

BURGOS

MACUSPANA

CUENCA DE

VERACRUZ

POZA RICA

Figure 6-6. Major Natural Gas Producing Areas in Mexico


0

1

2

3

4

5

6

1980 1985 1990

YEAR

1995 2000

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

NON-ASSOCIATED

ASSOCIATED

Source: SENER (Mexico's Energy Ministry).

Figure 6-7. Historical Natural Gas Production in Mexico

0

1

2

3

4

5

6

7

8

9

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

3% GROWTH(NPC ASSUMPTION)

1995 2000YEAR

Source: SENER (Mexico's Energy Ministry).

2005 2010

SENER FORECAST

Figure 6-8. Production Outlook for Mexico

aforementioned cases (“High Case,” “ContinuedTrends,” and “Low Case”). Figure 6-8 shows theNPC projection of 3% per year overlaid on theprojection of the SENER, which reflects anincrease in natural gas production of 7% per yearthrough 2012. The SENER projection is doublethe historical rate and will likely require a signifi-cant increase in the rate of investment. In thisregard, the Fox Administration has doubled theresources channeled to exploration and produc-tion within the past two years.

Resource and Reserves Potential

Mexico has five regions with significant undiscov-ered gas: Sabinas, Burgos, Tampico-Misantla,Veracruz, and Sureste. All of the regions haveonshore, shelf, and slope components except forSabinas, which is exclusively onshore. These regionsare shown in Figure 6-9. Water depth and drillingdepth are keys to the economic potential of theseareas.

The NPC estimates 70 trillion cubic feet (TCF) ofundiscovered gas in Mexico. The onshore and off-shore Burgos Basin located adjacent to southern Texasis the most important non-associated gas basin in

Mexico with undiscovered gas potential of 26 TCF.There is significant potential for growth of existing gasfields by infill drilling and reduced spacing in theonshore Burgos Basin. The other significant area isSureste onshore and offshore with undiscovered gas of23 TCF, which is mostly associated gas. The offshoreGulf of Mexico is lightly drilled compared to the U.S.Gulf of Mexico, particularly in the deepwater areas.Limited capital availability and long lead times todevelop production tend to limit deepwater gas devel-opment activity.

Mexico has current proved gas reserves of 28 TCFand annual production of 1.3 TCF per year. Totalremaining reserves including proved, probable, andpossible reserves, are 51 TCF; of this total, 41 TCF isassociated gas and 10 TCF is non-associated.

Natural Gas Import/Export Balance

The Mexican natural gas import/export balancehas deteriorated over the past 10 years becauseindigenous supply has been unable to keep up withincreased demand. As shown in Figure 6-10, the netbalance increased to almost 800 million cubic feet perday (MMCF/D) in 2002 from the U.S. to Mexico.


BURGOS

SURESTE

VERACRUZ

TAMPICO

SABINAS

Figure 6-9. Major Natural Gas Resource Areas in Mexico

Major expansions of gas pipeline import capacityhave been implemented with current capacity of2.5 BCF/D.

Three cases have been developed to address poten-tial pipeline and LNG imports under the threedemand outlooks – High Demand, ContinuedTrends, and Low Demand cases. As discussed earlier,production is assumed to grow at the historical rateof 3% per year in each of these cases. Pipeline andLNG imports were varied to balance supply anddemand. For the Continued Trends case, Mexico willlikely have to rely on the United States to meet rap-idly growing demand through 2005 with pipelineimports reaching 1.6 BCF/D. Longer term, pipelineimports from the United States decline to 1.1 BCF/D,reflecting declining South Texas supplies and theincentive to reduce foreign imports to minimum lev-els. Pipeline capacity limits of about 2.5 BCF/Dwould not appear to be a limiting factor.Additionally, pipeline imports could be expanded ifsupplies were available and Mexico were willing tomanage the related trade balance with the UnitedStates. Figure 6-11 shows the U.S./Mexicoimport/export balance for the Continued Trends

case. Significant LNG imports are implied by thiscase, and could be available.

Table 6-1 summarizes the pipeline imports assumedin each demand case.

Potential LNG Import Volumes

LNG imports are likely the least expensive of thesupply options in the longer term. The ContinuedTrends case has been adopted as the base for


-0.2

0

0.2

0.4

0.6

0.8

1991 1993 1995

YEAR

1997 1999 2001

IMPORTS TO MEXICO

EXPORTS FROM MEXICO

NET IMPORTS

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y


Figure 6-10. Historical U.S./Mexico Natural Gas Import/Export Balance

Case 2010 2020

High Demand 1.1 1.1

Continued Trends 1.1 1.1

Low Demand 0.8 0.9

Table 6-1. Pipeline Imports from United States to Mexico, Excluding Baja

(Billion Cubic Feet per Day)

U.S./Mexico import/export imbalances and requiresup to 2.9 BCF/D of LNG imports. The HighDemand case would require up to 5.5 BCF/D ofLNG imports. The Low Demand case requires noLNG imports due to the production growing at thesame rate as demand growth. Figures 6-12, 6-13,and 6-14 show the pipeline and LNG imports for thevarious cases.

Pipeline Interconnection Capacities

There is 2.6 BCF/D of pipeline import capacitybetween the United States and Mexico. This is suffi-cient capacity to meet the forecast Mexico importsfrom the United States unless new LNG import facili-ties are not constructed. In the “No LNG Imports”case, Mexico could need up to 6 BCF/D of pipelinecapacity. Figure 6-15 shows the pipeline interconnects.

LNG Import Terminals

LNG is expected to meet some of the growingMexican demand, particularly for power generationin the eastern and Baja areas. There are proposals todevelop LNG import facilities on the east and west

coasts of Mexico. The CFE (Mexican FederalElectricity Commission) has proposed the con-struction of import facilities in Altamira for a majorIndependent Power Plant development. Foreignfirms have shown interest in constructing importfacilities on the Pacific coast to serve markets inMexico and the Southwest U.S. LNG imports areimportant in that they will be designed to meetgrowing demand while not necessarily dependingon the Pemex transmission system. For instance,the Baja LNG terminal will not necessarily be con-nected to the Pemex system. Also, the Altamira ter-minal will primarily serve a single generation facil-ity. Figure 6-16 shows the proposed LNG importterminals.

The overall supply/demand balance could be sig-nificantly altered with varying LNG import profiles.LNG will have to compete with indigenous supplyfrom Mexico and with U.S. imports. As such, it seemsunlikely that Mexico will significantly increase itspipeline gas imports from the United States in thelonger term while pursuing LNG imports. Also therewill be large incentives for Mexico to fully develop itsindigenous supply.


-0.5

0

0.5

1.0

1.5

2.0

1990 1995 2000 2005 2010 2015 2020

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

EXCLUDING BAJA

LNG SHIPPED TO U.S.

NET INCLUDING BAJA

LNG SHIPPED TO U.S.

NET IMPORTS

NET IMPORTS TO MEXICO

NET EXPORTS TO U.S.

YEAR


Source: Historical data from Energy Information Administration.

Figure 6-11. U.S./Mexico Natural Gas Import/Export Balance Assumptions

Strategic Implications

Mexico remains an uncertainty in the NorthAmerican gas balance. Unknowns related to bothdemand and supply give rise to widely varyingpotentials for import and exports to the UnitedStates. This balance ultimately rests on Mexico’sstance relative to foreign participation in its explo-ration and production industries and its ability toattract LNG imports. Only with major increases ininvestment will Mexico be able to develop its sup-ply resources at a pace different from history.Without new indigenous supplies, demand growthwill be limited to what gas is produced andimported.

The NPC North American gas balance assumes theContinued Trends scenario with net pipeline importsfrom the United States to Mexico of approximately1.1 BCF/D (excluding Baja LNG shipments of 400MMCF/D to the United States via the Mexicanpeninsula).


0

2

4

6

2000 2005YEAR

2010 2015 2020

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

PIPELINE IMPORTS

Note: Demand grows at 3.0% per year.Production grows at historical 3% per year.

Figure 6-14. Mexico Imports – Low Demand Case

0

2

4

6

8

2000 2005 2010 2015 2020

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

LNG IMPORTS

PIPELINE IMPORTS

YEAR

Note: Demand grows at 5-7% per year.

Production grows at historical 3% per year.

Figure 6-12. Mexico Imports – High Demand Case

0

2

4

6

2000 2005YEAR

2010 2015 2020

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

LNGIMPORTS

PIPELINE IMPORTS

Note: Demand grows at 4.7% per year.Production grows at historical 3% per year.

Figure 6-13. Mexico Imports – Continued Trends


EL PASO - 150 MMCF/D

POTENTIAL

INTERCONNECTION POINTS

INTERCONNECTION POINTS

TOTAL INTERCONNECTION

CAPACITY = 2.6 BCF/D

TETCO - 150 MMCF/D

TENNESSE - 235 MMCF/D

MEXICALI - 28 MMCF/D

SAMALAYUCA - 272 MMCF/D

KINDER MORGAN - 300 MMCF/D

NACOZARI DE GARCIA - 85 MMCF/D

EXISTING PIPELINESROSARITO - 809 MMCF/D

NACO - 90 MMCF/D

CIUDAD JUAREZ - 80 MMCF/D

PIEDRAS NEGRAS - 38 MMCF/D

Note: MMCF/D = million cubic feet per day.

BCF/D = billion cubic feet per day.


Figure 6-15. Current Mexican Interconnection Capacities

EXISTING PIPELINES BAJA

0.7 - 1.4 BCF/D

TOPOLOBAMPO

MANZANILLO

ALTAMIRA

0.8 - 1

BCF/D

LÁZARO

CÁRDENAS

0.5 BCF/D

PROPOSED PIPELINES

Note: BCF/D = billion cubic feet per day.


Figure 6-16. Proposed Mexican LNG Terminals

CHAPTER 7 - NATURAL GAS MARKETS 281

Market-Related Issues

The North American natural gas market is thelargest and most liquid gas market in the world, withhundreds of suppliers and thousands of major con-sumers including local distribution companies(LDCs), industrials, and power generators. The mar-ket is functioning efficiently with lessened governmentinvolvement following years of regulatory reform. Thewholesale commodity and related financial instru-ments are not price regulated while the Federal EnergyRegulatory Commission (FERC) does regulate inter-state gas transmission and storage services.Furthermore, state public utilities commissions regu-late the prudency of gas commercial practices and ratesof local gas distribution companies.

Key characteristics of a healthy and well-functioningcompetitive North American gas market are a highdegree of price transparency and overall market liquid-ity. Market participants must have reliable supply,demand, storage, pipeline capacity, and price informa-tion. They should be encouraged and be able to takeappropriate actions in response to market signals,thereby optimizing the supply/demand balance fromtheir individual perspective.

There have been significant changes in gas marketparticipants since 2001. Several large marketing com-panies have exited the physical and financial gas trad-ing business, and on-line trading operations havedeclined. The broad portfolio of financial productsoffered by these players has been reduced and the needto trade with creditworthy entities has been reinforced.These changes have highlighted a potential decline inmarket depth (e.g., number of players) particularly forlong-term hedges, and therefore have contributed to a

reduction in some customers’ ability to manage long-term price volatility.

Price Transparency

Industry publications have reported monthly, week-ly, and daily price indices for physical trades of gas atapproximately 100 different locations for more than 10years. These indices are based on the reported volumestraded at each location for actual fixed price physicalcontracts. In addition, the highly transparent NewYork Mercantile Exchange (NYMEX) futures contractfor the Henry Hub, Louisiana physical delivery point isavailable for forward financial pricing and gas-relatedfinancial instruments. Industry and regulators haverecently questioned the integrity of the physical indexreported trade information. Some market participantshave acknowledged improper reporting of trade infor-mation. There is a major effort headed by FERC andindustry to improve the integrity of the process and toclarify and encourage accurate and adequate pricereporting. The NPC supports FERC’s efforts toincrease market transparency and price reliability withminimal government intervention.

Physical Market Liquidity

Liquidity of the physical gas markets must be exam-ined separately from that of the financial markets.Physically traded volumes of gas (first sales) haveincreased as supply and demand have grown over the1986 to 2001 time period. In parallel, reported physi-cal trade volumes of the largest natural gas marketersgrew throughout the 1990s. At the peak in 2001-2002,the top ten marketer’s volumes represented more thantwice the average daily physical consumption of gas, asshown in Figure 7-1. This buying and reselling of gastends to add liquidity and flexibility to the market. Gas

NATURAL GAS MARKETSCHAPTER 7

marketer volumes declined in 2002 and continued todecline in 2003 with the restructuring of the merchanttrading business.

Although physical flows have remained relativelyconstant, liquidity at some locations other than themajor hubs is reduced from that of recent years, andreported trading volumes have declined from recentpeaks. A number of factors may be contributing tothis phenomenon, including the lack of a safe harborfor inadvertent mis-reporting of gas trade informa-tion. Industry and FERC efforts to improve reportingof trades are targeted to improve this issue. Whileindustry lacks comprehensive historical data for trad-ing liquidity, anecdotal evidence indicates that avail-ability of buyers and sellers at various market hubsvaries significantly depending on supply and demandfactors. Some points are known to be highly liquidwith very large volumes and hundreds of participants,while others are not. Prudent market participantsshould continue to develop sufficient market intelli-gence to guide commercial decisions and protect theirinterests. Buyers and sellers have the option to tradeat larger, liquid points and make appropriate loca-tional or price basis adjustments for other less liquidpoints.

Credit between counterparties has become a muchmore important issue than in the past. Market partic-ipants have high-graded their portfolio of trading par-ties and reduced overall credit risk.

Financial Market Liquidity

The rise in the use of financial products has beenfairly dramatic since 1990. The trend in the use ofNYMEX financial instruments is illustrated in Figure7-2 and shows increasing open interest in NYMEXcontracts through mid-2002. A decline is exhibited in2003. Open interest is a measure of activity onNYMEX and gives some indication of overall marketdepth and liquidity.1 Current levels of NYMEX trad-ing at the Henry Hub are below the peak but above theoverall range of the 1990s. Some of the peak activitywas likely exacerbated by credit issues between coun-terparties that pushed transactions to the NYMEX inorder to address the credit concerns as several largemarketers exited the business.


0

20

40

60

80

100

120

140

160

1Q 2Q2001

3Q 4Q 1Q 2Q2002

3Q 4Q 1Q2003

BIL

LIO

N C

UB

IC F

EE

T P

ER

DA

Y

AVERAGE DAILY CONSUMPTION

Figure 7-1. Quarterly Gas Volumes by Top Ten Marketers

1 Open interest is defined as the number of open or out-standing contracts for which an individual or entity isobligated to the NYMEX because that individual or enti-ty has not yet made an offsetting sale or purchase, anactual contract delivery, or, in the case of options, exer-cised the option.

Marketers have traditionally been the major marketmakers and counterparties for a broad suite ofNYMEX and over-the-counter financial tools (priceswaps, forward price options, basis swaps, etc.) inaddition to physical gas volumes. There are nowfewer marketing entities offering these comprehen-sive services. The creditworthiness of the remainingparties and new entities entering the market (prima-rily banks) however has improved. This restructur-ing of the financial markets should be viewed as apositive trend.

Despite the recent changes in market participants,overall liquidity remains sufficient for parties to trans-act at multiple physical trading hubs and to accesseffective financial markets. Physical volumes at majorhubs have been relatively constant throughout theperiod. The volume of financial trades has been vari-able and is down from peaks in 2001.

Continued enhancement of market liquidity andexpanded market depth remain goals for industry, andthe market is adjusting as appropriate. Governmentshould allow free market forces to work, and marketswill continue adjusting for an effective, efficient balance.

Natural Gas Price Volatility

Restructuring of the gas industry and the deregula-tion of the natural gas commodity has produced acompetitive market with lower natural gas prices toconsumers. Accompanying this deregulation has beengreater variability in natural gas prices as market forcesestablish prices in the monthly and daily markets.Price volatility is a natural dynamic in a commoditymarket where supply and demand vary. Natural gas,electricity, crude oil, and oil product markets have allexhibited price volatility to varying degrees. Relativelylarge price changes (spikes and declines) occur in nat-ural gas markets because supply and/or demand arenot able to adjust quickly enough to cause a smoothprice trend. Volatility tends to highlight inelasticity insome market segments.

The principal drivers behind volatility are supplyand demand fundamentals, which include growthtrends, weather, storage levels, and perceived markettrends. Price volatility has a wide range of impact onmarket participants and there are several tools to man-age the effects.


0

100

200

300

400

500

600

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

NU

MB

ER

OF

CO

NT

RA

CT

S (

TH

OU

SA

ND

S)

Open Interest: The number of open or outstanding contracts for which an entity is obligated to the Exchange because that entity

has not yet made an offsetting sale or purchase, or an actual contract delivery.

YEAR

Figure 7-2. NYMEX Open Interest – Natural Gas Contracts

Price Volatility in the North American Market

The vast majority (80-90% by volume) of naturalgas marketed in the United States and Canada is soldon a monthly basis. The remainder (10-20%) isbought and sold in the daily cash market and is prima-rily used to manage the overall supply/demand balanceduring the month. Volatility is a measure of the varia-tion of price from its mean value over a period of inter-est (daily, monthly, or yearly). Volatility in the broadestsense is the “noise” around the long-term movement ofprice. Some industry participants tend to think ofvolatility either in terms of abnormally “high” or “low”prices, or specific upward or downward movement inprices. This is incorrect. Volatility is simply a measureof variability around a mean value, not a measure ofthe absolute price.

Price volatility is important to market participantsin optimizing near-term operating decisions becausethe level of volatility establishes the cost of options ingas futures contracts on NYMEX. The annual variabil-ity of gas price, if it is sufficiently large, creates a “sea-sonal spread” that produces an incentive for storage ofgas among merchant energy companies and producers.It is, however, the long-term price trend that drivesmajor investment decisions in both the consuming andsupply sectors.

Volatility Analysis

Gas prices exhibit a “log normal” distribution due tothe fact that prices have no upside constraint, but areconstrained on the downside by zero, as demonstratedin Figure 7-3. Therefore, a random distribution will beskewed positively around the mean price, the essenceof log normality.

Although commodity prices follow a log normal dis-tribution, changes in prices over specific periods can beeither positive or negative, and approximate a normaldistribution. Therefore the financial community looksat the log of the relative price changes to model histor-ical and future price variations (see calculationmethodology in box).


Finding: Price volatility is a fundamentalaspect of a free market, reflecting thevariable nature of demand and supply;physical and risk management tools allowmany market participants to moderate theeffects of volatility.

0

10

20

30

40

50

60

$0 TO $1 $1 TO $2 $2 TO $3 $3 TO $4 $4 TO $5 $5 TO $6 > $6

NU

MB

ER

OF

OB

SE

RV

AT

ION

S

Figure 7-3. Frequency of Monthly Prices (1995 to 2003), Number of Observations in Price Range

For the purposes of this discussion, volatility isexamined in a historical perspective. Implied for-ward volatility and forward NYMEX prices are finan-cial tools that may be used to understand where themarket is trading for future periods. The NPC rec-ognizes that market participants may use the forwardfinancial markets to buy and sell gas or enter intoother hedging activities (e.g., puts and calls) toobtain price certainty and mitigate the impact ofprice volatility.

Historical Natural Gas Prices and Volatility

Henry Hub is a pipeline interchange in Louisianawhere a number of interstate and intrastate pipelinesconnect through a header system. It is the standarddelivery point for the NYMEX natural gas futures con-tract. There are two common price bases quoted fornatural gas: 1) gas sold monthly and based on a first-of-month index price, and 2) gas sold on a daily cashbasis. Figure 7-4 shows Henry Hub natural gas pricesfor both price bases.

North American gas prices have ranged since 1994from less than $2.00/MMBtu to $10.00/MMBtu at theHenry Hub. The monthly index and daily cash pricesfollow each other closely. However, the daily cash priceshows wider variability than the monthly market. Thisis particularly evident in the winters of 1995-1996 and2000-2001.

Volatility of cash prices as calculated on a rolling 30-day basis has varied from 20% to 200% and has beenhighest during the late winter period, as shown inFigure 7-5. Periods of very high volatility reflect rela-tively inelastic demand during a peak winter period,usually exacerbated by abnormal weather, as shown inFigure 7-6. There is no correlation between volatilityand the absolute price, because there are volatile peri-ods with prices across the entire range.

Yearly average price volatility, as measured from themonthly index prices, is 60% over the 1995-2003 peri-od, as shown in Figure 7-7. This volatility measure isrelated to the range of monthly prices that could beexpected over a one-year period for longer-terminvestment decisions.

Comparison of Natural Gas Price Volatility vs.Crude Oil and Electricity

Figures 7-8 and 7-9 show the price trends for crudeoil and electricity, respectively. Electricity price hasvolatility in the range of 200% and higher. Volatility inelectricity price has been substantially higher thancrude oil or natural gas. The primary drivers are theinability to store electricity and a declining reservemargin in key areas of the United States. Recentdeclines in volatility show, in part, the impacts ofmajor gas-fired capacity additions creating a surplus ofgeneration capacity above consumption requirements.


Daily Gas Price Analysis

Pi = Price on a specific dayP i-1 = Price on prior dayPrice Change i = Return i = Ln(Pi / Pi-1)Return average = (∑ Return i )/n

Where: n = total number of price observationsLn = natural log∑ represents “the sum” from 1 to n observations

Standard Deviation = Square root of variance= SQRT[{∑(Return i – Return avg)* 2}/(n-1)]

Annualized Volatility = (Standard deviation) X (SQRT of # of prices in period)Volatility is expressed as percentage. By convention, the number of prices ortrading days in a year is 256 for daily prices.


0

2

4

6

8

10

12

0

50

100

150

200

WINTER (NOVEMBER-MARCH)

SUMMER/SHOULDER MONTHS

RECORD COLD NOV-DEC 2000

LOW STORAGE & SUPPLY ISSUES

EL NIÑO WINTER

1997-1998

WINTER PRICE SPIKE ON CONSECUTIVE HIGH

HEATING DEGREE DAYS IN THE NORTHEASTA

NN

UA

LIZ

ED

30-D

AY

RO

LLIN

G V

OLA

TIL

ITY

(P

ER

CE

NT

)

HE

NR

Y H

UB

DA

ILY

GA

S P

RIC

E (

DO

LLA

RS

PE

R M

ILLIO

N B

TU

)

EL NIÑO

2002-2003

HENRY HUB

CASH PRICE

30-DAY ROLLING

VOLATILITY

1995 1996 1997 1998 1999 2000 2001 2002

YEAR

2003FEB AUG FEB AUG FEB AUG FEB AUG FEB AUG FEB AUG FEB AUG FEB AUG FEB

Figure 7-5. Natural Gas Price and Volatility

0

2

4

1995 1996 1997 1998 1999YEAR

2000 2001 2002 2003

6

8

10

12D

OLL

AR

S P

ER

MIL

LIO

N B

TU

HENRY HUB DAILY CASH PRICE

HENRY HUB MONTHLY INDEX PRICE

Figure 7-4. Henry Hub Natural Gas Prices


0

25

50

75

100

125

150

40 45 50 55 60 65 70 75 80 85

JANUARY

FEBRUARY

MARCH

DECEMBER

NOVEMBER

AUGUST APRIL

OCTOBER

SEPTEMBER

JUNEJULY

MAY

MO

NT

HLY

AV

ER

AG

E 3

0-D

AY

VO

LA

TIL

ITY

(PE

RC

EN

T)

MONTHLY AVERAGE GAS DEMAND (BILLION CUBIC FEET PER DAY)

WINTER MONTHSSHOULDER MONTHSSUMMER MONTHS

Figure 7-6. Relationship of Monthly Average Volatility and Gas Demand

0

40

60

80

100

120

1994 1995 1996 1997 1998

YEAR

1999 2000 2001 2002

PE

RC

EN

T

Figure 7-7. Henry Hub First of Month Index Volatility (Full Year Annualized)


0

50

100

150

200

250

300

350

400

DO

LLA

RS

PE

R M

EG

AW

AT

T H

OU

R

1997 1998 1999

YEARNote: ERCOT = Electric Reliability Council of Texas, Inc.

2000 2001 2002

Figure 7-9. ERCOT Daily Peak Electricity Price

0

5

10

SEP MAR SEP MAR SEP MAR SEP MAR

15

20

25

30

35

40D

OLL

AR

S P

ER

BA

RR

EL

SEP MAR SEP MAR SEP MAR SEP SEPMAR

1994 1995 19971996 1998 2000 2001 20021999

Figure 7-8. West Texas Intermediate Crude Oil Daily Cash Price

As shown in Figure 7-10, crude oil prices have exhib-ited lower volatility on average than natural gas, withyearly volatility averaging 40%. The stabilizing effectof OPEC and spare production capacity are the pri-mary keys for the lower volatility.

Key Drivers of Natural Gas Price Volatility

Gas consumption variability and inelasticity in theUnited States are primary drivers behind price fluctua-tions. The winter peak of 80+ BCF/D can be comparedto the summer low of approximately 45 BCF/D. Gasstorage facilities have been developed all over theUnited States to balance this market. During the sum-mer period, gas is stored for use during the winter. TheEnergy Information Administration (EIA) estimatesthat about 4.2 TCF of gas storage capacity exists in theUnited States. On an annual basis, about 2 to 2.5 TCFof “working” gas is used to keep the market in balance,thereby mitigating seasonal price volatility.

Since the mid-1990s, the gas producers in NorthAmerica have been producing at maximum ratesthroughout the year. The production profile has beenrelatively flat, as seen in Figure 7-11. Gas production inexcess of demand is injected into storage in the sum-mer and pulled from storage to meet the winter peak asshown by the shaded areas.

Supply and Demand Elasticity Effects

The gas supply in North America is inelastic in theshort term. The ability to increase supply in the shortterm is limited to shutting down rich gas processing


0

50

100

150

200

250

300

350

400

1994 1995 1996 1997 1998

YEAR

1999 2000 2001 2002

WEST TEXAS INTERMEDIATE CRUDE

HENRY HUB GAS

ERCOT ELECTRICITY

PE

RC

EN

T

Figure 7-10. Energy Price Yearly Volatility Comparison – Daily Cash Prices (Yearly Period)

AffectsSupply

AffectsDemand

Weather √ √Inelasticity of Demand (during winter peaks) √Storage Levels √Pipeline Capacity √Operational Factors √Lack of Timely, Reliable Information √ √Alternate Fuel Price Volatility √

and/or gas injection for secondary oil recovery.Significant increases in supply have been difficult toachieve in recent times even with near record gas-directed drilling rig activity. Canadian gas importshave risen to 16% of total U.S. supply, as domesticproduction has not been able to keep up withdemand. This short-term supply inelasticity con-tributes to price volatility. Supply is more elastic inthe longer term with the potential to explore anddevelop new large supplies (e.g., deepwater Gulf ofMexico, Arctic, unconventional). However, the longlead times and large investments make short-termchanges difficult.

For gas demand, the primary driver behind the sea-sonal consumption profile is space heating for the res-idential and commercial customers. This “LDC”demand is driven by the weather. In effect, the demandcurve shifts to the right from summer to winter asshown in Figure 7-12. This dynamic shift in seasonaldemand moves the equilibrium point between sup-ply/demand upward and toward the steeper, less elasticportion of the demand curve. As a consequence, dur-ing a cold period in the winter when demand peaks, gasprice can change very quickly as the market provides aprice signal to consumers to curtail use or to switch to

an alternate fuel (if possible). The rapid change inprice leads to high volatility.

Pipeline Capacity and Operational Factors

Although the North American gas pipeline grid iswell interconnected, there are constraints on theamount of gas that can be transported between the


PR

ICE

QUANTITY

SUPPLY

DEMAND(SUMMER)

DEMAND(WINTER)

P1

P2

Figure 7-12. Price Elasticity Effects

PRODUCTION INJECTED INTO STORAGESTORAGE WITHDRAWALS

0

40

50

60

70

80

90B

ILLI

ON

CU

BIC

FE

ET

PE

R D

AY

1998

GAS DEMANDGAS PRODUCTION PLUS IMPORTS

1Q 2Q 3Q 4Q1999

1Q 2Q 3Q 4Q2000

1Q 2Q 3Q 4Q2001

1Q 2Q 3Q 4Q2002

1Q 2Q

Figure 7-11. U.S. Natural Gas Consumption and Production

supply areas and demand centers, particularly duringthe winter peak season. Therefore price differentialsbetween areas, or a “basis,” sometimes widen (orshrink) reflecting the availability of pipeline capacity.Pipeline capacity relative to demand impacts the deliv-ered price and affects price volatility.

Price differentials reflect the value of transportinggas between regions and provide market signals andincentives for new pipeline capacity additions. Inregions with excess capacity, the price basis may dropbelow the pipelines’ published tariff rates for firmtransportation. In regions where capacity is tight, theprice basis may exceed the published tariff rates.Figure 7-13 shows the difference in price versus theHenry Hub for New York and Chicago citygates andthe Rockies production area from 1998 to early 2003.

Citygate prices generally exceed wellhead prices,reflecting the value of transportation capacity betweenthe production area and the market area. Between theGulf Coast and New York, the basis variation hasranged as low as a few cents to over $2.00/MMBtu.Winter periods generally show the highest basis differ-entials due to pipeline capacity constraints. Chicagobasis is lower than New York basis due to excess

pipeline capacity from the Gulf Coast and Canadianproducing areas to the Midwest. Prices in the Rockiesproduction area are up to $2.00 less than the HenryHub price, reflecting insufficient pipeline capacity tothe market.

Lack of Timely, Reliable Information

The FERC and EIA publish demand and supply dataon a monthly basis. EIA monthly reports attempt todocument the overall U.S. supply and demand balance.However, due to the lack of complete data, the nearest6 months are estimates. Information out to about 18months is a combination of estimates and actual data,which are frequently revised (with large variations).The lack of reliable and timely information results inmarket uncertainty. On a daily basis, the marketsearches for the right clearing price. Uncertainty aboutdemand and supply of gas is a contributing factor tothese daily changes. The market has developed otherindirect measurements of supply and demand to assistin understanding trends. For example, the marketclosely monitors gas in storage, drilling rig activity, andheating and cooling degree days, among other funda-mentals. While helpful in some respects, these sources


-4

-2

0

2

4

6

8

10

1998 1999 2000

YEAR

2001 2002 2003

DO

LLA

RS

PE

R M

ILLIO

N B

TU

ROCKIES PRODUCTION AREA BASIS

CHICAGO CITYGATE BASIS

NEW YORK CITYGATE BASIS HENRY HUB

PRICE

Figure 7-13. Market Basis to Henry Hub

of information are, at best, second-hand indications oftrue supply and demand trends.

Alternate Fuel Price Effects

Many industrial and utility customers have devel-oped the ability to switch to an alternate fuel. Theavailability of this switchable load potentially decreas-es the upward price movement of gas for these cus-tomers when gas price exceeds alternate fuel parity anddecreases downward price movement when belowalternate fuel parity. Overall this has the potential todecrease natural gas price volatility during peakdemand periods.

Factors that Mitigate Gas Price Volatility

� Gas storage

� Fuel switching

� Financial hedging (does not eliminate risk but doescreate price certainty)

� Excess production capability and pipeline capacity

� Long-term contracting

� Timely and reliable information

Exposure to Price Volatility and Its Effects

Retail and Commercial Customers

Most LDC firm-service customers are insulatedfrom the day-to-day volatility in natural gas prices.Residential deliveries and approximately 60% of totalcommercial customers purchase natural gas at regu-lated rates from an LDC. The cost of natural gas tothese customers is controlled by regulation, and gen-erally reflects the rolled-in average cost of natural gasat the LDC citygate, plus the LDC distributioncharge. The average cost of gas is adjusted on agoing-forward basis, typically delayed by one to threemonths. In addition, many LDCs hedge gas prices ona portion of their requirements, either through phys-ical means via natural gas storage, contractual meansvia longer-term (monthly and seasonal) gas purchasecontracts, or a financial hedge. As a result, the gasprices faced by these users generally do not vary withshort-term changes in energy market prices.However, persistent price changes do result in sub-stantial price effects. Although prices to retail cus-tomers have varied over the past 10 years, only the

upward movements tend to receive significant regu-lator and customer attention.

Industrial Customers

Industrial customers tend to be more exposed tovolatility in energy prices. A vast majority of industri-als (more than 80%) purchase gas in the daily ormonthly markets and transport the gas to their facili-ties. The natural gas commodity is purchased either atmarket prices, or hedged through a third party.2 Ineither case, industrial customers are exposed to marketprices. Sales to industrial customers via LDCs at regu-lated prices account for only a small percentage(approximately 17%) of total sector requirements.

Industrial customers tend to have more options forreducing gas usage in response to price increases.Some industrial applications have dual-fuel capability,and can switch to residual fuel oil or distillate fuel oilwhen natural gas prices exceed fuel oil prices. Whengas prices rise, industrial facilities may also choose toshut down production rather than use natural gas.During the peak price periods from 2000 to 2002, largeamounts of ammonia production capacity shut downin response to higher natural gas prices.3

As a result, industrial customers tend to be moreprice sensitive than commercial or residential cus-tomers. This price sensitivity is reflected in both oper-ational day-to-day decisions, and in long-terminvestment decisions in energy technologies. Pricevolatility can impact profitability for the industrial sec-tor in positive and negative ways depending on thedirection of natural gas price movement. Sustained(multi-month) price spikes may also cause businessrationalization that cannot be easily or cost-effectivelyreversed. However, it is high absolute natural gas


2 The larger industrial consumers can consume enoughnatural gas to make direct price hedging attractive, henceproviding some insulation from price changes.

3 Examples in 2002 include: Mississippi Chemicalannounced the permanent shutdown of its Donaldson,Louisiana urea facility because of pricing pressures – thecomplex has an annual capacity of one million tons ofammonia and 578,000 tons of urea synthesis. Missouri-based Farmland Industries indicated the prolonged down-turn in fertilizer manufacturing resulted in a $183 millionloss in 2002 and a Chapter 11 filing on May 31, 2002.Pennsylvania-based Air Products and Chemicals is plan-ning to cease production of ammonia and methanol at itsPace Florida plant site, indicating 80% of ammonia andnitrogen feedstock costs are tied to natural gas prices. —Data per Natural Gas Week report on December 30, 2002.

prices (relative to its product sales prices) that tend tocause industrial customers to consider relocating fromthe United States to lower-cost supply regions else-where in the world.

Electric Power Generation

Natural gas has become a fuel of choice for newpower generation because it optimizes installed costand air emissions performance. Natural gas-fired gen-eration is currently capturing almost 100% of newpower capacity. Natural gas-fired combustion turbinescan be installed more quickly, and have a lower up-front capital cost but higher variable cost (primarilyfuel) relative to other technologies such as coal plants,and produce significantly lower CO2 emissions thancoal. The economics of natural gas-fired power gener-ation, however, depend on future natural gas prices. Asgas price and price volatility increases, the risks inmajor investments in gas-fired capacity increase rela-tive to other fuels. Coal, for example, is expected toenjoy more stable fuel costs.

Relatively few new gas-fired power plants have dual-fuel capability, due in part to air emissions permittingconstraints. Since the new gas-fired generation ismore efficient than older plants, some of these less-efficient plants have been shut down. The older steamplants had liquid fuel alternatives (low-sulfur fuel oiland distillate), therefore the overall switching capabili-ty in the system has been reduced. This tends todecrease gas demand elasticity and increase pricevolatility.

Volatility in electricity price has the same impact asnatural gas price volatility. Investors in potential pow-erplants must factor this risk into their “hurdle rate” 4

and adjust their investment decisions accordingly. Inaddition, volatility in gas prices – up or down – createsuncertainty in the planning process for both regulatedutilities and merchant power companies.

Natural Gas Producers

Energy price volatility presents a number of signifi-cant challenges to the natural gas producers. Naturalgas price volatility creates uncertainty around thefuture revenue of exploration or development projects.The primary risk to producers is the longer-termmovement of gas prices and potential “boom-bust”

investment cycles, rather than seasonal weather pat-terns or seasonal pricing variations.

These longer-term price risks for the producer andinvestors are incorporated into the effective financial“hurdle rate” for gas exploration and production proj-ects. Thus, a typical gas producer will invest in newexploration and production projects only when theproducer’s expectation of the gas price rises to a levelhigh enough to make the chances of reaching the tar-get financial criteria acceptable. However, no investor’sforecast is perfect, and the possibility of boom-bust gasprice cycles remains.

While all energy prices will fluctuate, the impact isparticularly significant to independent producers thatdo not have diversified sources of internally generatedfunds. A major investment decision taken in anticipa-tion of future higher demand and higher prices canresult in severe financial distress if the timing turns outto be incorrect.

Conclusions

� Price volatility is a natural dynamic in commoditymarkets where supply and demand vary. Gas pricevolatility has increased since deregulation. Theoverall tighter supply and demand balance and rela-tive inelasticity of demand in the winter is the pri-mary factor driving current volatility.

� Price levels provide consumers and suppliers withappropriate signals, and therefore cause rationalactions. High volatility tends to increase uncertain-ty and decrease market efficiency (increased capitalcosts).

� Consumers and producers have a broad range ofphysical and financial tools to mitigate the effects ofprice volatility if they so choose. Many of the toolscome at a cost. Use of financial tools may or may notreduce the cost or value of the natural gas product.

� Government policies should:

– Promote free-market solutions to market issues.

– Support transparency in market transactions.

– Adopt emission regulations that promote cus-tomer alternate fuel options and switchability(particularly for new powerplant installations).

– Provide safeguards against noncompetitivebehavior and unfair market manipulation.

– Foster timely and accurate information regardingsupply, demand, and storage.


4 The “hurdle rate” is the minimum acceptable expectedreturn needed for a project to proceed.

CHAPTER 8 - CAPITAL REQUIREMENTS 295

The NPC outlook incorporates all the expansionand efficiency projects for exploration, produc-tion, and infrastructure that are expected to

provide acceptable financial returns based on free-market price signals. The investment criteria take intoaccount today’s low interest rate environment and WallStreet analysts’ expectations of required returns oncapital employed for various segments of the energymarket. The cumulative amount of gas-related capitalexpenditures for the Reactive Path and BalancedFuture scenarios are shown in Table 8-1. For the 2003-2025 period, total North American capital expendi-tures are $1.55 trillion in the Reactive Path scenarioversus $1.45 trillion in the Balanced Future scenario.

The wide variety of capital spending envisioned inthis outlook provides opportunity for a wide range ofcompanies including small, private companies andlarge multinationals. Although there have been recent,notable bankruptcies and credit rating downgrades forcompanies linked to energy trading and merchantpower activities, there is more than sufficient capitalavailability, liquidity, and participation from creditwor-thy companies to complete the projects with acceptableeconomic returns. It should be understood that the gasmarket is likely to experience periods of high/lowprices, and there will be various degrees of risk andreturn inherent in these different types of investments.In markets where there is sufficient investor or con-sumer interest, there are financial mechanisms availableto mitigate a significant portion of price risk.

Capital Investment

Financial Assumptions, Methodology

Long-term Wall Street projections currently sug-gest that Corporate America could grow earnings 6%-

7% annually, have an approximate 40% dividend pay-out ratio, and realize average returns on equity of13.0%-13.5%, with ongoing 50-50 debt-equity capitalstructures. In the opinion of Wall Street analysts, thenatural gas industry needs to maintain these levels inorder to remain competitive in the capital markets.We have structured our modeling assumptionsaccordingly.

A key set of assumptions to model this behavior isthe expected cost of debt, the desired return onequity of the various market participants, and theirdesired debt-to-equity financing ratios. Theseassumed profitability levels are generally lower thanin past years, reflecting in large part the lower infla-tion outlook.

Exploration and Production

Figure 8-1 shows that North American explorationand production capital expenditures are higher thanthey were in the past decade. The average annualexpenditure for the forecast period (2003 to 2025) isapproximately $15 billion higher than in the 1990s. Inthe early 1990s, the upstream sector was faced withlower gas prices resulting from a surplus of productivecapacity. Increased gas demand over the past severalyears has used up this surplus, thus increasing prices.As a result, the level of exploration and productionexpenditures has increased, as shown in Figure 8-1, andis projected to stay in the $50-$60 billion per year range.

Natural Gas Infrastructure

Figure 8-2 shows that capital expenditures forinfrastructure are expected to be consistent with thepast several years. However, there will be a markedincrease in certain years as important new supplyareas in the Rockies and Arctic are brought on line.

CAPITAL REQUIREMENTSCHAPTER 8

Over time, a higher percentage of pipeline expendi-tures will be devoted to maintenance and enhance-ment of existing infrastructure in order to maintainsystem integrity.

As shown in Figure 8-3, the average of projectedstorage capital expenditures will be consistent with his-torical averages. The projected growth in gas firedpower generation has resulted in greater need for stor-age capacity to meet peak power needs during periods

of high seasonal demand. From 2003 to 2025, workinggas storage increases by 700 BCF, or approximately16%.

As shown in Figure 8-4, capital expenditures forLNG regasification in the United States increase signif-icantly in both the Reactive Path and Balanced Futurescenarios. LNG imports grow to 15 BCF/D in theBalanced Future scenario versus almost 12.5 BCF/D inthe Reactive Path scenario.



United States Canada North America†

Supply 1,010,214 314,915 1,325,129

Gathering 15,999 5,888 21,886

Interstate & Intrastate Pipelines 48,872 14,788 63,660

LDC 110,887 10,759 121,646

Storage 8,441 1,045 9,486

LNG* 5,058 5,058

Total CapitalExpenditures 1,199,471 347,395 1,546,866


United States Canada North America†

Supply 991,442 239,034 1,230,477

Gathering 14,450 4,597 19,047

Interstate & Intrastate Pipelines 50,632 13,929 64,561

LDC 110,887 10,759 121,646

Storage 8,441 1,045 9,486

LNG* 6,558 6,558

Total Capital Expenditures 1,182,410 269,364 1,451,775

*LNG figures reflect only U.S. regasification costs.†Does not include Mexico.

Table 8-1. Total Capital Expenditures from 2003 to 2025(Millions of 2002 Dollars)


0

10

20

30

40

50

60

70

80

1990 1995 2000 2005 2010 2015 2020 2025

YEAR

BIL

LIO

NS

OF

2002 D

OLLA

RS

0

10

20

BIL

LIO

NS

OF

2002 D

OLLA

RS STORAGE

PIPELINES

LDC

1990 1995 2000 2005 2010 2015 2020 2025

YEAR

Note: This figure is a revision of Figure 53 in Volume I of this report.

Figure 8-1. North American Upstream Expenditures – Balanced Future Scenario

Figure 8-2. North American Infrastructure Expenditures – Balanced Future Scenario


0

100

200

300

400

500

600

700

800

900M

ILLIO

NS

OF

2002 D

OLLA

RS


1990 1995 2000 2005

YEAR

2010 2015 2020 2025

Figure 8-3. North American Storage Capital Expenditures

0

1

2

3

4

5

6

7

2003 2008 2013

YEAR

2018 2023

BIL

LIO

NS

OF

2002 D

OLLA

RS

BALANCED FUTURE

REACTIVE PATH

Figure 8-4. Cumulative U.S. LNG Regasification Expenditures

CHAPTER 9 - SENSITIVITY ANALYSIS 299

The NPC considered the many factors that driveoverall supply and demand and created consis-tent input data sets for the forecasting models.

The NPC also evaluated alternative assumptions to testhow the natural gas market might` evolve under differ-ent conditions. Those alternative assumptionsaddressed the general economic environment, govern-ment policies affecting supply and demand, naturalresource size, upstream technological trends, weather,end-use efficiency improvements, and other factors.

Figure 9-1 is a highly simplified schematic showingsupply and demand versus price for North America.The blue circle in the middle represents an initial pro-jection of natural gas prices and volumes for a givenyear or period. That “solution point” is the combinedresult of many assumptions that affect natural gassupply and other assumptions that affect the demandfor natural gas. Alternative views of those factorswould lead to a different solution point. There are, ofcourse, thousands of potential solution points that

SENSITIVITY ANALYSISCHAPTER 9

SUPPLY

DEMAND

VOLUME

PR

ICE

INCREASED GAS DEMAND

AND

REDUCED GAS SUPPLY

INCREASED GAS DEMAND

AND

INCREASED GAS SUPPLY

REDUCED GAS DEMAND

AND

REDUCED GAS SUPPLY

REDUCED GAS DEMAND AND

INCREASED GAS SUPPLY

REACTIVE PATH BALANCED FUTURE STATUS QUO

Figure 9-1. Price-Volume Schematic

could be created with different combinations ofassumptions.

As is shown in Figure 9-1, those alternative pointscan be represented as falling around the initial solutionpoint depending on whether and by how much varioussupply/demand assumptions are changed. If the bluecircle is taken to represent the Reactive Path scenario,then the orange circle can represent the BalancedFuture scenario in which additional supplies and moreflexible uses of fuels combine to reduce natural gasprices. Likewise the gold circle can represent the StatusQuo scenario in which supply is less available, thusincreasing prices, even though natural gas demand isreduced to a certain degree.

Examples of the model “factors” that were adjustedin creating NPC scenarios and sensitivities are shownin Table 9-1. Among the factors that increase naturalgas demand are gas-favoring environmental regula-tions, higher economic growth, less end-use fuel effi-ciency, less fuel flexibility, higher oil prices, and moreextreme weather. Those same factors can be set in theopposite direction to create cases in which natural gasdemand could be lower. For example, slower economicgrowth would tend to reduce overall energy needs and

the demand for natural gas. The key gas supply driversshown in Table 9-1 include integration of Arctic gas,LNG imports, Rockies and offshore land access,upstream technological advances, the underlyingresource endowment, and tax and royalty policies.Each of these factors can be set in a manner that tendsto increase natural gas availability or reduce it.

In the preparation of this report, approximately 100integrated supply/demand model runs were madeusing the EEA models. There were several sets of pre-liminary or “strawman” cases that were prepared as thestudy group members investigated various issues andsettled on reasonable and consistent assumptions. Outof this process came the over thirty “final” cases thatare presented in the report and in supporting docu-mentation. This collection of case analyses servesmany purposes, including:

� Examining the effects of government policies onsupply and demand

� Quantifying the effects of “fuel flexibility” policieson consumer costs and other outcomes

� Quantifying the effects of land access polices onconsumer costs and other outcomes


Table 9-1. Sensitivity Factors

GAS DEMAND

Increased Demand

� Gas-Favoring Environmental Regulations

� Higher Economic Growth

� Less End-Use Efficiency

� Less Fuel Flexibility

� Higher Oil Prices

� Colder Winters/Hotter Summers

Reduced Demand

� Gas-Neutral Environmental Regulations

� Lower Economic Growth

� Greater End-Use Efficiency

� Enhanced Fuel Diversity

� Lower Oil Prices

� Warmer Winters/Cooler Summers

GAS SUPPLY

Reduced Supply

� No/Less Arctic Gas

� Reduced LNG Imports

� Reduced Access

� Lower Technology Evolution

� Lower Resource Potential

Increased Supply

� Increased Arctic Gas

� Greater LNG Imports

� Increased Access

� Greater Technology Evolution

� Increased Resource Potential

� Higher Oil Prices

� Illustrating the growing importance of large-scalepipeline and LNG projects on natural gas markets

� Exploring the critical linkages between economicactivity and the demand for electricity and naturalgas

� Measuring the impacts of end-use efficiencies onelectricity demand and natural gas markets

� Illustrating the uncertainty inherent in factors suchas resource endowment and upstream technologicaladvances

� Exploring the effects of weather variability ondemand swings and price volatility.

The major elements that went into the various sce-narios and sensitivities presented in this report are dis-cussed below by subject area starting with general eco-nomic factors, going then to demand drivers, supplydrivers, pipeline and other infrastructure assumptions,and finally liquefied natural gas investments.

Factors Related to the Economic Environment

The general economic environment is represented inEEA’s Gas Market Data and Forecasting Systemthrough inputs such GDP growth rates and industrialproduction growth rates in the United States andCanada and prevailing world oil prices. Alternativescenarios and sensitivities were created by varying gen-eral (GDP) growth rates and additional sensitivitieswere made in which just the industrial sector’s growthrates were adjusted. There was also one sensitivitymade with higher oil prices set at $28.00 per barrel forWest Texas Intermediate crude oil.

Demand-Related Factors

Many of the demand drivers adjusted by the NPC tocreate alternative scenarios related to electricitydemand and power generation. Adjustments were alsomade to inputs for the industrial, residential, and com-mercial sectors.

Electricity Sales as a Function ofEconomic Growth

For the purpose of the study, electricity sales weredefined as on-grid deliveries of electricity to retail cus-tomers, and do not include electricity consumed at thesource of generation or direct sales. The projectionsfor electricity sales were base on an assumed elasticity

relationship between GDP and electricity sales. Thestarting elasticity for all cases was 72%; that is, forevery one percent increase in GDP, electricity salesincrease by 0.72%. For the base projection, it wasassumed that the elasticity will gradually decrease dueto increasing energy efficiency. By 2025, the GDP-to-electricity sales elasticity decreases to 62% in theReactive Path projection. In the Balanced Future sce-nario, it decreases at a slightly faster rate to 60% by2025. In the Low Electricity Elasticity case, elasticitywas decreased to 52%, and in the High ElectricityElasticity case it was held constant at 72%.

Industrial/Power Generation Fuel Switching

The Reactive Path assumption for the industrial sec-tor is that there is limited expansion of oil/gas switch-ing capability. In the Balanced Future scenario, there isan increase in the switching capability of industrialboilers from the 2003 level of about 5% to 28%, as wellas a doubling of gas price elasticities for industrialprocess heat by 2025.

The Reactive Path assumption for power generationfuel switching was that the amount of switching fromgas to oil would be limited by environmental con-straints. Fuel switching was severely limited in theNortheast and West Coast, and allowed to expandslightly in other areas. In the Balanced Future scenario,the amount of time that dual-fuel units were allowedto switch to oil was increased by approximately 7%.

Existing Fossil-Fuel Generating Capacity

The Reactive Path assumption was that the majorityof existing fossil capacity would continue to operatethroughout the forecast time frame. It was assumedthat due to economic competition with newly con-structed combined-cycle and combustion turbinecapacity, 21.5 gigawatts of oil/gas steam capacity wouldbe retired between 2003 and 2008. Oil/gas steamcapacity was held constant after 2008. To meet upcom-ing restrictions on mercury emissions, it was assumedthat 21.1 gigawatts of coal capacity would be retiredbetween 2008 and 2010. In the Balanced Future sce-nario, there were no retirements of either coal oroil/gas steam capacity between 2003 and 2010.

New Fossil-Fuel Generating Capacity

In all cases, construction of new capacity was deter-mined by an analysis using busbar cost curves and pro-duction simulations. The busbar cost curves were usedto determine appropriate capacity factor operatingranges for each type of capacity. The costs include fuel


costs, construction costs, financing costs, taxes, opera-tions and maintenance expenses, and all the other costsof owning and operating a power plant. The produc-tion simulations determined if the newly added unitsare operating in those ranges. In the production simu-lation, the newly added units were integrated with theexisting fleet and all units were dispatched to meetload. In the base case, additional restrictions wereapplied as to where certain types of capacity could beconstruction and how much could be constructed peryear. For example, no new coal capacity could be builtin the Northeast or West Coast, and coal capacity addi-tions were limited to 12 gigawatts per year. In theBalanced Future scenario, some of these restrictionswere relaxed to allow a more favorable environment forconstructing coal- and oil-fired capacity.

Nuclear Capacity

The Reactive Path assumptions for existing nuclearcapacity were that all currently operating plants wouldreceive one 20-year extension of their operatinglicense, and that incremental improvements wouldincrease the operating capacity of the existing fleet by atotal of 2% by 2011. In the Balanced Future scenario,the operating capacity of the existing fleet wasincreased by a total of 10% by 2013.

Nuclear capacity was allowed to compete as anoption for new capacity in all cases. However, the longlead-time for planning and construction of nuclearplants made them economically uncompetitive in allbut the Carbon Reduction sensitivity case.

Renewable Capacity

Costs for new renewable capacity were representedby wind turbines, which is currently the most eco-nomic non-hydroelectric renewable energy technol-ogy. In the Reactive Path scenario, renewable genera-tion was assumed to increase at a rate of 11% per year.In the Balanced Future scenario, renewable genera-tion was assumed to increase at approximately 15%per year.

Industrial Production and Energy Intensity

Alternative cases were created in which the efficiencyimprovement by industrial sectors was varied. In onecase, the growth rate of energy-intensive industries wasincreased and in a second case it was lowered. Thesecases were intended to illustrate how energy use couldvary in the industrial sectors based on which industrygroups were expanding.

Residential/Commercial Energy Efficiency

The Balanced Future scenario and several othercases were run with assumptions of greater energyefficiency in the residential and commercial use ofnatural gas.

Domestic Supply-Related Factors

The main factors that were adjusted to create thealternative cases were resource access, resource basesize, and upstream technological advances.

Resource Access

Access to indigenous resources is essential for reach-ing North America’s full supply potential. New discov-eries in mature North American basins represent thelargest component of the future supply outlook,including potential contributions from imports andAlaska. However, the trend towards increasing leasingand regulatory land restrictions in the RockyMountain region and the Outer Continental Shelf(OCS) is occurring in precisely the areas that hold sig-nificant potential for natural gas production. In theRocky Mountain areas, previous studies have evaluatedthe effects of federal leasing stipulations.

Leasing moratoria in the Eastern Gulf of Mexico,Atlantic Coast, and the Pacific Coast currently prohibitaccess to these areas of the OCS.

Resources are classified in EEA’s HydrocarbonSupply Model under three categories:

� Accessible under standard lease terms

� Accessible but higher cost due to restrictions

� Inaccessible.

Seven resource access cases were developed. Fourcases evaluated increased access and three cases lookedat decreased access. Rockies and offshore access sce-narios were evaluated separately, and then together, inboth increased and decreased access scenarios.

Resource Base Sensitivities

The quantity of available undiscovered oil and gas isone of the key factors in modeling future activity andproduction. Because there is inherent uncertainty inassessing undiscovered resources, it is necessary toevaluate this range of uncertainty through model sen-sitivities. Two resource base sensitivity runs were eval-uated: High Resource Base sensitivity (also called P10)


and Low Resource Base sensitivity (also called P90).1

The Supply Subgroup developed resource base specifi-cations for these model runs and examined a range ofuncertainty from minus 30% to plus 35% of the refer-ence resource. This range was determined throughindustry discussions and comparisons with other pub-lished assessments.

A uniform change to all resources in the model wasused for both cases. For the Low Resource Base case, allundeveloped and undiscovered resources were reducedby 30% relative to the Reactive Path resource base. Forthe High Resource Base case, all undeveloped andundiscovered resources were increased by 35%. Thesemultipliers were applied to Canada as well as theUnited States. This approach differs somewhat fromthat used in the 1999 study, in which only the resourcesin specific regions were varied to create the low andhigh resource base cases.

Upstream Technological Advancements

Technology is a critical driver for the growth of thegas industry in North America. This is dictated by thenature and complexity of the undiscovered resourcebase, which is generally characterized by deeperdrilling, deepwater, and nonconventional reservoirs.Continued development of improved exploration anddevelopment technologies and cost reductions fordrilling and platform construction will be critical toimproving the economics of future gas supply.

In EEA’s Hydrocarbon Supply Model, supply tech-nologies are represented in three categories:

� Improved exploratory success rates

� Cost reductions in platform, drilling, and other costs

� Improved recovery per well.

These factors are input into the model by region andtype of gas and represent several dozen actual parame-ters. The Reactive Path scenario includes assumptions

for all upstream technology parameters based uponanalysis of past industry trends. Three technology sen-sitivities were developed versus the base technologyassumptions: High Supply Technology, Low SupplyTechnology, and Static Supply Technology (noimprovement).

LNG Assumptions

It was determined early in the study process thatLNG import projects would be highly influenced byU.S. regulatory processes, the policies of the sourcecountries, and long-term investment decisions by pri-vate industry. Therefore LNG volume and timingassumptions were developed collectively and inputexogenously to the model.

The Reactive Path assumption for LNG imports hasseven new terminals added and imports reach 4.6 TCFper year (12.5 BCF/D) by 2025. After existing importfacilities are utilized to full capacity and thenexpanded, new import facilities are built in the Gulf ofMexico and the Northeast.

The High LNG Imports case adds nine new importterminals and imports reach a total of 5.4 TCF per year(14.8 BCF/D) by 2025. The High LNG Imports caseadds facilities in south Florida, offshore Texas, andnorthern California in addition to those already builtin Northeast and Gulf of Mexico in the base scenario.The Balanced Future scenario uses the High LNGImports assumption of 5.4 TCF per year, as do manycases that use the Balanced Future scenario as a start-ing point.

The Low LNG Imports case adds only two newimport terminals for a total of 2.4 TCF per year (6.5BCF/D) by 2025. All existing LNG facilities still arefully utilized but there are no new import facilitiesbuilt in the Northeast and only one new Gulf ofMexico import facility is added.

The LNG Stress Test case for Balanced Future addsan additional 547 BCF (1.5 BCF/D) to the High LNGlevel for a total of 6.0 TCF LNG imports by 2025. Thisadditional LNG is added in the Gulf of Mexico, wheresufficient pipeline capacity exists to move it to market.

Assumptions for Pipeline Construction

The cost of expanding pipelines throughout theUnited States and Canada are inputs into the model.


1 Resource base estimates are often represented on a cumu-lative probability scale that goes from 100 percent chanceof X TCF or more resource to a near-zero percent chanceof Y TCF or more resource. The resource read at the point“P10” means that there is a 10% probability that theactual resource endowment is that much or more. Theresource read at point “P90” means that there is 90%chance that the actual resource base is that big or less. Theresource base read at “P50” indicates the point at whichthere is an equal likelihood that the real resource base ishigher as it is lower.

These data are used along with “rules” by which thedecision to expand capacity will be made as the modelcase is developed. The rules specify what basis defer-ential must be realized before a pipeline project will bestarted and how many years it will take to permit andimplement the project. These costs and rules were notvaried among the cases except for the Accelerated andDelayed Rockies Pipeline Development cases, in whichpipelines out of the Rockies were, respectively, acceler-ated or delayed.

A natural gas pipeline from the Alaska North Slopeto serve lower-48 demand was not modeled using deci-sion rules. Instead, its timing and size were set byArctic Subgroup guidance. Most cases were runassuming a 2013 in-service date for the Alaska pipelineand a throughput (after an initial ramp-up period) of4 BCF/D. There was also a case in which the Alaskapipeline was delayed for five years and two cases inwhich the Alaska natural gas pipeline was not built atall. One case was also run assuming that the Alaskanatural gas pipeline is expanded by 1 BCF/D to a totalof 5 BCF/D in 2020.

Construction of the NPC Cases

Cases were constructed by selecting and combiningthe factors discussed above. Table 9-2 shows in a sum-mary fashion how the 32 main cases were constructed.Each row on the table represents a separate case andeach column represents a category of assumptions thatcould be changed among the cases. The assumptionsare shown in the table as changes relative to theReactive Path scenario. Therefore, when assumptionsare the same as the Reactive Path scenario, that will beindicated as a blank box in the figure.

The Balanced Future scenario is presented in thisreport as the primary alternative to the Reactive Pathscenario. As is shown in the second row of Table 9-2,the Balanced Future scenario contains differentassumptions:

� Increased access to Rockies and offshore lands for oiland gas development

� A more favorable regulatory environment for LNGand greater amounts of investment in LNG for U.S.markets

� More efficient use of natural gas in residential andcommercial sectors

� More efficient use of electricity in all sectors

� Greater fuel flexibility in the industrial and powergeneration sectors

� A more favorable regulatory environment to pre-serve existing oil- and coal-fired steam powerplantsand to build new ones

� Greater increases in the capacity of existing nuclearpower plants

� More renewable capacity for power generation.

These assumptions are indicated by the gray boxes inTable 9-2. Some of these same assumptions were usedin other cases. For example, the case appearing in row24 is the Fuel Flexibility case. It contains all of the sameassumptions as the Balanced Future scenario, with theexception of land access, which is kept the same as inthe Reactive Path scenario. Row 25 contains the LNGStress Test case, which has the same assumptions as theBalanced Future scenario with the exceptions of higherLNG imports.

Most of the sensitivity cases appearing in this reportwere run off of the Reactive Path scenario. For exam-ple, rows 3 and 4 of Table 9-2 show the economicgrowth sensitivities, which contain the same assump-tions as the Reactive Path scenario for everything butthe economic environment. The same pattern ofchange in only one column appears for:

� Two cases related to the income elasticity of electric-ity demand (rows 5 and 6)

� High and low LNG cases (rows 7 and 8)

� Seven land access cases (rows 9 to 15)

� Three Alaska natural gas pipeline cases (rows 16 to 18)

� Three upstream technology sensitivities (rows 19 to 21)

� Two industrial production cases (rows 22 and 23)

� Accelerated and delayed Rockies pipeline develop-ment cases (rows 26 and 27)

� High and low resource base sensitivities (rows 28and 29).

Two other scenarios were also created and are shownin rows 30 and 31 of Table 9-2. The Carbon Reductioncase contained limitations of carbon emissions frompowerplants that were met by reduced use of coal andmore use of natural gas and nuclear power. The StatusQuo scenario represented what might happen if someof the regulatory hurdles to additional gas supplies andalternative fuel use were not eased even to the degreeanticipated in the Reactive Path scenario. This was rep-



resented in the EEA models as the combined effect ofdecreased access to land, no Alaska natural gaspipeline, low LNG imports, and delayed constructionof coal powerplants. Because of the high natural gasand electricity prices produced in this case, expectedfeedback effects of reduced electricity consumptionand more renewables use were also made part of theStatus Quo scenario.

In addition to the cases represented in Table 9-2,EEA ran a large number of weather sensitivity casesbased on the Reactive Path and Balanced Future sce-narios. This was done by changing the heating degreedays and cooling degree days by region and month invarious future years. The purpose of these weathercases was to measure the degree of stress on the natu-ral gas transmission and storage infrastructure causedby extreme weather and how natural gas price levelsand volatility are affected.

Summary Results of the NPC Cases

Some of the results of the NPC scenarios and sensi-tivities are presented elsewhere in this IntegratedReport and more details are provided in the Task

Group Reports. This section provides a broadoverview of the major NPC cases.

Figure 9-2 presents some of the supply-side sensitiv-ity cases that were run based on the Reactive Path sce-nario. It shows the Henry Hub price differences (Y-axis) and the U.S. plus Canadian gas supply/demanddifferences (X-axis) averaged over the 15-year periodof 2011 to 2025. The black circle at the center is theReactive Path scenario and, by definition, is at zero onboth axes. Results to the upper left of that point arecases for which available suppliers are reduced. Thereduced supplies tend to drop the overallsupply/demand balance point and increase prices.Cases that appear to the right and below the ReactivePath are those for which supplies are increased, leadingto lower natural gas prices. The two most extremepoints in Figure 9-2 are defined by the two resourcebase sensitivities: the Low Resource Base sensitivityproduced the greatest price increases and largestdemand/supply drops, while the greatest price reduc-tions and largest demand/supply gains resulted fromthe High Resource Base sensitivity.

The combined results of the supply-side sensitivitiesto the Reactive Path scenario trace out a “demand

-3,000 -2,000 -1,000 0 1,000 2,000 3,000 4,000-4,000

HE

NR

Y H

UB

DIF

FE

RE

NC

E

(DO

LLA

RS

PE

R M

ILLIO

N B

TU

)

-3.00

-2.00

-1.00

0.00

1.00

2.00

$3.00

U.S. & CANADIAN SUPPLY/DEMAND DIFFERENCE (BILLION CUBIC FEET PER YEAR)

“DEMAND CURVE” TRACED BY RESULTS




INCREASED ACCESS (COMBINED)

LOW LNG IMPORTS

REACTIVE PATH

HIGH LNG IMPORTS


DECREASED ACCESS (COMBINED)

NO ALASKA PIPELINE

LOW SUPPLY TECHNOLOGY

Figure 9-2. Selected Supply Sensitivities – United States and Canada (2011-2025 Averages)

CH

APTER 9 - SEN

SITIVITY

AN

ALYSIS

306

Table 9-2. Summary of Cases (Shown as Changes from Reactive Path Assumptions)

Economic Land Access Gas Supply Gas LNG Residential/Case Environment for Gas Technology & Transmission Investment Commercial

Production Resource Base Infrastructure Efficiency

1 Reactive Path

2 Balanced Future Increased Access Higher LNG Greater Efficiency

3 Low Economic Growth Slower GDP Growth

4 High Economic Growth Faster GDP Growth

5 Low Electricity Elasticity

6 High Electricity Elasticity

7 High LNG Imports Higher LNG

8 Low LNG Imports Low LNG

9 Increased Access (Combined) Increased Access (Rockies & Offshore)

10 Increased Offshore Access Increased Offshore Access

11 Increased Rockies Access (Gradual) Gradual Rockies Access

12 Increased Rockies Access Full Effect Rockies Access(Full Effect)

13 Decreased Rockies Access Decreased Rockies Access

14 Decreased Offshore Access Decreased Offshore Access

15 Decreased Access (Combined) Decreased Access (Rockies & Offshore)

16 No Alaska Pipeline No Alaska Pipeline Built

17 Delayed Alaska Pipeline Alaska Pipeline Delayed 5 Years

18 Expanded Alaska Pipeline Expanded Alaska Pipeline in 2020

19 Low Supply Technology Low Technological Advances

20 High Supply Technology High Technological Advances

21 Static Supply Technology No Technological Advances

22 Low Industrial Production Lower Industrial Production Growth

23 High Industrial Production Higher IndustrialProduction Growth

24 Fuel Flexibility Greater Efficiency

25 LNG Stress Test – Balanced Future Increased Access Highest LNG Greater Efficiency

26 Accelerated Rockies Quick Rockies Build LogicPipeline Development

27 Delayed Rockies Slow Rockies Build LogicPipeline Development

28 High Resource Base P10 High Resource Base

29 Low Resource Base P90 Low Resource Base

30 Carbon Reduction

31 Status Quo Decreased Rockies Access No Alaska Pipeline Built Low LNG

32 WTI $28 Oil Price

CH

APTER 9 - SEN

SITIVITY

AN

ALYSIS

307 Table 9-2. Summary of Cases (Continued)

Income Elasticity Industrial & Fossil Nuclear Renewable Other Case of Electricity Power Generation Generation Capacity Capacity and Items

Sales Fuel Switching Generation

1 Reactive Path

2 Balanced Future Increased Efficiency Yields Lower Income Elasticity Great Flexibility More Favorable Increased Uprates More Growthto Coal & Oil of Existing Units in Capacity

3 Low Economic Growth

4 High Economic Growth

5 Low Electricity Elasticity Increased Efficiency Yields Lower Income Elasticity

6 High Electricity Elasticity Less Efficiency Yields Higher Income Elasticity

7 High LNG Imports

8 Low LNG Imports

9 Increased Access (Combined)

10 Increased Offshore Access

11 Increased Rockies Access (Gradual)

12 Increased Rockies Access (Full Effect)

13 Decreased Rockies Access

14 Decreased Offshore Access

15 Decreased Access (Combined)

16 No Alaska Pipeline

17 Delayed Alaska Pipeline

18 Expanded Alaska Pipeline

19 Low Supply Technology

20 High Supply Technology

21 Static Supply Technology

22 Low Industrial Production

23 High Industrial Production

24 Fuel Flexibility Increased Efficiency Yields Lower Income Elasticity Greater Flexibility More Favorable Increased Uprates More Growthto Coal & Oil of Existing Units in Capacity

25 LNG Stress Test – Increased Efficiency Yields Lower Income Elasticity Greater Flexibility More Favorable Increased Uprates More GrowthBalanced Future to Coal & Oil of Existing Units in Capacity

26 Accelerated RockiesPipeline Development

27 Delayed RockiesPipeline Development

28 High Resource Base P10

29 Low Resource Base P90

30 Carbon Reduction High Retirement Rates New Nuclear Units More Growth Carbonfor Steam; No New Added After 2012 in Capacity Emissions

Conventional Coal Plants Constrained

31 Status Quo Increased Efficiency Yields Lower Income Elasticity Delayed Construction of More GrowthNew Coal Capacity in Capacity

32 WTI $28 Oil Price Oil Price is$28 for WTI

curve” of sorts, which shows how much natural gasdemand would exist in the United States and Canada atvarious price levels. This is shown as a black line inFigure 9-2. The solution point on that curve that isreached by any case depends on what supply-sideassumptions are made, since price will adjust in a freemarket to make demand equal supply. However, the“demand curve” that is traced by the cases is not per-fectly smooth because there are time-dependant per-turbations that result from the various assumptionsand regional price variability that cannot be capturedin a simple two-dimensional plot.

Figure 9-3 presents the same type of informationfor the demand-side sensitivities. It is possible to traceout a “supply curve” off of the Reactive Path demand-side assumptions. The black line in Figure 9-3 showsthe points that result from keeping the supply assump-tions the same, but varying demand-side factors. Forexample, the solution point furthest down on thecurve is the Fuel Flexibility case. By allowing otherfuels to compete more effectively with natural gas, theFuel Flexibility case achieves the largest gas pricereductions given the Reactive Path supply-sideassumptions. The point furthest up on the curve is theHigh Economic Growth case. This adds the greatest

amount of natural gas demand and reaches the high-est price increase given Reactive Path supply-sideassumptions.

All 32 of the major NPC cases are plotted in Figure9-4 using the same conventions as Figure 9-2 and 9-3.The implied demand and supply curves from theReactive Path scenario are shown as blue lines labeledD1 and S1. The solution point for the BalancedFuture scenario is Point 2 at the intersection of thetwo green lines that represent supply and demandcurves for that scenario. Likewise, the solution pointfor the Status Quo scenario is Point 31 at the inter-section of the two red lines. Because only one supply-side sensitivity was run based on the Balanced Futurescenario (LNG Stress Test), the demand curve for thatcase cannot be traced very fully. However, one cansurmise from Figure 9-4 that additional supply-enhancing assumptions, such as improved upstreamtechnologies, would drop prices further as the solu-tion point moved further down from Point 2 on theD2 curve.

Figure 9-5 is a graphical representation of theimpact of selected sensitivity cases.



"SUPPLY CURVE" TRACED BY RESULTS

-3,000 -2,000 -1,000 0 1,000 2,000 3,000 4,000

-3.00

-2.00

-1.00

0.00

1.00

2.00

$3.00

LOW ELECTRICITY ELASTICITY


REACTIVE PATH

LOW ECONOMIC GROWTHFUEL FLEXIBILITY

HIGH ELECTRICITY ELASTICITY

LOW INDUSTRIAL PRODUCTION

HIGH INDUSTRIAL GROWTH

-4,000

HE

NR

Y H

UB

DIF

FE

RE

NC

E

(DO

LLA

RS

PE

R M

ILLI

ON

BT

U)

Figure 9-3. Selected Demand Sensitivities – United States and Canada (2011-2025 Averages)



-2.00

-1.00

0.00

1.00

2.00

$3.00

-4,000 -3,000 -2,000 -1,000 0 1,000 2,000 3,000 4,000

-3.00

D2 D3 D1

S2

S1

S3

REACTIVE PATH

2 BALANCED FUTURE

3 LOW ECONOMIC GROWTH


5 LOW ELECTRICITY ELASTICITY


HIGH LNG IMPORTS

8 LOW LNG IMPORTS

INCREASED ACCESS (COMBINED)

INCREASED OFFSHORE ACCESS

INCREASED ROCKIES ACCESS (GRADUAL)

INCREASED ROCKIES ACCESS (FULL EFFECT)

DECREASED ROCKIES ACCESS

DECREASED OFFSHORE ACCESS


DELAYED ALASKA PIPELINE

EXPANDED ALASKA PIPELINE

LOW SUPPLY TECHNOLOGY



LOW INDUSTRIAL PRODUCTION

HIGH INDUSTRIAL PRODUCTION

FUEL FLEXIBILITY

LNG STRESS TEST – BALANCED FUTURE

ACCELERATED ROCKIES PIPELINE DEVELOPMENT

DELAYED ROCKIES PIPELINE DEVELOPMENT



CARBON REDUCTION

STATUS QUO

WTI $28 OIL PRICE

21

29

198

16

1

7

20

28

122

24

3

5

64

23

129

31

252

121011

1713

30

14 18

26 27

1

15

7

22

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

32

NO ALASKA PIPELINE16

HE

NR

Y H

UB

DIF

FE

RE

NC

E (

DO

LLA

RS

PE

R M

ILLIO

N B

TU

)

Figure 9-4. Selected Cases – United States and Canada (2011-2025 Averages)

Further Information on Cases

This chapter was intended only to introduce the fullrange of cases explored in this study. Please see the

individual Task Group Reports for additional details onthese cases including regional production and demandtrends, pipeline construction and flow patterns, chang-ing basis differentials, and consumer cost impacts.


Note: Values shown are averages for the 2011 to 2025 period.

-2.00 0.0 2.00 4.00 -4,000 -2,000 0 2,000

CHANGE IN VOLUMES(BILLION CUBIC FEET PER YEAR)

VS. REACTIVE PATH

CHANGE IN PRICE(2002 DOLLARS)

VS. REACTIVE PATH


FUEL FLEXIBILITY


LOW ECONOMIC GROWTH

INCREASED ACCESS

HIGH LNG IMPORTS



HIGH ECONOMIC ELASTICITY

WTI $28 OIL PRICE

NO ALASKA PIPELINE

LOW LNG IMPORTS



Figure 9-5. Price and Volume Impacts of Selected Sensitivities

CHAPTER 10 - MODELING METHODOLOGIES 311

This chapter describes the natural gas marketmodeling methodologies used by the NPC. Themodeling framework developed and maintained

by Energy and Environmental Analysis, Inc. (EEA)formed the basis of gas market outlooks in the currentstudy. Additional work was conducted to apply dataand develop underlying assumptions for models byAltos Management Partners. Further information isavailable in the Task Group Reports.

In the course of this study, the NPC developed data-bases related to resource base quantities and develop-ing and operating costs, gas pipeline capacity and rates,and characterizations of gas demand volumes versusprice. It is the intent of the NPC to make these dataavailable to government agencies and other interestedparties. The NPC also will continue working with government agencies, such as the USGS, to determine the feasibility of updating, utilizing, and maintainingthe resource, engineering, and cost data developed bythe NPC.

The EEA Models

Models licensed from Energy and EnvironmentalAnalysis, Inc. (EEA) for this study included theHydrocarbon Supply Model (HSM) and the GasMarket Data and Forecasting System (GMDFS). TheHSM models supply on an annual basis, while theGMDFS simulates monthly market behavior, and themodels are operated in an integrated manner. The pri-mary inputs from the HSM into GMDFS are gas deliv-erability data, and the primary data going back fromGMDFS to the HSM are gas production levels andprices.

The EEA models solve using a “market simulation”methodology, meaning the decisions are simulated on

a period-by-period basis using foresight assumptionsset by the user. As such, the EEA models produceresults that match history and provide insight into thefuture natural gas market.

Earlier versions of the HSM were used in both the1992 and 1999 NPC studies. The GMDFS also wasused in the 1999 study. Several changes were made tothose models since the 1999 study, both independentlyby EEA and in consultation with the NPC. The majorchanges for the 2003 NPC study, in contrast to earlierstudies, include:

• New, more disaggregated regions in the supplymodel

• Use of play-level resource description for new fieldresources

• Revised field appreciation (growth) data andresource estimate

• Revised data and more flexible methodology fortreating land access in the United States and Canada

• Updated performance parameters for well recover-ies, find rates, etc.

• Revised decline rates for existing and new reserves

• New upstream factor costs for wells, platforms, etc.

• Modeling of ethane rejection in gas processing as asupplement to natural gas supplies

• New gas pipeline corridors to accommodate newregions

MODELING METHODOLOGIESCHAPTER 10

• Updated existing pipeline capacities

• New costs for pipeline construction

• Expanded industrial gas demand model

• New cost and performance factors for multiplepower generation technologies

• Extended forecast horizon to the year 2030

• Enhanced outputs in Microsoft Excel and Access format.

Gas Market Data and Forecasting System

The Gas Market Data and Forecasting System hasthe capability to track and analyze the performance ofNorth American natural gas markets on a monthlybasis. At the heart of the system is a comprehensive gastransmission network that solves for natural gas supplyand demand in the United States, Canada, and north-ern Mexico. Specifically, the model solves for monthlynatural gas production and demand, storage injectionsand withdrawals, pipeline flows, natural gas prices,location, and seasonal basis for a very detailed naturalgas pipeline network comprised of over 100 nodes (ormarket hubs). Results are described at the node level.In the power generation sector, the GMDFS solves formonthly U.S. electricity demand, power generation bytype of fuel, and fuel use.

The GMDFS model simulates monthly gas marketperformance to 2030, considering the impact of a widerange of variables. Inputs include: growth rates foreconomic drivers, such as GDP and industrial produc-tion; projected prices of crude oil and alternative fuels;power generating capacity by technology and fuel sup-ply; weather and hydrological conditions; pipeline andstorage expansions; LNG imports and exports; andother annual and seasonal factors.

Overall, the model solves for monthly natural gasmarket clearing prices by considering the interactionbetween supply and demand relationships at each ofthe model’s nodes. On the supply side, prices are deter-mined by short-term production and storage pricecurves that reflect prices as a function of productionand storage utilization. Prices are also influenced by“pipeline discount” curves, which reflect the change inbasis or the marginal value of gas transmission as afunction of load factor. On the demand side, prices arerepresented by a curve that captures the fuel-switching

behavior of end-users at different price levels. Themodel balances supply and demand at all nodes in themodel at the market clearing prices determined by theshape of the supply and demand curves. EEA main-tains this model by doing significant “backcasting”(calibration) of the model’s curves and relationshipson a monthly basis to make sure that the model reliablyreflects historical gas market behavior, instilling confi-dence in the projected results.

The NPC provided input assumptions for weather,economic growth, and oil prices, among other vari-ables. EEA performs market reconnaissance and keepsthe model up to date with generating capacity, near-term gas supply deliverability, storage and pipelineexpansions, and the impact of regulatory changes ingas transmission.

Since the GMDFS solves on a monthly basis, EEA’sDaily Demand model was used to determine daily gasdemands. The Daily Demand model is an offshoot ofthe GMDFS. It is based on the same nodal structureand demand modules used by the GMDFS, but runsfor each day of a given historical or future year. Theoutput of the Daily Demand model is daily residential,commercial, industrial, and power generation gasdemand at each model node, and daily fossil genera-tion for each of the model’s power dispatch regions.The Daily Demand model was used by the NPC toproject peak-day demands, assess the need for high-deliverability storage, and identify possible pipelineconstraints.

In contrast to the GMDFS, which solves for the fullsupply and demand balance at each node to arrive atnatural gas market clearing prices, the Daily Demandmodel uses the gas prices from a GMDFS model run todetermine daily demand at each node. Once a GMDFSmodel run is complete, the Daily Demand routine isrun using the same inputs as the GMDFS, plus the gasprice outputs from the GMDFS model run. The dailytemperatures used in the Daily Demand model areadjusted to match the total monthly heating and cool-ing degree-day values used as input for the GMDFS.This allowed the NPC to model the demand variabilitywithin each month and distribute the monthly loadover the days of the month.

Hydrocarbon Supply Model

The Hydrocarbon Supply Model is an analyticalframework designed for the simulation, forecasting,

CHAPTER 10 - MODELING METHODOLOGIES312

and analysis of natural gas, crude oil, and natural gasliquids supply and cost trends in the United States andCanada. It is a process-engineering model with adetailed representation of potential gas resources andthe technologies with which those resources can beproven and produced. The degree and timing by whichresources are proven and produced are determined inthe model through discounted cash flow analyses ofalternative investment options and behavioral assump-tions in the form of inertial and cash flow constraintsand the logic for setting producers’ market expecta-tions (i.e., future gas prices).

The model covers the U.S. lower-48, Alaska, andCanada. The lower-48 states are represented in 28onshore regions and 11 offshore regions. Alaska isdivided into seven regions, and Canada is divided intoten regions. All regions are further broken out intosubregions or “intervals.” Each of these “intervals”represents some combination of drilling depths, waterdepth, or geographic areas.

Resources in the Hydrocarbon Supply Model aredivided into three general categories: new fields/newpools, field appreciation (growth), and nonconven-tional gas. For conventional resources in the UnitedStates, there are 220 region/interval categories that aremodeled with over 10,000 prototypical field develop-ment plans. Old-field appreciation is modeled inapproximately 525 categories. Nonconventional gas isrepresented by 261 “cells” that, for the United States,correspond to the “continuous plays” of USGS resourceassessments.

The Hydrocarbon Supply Model has a large numberof factors that can be changed to produce alternativecases. These include:

Resource Base

• Undiscovered New Field Resources

• Initial Old Field Growth

• Unconventional Gas Resources

Policies

• Taxes and Royalties

• Land Access

• Environmental Regulations

Exploration, Development, and Production Costs

• Drilling Costs

• Environmental Compliance Costs

Technologies

• New Field Exploration Efficiency

• Field Development, Gas Processing

• Recovery Factor Improvement

Producer Behavior

• Oil and Gas Price Expectations

• Rate-of-Return Criteria

Model output includes annual forecasted number ofwells drilled by type, reserve additions, production,end-of-year reserves, and various cash flow accounts.Outputs can be in viewed in Microsoft Excel andAccess format and include details by type of natural gasand by model regions/intervals.

The Altos Models

In order to develop additional tools to supplementthe EEA forecasting model, the NPC also licensed theNorth American Regional Gas (NARG) model and theNorth American Regional Electricity (NARE) modelfrom Altos Management Partners. All characteriza-tions of North American gas and electric power mar-kets incorporated into the Altos model were developedby a modeling team that was part of the NPC naturalgas study. Although significant progress was madedeveloping input to the model, there was insufficienttime during this study to thoroughly review the resultswith industry representatives and build this feedbackinto the model. At the time of publication, somechanges were felt to be appropriate. The modelingframeworks and their associated algorithms aredescribed below.

The NARG Model

For the purposes of the 2003 NPC study, 230 natu-ral gas supply nodes and 72 demand regions weredefined. The supply regions were aggregated from over700 plays, and the demand regions were segmentedinto residential, commercial, chemical industry, other


industry, and power generation. The model was con-figured through 2045. Because results are computedand reported annually, NARG does not incorporate gasstorage, or the seasonal load variations.

The NARG model assumes the existence of a com-petitive, transparent market, and seeks an economi-cally optimum solution over the investment life cyclethat allows investors in all segments of the market toachieve their target rates of return.

Users of NARG must specify supply and demandnodes, and pipeline infrastructure connections, calledlinks, as follows:

• Supply, or “resource,” nodes require cost-of-supplycurves, which characterize the increasing marginalcost of bringing new gas to market. Cost-of-supplycurves are derived from resource availability, pro-duction decline profiles, and cost data.

• Demand nodes require specification of demand vol-umes, price elasticity, and other variables such asincome, weather, and population.

• Pipeline links, processing plants, and LNG terminalsrequire capacity and tariff specification. The modelconstructs new capacity when user-defined eco-nomic criteria are met.

The NARE Model

The NARE model includes a database of every gen-erating unit in North America (capacity, fuel-type,costs, etc.), along with utility demand projections. Ituses a “zero-arbitrage” solution to dispatch unitswithin each region and across regional boundaries.Users must define the capital and operating costs ofnew capacity, which will be allowed to compete withexisting capacity. Dispatch of existing fleet capacity isbased on competitive economics.

Gas prices reported as output from the NARGmodel were input to the NARE model to project gasconsumption based on economic dispatch of the gen-eration. An iterative process was used to reach conver-gence of price and demand between the two models.

Modeling Methodology

The NPC modeling team comprehensively cus-tomized the NARG and NARE models. The mostimportant features incorporated into the NPC versionof the models are outlined as follows.

Resources and Supply

Initially, the team modeled cost-of-supply curves forall geologic plays in North America. However, in orderto simplify the computational load and improve modelperformance, the cost-of-supply data for provedreserves, proved growth, and undiscovered resourcesfrom over 700 plays were aggregated into 230 supplynodes.

Technical Resource. The technical resourceincluded proved reserves, growth, and undiscoveredresource assessments. The resource was also seg-mented into conventional gas (non-associated andassociated gas) and nonconventional gas (shale gas,coal bed methane, and tight gas). Total resource assess-ments were adjusted to reflect current access restric-tions.

Cost-of-Supply Curves. For each node and gasreservoir type, full development and operating costswere used for proved reserves growth and undiscov-ered resources, while only operating cost curves wereused to define continued production of provedreserves. All costs were provided by an upstream costestimating team within the Supply Task Group. As ofthe release of this report, the cost-of-supply curves hadnot been reviewed and analyzed by the Supply TaskGroup, and some changes were viewed to be appropri-ate to achieve consensus within the Supply Task Groupon the cost-of-supply input to the Altos model.Detailed pre-processing spreadsheet models weredeveloped to calculate the cost-of-supply input neededby the NARG model. The NRG Associates databasewas used to statistically estimate the EstimatedUltimate Recovery per well for each play.

The team developed a Monte Carlo simulation toforecast the size order of future discoveries. This “dis-covery process model” was used to estimate theexpected size of the three representative undiscoveredfields in each play needed to define a cost-of-supplycurve. Technology improvement factors were used forboth the capital and operating cost curves.

LNG Imports. Nodes were defined to representexisting and new LNG terminal capacity; the costs oflanding and regasification were defined for each ofthese nodes. The model then computed import vol-umes that would be economic depending on local mar-ket price.

CHAPTER 10 - MODELING METHODOLOGIES314

Arctic Gas. Alaska gas was assumed to be availablein 2013 and Mackenzie Delta gas was assumed to beavailable in 2009, with initial pipeline capacities of4 BCF/D and 1 BCF/D, respectively.

Pipelines

Existing Pipelines. Every major pipeline in NorthAmerica was characterized by its capacity and fixedand variable tariff rates. Discounting based on loadfactor was incorporated.

Expansions and New Capacity. Potential expan-sions or new-build capacity were specified by year ofavailability, capital costs, operating costs, and capaci-ties. The model then computed whether to utilizecapacity based on the overall economics of transporta-tion and regional price differentials. Near-termplanned expansions were explicitly modeled.

Demand

Dynamic elasticity of demand was incorporated inthe residential, commercial, chemical, and other indus-trial sectors. The power generation sector was mod-eled using NARE, and thus was characterized struc-turally rather than econometrically.

• The demand model incorporated exogenous factorssuch as economic growth, weather, and population.

• Gas demand was a model output for the residential,commercial, electric power generation, chemical,and other industrial sectors.

• For residential and commercial demand, price andincome elasticity, as well as the influence of popula-tion and weather, were econometrically estimatedfrom historical data.

• For chemicals and other industrial demand, priceand income elasticity were estimated from projectedprice and consumption consistent with the detailedanalysis undertaken by the NPC using expandedindustrial model developed by EEA.

• For electric power gas demand, the results of NAREwere input into NARG. The NARE model containednew-build power generation capacity assumptions.

• Specific model upgrades, such as integration into theWorld Gas model, increased industrial demand gran-ularity, incorporating a price/income feedback loop,etc., are addressed in the Demand Task Group Report.

Further Altos Modeling Work

The NPC modeling team will continue working withthe USGS to determine the feasibility of the USGSupdating and maintaining the resource, engineering,and cost data used in the cost-of-supply pre-processordeveloped by the NPC.


45393_apend_divider_ags 3/5/04 1:13 PM Page 1

INTEGRATED REPORT - APPENDIX A A-1

APPENDIX A

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 1

INTEGRATED REPORT - APPENDIX AA-2

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 2


DESCRIPTION OF THE NATIONAL PETROLEUM COUNCIL

In May 1946, the President stated in a letter to the Secretary of the Interior that he had been impressed by the contri-bution made through government/industry cooperation to the success of the World War II petroleum program. Hefelt that it would be beneficial if this close relationship were to be continued and suggested that the Secretary of theInterior establish an industry organization to advise the Secretary on oil and natural gas matters.

Pursuant to this request, Interior Secretary J. A. Krug established the National Petroleum Council (NPC) on June 18,1946. In October 1977, the Department of Energy was established and the Council was transferred to the new de-partment.

The purpose of the NPC is solely to advise, inform, and make recommendations to the Secretary ofEnergy on any matter, requested by the Secretary, relating to oil and natural gas or the oil and gas industries. Matters that the Secretary of Energy would like to have considered by the Council are submitted in theform of a letter outlining the nature and scope of the study. The Council reserves the right to decide whether it willconsider any matter referred to it.

Examples of studies undertaken by the NPC at the request of the Secretary of Energy include:

The NPC does not concern itself with trade practices, nor does it engage in any of the usual trade association activi-ties. The Council is subject to the provisions of the Federal Advisory Committee Act of 1972.

Members of the National Petroleum Council are appointed by the Secretary of Energy and represent all segments ofthe oil and gas industries and related interests. The NPC is headed by a Chair and a Vice Chair, who are elected bythe Council. The Council is supported entirely by voluntary contributions from its members.

•Factors Affecting U.S. Oil & Gas Outlook (1987)

• Integrating R&D Efforts (1988)

•Petroleum Storage & Transportation (1989)

• Industry Assistance to Government – Methods for Providing Petroleum Industry Expertise During Emergencies (1991)

•Short-Term Petroleum Outlook – An Examination of Issues and Projections (1991)

•Petroleum Refining in the 1990s – Meeting the Challenges of the Clean Air Act (1991)

•The Potential for Natural Gas in the United States (1992)

•U.S. Petroleum Refining – Meeting Requirements for Cleaner Fuels and Refineries (1993)

•The Oil Pollution Act of 1990: Issues and Solutions (1994)

•Marginal Wells (1994)

•Research, Development, and Demonstration Needs of the Oil and Gas Industry (1995)

•Future Issues – A View of U.S. Oil & Natural Gas to 2020 (1995)

• Issues for Interagency Consideration – A Supplement to the NPC’s Report: Future Issues – A View of U.S. Oil & Natural Gas to 2020 (1996)

•U.S. Petroleum Product Supply – Inventory Dynamics (1998)

•Meeting the Challenges of the Nation’s Growing Natural Gas Demand (1999)

•U.S. Petroleum Refining – Assuring the Adequacy and Affordability of Cleaner Fuels (2000)

•Securing Oil and Natural Gas Infrastructures in the New Economy (2001).

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 3

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 4



MEMBERSHIP

2002/2003

Jacob AdamsPresidentArctic Slope Regional Corporation

George A. Alcorn, Sr.PresidentAlcorn Exploration, Inc.

Conrad K. AllenPresidentNational Association of Black Geologists and Geophysicists

Robert J. Allison, Jr.Chairman, President and Chief Executive OfficerAnadarko Petroleum Corporation

Robert O. AndersonRoswell, New Mexico

Philip F. AnschutzPresidentThe Anschutz Corporation

Gregory L. ArmstrongChairman and Chief Executive OfficerPlains All American

Robert G. ArmstrongPresidentArmstrong Energy Corporation

Gregory A. ArnoldPresident and Chief Operating OfficerTruman Arnold Companies

Ralph E. BaileyChairman and Chief Executive OfficerAmerican Bailey Inc.

Robert W. BestChairman of the Board, President and Chief Executive OfficerAtmos Energy Corporation

M. Frank BishopExecutive DirectorNational Association of State Energy Officials

Alan L. BoeckmannChairman and Chief Executive OfficerFluor Corporation

Carl E. Bolch, Jr.Chairman and Chief Executive OfficerRacetrac Petroleum, Inc.

Donald T. BollingerChairman of the Board and Chief Executive OfficerBollinger Shipyards, Inc.

John F. BookoutHouston, Texas

Wayne H. BrunettiChairman, President and Chief Executive OfficerXcel Energy Inc.

Philip J. BurguieresChief Executive OfficerEMC Holdings, L.L.C.

Victor A. BurkManaging PartnerOil & Gas DivisionDeloitte & Touche LLP

Frank M. Burke, Jr.Chairman and Chief Executive OfficerBurke, Mayborn Company, Ltd.

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 5



Karl R. ButlerPresident and Chief Executive OfficerICC Energy Corporation

Thos. E. CappsChairman, President and Chief Executive OfficerDominion

Robert B. CatellChairman and Chief Executive OfficerKeySpan

Clarence P. Cazalot, Jr.PresidentMarathon Oil Company

Luke R. CorbettChairman and Chief Executive OfficerKerr-McGee Corporation

Michael B. CoulsonPresidentCoulson Oil Group

Gregory L. CraigPresidentCook Inlet Energy Supply

William A. CustardPresident and Chief Executive OfficerDallas Production, Inc.

Robert DarbelnetPresident and Chief Executive OfficerAAA

Charles D. DavidsonChairman, President and Chief Executive OfficerNoble Energy, Inc.

Claiborne P. DemingPresident and Chief Executive OfficerMurphy Oil Corporation

Cortlandt S. DietlerPresident and Chief Executive OfficerTransMontaigne Oil Company

Dan O. DingesChairman, President and Chief Executive OfficerCabot Oil & Gas Corporation

David F. DornChairman EmeritusForest Oil Corporation

E. Linn Draper, Jr.Chairman, President and Chief Executive OfficerAmerican Electric Power Co., Inc.

John G. DrosdickChairman, President and Chief Executive OfficerSunoco, Inc.

Archie W. DunhamChairman of the BoardConocoPhillips

W. Byron DunnPresident and Chief Executive OfficerLone Star Steel Company

Daniel C. EckermannPresident and Chief Executive OfficerLeTourneau, Inc.

James C. EllingtonChairmanThe Energy Council

James W. EmisonChairman and Chief Executive OfficerWestern Petroleum Company

Ronald A. EricksonChief Executive OfficerHoliday Companies

Sheldon R. EriksonChairman of the Board, President and Chief Executive OfficerCooper Cameron Corporation

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 6



Stephen E. EwingPresident and Chief Operating OfficerDTE Energy Gas

John G. FarbesPresidentBig Lake Corporation

Claire Scobee FarleyChief Executive OfficerRandall & Dewey, Inc.

G. Steven FarrisPresident and Chief Executive OfficerApache Corporation

William L. FisherBarrow Chair in Mineral Resources Department of Geological Sciences andDirector of the Jackson School of GeoscienceUniversity of Texas at Austin

James C. FloresChairman and Chief Executive OfficerPlains Exploration & Production Company

Eric O. FornellManaging Director and Group ExecutiveGlobal Natural Resources GroupJ. P. Morgan Securities Inc.

Joe B. FosterNon-executive ChairmanNewfield Exploration Company

Robert W. FriVisiting ScholarResources For the Future Inc.

Murry S. GerberPresident and Chief Executive OfficerEquitable Resources, Inc.

James A. GibbsChairmanFive States Energy Company

Rufus D. GladneyChairmanAmerican Association of Blacks in Energy

Lawrence J. GoldsteinPresidentPetroleum Industry Research Foundation, Inc.

Charles W. GoodyearChief Executive OfficerBHP Billiton Plc

Bruce C. GottwaldChairman of the BoardEthyl Corporation

Andrew GouldChairman and Chief Executive OfficerSchlumberger Limited

S. Diane GrahamChairman and Chief Executive OfficerSTRATCO, Inc.

William E. GreeheyChairman of the Board and Chief Executive OfficerValero Energy Corporation

Robbie Rice GriesPresidentAmerican Association of Petroleum Geologists

James T. HackettPresident and Chief Operating OfficerDevon Energy Corporation

Frederic C. HamiltonChairmanThe Hamilton Companies

Christine HansenExecutive DirectorInterstate Oil and Gas Compact Commission

Angela E. HarrisonChairman and Chief Executive OfficerWELSCO, Inc.

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 7



Lewis Hay, IIIChairman, President and Chief Executive OfficerFPL Group

Frank O. HeintzPresident and Chief Executive OfficerBaltimore Gas and Electric Company

John B. HessChairman, President and Chief Executive OfficerAmerada Hess Corporation

Jack D. HightowerPresident and Chief Executive OfficerCelero Energy LLC

Jerry V. HoffmanChairman, President and Chief Executive OfficerBerry Petroleum Company

Roy M. HuffingtonChairman of the Board and Chief Executive OfficerRoy M. Huffington, Inc.

Dudley J. HughesPresidentHughes South Corporation

Ray L. HuntChairman of the BoardHunt Oil Company

Hillard HuntingtonExecutive DirectorEnergy Modeling ForumStanford University

Frank J. IarossiChairmanAmerican Bureau of Shipping & Affiliated Companies

Ray R. IraniChairman and Chief Executive OfficerOccidental Petroleum Corporation

Eugene M. IsenbergChairman and Chief Executive OfficerNabors Industries, Inc.

Francis D. JohnChairman, President and Chief Executive OfficerKey Energy Services, Inc.

A. V. Jones, Jr.ChairmanVan Operating, Ltd.

Jon Rex JonesChairmanEnerVest Management Company, L. C.

Jerry D. JordanPresidentJordan Energy Inc.

Fred C. JulanderPresidentJulander Energy Company

John A. KanebChief Executive OfficerGulf Oil Limited Partnership

W. Robert KeatingCommissionerDepartment of Telecommunications and EnergyCommonwealth of Massachusetts

Bernard J. KennedyChairman EmeritusNational Fuel Gas Company

James W. KeyesPresident and Chief Executive Officer7-Eleven, Inc.

Richard D. KinderChairman and Chief Executive OfficerKinder Morgan Energy Partners, L.P.

Susan M. LandonPetroleum GeologistDenver, Colorado

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 8



Stephen D. LaytonPresidentE&B Natural Resources

Virginia B. LazenbyChairman and Chief Executive OfficerBretagne G.P.

Kathy Prasnicki LehneFounder, President and Chief Executive OfficerSun Coast Resources, Inc.

David J. LesarChairman of the Board, President and Chief Executive OfficerHalliburton Company

James D. LightnerChairman, President and Chief Executive OfficerTom Brown, Inc.

Michael C. LinnPresidentLinn Energy, LLC

Daniel H. LopezPresidentNew Mexico Institute of Mining and Technology

Thomas E. LoveChairman and Chief Executive OfficerLove’s Country Stores, Inc.

William D. McCabeVice PresidentEnergy ResourcesThermoEnergy Corporation

Aubrey K. McClendonChairman of the Board and Chief Executive OfficerChesapeake Energy Corporation

W. Gary McGilvrayPresident and Chief Executive OfficerDeGolyer and MacNaughton

Cary M. MaguirePresidentMaguire Oil Company

Steven J. MalcolmPresident and Chief Executive OfficerThe Williams Companies, Inc.

Timothy M. MarquezChief Executive OfficerMarqon Energy L.L.C.

Frederick R. MayerChairmanCaptiva Resources, Inc.

F. H. MerelliChairman, President and Chief Executive OfficerCimarex Energy Co.

C. John MillerChief Executive OfficerMiller Energy, Inc.

Merrill A. Miller, Jr.Chairman, President and Chief Executive OfficerNational Oilwell, Inc.

Herman Morris, Jr.President and Chief Executive OfficerMemphis Light, Gas & Water Division

Robert A. MosbacherChairmanMosbacher Energy Company

James J. MulvaPresident and Chief Executive OfficerConocoPhillips

John Thomas MunroPresidentMunro Petroleum & Terminal Corporation

David L. MurfinPresidentMurfin Drilling Co., Inc.

Marquez Energy L.L.C.

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 9



Mark B. MurphyPresidentStrata Production Company

William C. Myler, Jr.PresidentThe Muskegon Development Company

Gary L. NealeChairman, President and Chief Executive OfficerNiSource Inc.

J. Larry NicholsChairman of the Board and Chief Executive OfficerDevon Energy Corporation

John W. B. NorthingtonVice PresidentNational Environmental Strategies Inc.

Erle NyeChairman of the Board and Chief ExecutiveTXU Corp.

Christine J. OlsonChairman and Chief Executive OfficerS. W. Jack Drilling Company

David J. O'ReillyChairman of the Board and Chief Executive OfficerChevronTexaco Corporation

C. R. PalmerChairman of the BoardRowan Companies, Inc.

Mark G. PapaChairman and Chief Executive OfficerEOG Resources, Inc.

Paul H. ParkerVice PresidentCenter for Resource Management

Robert L. Parker, Sr.Chairman of the BoardParker Drilling Company

A. Glenn PattersonPresident and Chief Operating OfficerPatterson–UTI Energy Inc.

Ross J. PillariPresidentBP America Inc.

L. Frank PittsOwnerPitts Energy Group

Richard B. PrioryChairman, President and Chief Executive OfficerDuke Energy Corporation

Caroline QuinnMount Vernon, Illinois

Keith O. RattiePresident and Chief Executive OfficerQuestar Corporation

Lee R. RaymondChairman and Chief Executive OfficerExxon Mobil Corporation

John G. RicePresident and Chief Executive OfficerGE Power Systems

Corbin J. Robertson, Jr.PresidentQuintana Minerals Corporation

Robert E. RoseChairman of the BoardGlobalSantaFe Corporation

Henry A. Rosenberg, Jr.Chairman of the BoardCrown Central Petroleum Corporation

Robert J. RoutsFormer President and Country ChairmanShell Oil Company

Former Chairman and

Keith O. RattieChairman, President and

Chief Executive OfficerQuestar Corporation

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 10



Mark A. RubinExecutive DirectorSociety of Petroleum Engineers

Robert SantistevanDirectorSouthern Ute Indian Tribe Growth Fund

S. Scott SewellPresidentDelta Energy Management, Inc.

Bobby S. ShackoulsChairman, President and Chief Executive OfficerBurlington Resources Inc.

Scott D. SheffieldChairman, President and Chief Executive OfficerPioneer Natural Resources Company

Matthew R. SimmonsChairman and Chief Executive OfficerSimmons and Company International

Sam R. SimonPresident and Chief Executive OfficerAtlas Oil Company

Bob R. SimpsonChairman and Chief Executive OfficerXTO Energy Inc.

Bruce A. SmithChairman, President and Chief Executive OfficerTesoro Petroleum Corporation

Charles C. Stephenson, Jr.Chairman of the BoardVintage Petroleum, Inc.

J. W. StewartChairman, President and Chief Executive OfficerBJ Services Company

James H. StoneChairman of the BoardStone Energy Corporation

Carroll W. SuggsNew Orleans, Louisiana

Patrick F. TaylorChairman and Chief Executive OfficerTaylor Energy Company

James Cleo Thompson, Jr.PresidentThompson Petroleum Corporation

Gerald TorresAssociate Dean for Academic Affairs University of Texas School of Law andVice ProvostUniversity of Texas at Austin

Diemer TruePartnerTrue Companies, Inc.

H. A. True, IIIPartnerTrue Oil Company

Paul G. Van WagenenChairman, President and Chief Executive OfficerPogo Producing Company

Randy E. VelardePresidentThe Plaza Group

Thurman VelardeAdministratorOil and Gas AdministrationJicarilla Apache Tribe

Philip K. Verleger, Jr.PKVerleger, L.L.C.

Vincent ViolaChairman of the BoardNew York Mercantile Exchange

Joseph C. Walter, IIIPresidentWalter Oil & Gas Corporation

L. O. WardChairman and Chief Executive OfficerWard Petroleum Corporation

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 11



Wm. Michael Warren, Jr.Chairman, President and Chief Executive OfficerEnergen Corporation

J. Robinson WestChairman of the BoardThe Petroleum Finance Company

Michael E. WileyChairman, President and Chief Executive OfficerBaker Hughes Incorporated

Bruce W. WilkinsonChairman of the Board and Chief Executive OfficerMcDermott International, Inc.

Charles R. WilliamsonChairman of the Board and Chief Executive OfficerUnocal Corporation

Mary Jane WilsonPresident and Chief Executive OfficerWZI Inc.

Brion G. WiseChairman and Chief Executive OfficerWestern Gas Resources, Inc.

William A. WiseRetired ChairmanEl Paso Corporation

Donald D. WolfChairman of the Board and Chief Executive OfficerWestport Resources Corp.

George M. YatesPresident and Chief Executive OfficerHarvey E. Yates Company

John A. YatesPresidentYates Petroleum Corporation

Daniel H. YerginChairmanCambridge Energy Research Associates

Henry ZarrowVice Chairman

Sooner Pipe & Supply Corporation

Henry ZarrowVice Chairman

Sooner Pipe & Supply Corporation

45393_Ap1_Ap12_AGS 3/5/04 1:15 PM Page 12

STUDY GROUP ROSTERSINTEGRATED REPORT

INTEGRATED REPORT - APPENDIX B B-1

NPC Committee on Natural Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2

Coordinating Subcommittee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-5

Financial Team. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-7

Price Volatility Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-7

Modeling Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-8

Communications Team. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-9

Demand Task Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-10

Economics and Demographics Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . B-12

Power Generation Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-12

Residential and Commercial Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-13

Industrial Utilization Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-13

Supply Task Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-15

Resource Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-16

Technology Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-19

Environmental/Regulatory/Access Subgroup . . . . . . . . . . . . . . . . . . . . . . . . B-19

LNG Subgroup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-21

Arctic Subgroup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-22

Transmission & Distribution Task Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-23

Transmission Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-24

Distribution Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-25

Storage Subgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-26

Additional Study Participants

The National Petroleum Council wishes to acknowledge the numerous other individuals andorganizations who participated in some aspects of the work effort through workshops, outreachmeetings and other contacts. Their time, energy, and commitment significantly enhanced thestudy and their contributions are greatly appreciated.

APPENDIX B

Supplemental Modeling Team

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-1

INTEGRATED REPORT - APPENDIX BB-2


COMMITTEE ON NATURAL GAS

CHAIR

Bobby S. ShackoulsChairman of the Board, President and Chief Executive OfficerBurlington Resources Inc.

VICE CHAIR, DEMAND

Robert B. CatellChairman and Chief Executive OfficerKeySpan

VICE CHAIR, SUPPLY

Lee R. RaymondChairman and Chief Executive OfficerExxon Mobil Corporation

GOVERNMENT COCHAIR

Robert G. CardUnder Secretary of Energy

VICE CHAIR, MIDSTREAM

Richard D. KinderChairman and Chief Executive OfficerKinder Morgan Energy Partners, L.P.

SECRETARY

Marshall W. NicholsExecutive DirectorNational Petroleum Council

* * *


Conrad K. AllenPresidentNational Association of Black Geologists and Geophysicists

Robert J. Allison, Jr.Chairman, President and Chief Executive OfficerAnadarko Petroleum Corporation

Alan L. BoeckmannChairman and Chief Executive OfficerFluor Corporation

Wayne H. BrunettiChairman, President and Chief Executive OfficerXcel Energy Inc.

Thos. E. CappsChairman, President and Chief Executive OfficerDominion

Clarence P. Cazalot, Jr.PresidentMarathon Oil Company

Luke R. CorbettChairman and Chief Executive OfficerKerr-McGee Corporation

E. Linn Draper, Jr.Chairman, President and Chief Executive OfficerAmerican Electric Power Co., Inc.

Archie W. DunhamChairman of the BoardConocoPhillips

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-2


NPC COMMITTEE ON NATURAL GAS

Stephen E. EwingPresident and Chief Operating OfficerDTE Energy Gas

William L. FisherBarrow Chair in Mineral Resources Department of Geological Sciences andDirector of the Jackson School of GeoscienceUniversity of Texas at Austin

Eric O. FornellManaging Director and Group ExecutiveGlobal Natural Resources GroupJ. P. Morgan Securities Inc.

Robert W. FriVisiting ScholarResources For the Future Inc.

James A. GibbsChairmanFive States Energy Company

Andrew GouldChairman and Chief Executive OfficerSchlumberger Limited

James T. HackettPresident and Chief Operating OfficerDevon Energy Corporation

Lewis Hay, IIIChairman, President and Chief Executive OfficerFPL Group

Frank O. HeintzPresident and Chief Executive OfficerBaltimore Gas and Electric Company

Ray R. IraniChairman and Chief Executive OfficerOccidental Petroleum Corporation

Jon Rex JonesChairmanEnerVest Management Company, L. C.

Jerry D. JordanPresidentJordan Energy Inc.

Fred C. JulanderPresidentJulander Energy Company


W. Gary McGilvrayPresident and Chief Executive OfficerDeGolyer and MacNaughton

Steven J. MalcolmPresident and Chief Executive OfficerThe Williams Companies, Inc.

James J. MulvaPresident and Chief Executive OfficerConocoPhillips

Gary L. NealeChairman, President and Chief Executive OfficerNiSource Inc.

J. Larry NicholsChairman of the Board and Chief Executive OfficerDevon Energy Corporation

Erle NyeChairman of the Board and Chief ExecutiveTXU Corp.

Christine J. OlsonChairman and Chief Executive OfficerS. W. Jack Drilling Company

David J. O'ReillyChairman of the Board and Chief Executive OfficerChevronTexaco Corporation

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-3



Mark G. PapaChairman and Chief Executive OfficerEOG Resources, Inc.

Paul H. ParkerVice PresidentCenter for Resource Management

Robert L. Parker, Sr.Chairman of the BoardParker Drilling Company

Ross J. PillariPresidentBP America Inc.

Keith O. RattiePresident and Chief Executive OfficerQuestar Corporation

John G. RicePresident and Chief Executive OfficerGE Power Systems

Robert E. RoseChairman of the BoardGlobalSantaFe Corporation

Robert J. RoutsFormer President and Country ChairmanShell Oil Company

S. Scott SewellPresidentDelta Energy Management, Inc.

Matthew R. SimmonsChairman and Chief Executive OfficerSimmons and Company International

J. W. StewartChairman, President and Chief Executive OfficerBJ Services Company

Diemer TruePartnerTrue Companies, Inc.

Vincent ViolaChairman of the BoardNew York Mercantile Exchange

Wm. Michael Warren, Jr.Chairman, President and Chief Executive OfficerEnergen Corporation

J. Robinson WestChairman of the BoardThe Petroleum Finance Company

Charles R. WilliamsonChairman of the Board and Chief Executive OfficerUnocal Corporation

William A. WiseRetired ChairmanEl Paso Corporation

Daniel H. YerginChairmanCambridge Energy Research Associates

Keith O. RattieChairman, President and

Chief Executive OfficerQuestar Corporation

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-4



COORDINATING SUBCOMMITTEEOF THE


CHAIR

Jerry J. LangdonExecutive Vice President and Chief Administrative OfficerReliant Resources, Inc.

ASSISTANT TO THE CHAIR

Byron S. WrightVice PresidentStrategy and Capacity PricingEl Paso Pipeline Group

GOVERNMENT COCHAIR

Carl Michael SmithAssistant SecretaryFossil EnergyU.S. Department of Energy

ALTERNATE GOVERNMENT COCHAIR

James A. SlutzDeputy Assistant SecretaryNatural Gas and Petroleum TechnologyU.S. Department of Energy

SECRETARY

John H. Guy, IVDeputy Executive Director

National Petroleum Council


Keith BarnettVice PresidentFundamental AnalysisAmerican Electric Power Co., Inc.

Mark A. BorerExecutive Vice PresidentDuke Energy Field Services

Jon C. ColeVice PresidentBusiness Development and MarketingENSCO International Inc.

V. W. HoltVice PresidentOnshore U.S.BP America Production Inc.

Robert G. Howard, Jr.Vice PresidentNorth America UpstreamChevronTexaco Corporation

John Hritcko, Jr.Vice PresidentShell NA, LNG, Inc.Shell US Gas & Power Company


Paul D. KoonceSenior Vice PresidentPortfolio ManagementDominion Energy, Inc.

Patrick J. KuntzVice PresidentNatural Gas and Crude Oil SalesMarathon Oil Company

Paul D. KoonceChief Executive OfficerPortfolio ManagementDominion Energy, Inc.

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-5


Storage Management & System DesignKinder Morgan Inc.

Energy PolicyKeySpan

William N. StrawbridgeSenior Advisor

ExxonMobil Production Company

Mark T. MaasselVice PresidentRegulatory & Governmental PolicyNiSource Inc.

David J. ManningSenior Vice PresidentCorporate AffairsKeySpan

Thomas B. NuszVice PresidentAcquisitionsBurlington Resources Inc.

John E. OlsonSenior Vice President and Director of ResearchSanders Morris Harris

Scott E. ParkerPresidentNatural Gas Pipeline Company of America

Mark L. PeaseVice PresidentU.S. Onshore & OffshoreAnadarko Petroleum Corporation

Mark A. SikkelVice PresidentExxonMobil Production Company

Andrew K. SotoAdvisor to the ChairmanFederal Energy Regulatory Commission

Lee M. StewartSenior Vice PresidentGas TransmissionSouthern California Gas Company and San Diego Gas & Electric CompanySempra Energy Utilities

Lawrence H. TowellVice President, AcquisitionsKerr-McGee Corporation

Rebecca W. WatsonAssistant SecretaryLand and Minerals ManagementU.S. Department of the Interior

Dena E. WigginsGeneral CounselProcess Gas Consumers Group

Michael ZenkerDirectorNorth American Natural GasCambridge Energy Research Associates

SPECIAL ASSISTANTS

Ronald L. BrownVice PresidentStorage Management & System DesignKinder Morgan Inc.

Harlan ChappelleStaff Consultant – Energy PolicyKeySpan

William N. StrawbridgeSenior Advisor

ExxonMobil Production Company

COORDINATING SUBCOMMITTEE

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-6


COORDINATING SUBCOMMITTEE’S FINANCIAL TEAM

LEADERJohn E. Olson

Senior Vice President andDirector of Research

Sanders Morris Harris

Ronald S. BarrAdvisorExxonMobil Gas & Power Marketing Company

J. Michael BodellManagerStrategic Planning and Market AnalysisMidstream & TradeUnocal Corporation

Mark J. FinleySenior EconomistBP America Inc.

Edward J. GilliardSenior AdvisorPlanning and AcquisitionsBurlington Resources Inc.

Marianne S. KahChief EconomistConocoPhillips


John LutostanskiAssistant TreasurerExxonMobil Chemicals

Gary B. RennieVice PresidentTrading AnalyticsBP Energy Company

Barton D. SchouestManaging DirectorBanque BNP Paribas


COORDINATING SUBCOMMITTEE ‘S PRICE VOLATILITY TEAM

LEADERRonald S. Barr

AdvisorExxonMobil Gas &

Power Marketing Company

Keith BarnettVice PresidentFundamental AnalysisAmerican Electric Power Co., Inc.


COORDINATING SUBCOMMITTEE’S PRICE VOLATILITY TEAM


45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-7


COORDINATING SUBCOMMITTEE’S SUPPLEMENTAL MODELING TEAM

LEADERAndrew J. Slaughter

Senior Economics Advisor – EP AmericasShell Exploration & Production Company

Meg O’Connor GentleManagerEconomics and ForecastingAnadarko Petroleum Corporation

Michael S. HuppSenior Energy Market AnalystDominion Resources

Kenneth B. Medlock, IIIVisiting Professor, Department of Economics and Energy Consultant to the James A. Baker III Institute for Public PolicyRice University

Walter C. RieseConsulting GeologistReservoir/Wells Assurance GroupOnshore U.S. Business UnitBP America Production Company

Gary H. TsangDevelopment PlannerExxonMobil Development Company

Loring P. WhiteExploration AdvisorExxonMobil Exploration Company

Seth S. RobertsRisk Manager – EnergyDow Hydrocarbons and Resources Inc.

Andrew J. SlaughterSenior Economics Advisor – EP AmericasShell Exploration & Production Company

Alan R. WigginsDirectorGas and Power Regulatory AffairsConocoPhillips

Leslie J. DemanDirectorFundamental Research and AnalysisShell Trading Gas & Power


Donald R. KnopEconomic and Planning ConsultantWilliams Gas Pipelines

COORDINATING SUBCOMMITTEE’S PRICE VOLATILITY TEAM

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-8


COORDINATING SUBCOMMITTEE’S COMMUNICATIONS TEAM

LEADERBrian W. Miller

DirectorU.S. Government and International Affairs

BP America Inc.

G. David BlackmonManagerCorporate AffairsBurlington Resources Inc.


Ben J. DillonManagerGovernemnt AffairsShell Exploration & Production Company

Kathleen E. EcclestonCommunications SpecialistMarathon Oil Corporation

Gina M. GibbsPublic Affairs ManagerDow Chemical Company


Bryan S. LeeCommunications SpecialistOffice of External AffairsFederal Energy Regulatory Commission

Mark R. MaddoxPrincipal Deputy Assistant SecretaryFossil EnergyU.S. Department of Energy

Megan K. MurphySpecial AssistantNatural Gas and Petroleum TechnologyU.S. Department of Energy

Sarah G. NovoselAssistant CounselFederal PolicyAmerican Electric Power Company

Kyle M. SawyerConsultantStrategyEl Paso Pipeline Group

Tricia L. ThompsonFederal Issues Advisor–UpstreamExxon Mobil Corporation

William F. WhitsittPresidentDomestic Petroleum Council


Ben J. DillonManagerGovernment AffairsShell Exploration & Production Company

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-9



DEMAND TASK GROUPOF THE


CHAIR

David J. ManningSenior Vice PresidentCorporate AffairsKeySpan



GOVERNMENT COCHAIR


ALTERNATE GOVERNMENT COCHAIR

Wade W. MurphyExecutive AssistantNatural Gas & Petroleum TechnologiesOffice of Fossil EnergyU.S. Department of Energy

SECRETARY

Benjamin A. Oliver, Jr.Senior Committee Coordinator

National Petroleum Council

Barry D. AndersonOperations ManagerEnergy Marketing & TradingFlorida Power & Light Company

Dennis M. BaileyDirectorEnergy PurchasingPPG Industries Incorporated

Keith BarnettVice PresidentFundamental AnalysisAmerican Electric Power Co., Inc



Mark S. CrewsDirectorSouthern Company

Leslie J. DemanDirectorFundamental Research & AnalysisShell Trading Gas & Power



45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-10


Lee W. GoochVice PresidentNatural GasPCS Administration (U.S.A.), Inc.

R. William JewellBusiness Vice President – EnergyThe Dow Chemical Company



Diane G. LeopoldManaging DirectorBusiness Planning & Market AnalysisDominion Energy, Inc.

Charles W. LindermanDirectorEnergy Supply PolicyAlliance of Energy Suppliers(A Division of the Edison Electric Institute)

Ronald G. LukasVice PresidentTrading ServicesKeySpan

Steven W. MillerDirectorProject EvaluationSan Diego Gas & Electric

Vance C. MullisAssistant to the President and Chief Executive OfficerSouthern Company Gas


Charles Greer RossmannSenior Research EconomistResearch and Environmental AffairsSouthern Company

Mark C. SchroederVice PresidentRegulatory AffairsEl Paso Merchant Energy Group

Loren K. StarcherProject ManagerPGS – Power ProjectsExxonMobil Gas & Power Marketing Company

Jone-Lin WangDirectorNorth American Electric PowerCambridge Energy Research Associates


Byron S. WrightVice PresidentStrategy and Capacity PricingEl Paso Pipeline Group

DEMAND TASK GROUP


45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-11


DEMAND TASK GROUP’S POWER GENERATION SUBGROUP

LEADERKeith BarnettVice President

Fundamental AnalysisAmerican Electric Power Co., Inc.

Barry D. AndersonOperations ManagerEnergy Marketing & TradingFlorida Power & Light Company

Robert W. AndersonEconomistBonneville Power Administration

Lane T. MahaffeyDirectorCorporate PlanningSeminole Electric Cooperative, Inc.


DEMAND TASK GROUP’S ECONOMICS AND DEMOGRAPHICS SUBGROUP

LEADERLeslie J. Deman

DirectorFundamental Research and Analysis

Shell Trading Gas & Power







Charles Greer RossmannSenior Research EconomistResearch and Environmental AffairsSouthern Company

Alan R. WigginsDirectorGas and Power Regulatory AffairsConocoPhillips

Byron S. WrightVice President

Strategy and Capacity PricingEl Paso Pipeline Group



45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-12

DEMAND TASK GROUP’S INDUSTRIAL UTILIZATION SUBGROUP

LEADERDena E. WigginsGeneral Counsel

Process Gas Consumers Group

Dennis M. BaileyDirectorEnergy PurchasingPPG Industries Incorporated

Joseph G. BaranManagerStrategic Sourcing, EnergyPPG Industries Incorporated

Peggy R. ClaytorSenior Government Affairs SpecialistThe Timken Company



LEADERRonald G. LukasVice President

Trading ServicesKeySpan

Leslie J. DemanDirectorFundamental Research and AnalysisShell Trading Gas & Power



Laura E. TandyManagerStrategic PlanningKeySpan



Loren K. StarcherProject ManagerPGS – Power ProjectsExxonMobil Gas & Power Marketing Company

Jone-Lin WangDirectorNorth American Electric PowerCambridge Energy Research Associates

Lane T. MahaffeyDirectorCorporate PlanningSeminole Electric Cooperative, Inc.

DEMAND TASK GROUP’S RESIDENTIAL AND COMMERCIAL SUBGROUP


Terence J. BrennanPlanning AssociateFeedstock and EnergyExxonMobil Chemical Company


45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-13



Terence J. BrennanPlanning AssociateFeedstock and EnergyExxonMobil Chemical Company


Keith S. ClausonDirectorNatural Gas Services and ProcurementAlcoa Incorporated

R. William JewellBusiness Vice President – EnergyThe Dow Chemical Company

Vince J. KwasniewskiValue Chain AnalystBP Feedstocks Americas Business UnitBP

Richard O. NotteVice PresidentEnergy ServicesAlcoa Primary Metals


Barney J. SumrallSenior Commercial Manager, Energy

The Dow Chemical Company

Peggy R. ClaytorSenior Government Affairs SpecialistThe Timken Company



Keith S. ClausonDirectorNatural Gas Services and ProcurementAlcoa Incorporated

DEMAND TASK GROUP’S INDUSTRIAL UTILIZATION SUBGROUP

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-14



SUPPLY TASK GROUPOF THE


CHAIR

Mark A. SikkelVice PresidentExxonMobil Production Company


William N. StrawbridgeSenior AdvisorExxonMobil Production Company

GOVERNMENT COCHAIR

Elena S. MelchertProgram ManagerOil & Gas ProductionOffice of Fossil EnergyU.S. Department of Energy

SECRETARY

John H. Guy, IVDeputy Executive DirectorNational Petroleum Council

* * *



G. David BlackmonManagerCorporate AffairsBurlington Resources Inc.

Randall L. CouchGeneral ManagerEngineering & TechnologyAnadarko Petroleum Corporation

David J. CrowleyVice President, MarketingTODCO



John Hritcko, Jr.Vice PresidentShell NA, LNG, Inc.Shell US Gas & Power Company

Patrick J. KuntzVice PresidentNatural Gas and Crude Oil SalesMarathon Oil Company

Ryan M. LanceVice President, Lower 48ConocoPhillips

Mark O. ReidVice PresidentOffshore Exploitation/DevelopmentEl Paso Production Company

Robert D. SchilhabManagerAlaska Gas DevelopmentExxonMobil Production Company

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-15


Andrew J. SlaughterSenior Economics Advisor – EP AmericasShell Exploration & Production Company

Brent J. SmolickVice President and Chief EngineerBurlington Resources Inc.

Robert W. StancilChief GeologistAnadarko Petroleum Corporation

SUPPLY TASK GROUP’S RESOURCE SUBGROUP

LEADER

Gerry A. WorthingtonProject Lead

North America Resource AssessmentExxonMobil Exploration Company

George A. Alcorn, Jr.Vice PresidentAlcorn Development Company

R. Marc BustinGeoscientistDepartment of Earth and Ocean SciencesThe University of British Columbia

Audis C. ByrdManagerGlobal TechnologyHalliburton Energy Services

Randall D. ClarkVice PresidentMarketing and Business DevelopmentNabors Drilling USA, LP

Thierry M. DeCortGeophysicist/GeoscientistResource EvaluationMinerals Management ServiceU.S. Department of the Interior

Glynn EllisTeam LeaderRegional Gulf of Mexico ExplorationEPX-W GOM Greenfield ExplorationShell Exploration & Production Company

SUPPLY TASK GROUP

Gary C. StoneRegional Geology CoordinatorExxonMobil Exploration Company

Chad A. TidwellStrategy ManagerBP America Production Inc.

Gerry A. WorthingtonProject LeadNorth America Resource AssessmentExxonMobil Exploration Company

Gene A. AydinianGeoscientistExxonMobil Production Company

Kenneth J. BirdGeologist/GeoscientistEarth Surface Processes TeamGeologic DivisionU.S. Geological SurveyU.S. Department of the Interior

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-16


L. Wayne ElsnerSenior GeologistGeology and Reserves GroupAlberta Energy and Utilities Board

Gary M. ForsthoffAsset AdvisorMid-Continent Business UnitChevronTexaco Production Company

James B. FraserGeneral ManagerExplorationBurlington Resources Inc.

J. Michael GatensPetroleum EngineerMGV Energy Ltd.

Meg O’Connor GentleManagerEconomics and ForecastingAnadarko Petroleum Corporation

Mitchell E. HenryGeoscientistEnergy Resources TeamU.S. Geological Survey

Matthew HumphreysGeologistBusiness DevelopmentMarathon Oil Company

Kenneth B. Medlock, IIIVisiting Professor, Department of Economics and Energy Consultant to the James A. Baker III Institute for Public PolicyRice University

George PinckneyTeam LeaderNorth Heavy Oil GeoscienceExxonMobil Canada Ltd.

Richard M. ProcterSenior AnalystCanadian Gas Potential Committee

Richard W. MittlerPrincipal GeologistBusiness DevelopmentEl Paso Production Company

Richard D. NehringPresidentNehring Associates

Harry E. Newman, Jr.Operations Support ManagerDrilling TechnicalExxonMobil Development Company

Michael A. OestmannChief GeoscientistPermian Gas Asset ManagerPure Resources, Inc.

Lee E. PetersenGeoscientistTechnology and Exploration PlanningAnadarko Petroleum Corporation

J. David HughesGeoscientistGeological Survey of CanadaNatural Resources Canada

Peter A. LarabeeGeoscientistUpstream Technical ComputingExxonMobil Exploration Company

Mitchell E. HenryGeoscientistEnergy Resources TeamU.S. Geological SurveyU.S. Department of the Interior

R. Curtis PhillipsSenior Professional Petroleum EngineerStrategic Planning–Business DevelopmentKerr-McGee Corporation

Robert C. MiliciResearch GeologistEastern Energy Resources TeamU.S. Geological SurveyU.S. Department of the Interior

Robert A. MeneleyGeoscientistIndependent ConsultantCanadian Gas Potential Committee

Ray A. MissmanReservoir EngineerExxonMobil Production Company

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-17


Christopher J. SchenkSupervisor/GeoscientistU.S. Geological Survey

Pulak K. RayChief GeologistMinerals Management ServiceU.S. Department of the Interior

Mark O. ReidVice PresidentOffshore Exploitation/DevelopmentEl Paso Production Company

Eugene G. RhodesConsultant GeologistBusiness DevelopmentEl Paso Production Company

Walter C. RieseConsulting GeologistReservoir/Wells Assurance GroupOnshore U.S. Business UnitBP America Production Company

Earl J. RitchieVice President and General ManagerHouston DivisionEOG Resources, Incorporated

Robert T. RyderGeoscientistU.S. Geological Survey

Gary C. StoneRegional Geology CoordinatorExxonMobil Exploration Company

Gary H. TsangDevelopment PlannerExxonMobil Development Company

Loring P. WhiteExploration AdvisorExxonMobil Exploration Company

James B. WixtedTechnical DirectorDomestic Business DevelopmentEl Paso Production Company

Rob H. WoronukPresidentGasEnergy Strategies Inc.

Steven T. SchlotterbeckSenior Vice PresidentProduction ManagementEquitable Production Company

Kirk W. SherwoodGeologistMinerals Management ServiceU.S. Department of the Interior

Robert W. StancilChief GeologistAnadarko Petroleum Corporation

Lisa Marie SchronkDevelopment EngineerProject, Planning & Systems/

Cost & ScheduleExxonMobil Development Company

Grant D. ZimbrickGeoscientist

Assessment Core GroupExxonMobil Exploration Company

SUPPLY TASK GROUP’S RESOURCE SUBGROUP

U.S. Department of the Interior

U.S. Department of the Interior

Matthew A. SabiskyPlanning AdvisorExxonMobil Production Company

Consulting Geologist

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-18


SUPPLY TASK GROUP’S TECHNOLOGY SUBGROUP

LEADERRobert G. Howard, Jr.

Vice PresidentNorth America Upstream

ChevronTexaco Corporation

Peter S. AronstamDirector of TechnologyBaker Hughes Incorporated

Lana B. BilleaudOpportunity Assessment ManagerInternational Marketing and BusinessChevronTexaco Global Gas

Audis C. ByrdManagerGlobal TechnologyHalliburton Energy Services

Sheng DingReservoir EngineerEl Paso Energy Corporation

Gerard A. GabrielManagerTechnology Applications DivisionExxonMobil Upstream Research Company

Morris R. HastingManaging PartnerLandmark Graphics Corporation

Stephen A. HolditchSchlumberger FellowSchlumberger Oil Field Service


Kent F. PerryAssistant DirectorTight Sands and Gas Processing ResearchGas Research Institute

Jack C. RawdonEngineering AdvisorProduction OperationsDominion Exploration and Production, Inc.

David E. ReeseReservoir Engineering Fellow

Reservoir SciencesConocoPhillips

SUPPLY TASK GROUP’S ENVIRONMENTAL/REGULATORY/ACCESS SUBGROUP

LEADERG. David Blackmon

ManagerCorporate Affairs

Burlington Resources Inc.

Fernando BlackgoatUpstream Safety, Health and Environment AdvisorExxonMobil Production Company

Walter D. CruickshankDeputy DirectorMinerals Management ServiceU.S. Department of the Interior

LEADERG. David Blackmon

ManagerCorporate Affairs

Burlington Resources Inc.

Druann D. BowerVice PresidentPetroleum Association of Wyoming

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-19

David R. BrownManagerRegulatory AffairsHealth, Safety and EnviromentBP America Production Company

Norma L. CalvertGeneral ManagerState Government AffairsMarathon Oil Company

Bonnie L. CarsonEnvironmental EngineerO&G Environmental Consulting

Jeffrey S. ChapmanEnvironmental AdvisorUpstream Safety, Health and EnvironmentExxonMobil Production Company

J. Keith CouvillionLand ConsultantChevronTexaco Corporation

Wm. Dean CrandellProgram ManagerFluid MineralsForest ServiceU.S. Department of Agriculture


Claire M. MoseleyExecutive DirectorPublic Lands Advocacy

Neil R. LatimerEnvironmental AdvisorExxonMobil Production Company

Randall P. MeabonRegulatory CoordinatorRegulatory and Government ComplianceMarathon Oil Company



H. William HochheiserProgram ManagerOil and Gas Environmental ResearchOffice of Fossil EnergyU.S. Department of Energy

Gary L. HolsanOwner and ManagerGary Holsan Environmental Planning

John S. HullDirectorMarket IntelligenceChevronTexaco Global Trading

Erick V. KaarlelaManagerNational Energy OfficeBureau of Land ManagementU.S. Department of the Interior

Walter D. CruickshankDeputy DirectorMinerals Management ServiceU.S. Department of the Interior

Timothy A. DeinesPlanning ManagerWorldwide ProductionMarathon Oil Company

Eileen D. DeyRegulatory Compliance SupervisorMid-Continent DivisionBurlington Resources Oil & Gas Company, LP

Ben J. DillonManagerGovernment AffairsShell Exploration & Production Company

Robert J. SandilosSenior Government Relations AdvisorChevronTexaco Upstream

ENVIRONMENTAL/REGULATORY/ACCESS SUBGROUP

Environment

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-20


Edward J. ShawSpecial Assistant to the DirectorMinerals Management ServiceU.S. Department of the Interior

ENVIRONMENTAL/REGULATORY/ACCESS SUBGROUP

Kermit G. WitherbeeDeputy Group ManagerFluids GroupBureau of Land ManagementU.S. Department of the Interior

SUPPLY TASK GROUP’S LNG SUBGROUP

LEADERJohn Hritcko, Jr.Vice President

Shell NA, LNG, Inc.Shell US Gas & Power Company

Jayraj C. AminBusiness DevelopmentGlobal LNGBP

Karen N. BaileyManagerLNG Market DevelopmentExxonMobil Gas and Power Marketing

Sara J. BanaszakDirectorGas & PowerThe Petroleum Finance Company

James G. BuschDirectorEnergy Policy and RegulationGas and Power North AmericaBP Energy

Geoffrey C. CouperVice PresidentGlobal LNGSempra Energy Trading Corporation

David FrancoBusiness Developer, LNGConocoPhillips

Harvey L. HarmonConsultantShell US Gas & Power, LLC

Richard A. LammonsProject ManagerInternational GasChevronTexaco Overseas Petroleum

Geoff K. MitchellManaging DirectorMerrimack Energy Group

Raj K. MohindrooManagerLNG New VenturesConocoPhillips



45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-21


COLEADERSKenneth J. KonradSenior Vice PresidentAlaska GasBP Exploration Alaska Inc.

Joseph P. MarushackVice PresidentANS Gas DevelopmentConocoPhillips

Robert D. SchilhabManager

Alaska Gas DevelopmentExxonMobil Production Company

* *

Allan F. DriggsProject ManagerBasin StudiesAnadarko Petroleum Corporation


Angie L. KellyMarketing AnalystInternational CommercialDevelopmentAnadarko Petroleum Corporation

Richard J. LuckasavitchTechnical ManagerMackenzie Gas ProjectImperial Oil Resources

Michael J. McCarthyStaff Planning AdvisorAlaska Gas Development GroupExxonMobil Production Company

Colleen M. MukavitzCommercial ManagerAlaska Natural GasConocoPhillips

Randy J. OttenbreitDevelopment ExecutiveMackenzie Gas ProjectImperial Oil Resources

Steven P. SchwartzVenture ManagerNorth American West Coast LNGChevronTexaco Exploration & Production Company

David E. Van TuylCommercial Manager

Alaska GasBP America Production Inc.

SUPPLY TASK GROUP’S ARCTIC SUBGROUP

COLEADERSKenneth J. KonradSenior Vice PresidentAlaska GasBP Exploration Alaska Inc.

Joseph P. MarushackVice PresidentANS Gas DevelopmentConocoPhillips

Robert D. SchilhabManager

Alaska Gas DevelopmentExxonMobil Production Company

* *

Allan F. DriggsProject ManagerBasin StudiesAnadarko Petroleum Corporation


Angie L. KellyMarketing AnalystInternational CommercialDevelopmentAnadarko Petroleum Corporation


Michael J. McCarthyStaff Planning AdvisorAlaska Gas Development GroupExxonMobil Production Company

Colleen M. MukavitzCommercial ManagerAlaska Natural GasConocoPhillips

Randy J. OttenbreitDevelopment ExecutiveMackenzie Gas ProjectImperial Oil Resources

Steven P. SchwartzVenture ManagerNorth American West Coast LNGChevronTexaco Exploration & Production Company

David E. Van TuylCommercial Manager

Alaska GasBP America Production Inc.



Angie L. KellyMarketing AnalystInternational Commercial DevelopmentAnadarko Petroleum Corporation

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-22



TRANSMISSION & DISTRIBUTION TASK GROUPOF THE


CHAIR

Scott E. ParkerPresidentNatural Gas Pipeline Company of America



GOVERNMENT COCHAIR


SECRETARY

Benjamin A. Oliver, Jr.Senior Committee CoordinatorNational Petroleum Council

* * *

Stephen C. AllemanManagerBusiness OptimizationConocoPhillips

Steven D. BeckerVice PresidentGas DevelopmentTransCanada Pipelines Limited

James J. ClearyPresidentANR Pipeline Company

Dawn M. ConstantinLeaderTerm Fundamental AnalysisBP North America Gas & Power

Richard C. DanielSenior Vice PresidentEnCana Corporation


Jeffrey A. HolliganRegulatory ConsultantBP Energy Company


Terrance L. McGillVice PresidentCommerical Activity & Business DevelopmentEnbridge (U.S.) Inc.

Joseph L. RatiganVice President and Principal ConsultantPB Energy Storage Services, Inc.

Bradford D. ReeseSenior Vice President Northern Development and Co-Chief Executive Officer Foothills PipelineDuke Energy Gas Transmission – Canada


Terrance L. McGillVice PresidentCommerical Activity & Business DevelopmentEnbridge (U.S.) Inc.


Patrick A. JohnsonDirectorStrategy DepartmentEl Paso Corporation

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-23

TRANSMISSION & DISTRIBUTION TASK GROUP’S TRANSMISSION SUBGROUP

LEADERPatrick A. Johnson

DirectorStrategy DepartmentEl Paso Corporation

Michael K. AndersonAssociateStrategic InitiativesColumbia Gulf Transmission Co.

Nathan W. AndersonPrincipal StrategistEastern Pipeline GroupEl Paso Corporation

Sara J. BanaszakDirectorGas & PowerThe Petroleum Finance Company

Steven D. BeckerVice PresidentGas DevelopmentTransCanada Pipelines Limited

George R. FindlingManagerCommercial DevelopmentConocoPhillips Gas & Power


Carl W. LevanderVice PresidentRegulatory & Strategic InitiativesColumbia Gas Transmission Corp.

Francis C. PilleyDirectorGas StrategyTransCanada Pipelines Limited


Stephen T. RiesterTransportation AdvisorExxonMobil Gas Marketing Company


J. Gregory SnyderProject LeaderBusiness Planning and Market AnalysisDominion Energy, Inc.


TRANSMISSION & DISTRIBUTION TASK GROUP


James J. ClearyPresidentANR Pipeline Company

Dawn M. ConstantinLeaderTerm Fundamental AnalysisBP North America Gas & Power

George R. FindlingManagerCommercial DevelopmentConocoPhillips Gas & Power




Walter M. SimmonsVice President, Strategic PlanningGas Pipeline GroupKinder Morgan Inc.

Byron S. WrightVice-President

Strategy and Capacity PricingEl Paso Pipeline Group

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-24



Stephen T. RiesterTransportation AdvisorExxonMobil Gas Marketing Company


TRANSMISSION & DISTRIBUTION TASK GROUP’S DISTRIBUTION SUBGROUP

LEADERMark T. MaasselVice President

Regulatory and Governmental PolicyNiSource Inc.

Donald M. FieldExecutive Vice PresidentPeoples Energy Corporation

Randolph S. FriedmanManagerGas SupplyNW Natural Gas Company

Leonard J. PhillipsManagerGas OperationsMemphis Light, Gas & Water Division

Glen R. SchwalbachAssistant Vice PresidentCorporate PlanningWisconsin Public Service Corporation

Francis C. PilleyDirectorGas StrategyTransCanada Pipelines Limited


Carl W. LevanderVice PresidentRegulatory & Strategic InitiativesColumbia Gas Transmission Corp.


45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-25


TRANSMISSION & DISTRIBUTION TASK GROUP’S STORAGE SUBGROUP

LEADERRonald L. Brown

Vice PresidentStorage Management & System Design

Kinder Morgan Inc.

Mark D. CourtneyVice President, OriginationFalcon Gas Storage

Richard C. DanielSenior Vice PresidentGas StorageEnCana Corporation

Richard J. GentgesDirectorReservoir ServicesEl Paso Pipeline Group

David J. NightingaleGeneral Manager Storage Services & Producer Services andVice President Market Hub Partners, LPDuke Energy Gas Transmission




William A. TrapmannTeam LeaderNatural Gas Analysis TeamOffice of Oil and GasEnergy Information AdministrationU.S. Department of Energy

45393_Bp1_Bp26_AGS 3/5/04 1:17 PM Page B-26

INTEGRATED REPORT - ACRONYMS AND ABBREVIATIONS AC-1

AEO EIA’s Annual Energy Outlook

AFUE annual fuel utilization efficiency

AGA American Gas Association

ANGTS Alaska Natural Gas TransportationSystem

ANWR Arctic National Wildlife Refuge

API American Petroleum Institute

BCF billion cubic feet

BCF/D billion cubic feet per day

BLM U.S. Bureau of Land Management

Btu British thermal unit

CCGT combined-cycle gas turbines

CEQ Council on Environmental Quality

CERI Canadian Energy Research Institute

CFE Comision Federal de Electricidad(Mexico’s Federal ElectricityCommission)

CGPC Canadian Gas Potential Committee

CHP combined heat and power

CO2 carbon dioxide

COAs conditions of approval

CRE Comision Reguladora de Energia(Mexico’s Energy RegulatoryCommission)

CZM Coastal Zone Management

D&C drilling and completion

DOE U.S. Department of Energy

DOT U.S. Department of Transportation

E&P exploration and production

EEA Energy and Environmental Analysis, Inc.

EIA Energy Information Administration

EPA U.S. Environmental Protection Agency

ERCOT Electric Reliability Council of Texas

EUR estimated ultimate recovery

FERC Federal Energy Regulatory Commission

FPC Federal Power Commission (forerunner of FERC)

GDP gross domestic product

GMDFS EEA’s Gas Market Data and ForecastingSystem

GRI Gas Research Institute

GSR EEA’s Gas Supply Review

HSM EEA’s Hydrocarbon Supply Model

ACRONYMS AND ABBREVIATIONSINTEGRATED REPORT

45393_Acronyms_ags 3/5/04 1:18 PM Page 1

HVAC heating-ventilation-air conditioning systems

IHS IHS Energy Group

INGAA Interstate Natural Gas Association ofAmerica

IP initial production rate

ISTUM Industrial Sector Technology Use Model

JAS API’s Joint Association Survey

LDC local distribution company

LIHEAP Low Income Home Energy AssistanceProgram

LNG liquefied natural gas

MCF thousand cubic feet

MM million

MMBtu million British thermal units

MMCF million cubic feet

MMCF/D million cubic feet per day

MMS Minerals Management Service

MOU memorandum of understanding

MSC Multiple Services Contract

MTA million tons per annum

NAICS North American Industry ClassificationSystem

NEB National Energy Board of Canada

NEPA National Environmental Policy Act

NERC North American Electric ReliabilityCouncil

NGL natural gas liquid

NGPA National Gas Policy Act

NGV natural gas vehicle

NOx nitrogen oxides

NOAA National Oceanic and AtmosphericAdministration

NPC National Petroleum Council

NPRA National Petroleum Reserve, Alaska

NYMEX New York Mercantile Exchange

OCS Outer Continental Shelf

Pemex Petroleos Mexicanos

POLR provider of last resort

PSAC Petroleum Services Association ofCanada

psi pounds per square inch

PUC Public Utility Commission

PURPA Public Utility Regulatory Policies Act of 1978

quads quadrillion Btu

RACC refiner acquisition cost of crude oil

R&D research and development

ROE return on equity

R/P reserves to production (ratio)

RTOs Regional Transmission Organizations

RPS Renewable Portfolio Standards

SENER Secretaria de Energia (Mexico’s Energy Ministry)

SIC Standard Industrial Classification

SOLR supplier of last resort

SOx sulfur oxides

SO2 sulfur dioxide

TCF trillion cubic feet

USGS United States Geological Service

WCSB Western Canada Sedimentary Basin

WTI West Texas Intermediate crude oil

INTEGRATED REPORT - ACRONYMS AND ABBREVIATIONSAC-2

45393_Acronyms_ags 3/5/04 1:18 PM Page 2

INTEGRATED REPORT - GLOSSARY GL-1

AccessThe ability to drill and develop oil and natural gasresources, build associated production facilities, andconstruct transmission and distribution facilities oneither public and/or private land.

BasisThe difference in price for natural gas at two differ-ent geographical locations reported for the sametime period.

British Thermal Unit (Btu)A Btu is the amount of heat required to change thetemperature of one pound of water one degreeFahrenheit, and is the common energy measurementfor natural gas. One cubic foot of natural gas con-tains approximately 1,000 Btu.

Capacity, PeakingThe capacity of facilities or equipment normallyused to supply incremental gas or electricity underextreme demand conditions. Pipeline peakingcapacity is generally available for a limited numberof days at maximum flow rate while electric peakingcapacity is generally available whenever market priceconditions cover all variable costs and startupexpenses for such capacity.

Capacity, PipelineThe maximum physical throughput of natural gasover a specified period of time for which a pipelinesystem or portion thereof is designed or construct-ed, not limited by existing contract service condi-tions.

CitygateThe point at which interstate and intrastate pipelinessell and deliver natural gas to local distribution com-panies.

CogenerationThe production of electricity and useful thermalenergy from the same initial energy source. Naturalgas is a favored fuel for combined-cycle cogenerationunits, where it directly produces electricity from acombustion turbine and the resultant waste heat isconverted to steam for process use and for generat-ing electricity in a heat steam recovery generator(HSRG).

CommercialA sector of customers or service defined as non-manufacturing business establishments, includinghotels, motels, restaurants, wholesale businesses,retail stores, and health, social, and educational insti-tutions.

Compressed Natural Gas (CNG)Natural gas cooled to a temperature below 32°F andcompressed to a pressure ranging from 1,000 to3,000 pounds per square inch in order to allow thetransportation of large quantities of natural gas.

Cost RecoveryThe recovery of permitted costs, plus an acceptablerate of return, for an energy infrastructure projectsubject to rate regulations.

Cubic FootThe most common unit of measurement of gas vol-ume; the amount of gas required to fill a volume ofone cubic foot under standard conditions of tem-perature, pressure, and water vapor.

Distribution LineNatural gas pipeline system, typically operated by anLDC (local distribution company), for the deliveryof natural gas to end-users.

GLOSSARYINTEGRATED REPORT

45393_Glossary_ags 3/5/04 1:19 PM Page 1

ElasticityAn economic metric that typically measures themagnitude of changes in supply or demand as afunction of changes in price.

ElectricA sector of customers or service defined as genera-tion, transmission, distribution, or sale of electricenergy.

End-UserAn entity that actually consumes energy, as opposedto one who sells or re-sells it.

Federal Energy Regulatory Commission (FERC)The federal agency that regulates rates and terms ofservice for interstate gas pipelines and interstate gassales and for wholesale electric power transactionsunder federal energy statutes.

FeedstockThe use of one product as an ingredient to produceanother, such as using natural gas as a feedstock toproduce ammonia or methanol.

Firm CustomerA customer who has contracted for firm service.

Firm ServiceService offered to customers under schedules or con-tracts that anticipate no interruptions, except forforce majeure.

Fuel SwitchingSubstituting one fuel for another based on price andavailability. Large industries and power generatorsoften have the capability of using either oil or natu-ral gas to fuel their operation and of making theswitch on short notice.

Fuel-Switching CapabilityThe ability of an end-user to readily change fuel typeconsumed whenever a price or supply advantagedevelops for an alternative fuel.

GigawattsOne billion watts, or one thousand megawatts.

Gross Domestic Product (GDP)A dollar measure of total output of goods and serv-ices in the nation. Note that GDP can be measuredin nominal or current dollars or in real dollars,which removes the effects of inflation.

Henry HubA pipeline interchange near Erath, Louisiana, wherea number of interstate and intrastate pipelinesinterconnect through a header system operated by

Sabine Pipe Line. The standard delivery point forthe New York Mercantile Exchange natural gasfutures contract.

IndustrialA sector of customers or service defined as manufac-turing, construction, mining, agriculture, fishing,and forestry.

KilowattOne thousand watts.

Liquefied Natural Gas (LNG)The liquid form of natural gas, which has beencooled to a temperature –256°F or –161°C and ismaintained at atmospheric pressure. This liquefac-tion process reduces the volume of the gas byapproximately 600 times its original size.

Load ProfilesGas or electric power usage over a specific period oftime, usually displayed as a graphical plot.

Local Distribution Company (LDC)A company that obtains the major portion of its nat-ural gas revenues from the operations of a retail gasdistribution system and that operates no transmis-sion system other than incidental connections with-in its own or to the system of another company. AnLDC typically operates as a regulated utility within aspecified franchise area.

MegawattsOne million watts or one thousand kilowatts.

Marketer (natural gas)A company, other than the pipeline or LDC, thatbuys and resells gas or brokers gas for a profit.Marketers also perform a variety of related servic-es, including arranging transportation, monitor-ing deliveries and balancing. An independentmarketer is not affiliated with a pipeline, produc-er or LDC.

New FieldsA quantification of resources estimated to exist out-side of known fields on the basis of broad geologicknowledge and theory; in practical terms, these arestatistically determined resources likely to be discov-ered in additional geographic areas with geologiccharacteristics similar to known producing regions,but are untested by actual drilling.

Nominal DollarsDollars that have not been adjusted for inflation.

INTEGRATED REPORT - GLOSSARYGL-2


Nonconventional GasNatural gas produced from coalbeds, shales, and lowpermeability reservoirs. Development of thesereservoirs can require different technologies thanconventional reservoirs.

Peak-Day DemandThe maximum daily quantity of gas or power usedduring a specified 24-hour period and evaluatedover a specific period such as a year.

Peak ShavingMethods to reduce the peak demand for gas or elec-tricity or to meet those peaks with alternate deliv-ery sources or methods. Examples would beprice-controlled interruptions for demand reduc-tion or propane-air and distributed LNG for alter-nate resources.

Proved ReservesThe most certain of the resource base categoriesrepresenting estimated quantities that analysis ofgeological and engineering data demonstrate withreasonable certainty to be recoverable in futureyears from known reservoirs under existing eco-nomic and operating conditions. Generally, thesegas deposits have been “booked,” or accounted for asassets on the SEC financial statements of theirrespective companies.

Real DollarsDollars in a particular year that have been adjustedfor inflation to make financial comparisons in dif-ferent years more valid.

Refiner Acquisition Cost of Crude Oil (RACC) The cost of crude oil, including transportation andother fees paid by the refiner. The composite cost isthe weighted average of domestic and importedcrude oil costs. Note: The refiner acquisition costdoes not include the cost of crude oil purchased forthe Strategic Petroleum Reserve (SPR).

Regional Transmission Organization (RTO)A regulatory-recognized organization of electrictransmission owners, transmission users, and otherentities interested in coordinating transmissionplanning, expansion, and use on a regional andinterregional basis.

ResidentialThe residential sector is defined as private house-holds that consume energy primarily for space heat-ing, water heating, air conditioning, lightning,refrigeration, cooking, and clothes drying.

RevenueThe total amount of money received by a firm fromsales of its products and/or services.

ShipperOne who contracts with a pipeline for transporta-tion of natural gas and who retains title to the gaswhile it is being transported by the pipeline.

TerrawattsOne trillion watts.

WattThe common U.S. measure of electrical power.

INTEGRATED REPORT - GLOSSARY GL-3


national - npc · · 2006-02-28national petroleum council phone: (202) 393-6100 fax: (202)...

Documents