INTERNATIONAL ANTHROPOGENIC METHANE EMISSIONS:
ESTIMATES FOR 1990
REPORT TO CONGRESS
Editor: Michael J. Adler
U.S. Environmental Protection Agency
Office of Policy, Planning and Evaluation
January 1994
-------
This document has been reviewed in accordance with the U.S. Environmental
Protection Agency's and the Office of Management and Budget's peer and
administrative review policies and approved for publication. Mention of trade
names or commercial products does not constitute endorsement or
recommendation for use.
-------
TABLE OF CONTENTS
Acknowledgements xiii
EXECUTIVE SUMMARY ES-1
ES.1 GLOBAL METHANE EMISSIONS AND THE U.S. NATIONAL ACTION
PLAN ES-1
ES.2 KEY CONCLUSIONS OF THE REPORT ES-1
ES.3 OVERVIEW OF GLOBAL ANTHROPOGENIC CH4 EMISSIONS ES-2
ES.4 CURRENT ESTIMATES OF GLOBAL ANTHROPOGENIC METHANE
EMISSIONS ES-3
ES.4.1 Estimates by Source ES-3
ES.4.2 Regional Estimates ES-4
ES.4.3 Country Estimates ES-5
ES.5 REDUCING UNCERTAINTIES IN METHANE EMISSION ESTIMATES . . ES-7
ES.6 REFERENCES ES-7
1. INTRODUCTION 1-1
1.1 BACKGROUND 1-1
1.1.1 The Importance of Methane to Potential Greenhouse Warming .... 1-1
1.1.2 An Overview of Methane 1-2
1.1.3 The Atmospheric History and Geographic Distribution of Methane . . 1-4
1.1.4 The Role of Methane in Global Climate Change 1-8
1.1.5 The Methane Budget 1-10
1.1.6 Summary 1-12
1.2 OVERVIEW OF THIS REPORT 1-13
1.3 REFERENCES 1-14
2. METHANE EMISSIONS FROM THE DIGESTIVE PROCESSES OF LIVESTOCK .... 2-1
2.1 SUMMARY 2-1
2.2 BACKGROUND 2-1
2.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 2-4
2.3.1 Implementation of the Methodology 2-5
2.4 RESULTS 2-24
2.4.1 Cattle 2-24
2.4.2 Other Livestock 2-25
2.5 TRENDS 2-25
2.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 2-29
2.7 CONCLUSIONS 2-29
2.8 REFERENCES 2-30
APPENDIX A: EMISSION FACTORS FOR BUFFALO 2-34
A.1 FEED INTAKES 2-34
A.2 METHANE EMISSION FACTORS 2-38
A.3 REFERENCES 2-40
APPENDIX B: DERIVATION OF THE RELATIONSHIP BETWEEN NE AND DE . 2-42
B.1 REFERENCES 2-44
-------
TABLE OF CONTENTS (Continued)
3. METHANE EMISSIONS FROM RICE CULTIVATION 3-1
3.1 SUMMARY 3-1
3.2 BACKGROUND 3-1
3.2.1 Methane Emission Rates From Rice Fields 3-2
3.2.2 Factors Affecting Methane Emissions 3-3
3.2.3 Regional Estimates of Methane Emissions 3-13
3.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 3-16
3.3.1 Summary of Recommended Method 3-17
3.4 RESULTS 3-20
3.5 TRENDS 3-20
3.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 3-21
3.6.1 Specific Estimation Problems 3-21
3.6.2 Verification of Country Inventories 3-21
3.7 CONCLUSIONS 3-21
3.8 REFERENCES 3-25
4. METHANE EMISSIONS FROM ANTHROPOGENIC BIOMASS BURNING 4-1
4.1 SUMMARY 4-1
4.2 BACKGROUND 4-2
4.2.1 Introduction 4-2
4.2.2 Varieties of Biomass Burning 4-7
4.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 4-16
4.3.1 Review of IPCC-OECD Methodology 4-16
4.3.2 Methodology Used in this Chapter 4-18
4.4 RESULTS 4-22
4.5 TRENDS 4-24
4.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 4-25
4.7 CONCLUSIONS 4-27
4.8 REFERENCES 4-28
5. METHANE EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS 5-1
5.1 SUMMARY 5-1
5.2 BACKGROUND 5-1
5.2.1 Overview of Oil and Natural Gas Systems 5-1
5.2.2 Sources of Methane Emissions in Oil and Natural Gas Systems . . . 5-5
5.2.3 Emission Data 5-9
5.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 5-10
5.3.1 Definition of Regions 5-14
5.3.2 Emission Factors 5-14
5.3.3 Activity Levels 5-18
5.4 RESULTS 5-20
5.5 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 5-20
5.6 TRENDS 5-22
5.7 CONCLUSIONS 5-22
5.8 REFERENCES 5-23
-------
TABLE OF CONTENTS (Continued)
APPENDIX A: DERIVATION OF NON-U.S. EMISSION FACTORS FOR FUEL
COMBUSTION 5-25
APPENDIX B: EMISSION ESTIMATES BY REGION 5-28
6. METHANE EMISSIONS FROM THE COAL FUEL CYCLE 6-1
6.1 SUMMARY 6-1
6.2 BACKGROUND 6-3
6.2.1 Processes of Methane Generation and Emission in Coal 6-3
6.2.2 Key Emission Processes of the Coal Fuel Cycle 6-4
6.2.3 Previous Methane Emission Studies 6-5
6.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 6-8
6.3.1 Underground Mining 6-8
6.3.2 Treatment of Methane Utilization 6-14
6.3.3 Surface Mining 6-15
6.3.4 Post-Mining Activities 6-18
6.3.5 Coal Combustion 6-21
6.3.6 Treatment of Other Coal Fuel Cycle Emissions 6-21
6.4 RESULTS 6-22
6.4.1 General Findings 6-22
6.4.2 Country-Specific Information 6-25
6.5 TRENDS 6-33
6.5.1 General Factors Influencing Emissions 6-33
6.5.2 National Trends 6-34
6.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 6-37
6.6.1 Data Uncertainties 6-37
6.6.2 Methodological Uncertainties 6-38
6.6.3 Research Needs 6-40
6.7 CONCLUSIONS 6-41
6.8 REFERENCES 6-41
7. MINOR INDUSTRIAL SOURCES OF METHANE 7-1
7.1 SUMMARY .7-1
7.2 BACKGROUND 7-1
7.2.1 Recent Research 7-2
7.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 7-6
7.3.1 Peat Mining and Geothermal Power Production 7-7
7.3.2 Coke Production and Iron-and-Steel Production Processes 7-7
7.3.3 Miscellaneous Industrial Processes 7-8
7.3.4 Paper and Printing Processes 7-8
7.3.5 Chemical Manufacturing Processes 7-8
7.4 RESULTS 7-8
7.5 TRENDS 7-10
7.6 CONCLUSIONS 7-12
7.7 REFERENCES 7-13
-------
TABLE OF CONTENTS (Continued)
8. METHANE EMISSIONS FROM LANDFILLS AND OPEN DUMPS 8-1
8.1 SUMMARY 8-1
8.2 BACKGROUND 8-1
8.2.1 Methane Production from the Anaerobic Decomposition of Waste . . 8-1
8.2.2 Factors Affecting CH4 Potential from Landfilled Waste 8-3
8.2.3 Determination of the CH4 Potential of MSW 8-4
8.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 8-5
8.3.1 Estimation of Waste-Generation Rates 8-5
8.3.2. AEERL Regression Model Methodology 8-11
8.3.3 Draft IPCC/OECD Methodology 8-12
8.4 RESULTS 8-13
8.5 TRENDS 8-13
8.5.1 Europe 8-14
8.5.2 United States and Canada 8-14
8.5.3 Asia 8-15
8.5.4 Latin America and the Caribbean Islands 8-15
8.5.5 Africa and Middle East 8-15
8.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 8-15
8.6.1 Quantity and Composition of Waste-in-Place • • • • 8-"l5
8.6.2 CH4 Production of Waste in Open Dumps and Industrial Landfills . 8-16
8.6.3 Quantity of CH4 Emitted to the Atmosphere 8-16
8.7 CONCLUSIONS 8-17
8.8 REFERENCES 8-17
9. METHANE EMISSIONS FROM LIVESTOCK MANURE 9-1
9.1 SUMMARY 9-1
9.2 BACKGROUND 9-1
9.2.1 Characteristics of the Manure 9-2
9.2.2 Manure Management System 9-2
9.2.3 Climate 9-3
9.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS 9-3
9.3.1 Volatile Solids (VS) Production 9-5
9.3.2 Maximum Methane-Producing Capacity (B0) 9-6
9.3.3 Descriptions of Manure Management Systems 9-7
9.3.4 Methane Conversion Factors (MCFs) 9-9
9.3.5 Livestock Manure Management System Usage (MS%) 9-10
9.4 RESULTS 9-14
9.5 TRENDS 9-14
9.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 9-17
9.7 CONCLUSIONS 9-19
9.8 REFERENCES 9-20
10. METHANE EMISSIONS FROM WASTEWATER 10-1
10.1 SUMMARY 10-1
10.2 BACKGROUND 10-1
IV
-------
TABLE OF CONTENTS (Continued)
10.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS ... 10-3
10.3.1 Assumptions and Data Used to Eistimate Emissions 10-4
10.4 RESULTS 10-11
10.5 TRENDS 10-11
10.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES 10-14
10.6.1 Activity Data: Quantities of Wastewater Flows 10-14
10.6.2 Methane Yield per Unit BOD Loading 10-15
10.6.3 Efficiency and Output of Wastewater Treatment Facilities 10-15
10.6.4 BOD Values and Other Physical and Chemical Data 10-15
10.7 CONCLUSIONS : 10-16
10.8 REFERENCES 10-16
11. VERIFICATION OF METHANE EMISSIONS 11-1
11.1 INTRODUCTION 11-1
11.2 VERIFICATION METHODOLOGIES 11-3
11.2.1 Direct Verification 11-4
11.2.2 Indirect Verification 11-8
11.3 CONCLUSIONS 11-13
11.4 REFERENCES 11-14
-------
LIST OF FIGURES
EXECUTIVE SUMMARY
FIGURE ES-1: GLOBAL ANTHROPOGENIC'METHANE EMISSIONS BY
SOURCE ES-4
FIGURE ES-2: GLOBAL ANTHROPOGENIC METHANE EMISSIONS BY
REGION ES-5
FIGURE ES-3: METHANE EMISSIONS OF THE FIVE LARGEST EMITTING
COUNTRIES ES-6
1. INTRODUCTION
FIGURE 1-1: MEASUREMENTS OF GLOBAL METHANE CONCENTRATIONS . 1-6
FIGURE 1-2: ATMOSPHERIC METHANE AND AIR TEMPERATURE
RECORD DETERMINED FROM VOSTOK ICE CORE 1-7
FIGURE 1-3: SPATIAL AND TEMPORAL DISTRIBUTION OF ATMOSPHERIC
METHANE 1-9
2. METHANE EMISSIONS FROM THE DIGESTIVE PROCESSES OF LIVESTOCK
FIGURE B-1: NRC NE:DE RELATIONSHIP 2-43
FIGURE B-2: LINEAR EXTRAPOLATION OF THE NRC NE:DE RELATIONSHIP 2-44
3. METHANE EMISSIONS FROM RICE CULTIVATION
FIGURE 3-1: SCHEME OF PRODUCTION, REOXIDATION, AND EMISSION
OF CH4 IN A PADDY FIELD 3-10
FIGURE 3-2: RELATIONSHIP BETWEEN THE FLUX AND SOIL
TEMPERATURE IN A PADDY FIELD IN SICHUAN, CHINA 3-12
FIGURE 3-3: TIME SERIES OF GLOBAL CH4 EMISSIONS FROM RICE
AGRICULTURE 3-23
FIGURE 3-4: PER-CAPITA EMISSIONS OF METHANE (Tg CH4 PER BILLION
POPULATION (BP)) FROM RICE FIELDS IN CHINA DURING
THE LAST CENTURY 3-24
4. METHANE EMISSIONS FROM ANTHROPOGENIC BIOMASS BURNING
FIGURE 4-1: CARBONACEOUS PRODUCTS OF BIOMASS BURNING 4-5
FIGURE 4-2: ESTIMATES OF METHANE EMISSIONS FROM GLOBAL
BIOMASS BURNING 4-6
FIGURE 4-3: TYPES OF BIOMASS BURNING 4-8
FIGURE 4-4: ESTIMATING METHANE EMISSIONS FROM BIOMASS
BURNING 4-17
5. METHANE EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS
FIGURE 5-1: STAGES IN THE OIL & NATURAL GAS SYSTEM 5-2
VI
-------
LIST OF FIGURES (Continued)
6. METHANE EMISSIONS FROM THE COAL FUEL CYCLE
FIGURE 6-1: FLOW CHART FOR ESTIMATING METHANE EMISSION FROM
THE COAL CYCLE 6-9
FIGURE 6-2: 1990 METHANE EMISSIONS FROM THE COAL FUEL CYCLE . . 6-26
10. METHANE EMISSIONS FROM WASTEWATER
FIGURE 10-1: WASTEWATER TREATMENT SYSTEMS AND
METHANE PRODUCTION
10-2
VII
-------
LIST OF TABLES
ES. EXECUTIVE SUMMARY
TABLE ES-1: GLOBAL ANTHROPOGENIC METHANE EMISSIONS BY
SOURCE ES-9
TABLE ES-2: GLOBAL ANTHROPOGENIC METHANE EMISSIONS BY
REGION ES-10
TABLE ES-3: METHANE EMISSIONS OF THE FIVE LARGEST EMITTING
COUNTRIES ES-11
1. INTRODUCTION
TABLE 1-1: SUMMARY OF ATMOSPHERIC PARAMETERS OF METHANE
AND OTHER KEY GREENHOUSE GASES INFLUENCED BY
HUMAN ACTIVITIES 1-3
TABLE 1-2: ESTIMATED SOURCES AND SINKS OF METHANE 1-5
2. METHANE EMISSIONS FROM THE DIGESTIVE PROCESSES OF LIVESTOCK
TABLE 2-1: CATTLE POPULATIONS: 1990 2-6
TABLE 2-2: REPRESENTATIVE CATTLE TYPES 2-7
TABLE 2-3: DATA ON REPRESENTATIVE CATTLE TYPES 2-11
TABLE 2-4: NET ENERGY REQUIREMENTS FOR REPRESENTATIVE
CATTLE TYPES (MJ/day) 2-14
TABLE 2-5: GROSS ENERGY INTAKE FOR REPRESENTATIVE CATTLE
TYPES 2-18
TABLE 2-6: METHANE EMISSION FACTORS FOR CATTLE 2-21
TABLE 2-7: EMISSION FACTORS FOR LIVESTOCK 2-23
TABLE 2-8: REGIONAL AND GLOBAL CATTLE EMISSIONS: 1990 2-26
TABLE 2-9: GLOBAL LIVESTOCK METHANE EMISSIONS 2-27
TABLE A-1: DATA ON BUFFALO SIZE 2-35
TABLE A-2: DATA ON REPRESENTATIVE BUFFALO TYPES 2-37
TABLE A-3: NET ENERGY REQUIREMENTS FOR REPRESENTATION
BUFFALO TYPES (MJ/day) 2-37
TABLE A-4: GROSS ENERGY INTAKE FOR REPRESENTATIVE BUFFALO
TYPES 2-39
TABLE A-5: METHANE EMISSION FACTORS FOR BUFFALO 2-39
3. METHANE EMISSIONS FROM RICE CULTIVATION
TABLE 3-1: METHANE FLUX MEASUREMENTS FROM RICE PADDIES 3-4
TABLE 3-2: SEASONAL FACTORS FOR METHANE EMISSIONS FROM
RICE FIELDS 3-15
TABLE 3-3: ESTIMATES OF GLOBAL METHANE EMISSIONS FROM RICE
AGRICULTURE 3-16
TABLE 3-4: ESTIMATES OF METHANE EMISSIONS BY COUNTRY FOR
1990 USING FOUR SETS OF ASSUMPTIONS (Tg CH4/yr) 3-17
TABLE 3-5: RESULTS: AVERAGE ANNUAL (1990) METHANE EMISSION
ESTIMATES AND AREA, SEASON, AND FLUX DATA FOR
MAJOR RICE-PRODUCING COUNTRIES 3-22
VIII
-------
LIST OF TABLES (Continued)
4. METHANE EMISSIONS FROM ANTHROPOGENIC BIOMASS BURNING
TABLE 4-1: ESTIMATES OF BIOMASS BURNED GLOBALLY (Pg dm/yr) 4-3
TABLE 4-2: BIOMASS-BURNING METHANE EMISSION FACTORS 4-20
TABLE 4-3: DEFAULT VALUES OF PARAMETERS FOR USE IN
CALCULATING METHANE EMISSIONS FROM BIOMASS
BURNING 4-21
TABLE 4-4: ESTIMATES OF METHANE EMISSIONS FROM
ANTHROPOGENIC BIOMASS BURNING (Gg CH4/year) 4-23
5. METHANE EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS
TABLE 5-1: MAJOR OIL- AND GAS-PRODUCING AND OIL-REFINING
COUNTRIES 5-3
TABLE 5-2: EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS 5-8
TABLE 5-3: SUMMARY OF EMISSION FACTORS 5-11
TABLE 5-4: U.S. EMISSION FACTORS 5-15
TABLE 5-5: EASTERN EUROPE AND THE FORMER SOVIET UNION -
EMISSION FACTORS 5-15
TABLE 5-6: WESTERN EUROPE - EMISSION FACTORS 5-16
TABLE 5-7: OTHER OIL-EXPORTING COUNTRIES - EMISSION FACTORS . 5-17
TABLE 5-8: REST OF THE WORLD - EMISSION FACTORS 5-18
TABLE 5-9: ACTIVITY LEVELS FOR OIL AND NATURAL GAS SYSTEMS IN
1990 (petajoules/yr) 5-19
TABLE 5-10: GLOBAL EMISSIONS BY REGION AND INDUSTRY SEGMENT
(Tg of CH4 in 1990) 5-21
TABLE 5-11: GLOBAL EMISSIONS BY KEY COUNTRIES (Tg of CH4 in 1990) 5-21
TABLE A-1: DERIVATION OF STATIONARY-SOURCE, GAS-COMBUSTION
EMISSION FACTORS 5-26
TABLE A-2: DERIVATION OF STATIONARY-SOURCE, OIL-COMBUSTION
EMISSION FACTORS 5-27
TABLE B-1: U.S EMISSIONS 5-29
TABLE B-2: EMISSIONS FROM EASTERN EUROPE AND FORMER
SOVIET UNION : 5-30
TABLE B-3: EMISSIONS FROM WESTERN EUROPE 5-31
TABLE B-4: EMISSIONS FROM OTHER OIL EXPORTING COUNTRIES 5-32
TABLE B-5: EMISSIONS FROM THE REST OF THE WORLD 5-33
TABLE B-6: WORLD EMISSIONS 5-34
6. METHANE EMISSIONS FROM THE COAL FUEL CYCLE
TABLE 6-1: EMISSION ESTIMATES FROM SELECTED STUDIES 6-6
TABLE 6-2: 1990 COAL PRODUCTION AND CONSUMPTION DATA (million
tonnes) 6-10
TABLE 6-3: ESTIMATED UNDERGROUND MINING EMISSION FACTORS
FOR SELECTED COUNTRIES 6-13
TABLE 6-4: SELECTED COAL COMBUSTION EMISSION FACTORS, BY
SOURCE (nrVtonne of coal consumed) 6-21
ix
-------
LIST OF TABLES (Continued)
TABLE 6-5: ESTIMATED 1990 METHANE EMISSIONS FROM THE COAL
FUEL CYCLE (Tg) 6-24
TABLE 6-6: 1990 ESTIMATED METHANE EMISSIONS FROM THE COAL
FUEL CYCLE FOR THE TOP-10 EMITTING COUNTRIES 6-25
TABLE 6-7: METHANE RECOVERY BY D EG AS I Fl CATION SYSTEMS FOR
KEY COUNTRIES (Tg) 6-27
TABLE 6-8: SUMMARY OF GLOBAL AVERAGE EMISSION FACTORS FOR
THE COAL FUEL CYCLE 6-38
7, MINOR INDUSTRIAL SOURCES OF METHANE
TABLE 7-1: RANKED LISTING OF SIGNIFICANT INDUSTRIAL SOURCES
OF METHANE EMISSIONS IN THE UNITED STATES 7-3
TABLE 7-2: 1990 GLOBAL CH4 EMISSIONS ESTIMATED FOR SELECTED
MINOR SOURCES 7-5
TABLE 7-3: SELECTED MINOR SOURCE CATEGORIES AND EMISSION
ESTIMATES 7-6
TABLE 7-4: METHANE EMISSIONS FROM SELECTED MINOR SOURCES
(1,000 tonnes/yr) 7-9
8. METHANE EMISSIONS FROM LANDFILLS AND OPEN DUMPS
TABLE 8-1: TOTAL GLOBAL AND U.S. ESTIMATES OF METHANE
EMISSIONS FROM LANDFILLS AND OPEN DUMPS,
ACCORDING TO VARIOUS METHODS 8-2
TABLE 8-2: REFERENCES USED 8-6
TABLE 8-3: WASTE GENERATION RATES AND METHANE EMISSION
ESTIMATES USING IPCC/OECD AND AEERL METHODS 8-7
TABLE 8-4: ESTIMATES OF METHANE EMISSIONS FROM LANDFILLS
FROM OECD/IPCC AND EPA/AEERL METHODOLOGIES 8-17
9. METHANE EMISSIONS FROM LIVESTOCK MANURE
TABLE 9-1: LIVESTOCK MANURE PRODUCTION DATA FOR DEVELOPED
COUNTRIES 9-6
TABLE 9-2: LIVESTOCK MANURE PRODUCTION DATA FOR
DEVELOPING COUNTRIES 9-7
TABLE 9-3: B0 VALUES FOR DEVELOPED AND DEVELOPING
COUNTRIES 9-8
TABLE 9-4: METHANE CONVERSION FACTORS (MCFs) FOR LIVESTOCK
MANURE SYSTEMS 9-11
TABLE 9-5: LIVESTOCK MANURE SYSTEM USAGE 9-12
TABLE 9-6: METHANE EMISSIONS BY ANIMAL TYPE AND REGION FOR
1990 (Tg CH4) 9-15
TABLE 9-7: METHANE EMISSIONS BY REGION AND SYSTEM FOR 1990
(Tg CH4) 9-15
TABLE 9-8: METHANE EMISSIONS FOR MAJOR COUNTRIES FOR 1990
(Tg CH4) 9-16
-------
LIST OF TABLES (Continued)
TABLE 9-9: BASE-, HIGH-, AND LOW-CASE EMISSION ESTIMATE
ASSUMPTIONS 9-18
TABLE 9-10: EMISSION ESTIMATE RANGES FOR 1990 (Tg CH4) 9-19
10. METHANE EMISSIONS FROM WASTEWATER
TABLE 10-1: BIOCHEMICAL OXYGEN DEMAND (BOD) ESTIMATES FOR
VARIOUS INDUSTRIAL WASTEWATERS
TABLE 10-2: SELECTED WASTEWATER TREATMENT METHODS IN THE
UNITED STATES
TABLE 10-3: ESTIMATES OF GLOBAL AND COUNTRY-SPECIFIC
METHANE EMISSIONS FROM DOMESTIC WASTEWATER,
1990 (Tg/yr)
TABLE 10-4: ESTIMATE OF GLOBAL METHANE EMISSIONS FROM
INDUSTRIAL WASTEWATER 1990 (Tg/yr)
10-5
10-11
10-12
10-13
XI
-------
-------
ACKNOWLEDGEMENTS
This report represents the combined efforts of a large number of people who
contributed their skill, knowledge and sometimes great perseverance to see a long and
demanding process through to a successful completion.
EPA would first like to thank the lead authors of the individual source chapters. Each
has made major contributions to methane research and/or estimation methodologies, as
reflected in the high quality of this report. The lead authors are recognized as follows:
Introduction
Digestive Processes of Livestock
Rice Cultivation
Biomass Burning
Oil and Natural Gas Systems
Coal Fuel Cycle
Minor Industrial Sources
Landfills and Open Dumps
Livestock Manure
Waste water
Verification
Barbara Braatz
Michael Gibbs
Don Johnson
Aslam Khalil
Martha Shearer
Dilip Ahuja
Michael Gibbs
Craig Ebert
Dina Kruger
David Kirchgessner
Lee Beck
Steve Piccot
Sharon Kersteter
Susan Thorneloe1
Michiel Doom
Morton Barlaz
Jonathan Woodbury
Andy Hashimoto
Susan Thorneloe
Michiel Doom
Barbara Braatz
Inez Fung
Edward Dlugokencky
Special thanks goes to Barbara Braatz of ICF Incorporated who, in addition to serving
as a lead author, provided technical reviews and managed the report preparation and
production process. She approached each of these tasks with accuracy and thoroughness.
1 Susan Thorneloe has our highest esteem for having contributed her two chapters despite an extended illness.
She has made a remarkable recovery and has resumed her full responsibilities at EPA.
XIII
-------
She was ably assisted by Mary DePasquale, Megan Rush, Froilan Rosqueta, Pradeep
Hathiramani, and My Ton of ICF Incorporated, and Joan O'Callaghan of the Bruce Company.
Craig Ebert of ICF provided helpful oversight and direction.
A number of reviewers from industry, research institutions, and government contributed
valuable comments. Our thanks to the following individuals:
Meinrat Andreae (Max Planck Institute)
Natalia Andronova (Univ. of Illinois)
Dominique Bachelet (ManTech)
Morton Barlaz (NC State)
Jan Berdowski (TNO)
Jean Bogner (Argonne)
Clint Burklin (Radian)
William Chandler (Battelle)
Robert Delmas (Universite Paul Sabatier)
Philip Fearnside (INPA)
Ken Gregory (Hoskyns Group)
Robert Harriss (Univ. of New Hampshire)
Paul Hertz (Thermochem)
Kathleen Hogan (U.S. EPA)
Michael Jefferson (WEC)
Donald Johnson (Colorado State Univ.)
Chris Justice (NASA)
Daniel Kammen (Harvard)
Eugene Khartukov (Russian Ministry of
Foreign Affairs)
Penny Lassiter (U.S. EPA)
Joel Levine (NASA)
Elaine Mathews (NASA)
Katsu Minami (NIAES)
Catherine Mitchell (University of Sussex)
Heinz-Ulrich Neue (IRRI)
Jim Penman (UK Dept. of the Environment)
Helen Plume (New Zealand Ministry for the
Environment)
Keith Richards (UK Atomic Energy
Authority)
Kurt Roos (U.S. EPA)
L. M. Safley (NC State)
Peter Sage (British Coal)
R. L. Sass (Rice Univ.)
Masaki Shibata (NIAI)
Kirk Smith (East-West Center)
Susan L. Stefanek (U.S. EPA)
M. Tanaka (Inst. of Public Health)
Anne Thompson (NASA)
Alfred Vervaert (U.S. EPA)
Ming-Xing Wang (Inst. for Atmospheric
Physics)
Alan Williams (University of Leeds)
D. J. Williams (CSIRO)
Ted Williams (U.S. DOE)
Hugh Wise (U.S. EPA)
XIV
-------
EXECUTIVE SUMMARY
The Clean Air Act Amendments of 1990 (P.L. 101-549, Sec. 603) requires the U.S.
Environmental Protection Agency to prepare five reports on methane (CH4) emissions and
mitigation options. The topics to be covered by these reports include domestic and global
inventories of anthropogenic (Sec.603 b. and c.1) and natural sources (Sec.603 d.) of CH4,
and options for mitigating these emissions domestically and internationally (Sec.603 a. and
c.2).
This report addresses the request for a global inventory of the key anthropogenic
sources of CH4. As stated in the Clean Air Act Amendments, U.S. EPA is to prepare "a report
on methane emissions from countries other than the United States." (Sec. 603.C.1).
ES.1 GLOBAL METHANE EMISSIONS AND THE U.S. NATIONAL ACTION PLAN
On October 19, 1993, President Clinton issued the U.S. Climate Change Action Plan.
The Action Plan initiates specific measures to reduce greenhouse gas (GHG) emissions,
including methane, in the United States to 1990 levels by the year 2000. This report makes
several contributions to the general goals of the Action Plan:
The report provides information on inventory methods useful in measuring reductions
in methane emissions achieved under the Climate Change Action Plan. Accurate
information on inventory methods, including their limitations and research
requirements, will enable the United States to respond with appropriate policies in the
future.
The report also provides a baseline with which to assess trends in methane emissions
in other countries. Knowing where we stand in relation to other countries can help us
provide leadership in efforts to reduce global emissions. Having the necessary tools
allows us to track, analyze, and verify global emission data in the future.
The report contributes to international emission reductions by building confidence in
the state-of-the-art emission estimation methods. With more effective methods,
countries can prepare better national GHG inventories, contributing, in turn, to ensuring
that subsequent climate change actions plans are as effective as possible.
The report identifies general areas of opportunity for reducing methane emissions by
entities interested in joint implementation projects. It may also assist the International
Negotiating Committee and the Global Environment Facility in identifying and targeting
priority sources for financial support under the FCCC (Framework Convention on
Climate Change).
ES.2 KEY CONCLUSIONS OF THE REPORT
This report reaches several conclusions pertinent to U.S. climate change policy:
Page ES-1
-------
Total global emissions of CH4 from anthropogenic sources were estimated to be
slightly under 360 teragrams (Tg)1 in 1990, the same as the best-guess value of the
Intergovernmental Panel on Climate Change (IPCC, 1992). This is a favorable
comparison, but the correspondence belies the many differences in specific source
estimates.
The United States, with about 8% of global anthropogenic methane emissions, is
among the five countries with the highest emissions. The others are China (14%), the
former Soviet Union (12%), India (8%), and Brazil (4%).
The major sources of CH4 globally are livestock (80 Tg), rice cultivation (65 Tg), natural
gas and oil systems (51 Tg), and biomass burning (48 Tg), together contributing
approximately 69% of total anthropogenic CH4 emissions. Five sources, liquid wastes
(35 Tg), coal (30 Tg), landfills (27 Tg), livestock manure (14 Tg), and minor industrial
sources (4 Tg), comprise the remaining 31 % of emissions.
As the above figures indicate, energy and agricultural activities are thus the two key
sources of global anthropogenic CH4 emissions. Industrialized countries and countries
with economies in transition are major emitters of energy-related CH4 emissions, while
developing countries are major emitters of agriculture-related emissions.
Global anthropogenic CH4 emissions are expected to continue growing in the
foreseeable future largely because of population growth and economic development.
However, according to U.S. EPA's report to Congress on Options for Reducing
Methane Emissions Internationally (1993a), significantly slowing or halting the
increase in methane concentrations in the atmosphere appears to be achievable, cost-
effectively, over the next 10 years.
Of the four largest sources of CH4, natural gas and oil systems appear to have the
largest emission reduction potentials in the near term. Higher emission reductions
appear possible from the smaller sources, particularly landfills and coal mining. Large
opportunities to achieve reductions in specific countries, such as the countries of the
former Soviet Union, also exist (U.S. EPA, 1993a).
The quality of CH4 emission estimates varies considerably by source and by country.
Therefore, to accurately review progress toward national reduction goals under the
FCCC, the Conference of the Parties will need improved emission factors and activity
levels, national inventories, and data verification processes.
ES.3 OVERVIEW OF GLOBAL ANTHROPOGENIC CH4 EMISSIONS
Methane is a trace gas found in the atmosphere that results from various man-made
and natural activities. The primary sources of CH4 influenced by human activities
("anthropogenic" sources of CH4) account for about 70% of global emissions (Fung et al.,
1991; and IPCC, 1992). These sources include digestive processes of livestock, rice
cultivation, biomass burning, oil and natural gas systems, the coal fuel cycle, minor industrial
sources, landfills and open dumps, livestock manure, and waste water. Natural sources,
1 Teragram = 10s metric tonnes = 1012 grams.
Page ES-2
-------
which account for the remaining emissions, include wetlands, oceans and freshwater bodies,
termites, wildfires, and CH4 hydrates.
Methane concentrations are increasing roughly in parallel with world population growth
and associated industrial and agricultural activities. Concentrations have more than doubled
over the last 200 years and are currently rising by about 0.6% annually. In recent years, this
rate of increase has slowed slightly, however, it is not certain whether this slowing trend will
continue.
Increased CH4 concentrations are cause for significant concern because methane is a
potent greenhouse gas. Its direct radiative effect, averaged over a 100-year time frame, is at
least 11 times more effective at trapping heat in the atmosphere than carbon dioxide, the most
abundant greenhouse gas. When indirect effects2 are included, CH4 has approximately 20
times the global warming potential (GWP) of carbon dioxide (IPCC, 1992).3 The contribution
of anthropogenic CH4 to potential global warming, including its direct and indirect effects, is
approximately 18%. This share is uncertain, however, because of uncertainties in the
contribution of other gases, particularly CFCs.
The atmospheric lifetime of CH4 is relatively short, about 10 years (Lelieveld and
Crutzen, 1993), compared with about 120 years for carbon dioxide (IPCC, 1990). This short
lifetime means that programs and policies to reduce CH4 can help to mitigate the rate of
climate change faster than reductions in emission of carbon dioxide and other longer-lived
greenhouse gases. However, the cost of such actions must still be compared with other
options, such as energy-efficiency improvements to reduce carbon dioxide, to determine the
most cost-effective approaches available.
ES.4 CURRENT ESTIMATES OF GLOBAL ANTHROPOGENIC METHANE EMISSIONS
This report's best estimate of 354 Tg of methane is, given the large range of
uncertainty in the estimate (274-477 Tg/yr), essentially the same as the IPCC (1992) estimate
of 360 Tg. The similar estimates, however, belie important differences in individual source
categories. These are caused in part by different estimation methods and different source
definitions. For example, rice cultivation and biomass burning are between 8 and 20 percent
higher in this report than in the IPCC report. In the IPCC report, livestock manure, coal, and
oil and natural gas are higher than in this report, 80% higher in the case of livestock and an
uncertain but probably significant percent higher for the latter categories. In addition, the U.S.
EPA report covers industrial waste water emissions, which the IPCC report does not.
•v
ES.4.1 Estimates by Source
As shown in Figure ES-1 below, livestock (80 Tg), rice cultivation (65 Tg), natural gas
and oil systems (51 Tg), and biomass burning (48 Tg) are the most significant sources of CH4
2 The indirect effects of methane result from its role in atmospheric chemical reactions that increase the lifetimes
and abundances of methane and other greenhouse gases.
3 The GWP is a measure of the effect that one kilogram of methane would have on global warming, compared with
an equivalent amount of carbon dioxide, over a specified period of time.
Page ES-3
-------
globally. Together they account for about 69% of total anthropogenic CH4 emissions. The
other five sources contribute about 110 Tg. (See Table ES-1 at end of executive summary.)
FIGURE ES-1
GLOBAL ANTHROPOGENIC METHANE EMISSIONS BY SOURCE (1990)
(Midpoints of ranges given above bars)
80
65
O
IUW •
90 •
on ,
70 •
Rn
50 •
30 •
20 •
1O •
0.
ivestock
^ " '
i , . . ,
!»! • , „ , '"
CD
O
£
51
".'• '• • •-
u ,:,i, j
CO
(D
C35
48
I
I
Biomass
Burning
High Estimal
/
/ / Midpoint
/ /
35 30 27 // /
^-" / Low Estimat
»tw «UK>. < V. B "/• ' , ^1 /
— = , X 14
i 4-
11 ^ 1 11
(See notes for Table ES-1)
U.S. EPA (1993a) reports that significant global emission reductions are possible in oil
and natural gas systems, coal mining, livestock, livestock manure, landfills, and wastewater
treatment. The exact amount of possible reductions per source tend to be uncertain,
however. They vary from as little as 20% to as much as 90% per source, depending on the
availability of technology and capital, as well as other factors. Currently, the reduction
potential for rice cultivation appears to be relatively low, and that of biomass burning is
uncertain but also appears to be low.
ES.4.2 Regional Estimates
Regionally, CH4 emissions are widely distributed, due in large part to the varied
activities associated with CH4 production. In industrial countries, energy production and waste
management tend to be key sources. In developing countries, agricultural and land-use
activities, such as livestock, rice cultivation and biomass burning, tend to dominate. Figure
ES-2 presents anthropogenic CH4 emissions by region. (See Table ES-2 at end of executive
summary.)
Page ES-4
-------
FIGURE ES-2
GLOBAL ANTHROPOGENIC METHANE EMISSIONS BY REGION (1990)
70 T
x
o
O)
-------
FIGURE ES-3
METHANE EMISSIONS OF THE FIVE LARGEST EMITTING COUNTRIES
(1990)
50 T-
45 +
40 +
35 +
30 +
o 25 4-
20 +
15 4-
10 4-
5 4-
H Manure
D Minor Indus.
E3 Liquid Wastes
H Biomass Burning
RJSiig \_fO9l
H Landfills
M Nat. Gas & Oil
9 Rice
H Livestock
China FSU
(See notes for Table ES-3)
India
U.S.
Brazil
Page ES-6
-------
U.S. EPA (1993a) examined the potential for reducing emissions in different countries
for four sources: landfills, oil and natural gas, coal mining, and livestock (cattle and buffalo).
The report found that significant emission reductions could be achieved cost-effectively among
the five highest emitting countries. For example, 3-10 Tg of methane could be reduced per
year in the near term from the oil and gas sector in the former Soviet Union, 4-6 Tg/yr from
landfills in the United States, and 1.2-1.6 Tg/yr from coal mining in China.
ES.5 REDUCING UNCERTAINTIES IN METHANE EMISSION ESTIMATES
Resolving the relatively large uncertainties in global CH4 emission estimates should be
a priority of international cooperation on climate change issues. The accuracy of results is
limited by several uncertainties. One is the need for better emission factors for certain
categories, such as oil and natural gas production and use. Another is the need for better
information on emission factors and output for CH4-proclucing activities in developing
countries. For example, more information is needed on the viability of using emission factors
developed for industrialized-country activities to measure similar activities in developing
countries.
There are many international research efforts underway that are helping to resolve
these and other uncertainties in global GHG emission assessments. A key activity is the joint
effort of the IPCC and the OECD. The IPCC/OECD effort is developing an emission inventory
methods document. This document, when reviewed and adopted by the Conference of the
Parties of the FCCC, will be used by countries to prepare or strengthen national GHG
emission inventories, as required by the FCCC.
A number of bilateral and multilateral entities are directly assisting developing countries
in preparing GHG inventories. One of these is the Global Environment Facility, which is
funding GHG emission inventory assessments through the United Nations Development
Program (UNDP) and the United Nations Environment Program (UNEP). Another is the
United States Climate Country Study Program which is providing $25 million to developing
countries to assist them with GHG inventories and development of national climate plans.
In addition, several research organizations in the United States, including U.S. EPA,
the National Science Foundation, the Department of Energy, the National Aeronautic and
Space Administration and others, are sponsoring research to improve understanding of the
processes and activities leading to the production of methane from man-made sources.
ES.6 REFERENCES
Fung, I., J. John, J. Lerner, E. Matthews, M. Prather, L.P. Steele, and P.J. Fraser. 1991.
Three-dimensional model synthesis of the global methane cycle. Journal of Geophysical
Research 96:13,033-13,065.
IPCC (Intergovernmental Panel on Climate Change). 1990. Climate Change: The IPCC
Scientific Assessment. Cambridge University Press, Cambridge, United Kingdom.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
Page ES-7
-------
Lelleveld, J., and PJ. Crutzen. 1993. Methane emissions into the atmosphere, an overview.
In van Amstel, ed. Methane and Nitrous Oxide, Methods in National Emissions Inventories
and Options for Control. Proceedings of an International IPCC Workshop, 3-5 February 1993,
Amersfoort, Netherlands. RIVM, Bilthoven, Netherlands.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France.
U.S. EPA (U.S. Environmental Protection Agency). 1993a. Options for Reducing Methane
Emissions Internationally. Report to Congress. Office of Air and Radiation, Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). 1993b. Current and Future Methane
Emissions from Natural Sources. Report to Congress. Office of Air and Radiation,
Washington, DC.
Page ES-8
-------
TABLE ES-1
GLOBAL ANTHROPOGENIC METHANE
EMISSIONS BY SOURCE FOR 1990
Estimated Emissions
Category
Livestock
Rice Cultivation
Natural Gas and
Oil Systems
Biomass Burning
Liquid Wastes
Coal Fuel Cycle
Landfills
Livestock Manure
Minor Industrial
Sources"
Total
Low
65
60
33
29
30
24
19
10
4
274
(Tg CH.)
High
100
100
(38
68
40
40
39
18
4
477
Mid
80
65
51a
48
35
30
27
14
4
354
a. This figure is a mid-point of the estimated range.
A point estimate could not be determined with a
reasonable degree of confidence. A mid-point is
used here to help rank sources in their general
order of importance.
b. A range estimate for this source was not available
for this report.
Page ES-9
-------
TABLE ES-2
GLOBAL ANTHROPOGENIC METHANE
EMISSIONS BY REGION FOR 1990
Region
South and East Asia
Eastern Europe and the
Former Soviet Union
China and Centrally
Planned Asia
North America
Latin America
Africa
Western Europe
Oceania (incl. Japan)
Middle East
Rest of World3
Regional Totals"
Estimated Emissions
(Tg CHJ
Low High
66 69
40 66
55 64
28 49
34 38
28 32
17 24
6 8
5 8
10 15
289 373
a.
Applies only to minor industrial sources.
b. Totals differ from the world totals in Table
ES-1 because regional estimates were not
calculated for industrial liquid wastes.
Page ES-10
-------
TABLE ES-3
METHANE EMISSIONS OF THE FIVE LARGEST
EMITTING COUNTRIES, 1990
CH4)
Source
Category
Livestock
Rice Cultivation
Natural Gas and
Oil Systems"
Landfills
Coal Fuel Cycle3
Biomass Burning
Liquid Wastes"
Minor Industrial
Sources
Livestock Manure
Total
China
6
21
<1
1
13
7
<1
<1
2
50
Former
Soviet
Union
9
ns
26
1
5
<1
<1
1
2
44
United
States
6
<1
4
10
5
<1
<1
<1
2
27
India
10
16
ns
1
<1
3
ns
<1
<1
30
Brazil
7
<1
ns
1
ns
6
<0.1
<0.1
<1
14
a. Figures are mid-points of the estimated ranges and not
accurate point estimates. Point estimates could not be
determined with a reasonable degree of confidence. Mid-
points are used to help rank sources in their general
order of importance.
b. Includes domestic waste only.
ns Not significant — little or no emissions from this source.
PageES-11
-------
-------
CHAPTER 1
INTRODUCTION
Section 603 of the Clean Air Act Amendments of 1990 (Public Law 101-549) requires
the U.S. Environmental Protection Agency (U.S. EPA) to prepare and submit to Congress a
series of reports on issues concerning methane emissions from anthropogenic (human-
related) and natural sources. The topics for the five required reports are: (1) Anthropogenic
Methane Emissions in the United States; (2) Options for Reducing Anthropogenic Methane
Emissions in the United States; (3) International Anthropogenic Methane Emissions; (4)
Options for Reducing International Anthropogenic Methane Emissions; and (5) Methane
Emissions from Natural Sources. This report, which presents estimates of international
anthropogenic methane emissions for the year 1990, fulfills the third requirement.
1.1 BACKGROUND
1.1.1 The Importance of Methane to Potential Greenhouse Warming
Methane (CH4) is an important greenhouse gas whose concentration in the atmosphere
is increasing, although its rate of increase has slowed in recent years. Second only to carbon
dioxide (CO2) in terms of its contribution to potential global warming, CH4 is responsible for
about 17% of the total contribution of all greenhouse gas emissions in 1990 to future
warming.1 The atmospheric concentration of CH4 has more than doubled over the last 200
years (IPCC, 1992), but its rate of increase gradually slowed over the last decade or so, to
about 0.6%/year, or approximately 30 teragrams (Tg)2 CH4/year in 1990 (Steele et al., 1992).
This decline in growth rates could be due to a decrease in CH4 emission rates, or an increase
in CH4 destruction rates, or a combination of the two. The atmospheric concentration of CO2,
in comparison, is currently increasing at about 0.5%/year, or approximately 14,000 Tg
CO2/year (IPCC, 1990 and 1992). Although the amount of CO2 accumulating in the
atmosphere each year is orders of magnitude larger than that of CH4, each additional gram of
CH4 released to the atmosphere is as much as 22 times more effective in potentially warming
the Earth's surface over a 100-year period than each additional gram of CO2 (IPCC, 1992).
Compared with other greenhouse gases emitted by anthropogenic activities, CH4 has a
relatively short atmospheric lifetime. The lifetime of CH4 (defined as its atmospheric content
divided by its rate of removal) is approximately 10 years (Lelieveld and Crutzen, 1993),
1 This contribution to potential warming is based on converting 1990 emission estimates to a CO2-equivalent
basis using direct 100-year global warming potentials (GWPs) from IPCC (1992). When both direct and indirect
effects are considered, each additional kilogram of CH4 released to the atmosphere is as much as 22 times more
effective in warming the Earth than each additional kilogram of CO2. This estimate of CH4's contribution to warming
is quite uncertain because of scientific and methodological uncertainties associated with the GWPs, especially the
indirect GWPs for CH4 and other trace gases. The GWPs are continually being revised as these uncertainties are
analyzed and debated.
2 Teragram = 106 metric tonnes = 1012 grams.
Page 1-1
-------
whereas the atmospheric lifetime of CO2 is approximately 120 years (IPCC, 1990).3 A short
atmospheric lifetime of a gas means that the atmosphere will adjust relatively rapidly to
reductions in emissions. In other words, CH4 emission reductions will result relatively rapidly
in reductions in atmospheric concentrations of CH4.
Because of CH4's short lifetime and its potency as a greenhouse gas, reductions in
emissions of CH4 are a very effective means of mitigating future climate change. Reducing
annual CH4 emissions by 10-20% and holding emissions constant over the next century would
cut projected global warming by about 1QC (roughly 25% of the total) and might also result in
reduced atmospheric concentrations of other greenhouse gases because of chemical
feedbacks (Thompson et al., 1992). A number of technical options for reducing emissions
from different anthropogenic source categories have been identified, and work is ongoing to
better understand and implement these options (e.g., IPCC/RSWG, 1990; U.S. EPA, 1993).
Although total emissions of CH4 are fairly well constrained by measurements of CH4's
atmospheric concentrations and knowledge of its atmospheric lifetime (see section 1.1.4
below), the magnitudes of individual sources of CH4 are quite uncertain. Also, the
biogeochemical and socioeconomic factors that control CH4 emissions, and the geographic
distribution of emissions by source category, are poorly understood and/or poorly quantified.
Improving our understanding of the different sources of CH4 is a critical step in designing
appropriate and effective emission reduction strategies for the world. This report, by
quantifying anthropogenic emissions by source and by nation and/or region, contributes to this
understanding.
1.1.2 An Overview of Methane
Methane is a radiatively and chemically active atmospheric trace gas. It is called a
"trace gas" because its atmospheric concentration, approximately 1.72 parts per million by
volume (ppmv) in 1990 (IPCC, 1992), is minute in comparison with the atmospheric
concentrations of the gases that make up most of the atmosphere: nitrogen and oxygen (78%
and 21% by volume, respectively). Methane's atmospheric concentration lies within the broad
range of concentrations of the other trace gases that are also important greenhouse gases
(Table 1-1).
Methane is radiatively active in that it affects the radiative, or energy, balance of the
Earth. The Earth's radiative balance is controlled by the difference between the energy
absorbed by the Earth (from solar radiation) and the energy emitted by the Earth (in the form
of long-wave, infrared energy). By absorbing outgoing, infrared radiation emitted by the Earth,
and re-emitting some of that energy back toward the Earth, CH4 and other "greenhouse
gases" help to warm the Earth. This absorption of infrared radiation is called a "direct effect
on radiative forcing."
3 Carbon dioxide does not have any real sinks, i.e., it is not destroyed by chemical or biological processes in the
same manner as other trace gases, such as methane. Rather, CO2 is circulated between its various reservoirs (the
atmosphere, oceans, biota, and sediments). Therefore, the lifetime of CO2 cannot be defined as its atmospheric
content divided by its rate of removal. Instead, it is estimated here as the time necessary for the atmospheric
concentration to decline to 1/e, i.e., to decrease by 63%, in the absence of further emissions (IPCC, 1990).
Page 1-2
-------
TABLE 1-1
SUMMARY OF ATMOSPHERIC PARAMETERS OF METHANE
AND OTHER KEY GREENHOUSE GASES INFLUENCED BY HUMAN ACTIVITIES3
Parameters
Preindustrial atmospheric
concentration (1750-1800)
Current atmospheric concentration
(1990)°
Current rate of annual atmospheric
CO2
280 ppmv"
353 ppmv
1 .8 ppmv
CH4
0.8 ppmv
1.72 ppmv
0.010 ppmv
N2O
288 ppbvb
310 ppbv
0.8 ppbv
CFC-1 1
0
265 pptvb
9.5 pptv
CFC-1 2
0
470 pptv
16.5 pptv
accumulation
Atmospheric lifetime" (years)
(0.5%) (0.6%)
(50-200)
11
(0.25%)
130
65
130
a. Ozone, which is also an important greenhouse gas that is influenced by human activities, has not been
included in the table because of lack of precise data.
b. ppmv = parts per million by volume; ppbv = parts per billion by volume; pptv = parts per trillion by volume.
c. The current (1990) concentrations have been estimated based upon an extrapolation of measurements
reported for earlier years, assuming that the recent trends remained approximately constant.
d. For each gas in the table, except CO2, the "lifetime" is defined here as the ratio of the atmospheric content to
the total rate of removal. This time scale also characterizes the rate of adjustment of the atmospheric
concentrations if the emission rates are changed abruptly. Carbon dioxide is a special case since it has no
real sinks, but is merely circulated between various reserves (atmosphere, ocean, biota, sediments). The
"lifetime" of CO2 given in the table is a rough indication of the time it would take for the CO2 concentration to
adjust to changes in the emissions.
Sources: IPCC, 1990 and 1992.
Methane also has an indirect effect on radiative forcing because it is a chemically
active greenhouse gas. That is, CH4 enters into chemical reactions that indirectly alter the
Earth's radiative balance through increasing the lifetimes and abundances of CH4 and other
greenhouse gases. These positive feedbacks on the Earth's climate system augment CH4's
direct effect on radiative forcing, resulting in a greater total effect. (See section 1.1.4 of this
chapter for further discussion of CH4's direct and indirect impacts on radiative forcing.)
Methane is released to the atmosphere from a wide variety of anthropogenic and
natural sources. Anthropogenic sources, which are responsible for about 70% of total
emissions, include domestic livestock, rice cultivation, biomass burning, oil and natural gas
systems, coal systems (especially mining), landfills, domestic livestock wastes, wastewater
treatment, and a variety of minor industrial processes. Natural sources, responsible for about
30% of the total emissions from all sources (IPCC, 1992), include natural wetlands, termites,
oceans and fresh waters, wildfires, and CH4 hydrates.
Page 1-3
-------
The abundance of CH4 in the atmosphere is determined by the difference between the
rates of CH4 production by its sources and the rates of CH4 destruction by its sinks. The
primary sink, or removal mechanism, for CH4 is oxidation by hydroxyl (OH) radicals in the
troposphere.4 This chemical reaction, which accounts for about 90% of total CH4 destruction
(IPCC, 1992; Lelieveld and Crutzen, 1993), eventually yields carbon monoxide and water
vapor. The carbon monoxide then reacts with OH radicals to produce CO2. The remainder of
CH4 destruction occurs by chemical reactions in the stratosphere5 and by biological oxidation
in dry soils. Methane is destroyed in the stratosphere by reactions with OH, chlorine (Cl), and
excited atomic oxygen (O1(D)) radicals (Crutzen and Schmailzl, 1983). Reactions with OH
radicals are responsible for about three-quarters of the stratospheric destruction (Lelieveld and
Crutzen, 1993). The remaining sink is microbial oxidation of CH4 in dry (aerated) soils (Born
et al., 1990; Keller et al., 1983; Seller et al., 1984; and Seller and Conrad, 1987).
The current rate of increase in the atmospheric abundance of CH4 indicates that the
total rate of production by CH4 sources exceeds the total rate of destruction by CH4 sinks by
approximately 30 Tg/year. Table 1-2 presents a summary of CH4's sources and sinks.
1.1.3 The Atmospheric History and Geographic Distribution of Methane
Historic and paleo-atmospheric concentrations of CH4 have been measured using
various techniques, including direct atmospheric measurements, measurements of CH4 in air
extracted from air bubbles in polar ice cores, and analyses of solar infrared spectra. These
measurements reveal that current concentrations of CH4 and rates of growth over the last
century are higher than at any time during the past 160,000 years (e.g., Chappellaz et al.,
1990; Rasmussen and Khalil, 1984; Blake and Rowland, 1988; and Steele et al., 1992).
Figure 1-1 summarizes the inferred evolution of global average CH4 concentrations.
Analyses of air trapped in ice cores from Antarctica and Greenland (Raynaud et al.,
1988; Chappellaz et al., 1990; and Stauffer et al., 1988) indicate that CH4 levels in the
atmosphere rose rapidly from minimums of about 0.35 ppmv during the last two glacial
periods (about 150,000 and 18,000 years before present (YBP)) to maximums of about 0.65
ppmv during the subsequent interglacials (about 133,000 and 9,000 YBP). The highest
measured rates of growth during these glacial/interglacial transitions were about 0.2-0.3
ppbv/year. Methane concentrations were fairly constant between about 3000 and 200 YBP, at
about 0.7-0.8 ppmv, and began to increase rapidly between 100 and 180 years ago (Craig
and Chou, 1982; Rasmussen and Khalil, 1984; Stauffer et al., 1985; and Pearman et al.,
1986). Methane concentrations reached about 0.9 ppmv at the beginning of this century.
Analysis of the Vostok ice core from Antarctica (Chappellaz et al., 1990) indicates that
not only are high and low paleo-atmospheric concentrations of CH4 associated with interglacial
and glacial periods, but also CH4 concentrations closely follow higher-frequency fluctuations in
surface air temperature (Figure 1-2). This positive correlation suggests that higher surface
temperatures may increase the strength of CH4 sources, and/or that higher atmospheric CH4
levels may increase surface temperatures. Although it is uncertain which mechanism, or if
* The "troposphere" is defined as the layer of the atmosphere immediately above the earth's surface where the
temperature decreases with height. The maximum height of the troposphere varies from about 16 km near the
equator to about 8 km near the poles.
* The "stratosphere" is defined as the atmospheric layer immediately above the troposphere where temperature
Increases with height. It is approximately 30-40 km thick.
Page 1-4
-------
both mechanisms, were operative over the past 160,000 years, the CH4 concentration/-
temperature correlation indicates a fundamental link between temperature and the CH4 cycle.
TABLE 1-2
ESTIMATED SOURCES AND SINKS OF METHANE
(Tg CH4 per year)
Sources/Sinks
Anthropogenic Sources:
Rice
Domesticated Livestock
Oil/Gas Systems
Coal Mining
Biomass Burning
Landfills
Animal Wastes
Wastewater Treatment
Global Estimate
70
80
80
35
30
40
25
25
Global Range
20 - 120
60- 100
35 - 125
25-45
15-45
15-65
15-35
15-35
Natural Sources:
Wetlands
Termites
Oceans and Fresh Waters
Methane Hydrate Destabilization
Total Natural and Anthropogenic Sources:
Sinks:
Reaction with OH in the Troposphere
Reactions with OH, Cl, and O(1D)in the
Stratosphere
Soil Absorption
Total Sinks:
125
30
15
5
560
455
45
30
530
55 - 195
0-60
5-25
0- 10
470 - 650
405 - 505
35-55
5-55
445-615
Atmospheric Increase:
30
25-35
Source: Lelieveld and Crutzen, 1993.
Page 1-5
-------
FIGURE 1-1
MEASUREMENTS OF GLOBAL METHANE CONCENTRATIONS
1300-
a
o
800-
300-
KEYa
• Modem record
o Siple ice core
a Crete ice core
* Camp Century ice core
o Byrd 1 ice core
* Byrd 2 ice core
a Dye ice core
» Vostok ice core
o
o
CD
A
A
'«- a
B
. =
Op
106
105
10* 103 102
Years Before Present
10°
Annual atmospheric CH4 concentrations during the past 160,000 years (derived from ice cores and the
NOAA/CMDL flask sampling network).
Source: ORNL, 1991.
Page 1-6
-------
FBGURE 1-2
ATMOSPHERIC METHANE AND AIR TEMPERATURE RECORD
DETERMINED FROM VOSTOK ICE CORE
Depth (m)
1000 1500
CH4
(ppbv)
700
- 600
- 500
- 400
300
40 80 120
Age(kyrBP)
160
Methane concentrations (bottom) and estimated temperature changes (top) during the past 160,000 years, as
determined on the ice core from Vostok, Antarctica.
Sources: Chappellaz et al., 1990, as presented in IPCC, 1990.
Direct measurements of atmospheric CH4, which began in 1978, indicate that the
current global-average CH4 concentration is 1.72 ppmv (equivalent to about 4,900 Tg CH4) ~
more than double the preindustrial concentration (IPCC, 1992). Both direct atmospheric
measurements and analyses of solar spectra have documented that CH4 levels increased at
roughly 1%/year during the past 30-40 years (Blake and Rowland, 1986 and 1988; Ehhalt et
al., 1983; Fraser et al., 1981; Rinsland et al., 1985; Khalil and Rasmussen, 1982; Steele et al.,
1987; and Wallace and Livingston, 1990). Recent measurements, however, indicate that
CH4's rate of growth has slowed over the past decade or so, from about 18 ppbv/year during
the late 1970s (about 1.2%/year), to about 12-13 ppbv/year in the mid-1980s (about 0.75%/per
year), to 9.5 ppbv/year in 1990 (about 0.6%/year) (Blake and Rowland, 1986 and 1988; and
Steele et al., 1992). The decrease in the growth rate of atmospheric CH4 may be due to
reductions in CH4 emissions from natural or anthropogenic sources, increases in CH4 sinks, or
a combination of the two. An analysis of the trend in atmospheric concentrations of OH since
1978 indicates that increasing abundances of OH radicals may be largely responsible for this
observed decrease in atmospheric CH4 growth (Prinn et al., 1992). This inferred increase in
OH radical abundance may have several causes, including increased short-wave (ultraviolet,
or UV) radiation resulting from stratospheric ozone loss, and growing tropospheric
concentrations of gases that contribute to OH radical formation (especially nitrogen oxides
(NOX), ozone (O3), and water vapor) due to anthropogenic activities (Lelieveld and Crutzen,
Page 1-7
-------
1993). Most atmospheric chemistry models that compute OH, however, show OH
concentrations decreasing, rather than increasing, since preindustrial times (Thompson, 1992).
The reasons for CH4's slowing growth rate remain unresolved.
The spatial and temporal distribution of atmospheric CH4 is not uniform (Figure 1 -3), as
it reflects geographic and seasonal differences in source and sink strengths (Fung et al.,
1991). In the Southern Hemisphere, CH4 concentrations are approximately uniform from mid
to high latitudes and increase northward. The average concentration in the Northern
Hemisphere is approximately 100 ppbv higher than in the Southern Hemisphere, a reflection
of greater source strengths in the northern latitudes. The seasonal cycle in the Southern
Hemisphere shows a minimum in February (the Southern Hemisphere summer) and a
maximum in September/October, which is consistent with higher summer abundances of OH
and temperature-dependent destruction rates.6 The seasonal cycle is more complex in the
Northern Hemisphere and appears to be a function of seasonal fluctuations in sources as well
as sinks.
1.1.4 The Role of Methane in Global Climate Change
Increasing atmospheric concentrations of CH4 have important implications for global
climate change because CH4 is a potent greenhouse gas. If CH4's direct effect on radiative
forcing alone is taken into account, then over a 20-year period each additional kilogram of CH4
released to the atmosphere is about 35 times more effective in warming the Earth than each
additional kilogram of CO2. Over a 100-year period, CH4 is about 11 times more effective per
kilogram than CO2 (IPCC, 1992; and Lelieveld et al., 1993). The effectiveness of CH4 relative
to CO2 decreases as the time frame increases because of CH4's shorter atmospheric lifetime
(about 10 years, versus approximately 120 years for CO2).
If indirect chemical effects on greenhouse gases are taken into account, then CH4 is
even more effective in contributing to global warming. Although these indirect effects are both
positive and negative and are poorly quantified, the total indirect forcing of climate by CH4 is
positive and is likely to be comparable in magnitude to the direct effect (IPCC, 1992; and
Lelieveld et al., 1993). Therefore, when both direct and indirect effects are considered, each
additional kilogram of CH4 released to the atmosphere is as much as 22 times more effective
in warming the Earth than each additional kilogram of CO2.
Through reaction with OH radicals, CH4 influences the oxidizing capacity of the
troposphere and may cause its own concentration to increase. Hydroxyl radicals are
responsible for the oxidation of the majority of the gases emitted into the atmosphere,
including CH4, higher hydrocarbons, carbon monoxide, and hydrogenated CFCs. Therefore,
the atmospheric concentration of OH determines the lifetime and, hence, the abundance of
these gases. Since the oxidation of CH4 by OH is a sink not only for CH4, but also for OH
(Thompson and Cicerone, 1986; and Isaksen and Hov, 1987), increases in atmospheric CH4
may cause a depletion of OH in the troposphere. As OH is depleted, the lifetime of CH4
increases, causing a consequent increase in the atmospheric concentration of CH4. Depletion
of OH may also increase the lifetimes and, therefore, the concentrations of other greenhouse
gases that are oxidized by OH.
8 The OH + CH4 reaction rate is strongly dependent on temperature, increasing with increases in temperature
(Vaghjlani and Ravishankara, 1991).
Page 1-8
-------
FIGURE 1-3
SPATIAL AND TEMPORAL DISTRIBUTION OF ATMOSPHERIC METHANE
17QO .
1500
1987
Three-dimensional representation of the global distribution of atmospheric CH4 derived from direct measurements.
The arrow denotes the equator.
Source: Fung et al., 1991.
Page 1-9
-------
Increasing concentrations of CH4 in environments with high levels of NOX tend to
increase the amount of O3 in the troposphere and lower stratosphere, where O3 acts as a
greenhouse gas (Wuebbles and Tamaresis, 1993). High concentrations of NOX are primarily
found in the industrial regions of the Northern Hemisphere, since the major anthropogenic
source of NOX is fossil fuel combustion.
Oxidation of CH4 by OH and O1(D) in the stratosphere results in the production of
water vapor, which is also a greenhouse gas (Wuebbles and Tamaresis, 1993). An increase
of water vapor in the lower stratosphere may enhance global warming significantly, as the low
temperatures there augment the efficiency with which water vapor traps infrared energy.
Increases in stratospheric water vapor concentrations may also enhance the formation of polar
stratospheric clouds, which are thought to play an important role in the seasonal formation of
the stratospheric ozone "hole" over Antarctica (WMO, 1991).
Methane emissions also contribute to the atmospheric burden of CO2, since CH4
ultimately oxidizes to CO2. Only the fossil fuel-related CH4 emissions (i.e., emissions from
coal mining and from natural gas and oil systems), however, are net contributors to
atmospheric CO2, since biogenic CH4 emissions (i.e., emissions from rice cultivation, livestock,
solid and liquid wastes, and biomass burning) are recently recycled CO2 (Lelieveld and
Crutzen, 1992).
In sum, increased concentrations of CH4 can indirectly increase radiative forcing by
depleting tropospheric OH and by increasing tropospheric and lower stratospheric O3,
stratospheric water vapor, and CO2.
1.1.5 The Methane Budget
Because of CH4's significant role in global climate change, it is important to understand
CH4's atmospheric budget. Conservation of mass requires that CH4's total annual source
strength equal CH4's total annual sink strength plus the annual increase in atmospheric
concentrations of CH,, i.e.:
'4'
£ = S + dC/dt
where:
(1.1)
E = total emissions, or total sink strength (Tg CH4/yr)
S = total sink strength (Tg CH4/yr)
C = global-average atmospheric concentration (Tg CH4)
The total sink strength is equal to CH4's atmospheric concentration divided by CH4's lifetime.
Therefore, equation (1.1) can be rewritten as follows:
= C/L + dC/dt
(1.2)
where:
L =: atmospheric lifetime (yr)
The atmospheric concentration of CH4 and its annual rate of increase are well known from
high-precision atmospheric measurements. Because CH4 oxidation by OH is by far the
dominant sink for CH4, the atmospheric lifetime of CH4 is determined primarily by the reaction
rate of this process and the amount of OH in the troposphere, where the bulk of CH4 oxidation
Page 1-10
-------
occurs. There are significant uncertainties in the estimation of both these quantities. The
other CH4 sinks, oxidation in the stratosphere and removal by soils, must also be included in
the calculations. By including these two other sinks, the calculated lifetime of CH4 is
decreased (i.e., a greater sink strength implies a shorter lifetime).
The reaction rate of CH4 with OH has been determined by laboratory measurements,
although the presence of impurities and the occurrence of secondary reactions in laboratory
samples can result in significant errors in these measurements (DeMore et al., 1992). Recent
measurements by Vaghjiani and Ravishankara (1991), in which impurities and secondary
reactions were minimized, found that the CH4-OH reaction rate was about 20% lower than
indicated by preceding measurements. This lower reaction rate indicates that the atmospheric
lifetime of CH4 is longer than had been estimated previously.
Determining the global average distribution of OH in the troposphere by direct
measurement is difficult. Because the concentration of OH is extremely low (about one part
per 50 trillion), its measurement requires highly sensitive techniques. Also, because OH is so
short-lived (about 1-2 seconds), it is highly variable both temporally and spatially. Therefore,
indirect methods have been necessary to estimate OH concentrations. One method consists
of measuring atmospheric concentrations of methyl chloroform (CH3CCI3), a compound of
purely industrial origin whose sources are well known and whose major sink is reaction with
OH. Using CH3CCI3 emission estimates (based on industry data), measured CH3CCI3
concentrations, and an experimentally determined CH3CCI3-OH reaction rate, scientists can
infer the lifetime of CH3CCI3 and the atmospheric distribution of OH.7 Using this method, Prinn
et al. (1992) deduced a global-average tropospheric OH concentration of about 810,000
radicals per cubic centimeter. Laboratory measurements by Talukdar et al. (1992) indicate
that the CH3CCI3-OH reaction rate used by Prinn et al. in their calculations was too high;
scaling Prinn et al.'s estimated OH concentration upwards to account for this new reaction
rate results in an OH abundance of approximately 930,000 radicals per cubic centimeter
(Lelieveld and Crutzen, 1993).
Using their estimated global-average OH concentration and the CH4-OH rate coefficient
reported by Vaghjiani and Ravishankara (1991), Prinn et al. (1992) inferred a lifetime for
atmospheric CH4 of about 11.1 years. This estimate is likely about 10-15% too high because
of the implication of Talukdar et al.'s (1992) work on the OH concentration. If CH4 loss in the
stratosphere and removal by soils are also considered, the global atmospheric lifetime of CH4
reduces to about 9.5 years (Lelieveld and Crutzen, 1993). Inserting this lifetime into equation
(1.2), with the current atmospheric CH4 concentration (about 4,900 Tg) and annual rate of
increase (about 30 Tg/yr) yields an estimated total source strength of about 545 Tg CH4/yr.
Uncertainties in measured reaction rates and estimated OH concentrations indicate that the
estimated CH4 lifetime has an uncertainty of about + 25% (Fung et al., 1991). Therefore, the
estimated CH4 source strength ranges from approximately 450 to 750 Tg CH4/yr.
The technique of deducing CH4 emissions from the atmospheric lifetime and global-
average concentrations of CH4 can be used only to estimate total emissions; it cannot
distinguish among individual source components. Although the major sources of CH4 have
7 Methyl chloroform has one other sink: oceanic consumption (Eiutler et al., 1991). Although this sink is minor,
accounting for about 6% of the total sink strength, it must also be included in these calculations. Such inclusion
increases estimates of OH abundance by about 7% (Prinn et al., 1992), which results in a lower inferred CH4
lifetime.
Page 1-11
-------
been identified, their individual strengths are still highly uncertain. Estimates of individual
source strengths are based on CH4 flux measurements, which are highly variable, and activity
data (e.g., global area of wetland rice cultivation), which in many cases are uncertain.
Methane fluxes from an individual source can span several orders of magnitude, so
extrapolation from individual flux measurements to a global budget is very difficult.
Furthermore, emissions are controlled by a wide variety of both biogeochemical and
socioeconomic factors, which are neither well understood nor well quantified. Additional
information about individual source strengths can be obtained from isotopic composition
measurements of atmospheric CH4, but even with these additional constraints, CH4 source
estimates derived from estimated individual source strengths have uncertainties of + 20-40%
(e.g., Cicerone and Oremland, 1988; and IPCC, 1992). The total source strength estimated
from sink strength and atmospheric concentrations (via Equation 1.2) is thus better
constrained than that obtained by summing the contribution from individual sources.
Estimates of the strengths of individual CH4 sources have also been derived using
three-dimensional atmospheric chemistry models, with geographic and seasonal variations in
atmospheric concentrations of CH4 and other chemically active gases, as well as geographic
and temporal variations in the isotopic composition of atmospheric CH4 and in estimated CH4
sources, as input data. As described more fully in Chapter 11 of this report, these input data
place constraints on individual sources in the CH4 budget, but there are numerous possible
emission combinations that can satisfy these constraints (e.g., Fung et al., 1991; Taylor et al.,
1991; and Tie et al., 1991). Although scientific understanding of the CH4 budget has vastly
improved over the last few years, there remain large discrepancies among various authors'
estimates of the contributions of individual sources.
1.1.6 Summary
Methane is a radiatively and chemically important greenhouse gas, second only to CO2
in its contribution to future warming. Methane helps to warm the Earth by absorbing and re-
radiating outgoing energy from the Earth, and by entering into chemical reactions that
increase the atmospheric lifetimes and abundances of CH4 and other greenhouse gases.
Each additional gram of CH4 released to the atmosphere is as much as 22 times more
effective in potentially warming the Earth's surface over a 100-year period than each
additional gram of CO2 (IPCC, 1992). Because of CH4's potency as a greenhouse gas and its
short lifetime (about 10 years), reductions in emissions of CH4 are a very effective means of
mitigating future climate change.
The current atmospheric concentration of CH4 is greater than at any time during the
past 160,000 years. Methane's concentration more than doubled over the past 200 years,
although during the last decade or so, its rate of increase slowed from about 1%/year to about
0.6%/year (Steele et al., 1992). Since the atmospheric abundance of CH4 is determined by
the difference between rates of CH4 production by its sources and rates of destruction by its
sinks, this recent decline in atmospheric growth rates could be due to a decrease in emission
rates, an increase in destruction rates, or a combination of the two.
Atmospheric measurements of CH4 and information about its rate of destruction by its
sinks indicate that current emissions of CH4 range from approximately 450 to 750 Tg
CH4/year. Methane is released by both anthropogenic and natural sources, which are
responsible for about 70% and 30%, respectively, of total emissions (IPCC, 1992). Estimates
of individual source strengths, which are based on highly variable CH4 flux measurements
and, in some cases, inaccurate activity data, are very uncertain. Geographic and seasonal
Page 1-12
-------
variations in atmospheric concentrations of CH4 and in isbtopic composition data place
constraints on some of the source strengths, but estimates of emissions from individual
sources still vary by as much as an order of magnitude. Because of CH4's importance as a
greenhouse gas, improving our understanding of individual source strengths and their
geographic distribution is a critical step in designing appropriate emission reduction strategies.
1.2 OVERVIEW OF THIS REPORT
This report estimates the current magnitude and geographic distribution of each
anthropogenic source of CH4, based on the latest flux measurements and current scientific
understanding of the many factors that control emissions. In addition to estimating total
emissions by source category, the report provides country- or region-specific emission
estimates for each of the sources. Individual chapters on 'CH4 sources include background
information on the source (including a discussion of measured fluxes from the source, and the
biogeochemical and socioeconomic factors that control emissions), describe the methodology
used for estimating emissions, present estimates of 1990 national and/or regional CH4
emissions, discuss trends that could affect future emission levels, and describe uncertainties
associated with the emission estimates. These emission source chapters cover the following
topics:
Livestock Digestion: emissions from beef and dairy cattle and other
domestic animals;
Rice Cultivation: emissions from flooded rice fields;
Biomass Burning: emissions from open and confined burning of
biomass;
Natural Gas and Oil Systems: emissions from oil and gas production;
crude oil transportation and refining; natural gas processing,
transportation, and distribution; and fuel combustion;
Coal Fuel Cycle: emissions from surface and underground coal mining;
processing, transport, and storage of extracted coal; and coal
combustion;
Minor Industrial Sources: emissions from coke and steel production, chemical
manufacturing, and other miscellaneous sources;
Landfills and Open Dumps: emissions associated with disposal of solid
waste in landfills and dumps;
Livestock Manure: emissions from solid and liquid waste management
systems; and
Wastewater: emissions from domestic and industrial wastewater treatment.
The final chapter of the report discusses verification of CH4 emission inventories, particularly
the constraints placed on inventories by the available socioeconomic, biogeochemical, and
Page 1-13
-------
atmospheric data, and how uncertainties associated with CH4 emission estimates can be
reduced.
1.3 REFERENCES
Blake, D.R., and F.S. Rowland. 1986. World-wide increase in tropospheric methane, 1978-
1983. Journal of Atmospheric Chemistry 4:43-62.
Blake, D.R., and F.S. Rowland. 1988. Continuing worldwide increase in tropospheric
methane, 1978-1987. Science 239:1,129-1,131.
Born, M., H. Dorr, and I. Levin. 1990. Methane consumption in aerated soils of the temperate
zone. Tellus 42B:2-8.
Butler, J.H., J.W. Elkins, T.M. Thompson, and B.D. Hall. 1991. Oceanic consumption of
CH3CCI3: Implications for tropospheric OH. Journal of Geophysical Research 96:22,347-
22,355.
Chappellaz, J., J.M. Barnola, D. Raynaud, Y.S. Korotkevich, and C. Lorius. 1990. Ice-core
record of atmospheric methane over the past 160,000 years. Nature 345:127-131.
Cicerone, R.J., and R.S. Oremland. 1988. Biogeochemical aspects of atmospheric methane.
Global Biogeochemical Cycles 2:299-327.
Craig, H., and C.C. Chou. 1982. Methane: The record in polar ice cores. Geophysical
Research Letters 9:1,221-1,224.
Crutzen, P.J., and U. Schmailzl. 1983. Chemical budgets of the stratosphere. Planetary
Space Science 31:1,009-1,032.
DeMore, W.B., S.P. Sander, D.M. Golden, R.F. Hampson, M.J. Kurylo, C.J. Howard, A.R.
Ravishankara, C.E. Kolb, and M.J. Molina. 1992. Chemical Kinetics and Photochemical Data
for Use in Stratospheric Modeling, Evaluation Number 10. JPL Publication 92-20, National
Aeronautics and Space Administration, Jet Propulsion Laboratory, Pasadena, California.
Ehhalt, D.H., R.J. Zander, and R.A. Lamontagne. 1983. On the temporal increase of
tropospheric CH4. Journal of Geophysical Research 88:8,442-8,446.
Fraser, P.J., M.A.K. Khalil, R.A. Rasmussen, and A.J. Crawford. 1981. Trends of
atmospheric methane in the Southern Hemisphere. Geophysical Research Letters 8:1,063- .
1,066.
Fung, I., J. John, J. Lerner, E. Matthews, M. Prather, L.P. Steele, and P.J. Fraser. 1991.
Three-dimensional model synthesis of the global methane cycle. Journal of Geophysical
Research 96:13,033-13,065.
IPCC (Intergovernmental Panel on Climate Change). 1990. Climate Change: The IPCC
Scientific Assessment Cambridge University Press, Cambridge, United Kingdom.
Page 1-14
-------
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
IPCC/RSWG (Intergovernmental Panel on Climate Change/Response Strategies Working
Group). 1990. Methane Emissions and Opportunities for Control: Workshop Results of
Intergovernmental Panel on Climate Change Response Strategies Working Group.
Coordinated by Japan Environment Agency and U.S. Environmental Protection Agency,
Washington, D.C.
Isaksen, I.S.A., and O. Hov. 1987. Calculation of trends in the tropospheric concentration of
O3, OH, CO, CH4, and NOX. Tellus 396:122-139.
Keller, M., T.J. Goreau, S.C. Wofsy, W.A. Kaplan, and M.B. McElroy. 1983. Production of
nitrous oxide and consumption of methane by forest soils. Geophysical Research Letters
10:1,156-1,159.
Khalil, M.A.K., and R.A. Rasmussen. 1982. Secular trends of atmospheric methane (CH4).
Chemosphere 11:877-883.
Lelieveld, J., and P.J. Crutzen. 1992. Indirect chemical effects of methane on climate
warming. Nature 355:339-342.
Lelieveld, J., and P.J. Crutzen. 1993. 3. Methane emissions into the atmosphere: an
overview. In van Amstel, A.R., ed. Methane and Nitrous Oxide, Methods in National
Emissions Inventories and Options for Control. Proceedings of an International IPCC
Workshop, 3-5 February 1993, Amersfoort, Netherlands. RIVM, Bilthoven, Netherlands. 17-
25.
Lelieveld, J., P.J. Crutzen, and C. Bruhl. 1993. Climate effects of atmospheric methane.
Chemosphere 26(1-4):739-768.
ORNL (Oak Ridge National Laboratory). 1991. Trends '91, A Compendium of Data on Global
Change. Carbon Dioxide Information Analysis Center, ORNL, Oak Ridge, Tennessee.
Pearman, G.I., D. Etheridge, F. De Silva, and P.J. Fraser. 1986. Evidence of changing
concentrations of atmospheric CO2, N2O and CH4 from air bubbles in Antarctic ice. Nature
320:248-250.
Prinn, R., D. Cunnold, P. Simmonds, F. Alyea, R. Boldi, A. Crawford, P. Fraser, D. Gutzler, D.
Hartley, R. Rosen, and R. Rasmussen. 1992. Global average concentration and trend for
hydroxyl radicals deduced from ALE/GAGE trichloroethane (methyl chloroform) data for 1978-
1990. Journal of Geophysical Research 97:2,445-2,461.
Rasmussen, R.A., and M.A.K. Khalil. 1984. Atmospheric methane in the recent and ancient
atmospheres: Concentrations, trends, and interhemispheric gradient. Journal of Geophysical
Research 89:11,599-11,605.
Raynaud, D., J. Chappellaz, J.M. Barnola, Y.S. Korotkevich, and C. Lorius. 1988. Climatic
and CH4 cycle implications of glacial-interglacial CH4 change in the Vostok ice core. Nature
333:665-667.
Page 1-15
-------
Rinsland, C.P., J.S. Levine, and T. Miles. 1985. Concentration of methane in the troposphere
deduced from 1951 infrared solar spectra. Nature 318:245-249.
Seiler, W., and R. Conrad. 1987. Contribution of tropical ecosystems to the global budgets of
trace gases, especially CH4, H2, CO, and N2O. In Dickinson, R.E., ed. The Geophysiology of
Amazonia. John Wiley, New York, New York. 133-160.
Seiler, W., R. Conrad, and D. Scharffe. 1984. Field studies of methane emission from termite
nests into the atmosphere and measurements of methane uptake by tropical soils.
Atmospheric Chemistry 1:171-186.
Stauffer, B., G. Fischer, A. Neftel, and H. Oeschger. 1985. Increase of atmospheric methane
recorded in Antarctic ice core. Science 229:1,386-1,388.
Stauffer, B., E. Lochbronner, H. Oeschger, and J. Schwander. 1988. Methane concentration
in the glacial atmosphere was only half that of the preindustrial Holocene. Nature 332:812-
814.
Steele, L.P., E.J. Dlugokencky, P.M. Lang, P.P. Tans, R.C. Margin, and K.A. Masarie. 1992.
Slowing down of the global accumulation of atmospheric methane during the 1980s. Nature
358: 313-316.
Steele, L.P., P.J. Fraser, R.A. Rasmussen, M.A.K. Khalil, T.J. Conway, A.J. Crawford, R.H.
Gammon, K.A. Masarie, and K.W. Thoning. 1987. The global distribution of methane in the
troposphere. Journal of Atmospheric Chemistry 5:125-171.
Talukdar, R.K., A. Mellouki, A.-M. Schmoltner, T. Watson, S. Montzka, and A.R.
Ravishankara. 1992. Kinetics of the OH reaction with methyl chloroform and its atmospheric
implications. Science 257:227-230.
Taylor, J.A., G.P. Brasseur, P.R. Zimmerman, and R.J. Cicerone. 1991. A study of sources
and sinks of methane and methyl chloroform using a global three-dimensional Lagrangian
tropospheric tracer transport model. Journal of Geophysical Research 96(D2):3,013-3,044.
Thompson, A.M. 1992. The oxidizing capacity of the Earth's atmosphere: Probable past and
future changes. Science 256:1,157-1,165.
Thompson, A.M., and R.J. Cicerone. 1986. Atmospheric CH4, CO, and OH from 1960 to
1985. Nature 321:148-150.
Thompson, A.M., K.B. Hogan, and J.S. Hoffman. 1992. Methane reductions: Implications for
global warming and atmospheric chemical change. Atmospheric Environment 26 A(14):2,6G5-
2,668.
Tie, X.X., F.N. Alyea, D.M. Cunnold, and C.-Y.J. Kao. 1991. Atmospheric methane: A global
three-dimensional model study. Journal of Geophysical Research 96(D9):17,339-17,348.
U.S. EPA (U.S. Environmental Protection Agency). 1993. Options for Reducing Methane
Emissions Internationally: Report to Congress. Office of Air and Radiation, Global Change
Division, Washington, D.C.
Page 1-16
-------
Vaghjiani, G.L., and A.R. Ravishankara. 1991. New measurements of the rate coefficient for
the reaction of OH with methane. Nature 350:406-409.
Wallace, L, and W. Livingston. 1990. Spectroscopic observations of atmospheric trace
gases over Kitt Peak; 1. Carbon dioxide and methane from 1979 to 1985. Journal of
Geophysical Research 95:9,823-9,827.
WMO (World Meteorological Organization). 1991. Scientific Assessment of Stratospheric
Ozone Depletion: 1991. WMO Global Ozone Research and Monitoring Project - Report No.
25. Geneva, Switzerland.
Wuebbles, D.J., and J.S. Tamaresis. 1993. The role of methane in the global environment.
In Khalil, M.A., ed. Global Atmospheric Methane. NATO ARW Series. Springer-Verlag,
Heidelberg, Germany. In press.
Page 1-17
-------
-------
CHAPTER 2
METHANE EMISSIONS FROM THE DIGESTIVE PROCESSES OF LIVESTOCK
2.1 SUMMARY
Domesticated livestock are an important source of methane emissions, accounting for
about 80 teragrams (Tg)1 per year, or about 20-25% of global anthropogenic emissions.
Methane is produced as a part of the normal digestive processes of ruminant animals (cattle,
buffalo, sheep, goats, and camels), which account for nearly all the emissions from this
source. Cattle are the largest single source, responsible for nearly 75% of the total. Five
countries account for nearly 50% of the total emissions from livestock: India (13%), the
former Soviet Union (11%), Brazil (9%), China (8%), and the United States (7%). In the
future, emissions from livestock are expected to increase as animal populations and animal
production continue to grow.
A firm scientific basis exists for estimating the rate of methane emissions from
ruminant animals managed in temperate agricultural systems. Several decades of
measurements are available that relate methane emission rates to various feeding and
management practices. Very few measurements are available on the rate of rumen
methanogenesis under tropical livestock management systems. Consequently, additional data
are needed to verify the emission estimates that have been made to date. Additionally,
improved data for estimating animal populations and levels of production would help reduce
the uncertainty in the emission estimates for all regions. In particular, data are needed on
numbers of animals classified by age groups and nutrition or feeding conditions for individual
countries. Given the available data, the emission estimate is believed to be approximately
±25%, so that the total global emission estimate range is 80 ±20 Tg per year.
2.2 BACKGROUND
Methane (CH4) is produced as part of the normal digestive processes of animals.
Referred to as "enteric fermentation," these processes produce emissions that have been
estimated to account for a significant portion of the global CH4 budget, about 65-100 Tg
annually (IPCC, 1992). Of domesticated animals, ruminant animals (cattle, buffalo, sheep,
goats, and camels) are the major source of CH4 emissions, with cattle being the most
important source globally. Therefore, emissions from cattle are emphasized in this chapter.2
Although wild ruminant animals, such as deer, and wild nonruminant herbivores, such as
rabbits, also produce CH4, they are not included because emissions from these animals are
considered a natural source.
1 Teragram = 106 metric tonnes = 1012 grams.
2 Although nonruminant animals produce, only a small quantity of methane from enteric fermentation as
compared with ruminant animals, emissions from nonruminant animal wastes, especially swine wastes, may be
significant. Methane emissions from domestic animal wastes are discussed in Chapter 9 of this report.
Page 2-1
-------
Ruminant animals have a large "fore-stomach," or rumen, within which microbial
fermentation breaks down feed into soluble products that can be utilized by the animal.
Approximately 200 species and strains of microorganisms are present in the anaerobic rumen
environment, although only a small portion - about 10 to 20 species -- is believed to play an
important role in ruminant digestion (Baldwin and Allison, 1983). The microbial fermentation
that occurs in the rumen enables ruminant animals to digest coarse plant material that
monogastric animals,3 including humans, cannot digest.
Methane is produced in the rumen by bacteria as a by-product of the fermentation
process. This CH4 is exhaled or eructated by the animal and accounts for the majority of
emissions from ruminants. Methane also is produced in the large intestines of ruminants and
is excreted. Nonruminant herbivores, such as horses, mules, rabbits, pigs, and guinea pigs,
have a limited amount of fermentation in their large intestines or ceca. The CH4 produced in
this manner is quite small compared with the amount produced by ruminant animals.
A significant body of scientific literature exists that describes the quantity of CH4
produced by individual ruminant animals, particularly cattle. This literature results from
decades of research evaluating feeding practices for cattle and other ruminants.4 Over the
past 30 years, hundreds of CH4 measurements have been performed on a wide variety of
cattle and sheep diets typically found in the temperate countries of North America and Europe.
The main purpose of these measurements was to develop scientifically based feeding
standards for these animals. In the United States the standards are presented in a series of
National Research Council publications (e.g., NRC, 1984 and 1989). In the United Kingdom,
feeding standards are published in ARC (1980); other countries have developed similar
publications.
Based on the scientific information available, a variety of factors affect CH4 production
in ruminant animals. These include the physical and chemical characteristics of the feed, the
feeding level and schedule, the use of feed additives to promote production efficiency, and the
activity and health of the animal. It has also been suggested that there may be genetic
factors that affect CH4 production. Of these factors, the feed characteristics and level have
the most influence.
To describe the CH4 production by ruminant animals, it is convenient to refer to the
portion of feed energy intake that is converted to CH4. Higher levels of conversion translate
into higher emissions, given constant feed energy intake. Similarly, higher levels of intake
translate into higher emissions, given constant conversion. There are, however, interactions
between level of intake and conversion to CH4, so these values are not independent. Blaxter
and Clapperton (1965) developed a statistical relationship that correlates feed-intake level and
3 Monogastric animals have a mouth, esophagus, stomach, small intestines, large intestines, pancreas, and
liver (Ensrnlnger, 1983). Examples of monogastric animals include swine, dogs, monkeys, and humans.
4 Calorimetry is the laboratory technique currently used to conduct in-depth evaluations of the performance of
alternative feeding practices. This technique involves placing an animal in a chamber maintained under thermo-
neutral conditions for two days and measuring levels of input to the chamber (oxygen and carbon dioxide) and
output from the chamber (carbon dioxide, oxygen, and methane). Because methane is produced as part of the
normal digestive process of ruminant animals, it is measured as part of this feed-evaluation technique. The
technique is accompanied by a seven- to ten-day physical balance of feed nutrients consumed versus excreted
(faces and urine) and product nutrient outputs from the animals.
Page 2-2
-------
feed digestibility (%)5 with the level of feed-energy conversion to CH4. This relationship
captures the interdependent nature of these variables and may be used to evaluate the CH4
conversion rates for diets that fall within the range of diets evaluated by Blaxter and
Clapperton.
As a result of the various interrelationships among feed characteristics, feed intake,
and conversion rates to CH4, most well-fed ruminant animals in temperate agricultural systems
will convert about 5.5-6.5% of their feed-energy intake to CH4 (Johnson et al., 1991). With
only one exception, this range of CH4 production can be applied to the ruminant animals
typically found throughout North America and Europe, as well as Australia and New Zealand.
Given this range for the rate of CH4 formation, CH4 emissions can be estimated based on the
feed energy consumed by the animals. Because feed energy intake is related to production
level (e.g., weight gain or milk production), the feed energy intake can be estimated for these
regions based on production statistics.
The important exception to this general range of 5.5-6.5% in temperate agricultural
systems occurs when ruminant animals are fed diets comprised primarily of grains. In the
United States, cattle are commonly fed high-grain diets during the final stage prior to
slaughter. Similar systems are used in other countries, such as Canada, Australia, and
France, although to a lesser extent. In the United States, numerous measurements have
been conducted to estimate the conversion of feed energy to CH4 for these high-grain diets,
showing that CH4 formation is generally lower, ranging between 2.5% and 4.5%. A typical
value of 3.5% has been suggested (Johnson et al., 1991).6
Given the emission measurement data and the animal production data available for
temperate agricultural systems, there is a solid foundation for estimating CH4 emissions from
ruminant animals in these regions. However, the basis for estimating emissions from tropical
countries is less complete. Few measurements of CH4 production have been performed for
the types and levels of feeding found in many parts of Asia, the Indian subcontinent, Africa,
and Latin America. The feeds used in these areas are often more fibrous and less digestible
than the feeds typically encountered in temperate countries.
It has been suggested that CH4 conversion rates will be higher for these tropical feeds,
and a value of 7.5% has been suggested for situations in which feed quality and quantity are
relatively poor (Leng, 1991). The Blaxter and Clapperton (1965) relationship also produces
estimates in the range of 7-8% for feeding levels and feed digestibility commonly found in
tropical countries. However, tropical diets were not included in the Blaxter and Clapperton
data set, and the basis for the CH4 conversion estimates for these conditions needs to be
improved.
Margan et al. (1988) present measurements of CH4 production from sheep for two
tropical grasses typically found in Australia: Setaria spacelata and Digitaria decumbens.
5 The digestibility of energy in feed is defined as the proportion of energy in the feed that is not excreted in the
feces. Digestibility is commonly expressed as a percentage (%).
6 The value of 3.5% was chosen as representative of the 90% concentrate diets fed at 2.6 times maintenance
at typical U.S. feedlots (maintenance feed intake is that necessary to support the animal without growth). The very
high proportion of concentrate in the diet and the high feeding level combine to reduce the methane conversion
rate.
Page 2-3
-------
Compared with two temperate forages (Lolium perenne7 and Trifolium resupinaturrf), these
tropical forages converted more feed energy to CH4 at both maintenance and ad libitum
feeding levels. Although these estimates tend to support the thesis that tropical feeds have
higher CH4 conversion rates than temperate feeds, additional measurements are needed to
evaluate the wide diversity of feeds used in tropical agricultural systems. In particular, the
CH4 conversion characteristics of crop by-products, such as rice straw, must be measured
because they are important animal feeds in many countries (Huque and Stem, 1993;
Devendra, 1989; and Winrock, 1978). Also, in the work described by Margan et al., the
tropical forages were modified to remove part of the stems, whereas the temperate forages
were fed whole. It is not known what effect this selected plant part removal had on the CH4-
producing properties of the species studied.
The uncertainty in the CH4 emission estimates for tropical countries is due also to a
lack of data on feed-intake levels, or typical animal-growth and milk-production rates.
Improved characterizations of the animal populations in many tropical countries are needed to
improve the emission estimates.
The rates of conversion of feed energy to CH4 for the nonruminant animals are much
lower than those for ruminants. For swine on good-quality grain diets, found mostly in
developed countries, about 0.6% of feed consumed is converted to CH4. For swine on lower-
quality diets, found mostly in developing countries, about 1.3% is converted (Crutzen et al.,
1986). For horses, mules, and asses that estimate is about 2.5%. While these estimates are
also uncertain and are likely to vary among regions and countries, the global emissions from
these species are much smaller than the emissions from ruminant animals. Consequently, the
uncertainty in these values does not contribute significantly to the uncertainty in the estimates
of total global CH4 emissions from livestock digestion.
2.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The general approach described in OECD (1991) is recommended for estimating CH4
emissions from livestock. Each of the following steps may be performed for each animal
species:
(1) Define representative animal types. For example, cattle could be divided into
dairy cattle and beef cattle. Within these categories, further divisions by age
(e.g., calves and mature animals) and production level should be considered.
(2) Estimate feed energy intake for each representative animal type. The feed
requirements for each representative animal type should be calculated based
on data about their size and production level. For example, large dairy cows
with high levels of production require more feed than growing calves.
(3) Estimate the portion of feed-energy intake that is converted to CK, by the
animal. As discussed above, a general range of 5.5-6.5% is applicable to most
7 Ryegrass grown under the same conditions as the tropical grasses.
8 Temperate clover hay grown under the same conditions as the tropical grasses.
Page 2-4
-------
well-fed ruminant animals in temperate agricultural systems. Special cases,
such as cattle consuming high-grain diete, should use estimates that are
tailored to their specific feed-intake conditions. The Blaxter and Clapperton
(1965) relationship can be used to estimate this value in cases where diets fall
within the range of observations used to generate the relationship.
(4) Estimate emission factors per head for each animal type. The feed intake,
expressed in units of feed energy, is multiplied by the conversion percentage to
estimate CH4 energy produced per head. This value is then converted to
kilograms (kg) of CH4 per head per year, using a conversion of 55.65
megajoules (MJ) per 1,000 grams (g) of CH4. Separate emission factors are
estimated for each animal type.
(5) Multiply the emission factors by the appropriate animal populations. The total
emissions are computed as the emission factor per head times the appropriate
population of each animal type.9
The full implementation of this method requires detailed information about animal
populations and their feed and production characteristics. Because these characteristics vary
substantially for cattle among regions and countries, and even within countries, it is expected
that emission factors also will vary significantly. Therefore, it is recommended that this
method be implemented for cattle in as much detail as possible, given available data for those
countries for which cattle are an important source of emissions.
With the exception of two broad and general differences between developed and
developing countries (discussed below), the emission factors for the other animal species are
not expected to vary as much by country as the emission factors for cattle. Additionally,
emissions from the other species are not estimated to be a large portion of total CH4
emissions from livestock, with only several notable exceptions (such as sheep emissions in
New Zealand). Consequently, in general it is recommended that broadly representative
emission factors be used to estimate emissions from these other species and that the full
implementation of the detailed method is not necessary.
2.3.1 Implementation of the Methodology
Cattle
This section implements the methodology for cattle in nine regions of the world. These
regions and the six countries with the largest cattle populations are listed in Table 2-1. As
shown in the table, the total cattle population was about 1.3 billion in 1990, and the six
countries with the largest populations accounted for about one-half of the total.
9 The population of each animal type must reflect the time that an animal spends as that type. Although most
animal categories are based on a complete year, some categories are based on shorter time periods (e.g., U.S.
feedlots are based on a 140-day period).
Page 2-5
-------
TABLE 2-1
CATTLE POPULATIONS: 1990
Regions
North America
Western Europe
Oceania
Eastern Europe
China and Centrally Planned Asia
Middle East
Africa
Latin America
South and East Asia
World Total
Key Countries
India (71%)a
Brazil (45%)
Former Soviet Union (77%)
United States (89%)
China (89%)
Australia (64%)
Key Country Total (51%f
Population (millions)
110
100
36
154
87
13
188
314
278
7,279
Population (millions)
197
140
118
98
77
23
653
a. Percentage of regional cattle population accounted for
by this country.
b. Percentage of global cattle population accounted for by
the key countries.
Source: FAO, 1991.
Information characterizing the cattle in each of the major countries was reviewed to
estimate emission factors for those countries. These emission factors were used to estimate
emissions from each of the regions. As discussed previously, the characteristics of cattle in
many developing countries are not well documented. While the best readily available data
were used, the estimates could be improved by collecting more detailed information for the
countries with large cattle populations. Following are descriptions of each step of the method.
Page 2-6
-------
Step 1: Define Representative Animal Types
Developing precise estimates of emissions necessitates dividing cattle into categories
of relatively homogeneous groups. For each category, a representative animal is chosen and
characterized for purposes of estimating an emission factor. For purposes of this study, the
definition of "categories" is limited by the availability of data for characterizing the animals in
terms of the key variables needed to estimate an emission factor. This limitation is particularly
severe for many developing countries and regions for v/hich precise data were not readily
identified. Consequently, a relatively aggregate analysis is presented here for those countries
and regions. More detailed assessments performed by animal experts within the individual
countries should be performed to improve these estimates.
Table 2-2 presents a set of representative animal types that are recommended for
estimating emission factors for cattle. Two main categories, Mature Cattle and Young Cattle,
are recommended as the minimum set of representative types. The subcategories listed
should be used when data are available. In particular, the subpopulation of milking cows
should be identified because the feed intake necessary to support milk production can be
substantial. In the United States, the feedlot category is needed so that the implications of the
high-grain diets can be incorporated.
TABLE 2-2
REPRESENTATIVE CATTLE TYPES
Main Categories
Subcategories
Mature Animals
Mature Females:
Dairy Cows:
used principally for
commercial milk production
Beef Cows: used principally for
producing beef steers and
heifers
Multiple-Use Cows: used for milk production,
draft power, and other uses
Mature Males:
Breeding Bulls:
Draft Bullocks:
used principally for breeding
purposes
used principally for draft
power
Young Animals
Preweaned Calves
Growing Heifers, Steers/Bullocks and Bulls
Feedlot-Fed Steers and Heifers on High-Grain Diets
Page 2-7
-------
Step 2: Estimate Feed Intake
Feed intake is estimated based on the feed-energy requirements of the representative
animals, subject to feed-intake limitations. For this analysis, the net energy system described
in NRC (1984 and 1989) is used. Because this system was developed for feeding conditions
in temperate regions, several adjustments were made to avoid potential biases when applied
to evaluate feed-energy intakes for tropical cattle (see Appendix B).
The net energy system specifies the amount of feed-energy required for the
physiological functions of cattle, including maintenance, growth, and lactation. Feed-energy
requirements for work have also been estimated and are included in this analysis for the draft
animals in developing countries. Energy requirements for pregnancy have also been added
for the portion of cows that give birth in each year. The following information is required to
estimate feed-energy intakes:
Maintenance. Maintenance refers to the apparent feed energy required to keep
the animal in energy equilibrium, i.e., there is no gain or loss of energy in the
body tissues (Jurgens, 1988). For cattle, net energy for maintenance (NEJ has
been estimated to be a function of the weight of the animal raised to the 0.75
power (NRC, 1984):
NEm(MJIdafi = 0.322 x (weight in kg]
0.75
(2.1)
NRC (1989) recommends that lactating dairy cows be allowed a slightly higher
maintenance allowance:
NEm(MJIdafi = 0.335 x (weight In kg)0-75 {dairycows}
(2.1*)
Consequently, the weight of the animal is needed to estimate maintenance
requirements in megajoules per day (MJ/day). It has also been suggested that
additional energy is required for animals to obtain their food. Grazing animals
require more energy for this activity than do stall-fed animals. The following
energy requirements are added for this activity:10
Confined animals (pens and stalls): no additional NE
Animals grazing good-quality pasture: 17% of NEm;
Animals grazing over very large areas: 37% of NE
m>
Growth. The energy requirements for growth can be estimated as a function of
the weight of the animal and the rate of weight gain. NRC (1989) presents
formulae for large- and small-frame males and females. The equations for
large-frame females and small-frame males produce about the same values for
energy requirements per unit of weight gain. The equation for large-frame
males produces values about 25% lower per unit of weight gain, and the
equation for small-frame females produces values about 25% higher. For this
10 OECD (1991) recommends slightly higher energy additions. These revised figures are based on newly
published information in AAC (1990).
Page 2-8
-------
analysis, the equation for large-frame females is used, which is about the
average for the four types:
NEJMJfday) = 4.18 x (0.035 W°-7S x
.119
WG)
(2.2)
where:
W = animal weight in kilograms (kg); and
WG = weight gain in kg per day.
The relationships for NEg were developed for temperate agricultural conditions,
and may overestimate energy requirements for tropical conditions, particularly
for draft animals that may have a lower fat content in their weight gain
(Graham, 1985). However, no data are available for improving the estimates at
this time.
Lactation. Net energy for lactation has been expressed as a function of the
amount of milk produced and its fat content (NRC, 1989):
= kg of midday x (1 ,,47 + 0.40 x Fat%)
(2.3)
At 4.0% fat, the NE, in MJ/day is about 3.1 x kg of milk per day.
Draft Power. Various authors have summarized the energy intake requirements
for providing draft power (e.g., Lawrence, 1985; Bamualim and Kartiarso, 1985;
and Ibrahim, 1985). The strenuousness of the work performed by the animal
influences the energy requirements; consequently, a wide range of energy
requirements has been estimated. The values by Bamualim and Kartiarso
show that about 10% of NEm requirements are required per hour of typical work
for draft animals. This value is used in this analysis.
Pregnancy. Daily energy requirements for pregnancy are presented in NRC
(1984). Integrating these requirements over a 281-day gestation period yields
the following equation:
NE.
•pregnancy
(MJ/281 -day period) = 28 x calf birth weight in kg
(2.4)
The following equation can be used to estimate the approximate calf birth
weight as a function of the cow's weight:11
Calf birth weight (kg) = 0.266 x (cow weight in kg)0-79 (2-5)
Manipulating equations 2.4 and 2.5, in conjunction with equation 2.1, shows
that the NE required for pregnancy is about 7.5% of NEm for the range of cow
sizes considered in this analysis. Therefore, a factor of 7.5% of NEm is added
11 This species-specific equation from Bobbins and Bobbins (1979) was adjusted to the mean cow and calf
weight of a typical beef breed of cattle. This adjustment increases the coefficient in the equation from 0.214 to
0.266.
Page 2-9
-------
to account for the energy required for pregnancy for the portion of cows giving
birth each year.
Based on these equations, the following information is needed to estimate the
net energy intakes of the representative animal types: weight in kilograms;
feeding situation (stall- or pen-fed, grazing good pasture, or grazing over very
large areas); weight gain per day in kilograms; milk production in kilograms of
4% fat-corrected milk; number of hours of work performed per day; and portion
that give birth. Table 2-3 presents estimates of these values for the
representative animal types in each of the six key countries (equation 2.1* is
used only for mature dairy cows). Additionally, Table 2-3 (and subsequent
tables) includes representative cattle types for Africa and the Middle East and
Western Europe.
As shown in Table 2-3, four to eight representative animal types are defined for each
country or region. The data on animal weights come from a variety of sources.12 Because
data on cattle in developing countries are less available, judgment is required in some cases
to make the estimates. In many cases the estimates are based on case studies conducted in
individual regions or countries. Generally, the cattle in developing countries are smaller at all
ages than the cattle in developed countries.
The feeding situations for the animal types are assigned based on the types of animal-
production systems employed. For example, in India, due to limitations on the availability of
grazing land, nearly all animals are stall-fed to some degree. Alternatively, in Brazil, nearly all
the animals graze, some over large areas. In the United States, dairy cows and feedlot cattle
are stall- or pen-fed, while most other cattle graze. With the exception of Brazil, it was
assumed that all dairy cows are primarily stall-fed. Also, all draft animals were assumed to be
stall-fed. In China and Africa, separate animal types were defined to represent that some
cattle principally graze, while others are principally stall-fed.
Weight gain is estimated based on the mature weights of the animals and the
approximate age of maturity. For example, young cattle in India have three to four years to
gain about 50-200 kg. The average daily weight gain would be about 0.05-0.14 kg per day
(kg/day), and a middle value of 0.10 kg/day is used.13 Mature animals are assumed to have a
constant weight; consequently, no weight gain is reported.14
12 Estimates of cattle weights were developed from Reuss et al. (1990) and sources identified therein,
including: Pathak and Jakhomola (1983); Simpson and Farris (1982); CIAT (1975 and 1979); Simpson (1988); Rong
Yi (1983); Gryseels et al. (1988); Doppier (1980); and ADETEF (1982).
13 50 kg + (365 days x 3 years) = 0.05 kg/day. 200 kg -*- (365 days x 4 years) = 0.14 kg/day.
14 Even mature animals gain and lose weight during the year. In some cases, weight change between the wet
and dry seasons can be large. Reflecting these fluctuations in weight during the year would tend to increase the
estimate of the feed-energy intake requirements of the animals by a small amount. However, insufficient data are
available to quantify this effect for most regions.
Page 2-10
-------
TABLE 2-3
DATA ON REPRESENTATIVE CATTLE TYPES
Animal Type
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Animal Category
Dairy Cows
Other Cows
Draft Bullocks
All
Dairy Cows
Other Cows
Bulls
All
Dairy Cows
Other Cows
Bulls
All
Weight
(kg)
275
125
200
80
400
400
450
230
Former
550
500
600
230
Feeding
Situation
India
Stall Fed
Stall Fed
Stall Fed
Stall Fed
Brazil
Pasture/Range
Large Areas
Large Areas
Large Areas
Soviet Union
Stall Fed
Pasture/Range
Pasture/Range
Pasture/Range
Weight
Gain
(kg/day)
0.00
0.00
0.00
0.10
0.00
0.00
0.00
0.30
0.00
0.00
0.00
0.40
Milk
(kg/day)
2.5
0.6
0.0
0.0
2.2
1.1
0.0
0.0
7.0
3.3
0.0
0.0
Work
(hrs/day)
0.00
0.00
2.74
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Percent
Pregnant
50%
33%
0%
0%
80%
67%
0%
0%
80%
67%
0%
0%
United States
Mature Females
Mature Females
Mature Males
Young
Young
Young
Young
Feedlot
Dairy Cows
Other Cows
Bulls
Calves on Milk
Calves on Forage
Growing Heifers/Steers
Replacements
All
600
500
800
100
185
265
375
415
Stall Fed
Pasture/Range
Pasture/Range
Pasture/Range
Pasture/Range
Pasture/Range
Pasture/Range
Stall Fed
0.00
0.00
0.00
0.90
0.90
0.70
0.40
1.30
18.4
3.3
0.0
0.0
0.0
0.0
0.0
0.0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
90%
80%
0%
0%
0%
0%
0%.
0%
(continued)
Page 2-11
-------
TABLE 2-3
DATA ON REPRESENTATIVE CATTLE TYPES (Continued)
Animal Type
Mature Females
Mature Females
Mature Females
Mature Males
Mature Males
Young
Animal Category
Dairy Cows
Farming Region
Grazing Region
Farming Region
Grazing Region
All
Weight
(kg)
350
325
300
450
'400
200
Feeding
Situation
China
Stall Fed
Stall Fed
Pasture/Range
Stall Fed
Pasture/Range
Pasture/Range
Weight
Gain
(kg/day)
0.00
0.00
0.00
0.00
0.00
0.20
Milk
(kg/day)
4.5
1.1
1.1
0.0
0.0
0.0
Work
(hrs/day)
0.00
0.55
0.00
1.37
0.00
0.00
Percent
Pregnant
80%
33%
50%
0%
0%
0%
Australia
Mature Females
Mature Females
Mature Males
Young
Dairy Cows
Other Cows
Bulls
All
500
400
450
200
Stall Fed
Pasture/Range
Pasture/Range
Pasture/Range
0.00
0.00
0.00
0.30
4.7
2.4
0.0
0.0
0.00
0.00
0.00
0.00
80%
67%
0%
0%
Africa/Middle East
Mature Females
Mature Females
Mature Males
Mature Females
Mature Males
Young
Dairy Cows
Other Cows
Draft Bullocks
Other Cows: Grazing
Bulls: Grazing
All
275
200
275
200
275
75
Stall Fed
Stall Fed
Stall Fed
Large Areas
Large Areas
Pasture/Range
0.00
0.00
0.00
0.00
0.00
0.10
1.3
0.3
0.0
0.3
0.0
0.0
0.00
0.55
1.37
0.00
0.00
0.00
67%
33%
0%
33%
0%
0%
Western Europe
Mature Females
Mature Males
Young
Young
Young
Dairy Cows
Bulls
Replacement/
Growing
Calves on Milk
Calves on Forage
550
600
400
100
230
Stall Fed
Pasture/Range
Pasture/Range
Pasture/Range
Pasture/Range
0.00
0.00
0.40
0.30
0.30
11.5
0.0
0.0
0.0
0.0
0.00
0.00
0.00
0.00
0.00
90%
0%
0%
0%
0%
Page 2-12
-------
Milk production is estimated for daily cows and other mature females. Milk production
per dairy cow is taken from FAO (1991). Milk production for other mature females is
estimated as follows: for developing countries, milk production is assumed to be 25% of the
dairy cow value in the country; and for developed countries and Brazil (Latin America), milk
production is assumed to be 50% of the dairy cow value, up to a maximum of 1,200 kg of milk
per year per cow.15
Only draft bullocks and other mature females (nondairy cows) in developing countries
are estimated to perform work. In Asia and Africa, draft bullocks are estimated to work five
hours per day for 100 days a year, for a total of 500 hours per year, or 1.4 hours per day on
average. Other mature females are assumed to work for two hours per day for 100 days, for
an average of 0.55 hours per day for the entire year. In the Indian subcontinent, draft
bullocks are estimated to work for 10 hours per day for 100 days (2.7 hours per day on
average), and other mature females are assumed to perform no work at all. While it is true
that the duration and strenuousness of work vary, these values are believed to be broadly
representative of the work provided by these animals in the various regions.
Based on the data presented in Table 2-3, the net energy requirement of each of the
representative animal types was estimated, as shown in Table 2-4. As expected, the larger,
more productive cattle have higher feed-energy requirements.
These net energy requirements must be translated into gross energy intakes. Gross
energy-intake estimates also allow the net energy estimates to be checked for reasonableness
against expected ranges of feed intake as a percentage of animal weight. Estimating gross
energy intake requires considering the relationship between the net and gross energy values
of different feeds. This relationship can be summarized briefly as follows:
Digestible Energy
Metabolizable Energy
Net Energy
Gross Energy - Fecal Losses
Digestible Energy - Urinary and Combustible Gas Losses
Metabolizable Energy - Heat Increment
Net Energy
Gross hnergy - Fecal Losses - Urinary and Combustible
Gas Losses - Heat Increment
The quantitative relationship among these energy values varies among feed types.
Additionally, the values depend on how the feeds are prepared and fed, and the level at which
they are fed. For purposes of this analysis, simplifying assumptipris^are used to produce a
reasonably representative relationship for the range of diets typicallyled to cattle.
15 The value of 1,200 kg is typical for the milk production of a U.S. beef cow per lactation. Milk production for
nursing a calf is unlikely to exceed this level. Milk production in developing countries is expected to be lower due to
poorer feed resources.
Page 2-13
-------
TABLE 2-4
NET ENERGY REQUIREMENTS FOR REPRESENTATIVE CATTLE TYPES
(MJ/day)
Animal Type
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Animal Category NEm NE,ead
Dairy Cows
Other Cows
Draft Bullocks
All
Dairy Cows
Other Cows
Bulls
AH
India
22.6
12.0
17.1
8.6
Brazil
30.0
28.8
31.5
19.0
0.0
0.0
0.0
0.0
5.1
10.7
11.6
7.0
NEg
0.0
0.0
0.0
0.7
0.0
0.0
0.0
3.5
NE,
7.8
1.9
0.0
0.0
6.8
3.4
0.0
0.0
NEW
0.0
0.0
4.7
0.0
0.0
0.0
0.0
0.0
NEp
0.8
0.3
0.0
0.0
1.8
1.4
0.0
0.0
Former Soviet Union
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Young
Young
Young
Feedlot
Mature Females
Mature Females
Mature Females
Dairy Cows
Other Cows
Bulls
All
Dairy Cows
Other Cows
Bulls
Calves on Milk
Calves on Forage
38.0
34.0
39.0
19.0
United States
40.6
34.0
48.4
10.2
16.2
Growing Heifers/Steers 21 .1
Replacements
All
Dairy Cows
Farming Region
Grazing Region
27.4
29.6
China
26.8
24.6
23.2
0.0
5.8
6.6
3.2
0.0
5.8
8.2
1.7
2.7
3.6
4.7
0.0
0.0
0.0
3.9
0.0
0.0
0.0
4.8
0.0
0.0
0.0
7.9
10.3
9.4
6.1
23.5
0.0
0.0
0.0
21.7
10.2
0.0
0.0
57.0
10.2
0.0
0.0
0.0
0.0
0.0
0.0
14.0
3.4
3.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.4
0.0
2.3
1.7
0.0
0.0
2.7
2.0
0.0
0.0
0.0
0.0
0.0
0.0
1.6
0.6
0.9
(continued)
Page 2-1 4
-------
TABLE 2-4
NET ENERGY REQUIREMENTS FOR REPRESENTATIVE CATTLE TYPES
(MJ/day) (continued)
Animal Type Animal Category
NEm
NEteed
NEg
NE,
NEW
NEp
China (continued)
Mature Males Farming Region
Mature Males Grazing Region
Young All
31.5
28.8
17.1
0.0
4.9
2.9
0.0
0.0
2.1
0.0
0.0
0.0
4.3
0.0
0.0
0.0
0.0
0.0
Australia
Mature Females Dairy Cows
Mature Females Other Cows
Mature Males Bulls
Young All
35.4
28.8
31.5
17.1
0.0
4.9
5.3
2.9
0.0
0.0
0.0
3.3
14.6
7.4
0.0
0.0
0.0
0.0
0.0
0.0
2.1
1.4
0.0
0.0
Africa/Middle East
Mature Females Dairy Cows
Mature Females Other Cows
Mature Males Draft Bullocks
Mature Females Other Cows: Grazing
Mature Males Bulls: Grazing
Young All
Western
Mature Females Dairy Cows
Mature Males Bulls
Young Replacement/Growing
Young Calves on Milk
Young Calves on Forage
NEm = net energy for maintenance.
NE[Md = net energy for obtaining feed; stall fed, grazing,
NEg = net energy for growth.
NE, = net energy for lactation (milk production).
NEy, = net energy for work.
NEp = net energy for pregnancy.
22.6
17.1
21.7
17.1
21.7
8.2
Europe
38.0
39.0
28.8
19.0
19.0
or grazing
0.0
0.0
0.0
6.3
8.0
1.4
0.0
6.6
4.9
3.2
3.2
over very
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.0
6.4
3.5
3.5
large areas
4.0
0.9
0.0
0.9
0.0
0.0
35.7
0.0
0.0
0.0
0.0
0.0
0.9
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.1
0.4
0.0
0.4
0.0
0.0
2.6
0.0
0.0
0.0
0.0
Page 2-15
-------
Given the above relationship between net and gross energy, the digestibility of feeds
consumed by cattle was assigned to each of the key countries and regions by animal type as
follows:
Dairy cows generally get the best available feeds. Consequently, they have
feeds with high digestibilities. A value of 60% was assigned for all dairy cows,
except for India (where large amounts of crop by-products are fed) and the
United States (where grains are fed). An estimate of 55% is used for dairy
cows in India, and an estimate of 65% is used for dairy cows in the United
States.
Most of the grazing animals were assumed to consume feed with a digestibility
of 60%. Young cattle were assumed to have a higher digestibility (65%)
because they are nursing for a period of time, are often supplemented, and
preferentially consume forages with higher digestibilities. Because the
digestibilities of grazing resources in Africa were assumed to be of lower
quality, a 55% digestibility was assumed.
Stall-fed animals subsisting on crop by-products were assumed to have a
digestibility of 55%, except in India, where 50% was assumed. Feedlot cattle in
the United States that consume large amounts of grain were estimated to have
an 82% digestibility.
Given the digestibility of the feed, a general relationship between digestible energy and
metabolizable energy can be used as follows (NRC, 1984):
Metabolizable Energy (ME) = 0.82 x Digestible Energy (DE) (2-6)
Equation 2.6 is a simplified relationship; larger (smaller) CH4 conversion rates would tend to
reduce (increase) the coefficient to values below (above) 0.82.
NRC (1984) presents a quantitative relationship between metabolizable energy and net
energy. Separate relationships are presented for net energy used for growth versus net
energy used for other functions. However, the NRC (1984) relationships were developed
based on diets with digestibilities that are higher than most of the digestibilities discussed
previously. Consequently, the NRC (1984) relationship (which is nonlinear) may not be
appropriate for all of the digestibility values used in this analysis.
Other energy systems (e.g., ARC, 1980) use linear relationships between
metabolizable energy and net energy values for feeds. The nonlinear nature of the NRC
(1984) relationship could bias the estimates of feed intake upward, particularly for low-
digestibility feeds. Such an upward bias could also bias the emission estimate upward.
Consequently, as described in Appendix B, a linear relationship was adopted. The following
equations define the ratio of net energy to metabolizable energy when digestibility is 65% or
lower:
Page 2-16
-------
NE/ME = 0.364 + 0.00499 x ME%
NEJME = -0.044 + 0.00796 x ME%
where:
NE/ME
NEg/ME
ME%
the ratio for net energy consumed for maintenance, lactation,
work, and pregnancy;
the ratio for net energy consumed for growth; and
the metabolizable energy as a percentage of gross energy,
expressed in percent (e.g., 50%).
For example, for a digestibility of 60%, ME% equals 49.2%, NE/ME equals 0.61, and NEg/ME
equals 0.35.
Based on equation 2.6, these linear relationships can be expressed as a function of
digestible energy as follows:
NEIDE = 0.298 + 0.00335 x DE%
(2.7)
NEgIDE = -0.036 + 0.00535 x DE%
For example, at DE% of 60%, NE/DE equals 0.50, and NE/DE equals 0.28.
(2.8)
Given the estimates for feed digestibility and equations 2.7 and 2.8, the gross energy
intake (GE in MJ/day) can be estimated as follows:
GE = [(NEm H- NEfeed + NE, + NEW + NEp) + [NEfDE\ + (NEg +
+ (DE%/100) (2.9)
where:
{NE/DE} is computed from equation 2.7;
{NE^DE} is computed from equation 2.8; and
DE% is digestibility in percent (e.g., 60%).
Using equation 2.9 for the cases in which digestible energy is 65% or lower (which is nearly
all the cases), and the nonlinear NRC relationships for the case in which digestibility exceeds
65%, Table 2-5 lists the estimated gross energy intakes and intakes as a percentage of body
weight. Nearly all of the estimated daily intakes are between 1 .5% and 3.0% of body weight,
a reasonable range for the diverse cattle management conditions found globally.
These intake estimates also are consistent with previous estimates. For example, the
energy requirements and intakes listed in Tables 2-4 and 2-5 imply an average metabolizable
energy (ME) intake for Indian cattle of about 10,000 MJ per year. Winrock (1978) estimates
the average ME requirements for Indian cattle at 10,600 MJ per year. Similarly, the values
implied for the U.S. dairy and beef cows are 58,000 MJ and 31,000 MJ per year, respectively,
which are similar to the respective estimates of 62,000 MJ and 31 ,700 MJ derived in U.S.
EPA (1 993). Consequently, for the diverse set of conditions represented in the analysis, the
Page 2-17
-------
TABLE 2-5
GROSS ENERGY INTAKE FOR REPRESENTATIVE CATTLE TYPES
Animal Type
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Young
Young
Young
Feedlot
Mature Females
Mature Females
Mature Females
Animal Category Digest (%)a
India
Dairy Cows
Other Cows
Draft Bullocks
All
Brazil
Dairy Cows
Other Cows
Bulls
All
Former Soviet Union
Dairy Cows
Other Cows
Bulls
All
United States
Dairy Cows
Other Cows
Bulls
Calves on Milk
Calves on Forage
Growing Heifers/Steers
Replacements
All
China
Dairy Cows
Farming Region
Grazing Region
55
50
50
50
60
60
60
60
60
60
60
60
65
60
60
NA
65
65
60
75
60
55
60
GEb
117.7
61.0
93.7
43.2
145.9
148.0
144.0
107.5
207.2
172.9
152.5
102.2
299.5
174.0
189.9
NA
107.2
120.1
143.2
161.8
141.6
113.2
105.0
Percent BW°
2.3%
2.6%
2.5%
2.9%
2.0%
2.0%
1 .7%
2.5%
2.0%
1 .9%
1.4%
2.4%
2.7%
1.9%
1.3%
NA
3.1%
2.5%
2.1%
2.1%
2.2%
1.9%
1.9%
(continued)
Page 2-18
-------
TABLE 2-5
GROSS ENERGY INTAKE FOR REPRESENTATIVE CATTLE TYPES (Continued)
Animal Type
Mature Males
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Mature Females
Mature Males
Young
Mature Females
Mature Males
Young
Young
Young
Animal Category Digest (%)a
China (continued)
Farming Region
Grazing Region
All
Australia
Dairy Cows
Other Cows
Bulls
All
Africa/Middle East
Dairy Cows
Other Cows
Draft Bullocks
Other Cows: Grazing
Bulls: Grazing
All
Western Europe
Dairy Cows
Bulls
Replacement/Growing
Calves on Milk
Calves on Forage
55
60
60
60
55
55
55
60
55
55
55
55
60
60
60
60
NA
65
GEb
134.9
112.5
79.3
174.1
160.5
138.8
98.6
92.8
73.2
93.2
93.6
112.3
36.2
254.7
152.5
149.8
NA
83.7
Percent BWC
1.6%
1.5%
2.1%
1.9%
2.2%
1.7%
2.7%
1.8%
2.0%
1.8%
2.5%
2.2%
2.6%
2.5%
1.4%
2.0%
NA
2.0%
a. Digestibility in percentage on an energy basis.
b. Gross-energy intake, MJ/day.
c. Dry-matter intake as a percentage of body weight. Energy density of feed estimated
at 18.45 MJ/kilogram.
NA = values not estimated for nursing calves.
Page 2-19
-------
intake estimates correspond to reasonably expected ranges and previously published
estimates.
Step 3: Estimate Conversion of Feed to Methane
As discussed previously, the rate of conversion of feed energy to CH4 is relatively
constant across most feeding and management situations. The following estimates were
adopted for this analysis:
Developed Countries. A 6.0% conversion rate was assumed for all cattle in
developed countries, except for feedlot cattle in the United States, where a
3.5% rate was used.
Developing Countries. Several assumptions were made for different animal
management situations in developing countries:
All dairy cows and young cattle were assumed to have a conversion rate
of 6.0%. These cattle are generally the best-fed cattle in these regions.
All nondairy, nonyoung stall-fed animals consuming low-quality crop by-
products were assumed to have a conversion rate of 6.5%, except the
cattle in India, which were assigned a conversion rate of 7.5%, because
feed resources are particularly poor in many cases in this region.
All grazing cattle were assumed to have a conversion rate of 6.0%,
except for grazing cattle in Africa, which were assigned a rate of 7.5%
because of the forage characteristics found in many portions of tropical
Africa.
Step 4: Develop Emission Factors
Methane emission factors per head were developed for each of the representative
animal types by using the feed-intake and CH4-conversion-rate estimates. First, the daily
feed-energy intake was multiplied by the CH4 conversion rate for each animal type. Then, the
daily CH4 emission rate was estimated using the conversion factor of 55.65 MJ per 1,000 g of
CH4. The resulting daily estimates, shown in Table 2-6, vary from 0 g per day (for nursing
calves) to over 300 g per day across the animal types.
Other Livestock
Table 2-7 lists emission factors that may be used for the noncattle species. While
some variation is expected in the level of feed-energy intake among countries, the average
values reported are reasonable for most situations. For two species, buffalo and camels, the
estimates of the CH4 conversion rates are particularly uncertain because measurements are
not available for the types of diets they typically consume. Crutzen et al. (1986) assumed a
CH4 conversion rate of 9.0% for these animals. In this analysis, lower rates were used
because the poor-quality diets generally consumed by these animals would not likely support
a 9.0% conversion to CH4. For buffalo, a 7.5% rate was used, consistent with Leng (1991).
A recent report by Carmean et al. (1992) with the llama, a close relative of the camel,
found a 7.1% CH4 conversion rate with a mixed concentrate-forage diet fed at restricted levels
near maintenance. The significance of this result for choosing a CH4 conversion rate for
Page 2-20
-------
TABLE 2-6
METHANE EMISSION FACTORS FOR CATTLE
Animal Type
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Animal Category
Dairy Cows
Other Cows
Draft Bullocks
All
Dairy Cows
Other Cows
Bulls
All
CH4 CH4 CH4 Population
Percent (g/day) (kg/head/yr)a Mix (%)
India
6.0%
7.5%
7.5%
6.0%
Brazil
6.0%
6.0%
6.0%
6.0%
127
82
126
47
157
160
155
116
46.3
30.0
46.1
17.0
57.4
58.2
56.7
42.3
15%
34%
9%
43%
13%
32%
5%
50%
Average
Emission
Factor
(kg/head/yr)
28.3
50.1
Former Soviet Union
Mature Females
Mature Females
Mature Males
Young
Dairy Cows
Other Cows
Bulls
All
6.0%
6.5%
6.5%
6.0%
223
202
178
110
81.5
73.7
65.0
40.2
30%
21%
15%
34%
63.4
United States
Mature Females
Mature Females
Mature Males
Young
Young
Young
Young
Feedlot
Dairy Cows
Other Cows
Bulls
Calves on Milk
Calves on Forage
Growing Heifers/Steers
Replacements
All
6.0%
6.0%
6.0%
0.0%
6.0%
6.0%
6.0%
3.5%
323
188
204
0
116
130
154
102
117.9
68.5
74.5
0.0b
8.1°
19.4
56.3
14.3"
10%
32%
2%
14%
7%
15%
10%
10%
54.2
(continued)
Page 2-21
-------
TABLE 2-6
METHANE EMISSION FACTORS FOR CATTLE (Continued)
Animal Type
Mature Females
Mature Females
Mature Females
Mature Males
Mature Males
Young
Mature Females
Mature Females
Mature Males
Young
Mature Females
Mature Females
Mature Males
Mature Females
Mature Males
Young
Animal Category
Dairy Cows
Farming Region
Grazing Region
Farming Region
Grazing Region
All
Dairy Cows
Other Cows
Bulls
All
Dairy Cows
Other Cows
Draft Bullocks
Other Cows: Grazing
Bulls: Grazing
All
Average
Emission'
CH4 CH4 CH, Population Factor
Percent (g/day) (kg/head/yr)a Mix (%) (kg/head/yr)
China
6.0%
6.5%
6.0%
6.5%
6.0%
6.0%
Australia
6.0%
6.0%
6.0%
6.0%
Africa/Middle
6.0%
6.5%
6.5%
7.5%
7.5%
6.0%
153
132
113
158
121
86
188
173
150
106
East
100
86
109
126
151
39
55.7
48.3
41.3
57.5
44.3
31.2
68.5
63.2
54.6
38.8
36.5
31.2
39.7
46.0
55.2
14.2
3%
26%
9%
23%
8%
31% 44.5
7%
48%
' 10%
35% 54.2
20%
10%
10%
5%
20%
35% 32.7
Western Europe
Mature Females
Mature Males
Young
Young
Young
Dairy Cows
Bulls
Replacement/Growing
Calves on Milk
Calves on Forage
a. Estimates based on 365 days of feeding,
b. Based on 140
days of feeding.
6.0%
6.0%
6.0%
0.0%
6.0%
275
164
162
0
90
100.2
60.0
84.0
0.0C
6.3d
31%
15%
38%
11%
5% 64.2
unless otherwise noted.
c. Based on 70 days of feeding.
d. Based on 150
days of feeding.
Page 2-22
-------
TABLE 2-7
EMISSION FACTORS FOR LIVESTOCK
Animal
Buffaloa
India
Other Countries
Sheep
Developed Countries
Developing Countries
Goats
Camels
Pigs
Developed Countries
Developing Countries
Horses
Mules and Asses
Gross Energy Intake
(MJ/day)
108
119
20
13
14
100
38
13
110
60
Einergy Intake
Released as Methane
(%)
7.5"
7.5b
6.0
6.0
6.0
7.0C
0.6
1.3
2.5
2.5
Emission Factor
(kg/head/yr)
53.0
58.0
8.0
5.0
5.0
46.0d
1.5
1.0
18.0
10.0
a. The derivation of the feed-energy intake for buffalo is described in Appendix A. The value of 85
MJ/day used by Crutzen et al. was found to be too low based on the information reviewed for
this study.
b. Crutzen et al. assume that the developing country emission factor for sheep is
appropriate to use for Australia and New Zealand. In this analysis, the developed
country emission factor for sheep was found to be more representative of the emission
rate for sheep in these countries.
c. The CH4 conversion rates for buffalo and camels were reduced from the 9.0% reported
by Crutzen et al. Although available measurements are sparse, a 9.0% conversion
rate to CH4 is unlikely to be supported on the poor-quality feeds typically consumed by
buffalo and camels.
d. The emission factor was reduced to 46 from the 58 reported by Crutzen et al., to be
consistent with the 7.0% CH4 conversion rate in this analysis.
MJ = megajoules.
Source: Crutzen et al., 1986.
Page 2-23
-------
camels Is difficult to assess for two reasons: camels will typically be fed above maintenance
so they can provide draft power, and camels will not typically be fed concentrates. In the
absence of better information, an assumption of 7% was used in this analysis.
Based on a review of available data, the feed-energy intake levels assumed for buffalo
by Crutzen et al. (1986) appear to be underestimates. Consequently, the feed-intake levels
and emission factors for buffalo are derived separately in Appendix A. As discussed in the
appendix, the average feed-energy intake requirements for buffalo were calculated using data
and certain assumptions about their size, feeding situation, weight gain, lactation, -and typical
work hours. The feed-energy-intake estimates are on the order of 90-150 MJ per day, with
the averages reported in the table. In comparison with Crutzen et al., the buffalo feed-energy
intake level is higher in this study, and the CH4 conversion rate is lower. These two
differences offset each other to a large extent, so that the emission factors for buffalo listed in
Table 2-7 are quite similar to the 50 kg/head/yr reported by Crutzen et al.
The 6.0% CH4 conversion rate used for the sheep and goats falls within the 5.5-6.5%
range discussed previously. Sheep and goats in developing countries are assumed to have
this conversion rate, despite the relatively poorer-quality feeds found in this region because
sheep and possibly, to some extent, goats are less able to digest poor-quality forages as
contrasted with cattle (Terada et al., 1987; and Minson, 1990). The reduced digestibility, and
thus, reduced fermentation of these forages appear to be associated with the faster rate of
passage of the particles through the digestive tract of sheep (Poppi et al., 1981).
The values for the nonruminant animals are broadly representative. The nonruminant
animals do not contribute significantly to total emissions. Consequently, further refinement of
the estimates is not considered vital. The CH4 emissions for these species for a given country
are estimated by multiplying the emission factors in Table 2-7 by the animal populations. The
animal population data are reported by the Food and Agriculture Organization (FAO) in the
Production Yearbook series (FAO, 1991).
2.4 RESULTS
2.4.1 Cattle
The total cattle population and the dairy cow population for each country and region
were taken from FAO (1991). The makeup of the nondairy cow populations for each country
or region was developed from a variety of sources.16 These data were used to estimate an
average emission factor per head for each country and region that reflects the mix of animal
types within that country or region, as well as the number of days that animals spend in each
animal category in a year (Table 2-6). For example, the average emission factor for India was
estimated to be 28.3 kg per head per year (kg/h/yr). This value is lower than the emission
factor reported by Crutzen et al. (1986) for developing countries. Western Europe and the
former Soviet Union have relatively higher emission factors because a large portion of the
18 The population estimates for India were developed from Vaidyanathan (1988), which presented detailed
breakdowns of cattle populations by age and sex based on the national agriculture census. Sources for other
regions were Reuss et al. (1990) and sources listed therein, including: Copland (1985); Ward et al. (1986); CIAT
(1979); Swann (1987); Gryseels et al. (1988); and Politick and Bakker (1982).
Page 2-24
-------
cattle in these regions consists of dairy cows, which tend to have higher emissions per head.
The average emission factor for the United States, 54.2 kg/h/yr, is similar to the value
developed in greater detail in U.S. EPA (1993) and is slightly below the value implied by the
estimates in Johnson et al. (1991).
The emission factors for each country and region were applied to the total cattle
population in each region, as reported by FAO (1991), to estimate regional and global CH4
emissions from cattle. Table 2-8 presents the estimates by region and for the key countries.
As shown in the table, total emissions are estimated at about 58 Tg for 1990. The regions
with the largest emissions are Latin America (15.7 Tg) and Eastern Europe (9.8 Tg), which
includes the former Soviet Union. The former Soviet Union and Brazil are the countries with
the largest emissions, followed by India and the United States.
2.4.2 Other Livestock
Table 2-9 lists the emissions for the other livestock by region. Cattle estimates also
are shown for completeness. Both the animal populations and the emission estimates (in Tg
per year) are reported. As shown in the table, buffalo arid sheep are the second and third
largest sources of emissions, each contributing about 10% of the total emission estimate. The
emissions from the remaining livestock are relatively small, accounting for less than 10% of
total emissions. The table also lists the top 10 countries in terms of emissions. These
countries account for an estimated 49.1 Tg of emissions, or about 60% of the total.
2.5 TRENDS
Future CH4 emissions from livestock will be driven by future production levels and
practices. With increasing human population, the trend of increased production of milk, meat,
and hides from livestock is expected to continue. Because this trend is contingent on
continued economic growth, it may be influenced by a variety of factors. For example, Parikh
et al. (1988) project a 1.6% annual increase in milk and bovine/ovine meat production through
the year 2000. Growth rates are expected to be larger among developing countries because
of more rapid population increases and because of the high income elasticities of demand for
these products among the relatively poorer populations of most developing countries.
U.S. EPA (1989) also developed a scenario of future milk and meat production as part
of its evaluation of policies for stabilizing global climate. This scenario, based on the results
of an international agricultural model that incorporates demand and supply functions for the
major agricultural commodities (including grains, meat, and milk products), trade policies, and
domestic agricultural pricing and supply policies, projects slightly lower rates of increases in
production over the next 20 years.
Some have questioned whether developing countries can achieve the increases in
animal production projected by such studies as Parikh et al. (1988) and U.S. EPA (1989)
because feed availability may impose a constraint on future production increases. Parikh et
al. address this issue and expect feed costs to increase as supplies tighten in some regions.
However, assuming that economic development continues, the prices for animal products are
also expected to increase, substantially offsetting the increased cost of feed.
Page 2-25
-------
TABLE 2-8
REGIONAL AND GLOBAL CATTLE EMISSIONS: 1990
Region/Country
North America
Western Europe
Oceania
Eastern Europe
China and Centrally Planned Asia
Middle East8
Africa
Latin America
South and East Asia
World Total
India (71%)b
Brazil (45%)
Former Soviet Union (77%)
United States (89%)
China (89%)
Australia (64%)
Key Country Total (52%f
Population Emission Factor
(millions) (kg/h/yr)
Regions
110
100
36
154
87
13
188
314
278
1,279
Key Countries
197
140
118
98
77
23
653
54.2
64.2
54.2
63.4
44.5
32.7
32.7
50.1
28.3
45.4
28.3
50.1
63.4
54.2
44.5
54.2
46.0
Emissions
(Tg/yr)
6.0
6.4
2.0
9.8
3.9
0.4
6.1
15.7
7.9
58.1
5.6
7.0
7.5
5.3
3.4
1.2
30.0
a. Emissions estimated using the emission factor for Africa.
b. Percentage of regional emissions accounted for by this country.
c. Percentage of global emissions accounted for by the key countries.
Source: Population data from FAO (1991).
Page 2-26
-------
TABLE 2-9
GLOBAL LIVESTOCK METHANE EMISSIONS
Region/Country
North America
Western
Europe
Oceania
Eastern Europe
China and CP
Asia
Middle East
Africa
Latin America
South and East
Asia
World Total
Cattle
110,449
99,831
36,024
154,171
86,986
12,646
187,771
313,502
277,877
1,279,257
Buffalo
0
653
0
682
26,119
379
2,500
1,209
109,216
140,758
Sheep
Livestock
12,123
141,043
226,149
179,672
128,193
72,71 1
205,094
119,731
105,784
1, 190,500
Goats
Populations
1,927
25,771
1,996
9,257
103,660
25,895
173,944
35,302
179,278
557,030
Camels
(OOOs)
0
3
0
300
1,035
899
14,509
0
2,704
19,450
Swine
64,384
111,835
17,155
148,972
379,152
446
13,585
80,060
41,174
856,763
Horses
5,630
2,455
519
8,283
12,742
395
4,987
22,828
3.081
60,920
Mules/
Asses
56
2,342
9
797
16,532
3,383
14,445
14,613
6,219
58,396
Total
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1990 Emissions (Tg/yr)
North America
Western
Europe
Oceania
Eastern Europe
China and CP
Asia
Middle East
Africa
Latin America
South and East
Asia
World Total
6.0
6.4
2.0
9.8
3.9
0.4
6.1
15.7
7.9
58.1
0.0
0.0
0.0
0.0
1.5
0.0
0.1
0.1
5.8
7.7
0.1
1.1
1.8
1.4
0.6
0.4
1.0
0.6
0.5
7.6
0.0
0.1
0.0
0.0
0.5
0.1
0.9
0.2
09
2.8
0.0
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.1
0.9
0.1
0.2
0.0
0.2
0.4
0.0
0.0
0.1
0.0
1.0
0.1
0.0
0.0
0.1
0.2
0.0
0.1
0.4
0.1
1.1
0.0
0.0
0.0
0.0
0.2
0.0
0.1
0.1
0.1
0.6
6.3
7.9
3.8
11.7
7.4
1.0
9.1
17.2
15.4
79.8
(continued)
Page 2-27
-------
TABLE 2-9
GLOBAL LIVESTOCK METHANE EMISSIONS (continued)
Region/Country
Cattle
Buffalo
Sheep
1990 Emissions for the
India
Former Soviet
Union
Brazil
China
United States
Argentina
Australia
Pakistan
Mexico
France
5.59
7.50
7.01
3.42
5.32
2.53
1.22
0.50
1.41
1.36
4.00
0.02
0.07
1.25
0.00
0.00
0.00
0.78
0.00
0.00
0.27
1.10
0.11
0.57
0.09
0.14
1.34
0.15
0.03
0.10
Goats
Top 10
0.55
0.03
0.06
0.49
0.01
0.02
0.00
0.18
0.05
0.01
Camels
Countries
0.07
0.01
0.00
0.02
0.00
0.00
0.00
0.05
0.00
0.00
Swine
(Tg/yr)
0.01
0.12
0.03
0.36
0.08
0.00
0.00
0.00
0.02
0.02
Horses
0.02
0.11
0.11
0.19
0.09
0.05
0.01
0.01
0.11
0.00
Mules/
Asses
0.02
0.00
0.03
0.17
0.00
0.00
0.00
0.03
0.06
0.00
Total
10.52
8.90
7.42
6.47
5.60
2.75
2.58
1.69
1.68
1.49
N/A = Not applicable.
Sources: Population data from FAO (1991). Emission estimates for cattle described above. Emission
factors for buffalo are derived in Appendix A. Emission factors for camels based on a 7.0% CH4
conversion rate. Emission factors for all other noncattle livestock taken from Crutzen et al. (1986).
If production practices were to remain constant, emissions from animals involved in the
production of meat, milk, and hides would be expected to increase at the same rate as the
increase in production. However, several factors must be considered:
Improved animal management and breeding practices are continuing to
increase animal productivity in developed and developing countries. Increased
productivity tends to reduce CH4 emissions per unit of product produced. This
point is emphasized by the fact that in 1990 nearly 80% of the emissions from
cattle was estimated to be associated with animal maintenance, so that only
20% was associated with feed intake required for production. Because
improved productivity can have a substantial impact on emissions per unit of
product produced, emissions will increase more slowly than the increase in
production, and could actually decline.
Because of changes in animal productivity, emissions do not track trends in
animal populations. As shown previously, emission factors per head for cattle
vary substantially among countries. These factors increase over time as
production levels per animal increase. Therefore, emissions can increase more
Page 2-28
-------
rapidly than the rate of increase in animal populations. Additionally, emissions
can increase as production increases, even if animal populations remain
unchanged or decline.
The projections of milk and meat production do not allow for the fact that a
substantial number of cattle, buffalo, and camels are used as draft animals.
The number of draft animals used is driven by a range of interrelated and
complex factors, including the size of landholdings (smaller landholdings tend to
have a higher number of draft animals per unit area), the availability and
mechanization of irrigation systems, the cost and availability of feed resources,
and the types of crops grown (Vaidyanathan, 1988). Projections of global draft
animal requirements are not available at this time.
Based on these observations, it is likely that CH4 emissions from livestock will continue to
increase in the future, although at a rate less than the rate of increase in meat and milk
production.
2.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
The CH4 emission estimates are uncertain for a variety of reasons. First, more
detailed analyses of feed energy requirements could be conducted to improve the precision of
the estimates for particular countries. A more detailed analysis would require better data to
characterize the production levels and management conditions of the animals. Second, better
descriptions of the feed characteristics would improve the estimates. In particular, data
describing the energy, protein, and other characteristics of crop by-products would help
improve the basis for making the estimates.
Third, measurements are required to provide a basis for estimating the CH4 conversion
rates for feed types and feeding levels typically encountered in tropical developing countries.
While limited data and simple models indicate that the CH4 production rates for typical tropical
production systems may be higher than the rate measured for temperate systems, this
difference must be verified. If a 6.0% conversion rate were assumed for all cattle, including all
cattle in tropical management systems, the global emission estimate for cattle for 1990 would
be reduced from 58.1 Tg to 56.2 Tg.
Based on the available data, the global emissions estimate is likely to be within a
range of about ±25% of the central estimate. This range accounts for the uncertainty in the
animal populations as well as the emission rates for the wide variety of animal management
conditions that exist. The overall range for CH4 emissions from the digestive processes of
livestock is, therefore, about 60 to 100 Tg per year.
2.7 CONCLUSIONS
The amount of CH4 emitted by domesticated livestock is estimated to be about 80 Tg
per year, which is consistent with previous estimates. Cattle are the largest single source,
accounting for 75% of the total. Buffalo and sheep are the second and third largest sources,
accounting for about 10% each. Emissions from domesticated livestock are likely to increase
in the future as the production of animal products continues to increase.
Page 2-29
-------
These emission estimates were develbped using the general approach defined in
OECD (1991); adjustments were made to several assumptions to improve the estimates.
National inventories of emissions from this source are possible, given the available data and
understanding of the physical processes that control CH4 emissions from livestock. The
estimates presented here can be improved by incorporating additional country-specific
information.
2.8 REFERENCES
AAC (Australian Agricultural Council). 1990. Feed Standards for Australian Livestock.
Ruminants. Commonwealth Scientific and Industrial Research Organization (CSIRO)
Publications, East Melbourne, Victoria, Australia.
ADETEF (Association pour la diffusion a I'etranger des techniques de Pelevage francais).
1982. Bulletin de L'elevage Francais. vol. 15. ADETEF, Paris, France.
ARC (Agriculture Research Council). 1980. The Nutrient Requirements of Ruminant Livestock.
Commonwealth Agricultural Bureaux, The Lavenham Press, Ltd., United Kingdom.
Baldwin, R.L., and MJ. Allison. 1983. Rumen metabolism. Journal of Animal Science
57:461-477.
Bamualim, A., and Kartiarso. 1985. Nutrition of draught animals with special reference to
Indonesia. In Copland, J.W., ed. Draught Animal Power for Production. Australian Centre for
International Agricultural Research (ACIAR) Proceedings Series No. 10. ACIAR, Canberra,
A.C.T., Australia.
Blaxter, K.L., and J.L. Clapperton. 1965. Prediction of the amount of methane produced by
ruminants. British Journal of Nutrition 19:511 -522.
Carmean, B.R., K.A. Johnson, D.E. Johnson, and L.W. Johnson. 1992. Maintenance energy
requirement of llamas. Amer. J. Vet. Res. 53:1,696.
CIAT (Centra Internacional de Agriculture Tropical). 1975. Potential to Increase Beef
Production in Tropical America: Proceedings of Seminar in Colombia, February 1974. CIAT,
Cali, Colombia.
CIAT (Centra Internacional de Agricultura Tropical). 1979. Pasture Production in Acid Soils of
the Tropics: Proceedings of the Seminar. Sanchez, P., and L. Tergas, eds. CIAT, Cali,
Colombia.
Copland, J.W., ed. 1985. Draught Animal Power for Production: Proceedings of an
International Workshop in Townsville, Queensland, Australia, 1985. Australian Centre for
International Agricultural Research, Canberra, Australia.
Crutzen, P.J., I. Aselmann, and W. Seller. 1986. Methane production by domestic animals,
wild ruminants, other herbivorous fauna, and humans. Tellus 386:271 -284.
Page 2-30
-------
Devendra, C. 1989. Ruminant production systems in developing countries: Resource
utilization. In Feeding Strategies for Improving Productivity of Ruminant Livestock in
Developing Countries. International Atomic Energy Agency, Vienna, Austria.
Doppler, W. 1980. The Economics of Pasture Improvement and Beef Production in Semi-
Humid West Africa. Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ).
Ensminger, M.E. 1983. Animal Science. The Interstate Printers & Publishers, Inc., Danville,
Illinois.
FAO (Food and Agriculture Organization of the United Nations). 1991. FAO Yearbook:
Production 1990. vol. 44. FAO, Rome, Italy.
Graham, N.M. 1985. Relevance of the British metabolisable energy system to the feeding of
draught animals. In Copland, J.W., ed. Draught Animal Power for Production. Australian
Centre for International Agricultural Research (ACIAR) Proceedings Series No. 10. ACIAR,
Canberra, A.C.T., Australia.
Gryseels, G.G., F. Assamenew, F. Anderson, A. Misgina, B. Kidane, R. Sayers, and W.
Mariam. 1988. Role of Livestock on Mixed Farms in the Ethiopian Highlands: A Case Study
from the Baso and Worena Woredda Near Debre Berhan. International Livestock Centre for
Africa, Addis Ababa, Ethiopia.
Huque, Q.M.E., and C. Stem. 1993. Livestock Feeds of Bangladesh: Availability & Nutrient
Composition. Bangladesh Agricultural Research Council/U.S. Agency for International
Development, Dhaka, Bangladesh.
Ibrahim, M.N.M. 1985. Nutritional status of draught animals in Sri Lanka. In Copland, J.W., ed.
Draught Animal Power for Production. Australian Centre for International Agricultural Research
(ACIAR) Proceedings Series No. 10. ACIAR, Canberra, A.C.T., Australia.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment Cambridge University Press,
Cambridge, United Kingdom.
Johnson, D.E., M. Branine, G.M. Ward, B. Carmean, and D. Lodman. 1991. Livestock
methane emissions: Variation, comparative warming perspectives and amelioration potential.
In Proceedings of the Southwest Nutrition and Management Conference. Arizona State
University, Tempe, Arizona, 31 January -1 February 1991.
Jurgens, M.H. 1988. Animal Feeding and Nutrition. 6th Ed. Kendall/Hunt Publishing Company,
Dubuque, Iowa.
Lawrence, P.R. 1985. A review of nutrient requirements of draught oxen. In Copland, J.W., ed.
Draught Animal Power for Production. Australian Centre for International Agricultural Research
(ACIAR) Proceedings Series No. 10. ACIAR, Canberra, A.C.T., Australia.
Leng, R.A. 1991. Improving Ruminant Production and Reducing Methane Emissions from
Ruminants by Strategic Supplementation. EPA/400/1-91/004, U.S. Environmental Protection
Agency, Global Change Division, Office of Air and Radiation, Washington, D.C.
Page 2-31
-------
Margan, D.E., N. McC. Graham, D.J. Minson, and T.W. Searle. 1988. Energy and protein
values of four forages, including a comparison between tropical and temperate species.
Australian Journal of Experimental Agriculture 28:729-736.
Minson, DJ. 1990. Forage and Ruminant Nutrition. Academic Press. New York, New York.
Page 90.
NRG (National Research Council). 1984. Nutrient Requirements of Beef Cattle. National
Academy Press, Washington, D.C.
NRC (National Research Council). 1989. Nutrient Requirements of Dairy Cattle. National
Academy Press, Washington, D.C.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Parikh, K.S., G. Fischer, K. Frohberg, and O. Gulbrandsen. 1988. Towards Free Trade in
Agriculture. International Institute for Applied Systems Analysis, Martinus Nijhoff Publishers,
Boston, Massachusetts.
Pathak, N.N., and R.C. Jakhomola. 1983. Forage and Livestock Production. Vikas Publishing
House Pvt. Ltd., India.
Politiek, R.D., and J.J. Bakker. 1982. Livestock Production in Europe: Perspectives and
Prospects. Livestock Production Science, vol. 9 (1-2). Elsevier Scientific Publishing Company,
Amsterdam, the Netherlands.
Poppi, D.P., D.J. Minson, and J.H. Termouth. 1981. Studies of cattle and sheep eating leaf
and stem fractions of grasses (Digitaria decumbens, Chloris gayana). II. Factors controlling
the retention in the reticulo-rumen. Australian Journal of Agricultural Research 21(1):109-121.
Reuss, S.K., D.M. Swift, G. Ward, and J.E. Ellis. 1990. Global Ruminant Livestock Production
Systems: Estimated 1988 Methane Emissions. Natural Resource Ecology Laboratory,
Colorado State University, Fort Collins, Colorado.
Robbins, C.T., and D.L. Robbins. 1979. Fetal and neonatal growth patterns and maternal
reproductive effort in ungulates and sub-ungulates. Amer. Naturalist 114:101-116.
Rong Yi. 1983. Beef production in China. In Proceedings of First China-U.S. Symposium on
Ruminant Animal Productivity, September 1983. Beijing, China.
Simpson, J.R. 1988. China's dairy industry: Economic analysis and implications. Feedstuffs. 3
October 1988.
Simpson, J.R., and D.E. Farris. 1982. The World's Beef Business. Iowa State University
Press, Ames, Iowa.
Page 2-32
-------
Swann, N. 1987. Agricultural Statistics of Australia, 1960-1986. U.S. Department of
Agriculture, Washington, D.C.
Terada, F., K. Iwasaki, R. Tano, and K. Kameoka. 1987. Comparison of metabolizability and
metabolizable energy contents among cattle, sheep and goats fed the same diets. In Moe,
P.W., ed. Energy Metabolism of Farm Animals. European Association for Animal Production
(EAAP) Pub. 32:130-133, Rome, Italy.
U.S. EPA (U.S. Environmental Protection Agency). 1989. Policy Options for Stabilizing Global
Climate. Office of Policy, Planning, and Evaluation, Washington, D.C.
U.S. EPA (U.S. Environmental Protection Agency). 1993. Anthropogenic Methane Emissions
in the United States. Global Change Division, Office of Air and Radiation, Washington, D.C.
Vaidyanathan, A. 1988. Bovine Economy in India. Centre for Development Studies, Oxford &
IBM Publishing Co. Pvt. Ltd., New Delhi, India.
Ward, G., R. Oltjen, R. Temple, and D. Gates. 1986. New technologies needed to improve
animal production in China. Feedstuffs. 6 January 1986.
Winrock (Winrock International). 1978. The Role of Ruminants in Support of Man. Winrock
International, IVIorrilton, Arkansas.
Page 2-33
-------
APPENDIX A
EMISSION FACTORS FOR BUFFALO
This appendix summarizes the derivation of CH4 emission factors for buffalo. Crutzen
et al. (1986) estimated an emission factor of 50 kg of CH4 per animal per year based on a
gross energy intake of 85 MJ per animal per day (Pandey, 1981) and a CH4 yield of 9.0%.
Alternative emission factors for buffalo are presented here, using different estimates of feed
intake and a CH4 conversion rate of 7.5%.
A.1 FEED INTAKES
A.1.1 Buffalo Characteristics
As with cattle, the feed intakes for buffalo are driven by size (weight in kg), growth
rate, lactation, and work performed. Among these factors, size is by far the most important.
Table A-1 shows the data collected on buffalo weights for various types throughout the world.
The majority of these data are based on case studies of buffalo in specific countries or
regions of countries. For purposes of this study, the following weight ranges are used to
estimate feed intakes:
Adult Males: 350-550 kg;
Adult Females: 250-450 kg; and
Young: 100-300 kg.
The upper ends of these ranges represent the higher values listed in Table A-1. The lov/er
values of the ranges may be found in some areas with relatively small animals and poor/feed
resources.
The estimates of weight gain for young buffalo are based on the adult weight of buffalo
and the approximate age of maturity, which is about three to four years. The rate of weight
gain for young is estimated at about 0.15 kg/day.17 Mature buffalo are assumed to h£ve a
constant weight; consequently, no weight gain is reported.
Milk production per female buffalo is estimated from total commercial buffalo milk
production, reported in FAO (1990), and from estimates of the female buffalo population. Two
representative countries were used for the estimates: India, with about 75 million (buffalo, and
China, with about 21 million. Together, these two countries have nearly 70% of ail buffalo.
For India, the adult female population is assumed to be about 40% of the total buffalo
population (Agricultural Statistics Division, 1977); for China, the estimate is 45% (Reuss et al.,
1990). The resulting estimates of milk production per adult female are about 2.7 kg/day for
India and 0.65 kg/day for China. These values are used to represent milk production of
buffalo in the Indian Subcontinent and Other Countries, respectively.
As with cattle, draft male buffalo are estimated to work 5 hours per day for 100 days a
year, for a total of 500 hours per year, or 1.37 hours per day on average. Draft female buffalo
1? Growth rate estimated as: 150 kg * 3 years = 0.14 kg/day to 250 kg -*- 4 years = 0.17 kg/day.
Page 2-34
-------
TABLE A-1
DATA ON BUFFALO SIZE
Region Category
Indian Subcontinent Adult Males
Adult Females
Young
Dry Dairy
Milk Dairy
Southeast Asia Adult Males
Bullocks
Adult Females
All Adults
Young
North Africa & All Adults
Middle East
Young
Tropical America All Adults
Young
Eastern Europe & All Adults
the Former Soviet
Union You"9
China, Mongolia, Adult .Males
Korea
Adult Females
Weight
(kg)
550
450
300
350
350
540
525
525
395
400
435
530
500
637
98
170
305
360
500
300
700
385
530
115
515
410
Weight Source
Bangladesh/Foik (1988)
Bangladesh/Folk (1988)
Bangladesh/Foik (1988) and Pathak
and Jakhmola (1983)
Pathak and Jakhmola (1 983)
Pathak and Jakhmola (1 983)
Thailand/Chantalakhana (1985)
Copland (1985)
Thailand/Chantalakhana (1985)
Copland (1985)
Thailand/Chantalakhana (1985)
Malaysia/Liang et al. (1989)
Philippines/Eusebio (1984)
Devendra (1989)
Thailand/Bunyavejchewin et al. (1985)
Thailand/Chantalakhana (1985)
Copland (1985)
Malaysia/Liang et al. (1989)
Philippines/Eusebio (1984)
FAO(1977)
FAO(1977)
FAO(1977)
FAO(1977)
FAO (1977)
FAO(1977)
FAO (1977)
FAO (1977)
Page 2-35
-------
are assumed to work 2 hours per day for 100 days a year, for an average of 0.55 hours per
day for the entire year. These buffalo characteristics are summarized in Table A-2.
A.1.2 Energy Requirements
Based on the buffalo characteristics in Table A-2, the energy intake requirements of
the representative animal types were estimated (Table A-3). The energy relationships
described in the main text for cattle are applied here for buffalo.
Maintenance
Net energy for maintenance (NEJ for cattle is estimated as a function of the weight of
the animal. Table A-3 presents the estimates for NEm for the various animal sizes. The range
is from about 10 MJ/day for small young to about 37 MJ/day for large adult males. The net
energy required to obtain food is assumed to be zero for stall- and pen-fed animals.
Growth
The energy requirements for growth are estimated as a function of the weight of the
animal and the rate of weight gain. The NRC (1989) equation for large-frame female cattle,
as described in Section 2.4, is used to estimate the net energy requirement for growth (NEg).
As shown in Table A-3, the energy requirements for growth are relatively small compared with
the maintenance requirements because the animals grow very slowly. Consequently, these
estimates of energy requirements for growth do not have an important influence on the final
emission estimates.
Lactation
Net energy for lactation (NE,) is expressed as a function of the amount of milk
produced and its fat content (NRC, 1989). At 4.0% fat, the NE, in MJ/day is about 3.1 x kg of
milk produced per day. Although buffalo milk generally has a higher fat content than that of
cow milk, a 4% fat content is assumed for this analysis. The resulting NE, estimates range
from about 2.0 MJ/day to 8.4 MJ/day.
Draft Power
The values by Bamualin and Kartiarso (1985) show that about 10% of NEm
requirements are required per hour of typical work for draft animals. This value is used for
estimating the net energy required for work (NEJ for buffalo, as shown in Table A-3. As with
energy for growth, these values are relatively small compared with those for maintenance
requirements.
Pregnancy
Net energy for pregnancy (NEp) is computed as described for cattle in the main text.
One-third of mature female buffalo in India are assumed to give birth each year, and one-
fourth is the portion assumed for other countries. The resulting energy estimates for
pregnancy are small and do not have a significant influence on the emission estimates.
Page 2-36
-------
TABLE A-2
DATA ON REPRESENTATIVE BUFFALO TYPES
Region
Indian Subcontinent
Other Countries
Category
Adult Males
Adult Females
Young
Adult Males
Adult Females
Young
Weight
(kg)
350-
250-
100-
350-
250-
100-
550
450
300
550
450
300
Wt. Gain
(kg/day)
0.00
0.00
0.15
0.00
0.00
0.15
Milk
(kg/day)
0.00
2.70
0.00
0.00
0.65
0.00
Work
(hrs/day)
1.37
0.55
0.00
1.37
0.55
0.00
TABLE A-3
NET ENERGY REQUIREMENTS FOR REPRESENTATIVE BUFFALO TYPES
(MJ/day)
Region
Indian
Subcontinent
Category NEm NE.
Adult
Adult
Males
Females
Young.
Other Countries
Adult
Adult
Males
Females
Young
26.1
20.2
10.2
26.1
20.2
10.2
-36.6
-31.5
-23.2
-36.6
-31.5
-23.2
0.
0.
1.2-
0
0
1.9
0.0
0.0
1.2-
1.9
NE,
0.0
8.4
0.0
0.0
2.0
0.0
NEW
3.6-
1.1 -
0.
3.6-
1.1 -
5.0
1.7
0
5.0
1.7
0.0
NEp
0.0
0.5 - 0.
0.0
0.0
0.4 - 0.
0.0
8
.6
NEm = net energy for maintenance.
NEg = net energy for growth.
NE, = net energy for lactation.
NEW = net energy for work.
NEp = net energy for pregnancy.
Page 2-37
-------
Gross Energy Intake
As with cattle, the gross energy (GE) intakes of the buffalo are estimated and
compared to the weights of the representative animals. Because buffalo are almost entirely
found in developing countries where they subsist on crop by-products, a digestibility of 55% is
assumed. The resulting GE estimates are shown in Table A-4. All the GE estimates are
between 1.5% and 2.5% of body weight, which is reasonable.
A.2 METHANE EMISSION FACTORS
The CH4 conversion rate adopted for this analysis is 7.5%. Based on the feed-intake
and CH4 conversion rate estimates, CH4 emission factors per head are developed for each of
the representative buffalo types. First, the daily feed-energy intake is multiplied by the CH4
conversion rate of 7.5%. Then, the daily CH4 emission rate is estimated using the conversion
factor of 55.65 MJ per 1,000 g of CH4. The resulting daily estimates are shown in Table A-5.
Then the daily rates are converted to annual rates by multiplying them by 365.
As shown in the table, annual emission rates for the buffalo in the Indian Subcontinent
range from about 23 kg/yr for small young to about 80 kg/yr for large adult females. A similar
range is shown for Other Countries.
Average emission factors are estimated for the Indian Subcontinent and Other
Countries by taking into account the mix of animal types within the population of each region.
Based on the population mixes in India and China, the average emission factors per head for
the two regions are estimated at about 41 -66 kg and 47-70 kg, respectively. The mid-points
of these range estimates are used as the emission factors for buffalo in Tables 2-7 and 2-9 of
the main text.
Page 2-38
-------
TABLE A-4
GROSS ENERGY INTAKE FOR REPRESENTATIVE BUFFALO TYPES
Region
Indian Subcontinent
Other Countries
Category
Adult Males
Adult
Females
Young
Adult Males
Adult
Females
Young
Digest.
(percent)3
55
55
55
55
55
55
GE (MJ/dayf
112-
116-
47-
112-
91 -
47-
157
162
101
157
137
101
Percent BW°
1.5-
2.0-
1.8-
1.5-
1.6-
1.8-
1.7
2.5
2.5
1.7
2.0
2.5
a. Digestibility in percent on an energy basis.
b. Gross energy intake.
c. Daily dry-matter intake as a percent of body weight. Energy density of feed
estimated at 18.45 MJ/kg.
TABLE A-5
METHANE EMISSION FACTORS FOR BUFFALO
CH4
Region Category (%)
Indian Subcontinent Adult Males 7.5
Adult Females 7.5
Young 7.5
Other Countries Adult Males 7.5
Adult Females 7.5
Young 7.5
CH4 CH4
(g/day) (kg/head/yr)a
150- 54.9-77.1
211
156- 56.9-79.8
219
63-136 23.0-49.6
150- 54.9-77.1
211
122- 44.6-67.2
184
63-136 23.0 - 49.6
Average
Emission
Population Mix Factor
(%) (kg/head/yr)
14%
40% 41 .0 - 65.5
46%
45%
45% 47.1 - 69.9
10%
a. Estimates based on 365 days of feeding.
Page 2-39
-------
A.3 REFERENCES
Agricultural Statistics Division. 1977. Agricultural Statistics of Nepal, 1977. Department of
Food and Agricultural Marketing Services, Government of Nepal, Nepal.
Bamualin, A., and Kartiarso. 1985. Nutrition of draught animals with special reference to
Indonesia. In Copland, J.W., ed. Draught Animal Power for Production. Australian Centre for
International Agricultural Research (ACIAR) Proceedings Series No. 10. ACIAR, Canberra,
A.C.T., Australia.
Bunyavejchewin, P., P. Veerasit, P. Chaidirek, and C. Chantalakhana. 1985. Changes in body
temperature and working efficiency of Thai swamp buffalo. In Copland, J.W., ed. Draught
Animal Power for Production. Australian Centre for International Agricultural Research
(ACIAR) Proceedings Series No. 10. ACIAR, Canberra, A.C.T., Australia.
Chantalakhana, C. 1985. Beef cattle and buffalo breeding in Thailand. In Copland, J.W., ed.
Draught Animal Power for Production. Australian Centre for International Agricultural Research
(ACIAR) Proceedings Series No. 10. ACIAR, Canberra, A.C.T., Australia.
Chantalakhana, C. 1985. Breeding improvement of draught buffalo and cattle for small farms.
In Copland, J.W., ed. Draught Animal Power for Production. Australian Centre for International
Agricultural Research (ACIAR) Proceedings Series No. 10. ACIAR, Canberra, A.C.T.,
Australia.
Copland, J.W., ed. 1985. Draught Animal Power for Production: Proceedings of an
International Workshop in Townsville, Queensland, Australia, 1985. Australian Centre for
International Agricultural Research, Canberra, Australia.
Crutzen, P.J., I. Aselmann, and W. Seller. 1986. Methane production by domestic animals,
wild ruminants, other herbivorous fauna, and humans. Tellus 386:271-284.
Devendra, C. 1989. Ruminant production systems in developing countries: Resource
utilization. In Feeding Strategies for Improving Productivity of Ruminant Livestock in
Developing Countries. International Atomic Energy Agency, Vienna, Austria.
Eusebio, A.N. 1984. Strengthening carabao research in the Philippines. In Copland, J.W., ed.
Evaluation of Large Ruminants for the Tropics. Australian Centre for International Agricultural
Research (ACIAR), Canberra, A.C.T., Australia.
FAO (Food and Agriculture Organization of the United Nations). 1977. The Water Buffalo.
FAO Animal Production and Health Paper #4. FAO, Rome, Italy.
FAO (Food and Agriculture Organization of the United Nations). 1990. FAO Yearbook:
Production 1989. vol. 43. FAO, Rome, Italy.
Foik, I.M. 1988. Review of the Livestock Sector in Bangladesh. Ford Foundation Report. Ford
Foundation, Washington, D.C.
Liang, J.B., A.M. Nasir, A. Ismail, and R.S. Abdullah. 1989. Management of draught animals in
Malaysian oil palm estates. In Hoffmann, D., J. Nari, and RJ. Petheram, eds. Draught Animals
Page 2-40
-------
in Rural Development. Proceedings of an international research symposium, Cipanas,
Indonesia.
NRG (National Research Council). 1984. Nutrient Requirements of Beef Cattle. National
Academy Press, Washington, D.C.
NRG (National Research Council). 1989. Nutrient Requirements of Dairy Cattle. National
Academy Press, Washington, D.C.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from The OECD Experts Meeting, 18-21
February 1991, Paris, France.
Pandey, A.N. 1981. Vegetation and bovine population interactions in the savannah
grazinglands of Chandraprabha Sanctuary, Varanasi. II. Seasonal behavior of grazing animals
and an assessment of carrying capacity of the grazinglands. Tropical Ecology 22:170-186.
Pathak, N.N., and R.C. Jakhmola. 1983. Forage and Livestock Production. Vikas Publishing
House Pvt. Ltd., India.
Reuss, S.K., D.M. Swift, G. Ward, and J.E. Ellis. 1990. Global Ruminant Livestock Production
Systems: Estimated 1988 Methane Emissions. Natural Resource Ecology Laboratory,
Colorado State University, Fort Collins, Colorado.
Page 2-41
-------
APPENDIX B
DERIVATION OF THE RELATIONSHIP BETWEEN NE AND DE
This appendix summarizes the derivation of the relationship between net energy (NE)
and digestible energy (DE) that is used to estimate total feed-intake requirements for cattle.
As described in the main text, the relationship among the energy values of feed consumed by
cattle can be summarized as follows:
Digestible Energy
Metabolizable Energy
Net Energy
Gross Energy - Fecal Losses
Digestible Energy - Urinary and Combustible Gas Losses
Metabolizable Energy - Heat Increment
Net Energy
Gross Energy - Fecal Losses - Urinary and
Combustible Gas Losses - Heat Increment
NRC (1984) presents the following quantitative relationships among these energy values:
ME = 0.82 X DE (B.1)
NEm = 1.37 x ME - 0.138 x ME2
-0.0105 x ME3 - 1.12
(B.2)
= 1.42 X ME - 0.174 x ME2 + 0.0122 x ME3 - 1.65
(B-3)
where:
DE = digestible energy in Meal/kg (dry-matter basis);
ME = metabolizable energy in Meal/kg (dry-matter basis);
NEm = net energy for maintenance in Meal/kg (dry-matter basis); and
NEg = net energy for growth in Meal/kg (dry-matter basis).
The ratio of NEm and NEg to ME or DE can be estimated by using these relationships.
Figure B-1 shows that the ratio of NE to DE is nonlinear, with an increasing slope with
decreasing DE. These relationships imply that at lower values of DE, cattle can recover a
decreasing portion of the energy to use for maintenance or growth.
For purposes of estimating CH4 emissions from cattle, it was decided that applying
these relationships to cattle consuming relatively low-quality feeds (such as cattle in many
tropical countries) may be inappropriate because the relationships were developed based on
analyses of the higher-quality feeds typically found in the U.S. temperate agricultural system.
Consequently, the experimental basis for extrapolating the nonlinear relationships to low levels
of DE is not very strong.
An examination of other energy systems indicates that the rate of net energy retention
declines at lower values of digestible energy. Unlike the NRC system, however, many other
energy systems imply a linear relationship between NE and DE. The U.K. energy system
(ARC, 1980), which is typical of the energy systems used in Europe, has a slope for the linear
Page 2-42
-------
NEm:DE relationship that is similar to the slope of the nonlinear NRC relationship in the range
of 65-70% digestibility. Similarly, the slope of the U.K. NEg:DE relationship is similar to the
slope of the nonlinear NRC relationship in the range of 60-65% digestibility.
To avoid possible biases in estimating feed-intake requirements in this study, the
relationships were extrapolated linearly for DE values below 65%, using the average slopes of
the NRC relationships between 60% and 70% DE. Figure B-2 shows the extrapolated linear
relationships along with the nonlinear estimates. As expected, the linear extrapolations fall
above the original nonlinear estimates.
The implication of making this adjustment to the NRC (1984) relationship for the global
emissions estimate is relatively minor. Using the nonlinear relationship for the entire analysis
increases the 1990 emissions estimate for cattle by 1 Tg, from 58.1 Tg to 59.1 Tg.
Considering the wide range of factors that contribute to uncertainty in the estimates, including
characterization of animal populations, this adjustment has a minor influence on the estimates.
FIGURE B-1
NRC NE:DE RELATIONSHIP
0.54
0.52 -
0.50 -
0.48 -
0.46 -
0.44 -
0.42 -
0.40 -
0.38 -
0.36 -
0.34 -
0.32 -
0.30 -
o.aa -
0.26 -
0.24 -
0.22 -
0.30 -
0.18
44
NE to DE Ratio by DE
Derived from NRC C19843
48
—I—
52
56
—I—
60
—r~
64
—I—
68
—I—
72
76
DIgest i b I e Energy
Page 2-43
-------
FIGURE B-2
LINEAR EXTRAPOLATION OF THE NRC NE:DE RELATIONSHIP
,54 -y
,52 -
50 -
4B -
46 -
44 -
42 -
40 -
38 -
36 -
34 -
32 -
30 -
28 -
26 -
24 -
22 -
20 -
-IB - -
44
NE to DE Ratio by DE
Linear Extrapolation
Linear Extrapolatioi
48
52
56
60
68
72
76
Digestible Energy C5O
B.1 REFERENCES
ARC (Agriculture Research Council). 1980. The Nutrient Requirements of Ruminant Livestock.
Commonwealth Agricultural Bureaux, The Lavenham Press, Ltd., United Kingdom.
NRC (National Research Council). 1984. Nutrient Requirements of Beef Cattle. National
Academy Press, Washington, D.C.-38
Page 2-44
-------
CHAPTER 3
METHANE EMISSIONS FROM RICE CULTIVATION
3.1 SUMMARY
Rice cultivation is one of the major anthropogenic sources of methane. Rice has been
grown for millennia in Asia; it is the primary grain eaten by a third of the world's population.
Over 140 million hectares of rice were harvested in 1990, much of it from traditional flooded
rice fields. These artificial wetlands provide the habitat for methanogenic bacteria and
conditions that produce and emit methane at rates that greatly exceed fluxes from natural
wetland ecosystems. The increase in rice cultivation was probably one of the main
contributors to the growth in atmospheric methane concentration during the last century.
Current estimates of global methane emissions from rice cultivation range from 60 to
100 teragrams (Tg)1. Global estimates are made by summing regional estimates; regional
estimates are made by multiplying a seasonal average methane flux factor by the season
length and the area planted to rice. Because of the limited number and location of seasonal
flux measurements, correctly associating an average flux with a particular area and growing
season is the greatest uncertainty in making regional and global estimates. Soil temperature,
soil type, soil microbiology, soil redox (oxidation-reduction) potential, day length, rice variety,
fertilization, and irrigation all affect the methane flux from a particular area. This report
estimates that in 1990 the global methane emissions from rice cultivation were about 65 Tg.
Rice area harvested increased rapidly in the 1950s, and estimated methane emissions
from rice agriculture increased with it. In the last decade, the increase in area planted to rice
has slowed greatly. The two major rice-planting countries, India and China, have apparently
reached practical and economic limits of the available area for rice paddies. Per-capita rice
area harvested in China has been decreasing since the 1960s. During the same period, rice
yield has been increasing. Efforts to increase rice yield without increasing the area of rice
harvested, through irrigation, fertilizers, and high-yield rice cultivars, will affect the future
methane emissions from rice paddies.
3.2 BACKGROUND
Rice cultivation has long been recognized as a major source of methane (CH4). Global
budgets of CH4 have generally included emissions of about 100 Tg/yr from rice cultivation
(with a range of 50-300 Tg/yr), constituting about 20% of emissions from all sources (range
14-40%) (Ehhalt and Schmidt, 1978; Donahue, 1979; Khalil and Rasmussen, 1983; Blake,
1984; Bolle et al., 1986; Bingemer and Crutzen, 1987; Cicerone and Oremland, 1988; and
Khalil and Rasmussen, 1990b). The Intergovernmental Panel on Climate Change (IPCC,
1992) estimates a source of 60 Tg/yr (range 20-150 Tg/yr). Increases of atmospheric CH4
were first reported by Rasmussen and Khalil (1981 a and 1981b) and later by many other
authors (see Khalil and Rasmussen, 1983; Fraser et al., 1984; Blake and Rowland, 1986;
Steele et al., 1987; and Khalil and Rasmussen, 1990a). Rice cultivation is estimated to
contribute some 230 ppbv (parts per billion by volume) to the 1990 atmospheric burden of
1 Teragram = 106 metric tonnes = 1012 grams.
Page 3-1
-------
1,635 ppbv CH4, and may be responsible for some 20-30% of the increase of CH4 during the
last century (Khalil and Rasmussen, 1991).
Estimating the flux of CH4 from rice fields in various parts of the world requires
knowledge of two factors: the emission rates and the regional or global extrapolating factor
(extrapolant). The emission rate or flux depends on several internal and external variables.
Internal variables include soil characteristics, rice variety, and soil microbiology. External
factors include soil temperature, which is driven by solar radiation; meteorological conditions;
water level, which is affected by rainfall, availability of irrigation, and water management; and
treatments, such as the type and amount of fertilizers and organic amendments applied. The
global flux, FG, is usually calculated by equation 3.1:
FG = £ */? TR AR
R
(3.1)
where R is an accurate representation of flux for the entire area; and TR is
the growing season (days/year). The global extrapolant is AR TR. Any smaller region, such as
a country, may also be subdivided into similar regions, and the flux of CH4 from rice cultivation
from the country can be calculated based on an equation analogous to equation 3.1.
3.2.1 Methane Emission Rates From Rice Fields
Direct flux measurements over the entire growing period provide precise and accurate
values for R. Usually, AR and TR are determined from agricultural statistics and are also well
known. Problems with the extrapolation arise in associating a measured flux at a specific site
with an appropriate area and growing season and in the assumption that the measured flux is
representative of the associated area and season. This empirical approach requires whole-
season measurements of CH4 emission rates from as many regions as possible so that the
large variations of emissions from one region to another can be properly included in the
estimate of global or regional emissions.
Alternative approaches to global extrapolation also exist and are based on the
knowledge of the processes or factors that control emission rates. At present there are
insufficient data for such approaches to produce reliable estimates of global or regional
emissions.
Flux Measurements
During the last decade, a number of systematic experiments have been reported on
CH4 emissions from rice fields. All are based on static chamber methods. While there are
many variants, the method consists of enclosing a small part (0.1-100 m2) of the rice fields
within a chamber and taking periodic air samples. The samples are analyzed for CH4 content
usually by gas chromatography using flame-ionization detectors (GC/FID). Methane, emitted
from the enclosed area of the rice field, builds up in the chamber. The rate of accumulation is
directly proportional to the flux or the rate of emission from the area covered by the chamber.
Page 3-2
-------
It is usually assumed that there are no losses of CH4 from the chamber during the experiment,
which is usually short (< 15 minutes). Methane may be lost if the chamber is not sealed, or
by dissolving into the paddy water. The relationship is:
_ P
M V
NAA
10
-6
dC
dt
(3.2)
where § is the flux, p is the density of air (molecules/m3), M is the molecular weight of CH4, V
is the volume of the chamber (m3), A is the area covered by the chamber (m2), NA is
Avogadro's number, C is the concentration of CH4 (ppbv), and dC/dt is in ppbv/hr. The most
common units for reporting fluxes are mg/m2/hr.
The advantages of chamber methods are: they are inexpensive, they are easy to use
in remote locations, and they are coupled to a highly sensitive and precise measurement
method using GC/FID. Since the fluxes of CH4 are quite large, the plants need not be
exposed to the unnatural conditions of the chamber environment for very long; often
10-minute exposures are sufficient. This fact also makes chamber methods suitable for CH4
measurements, even though they may not be convenient for other gases.
One disadvantage is that placing the chamber can disturb the soil and release
abnormal amounts of CH4. Several methods have been devised to reduce, if not overcome,
this problem. In the studies of Khalil et al. (1991), a permanent aluminum base is installed in
the soil at the time rice is planted, and chambers fit into grooves in this base. In studies
reported by Schutz et al. (1989a and 1990), a large permanent chamber is used that has a lid
that opens and closes, but the chamber itself is not removed until the rice is harvested. Both
these methods create some feedbacks that may affect flux estimates. The chambers also
affect the immediate environment of the rice plant by causing heating and a buildup of carbon
dioxide, which may affect emissions of CH4. Finally, since most chambers are small, the
extrapolation of direct flux measurements to large regions may be unreliable because of
heterogeneities within fields, within local regions, and within different parts of the same
country.
In recent years, experiments have been designed to measure the flux throughout the
growing season. The earliest experiments of this type made it clear that there are large
systematic changes in CH4 emissions during the growing cycle. Such changes are driven by
several factors, including the growth of root mass, maturation process for the plants,
availability of nutrients and fertilization, and the seasonal change of temperature and length of
day (Schutz et al., 1989a; Yagi and Minami, 1990; and Khalil et al., 1991).
Direct flux measurements are summarized in Table 3-1, with a description of the nature
of the experiments. These studies were originally designed to measure the flux of CH4 frorn
rice paddies; later, to establish which factors affect CH4 emissions; and most recently, to
investigate treatments that may reduce CH4 emissions.
3.2.2 Factors Affecting Methane Emissions
Methane emissions from rice cultivation result from a combination of three processes:
CH4 production in the paddy soil, CH4 oxidation in the root zone and at the soil surface, and
CH4 transport from the soil to the atmosphere by diffusion through the floodwater, ebullition,
and plant-mediated transport.
Page 3-3
-------
TABLE 3-1
METHANE FLUX MEASUREMENTS FROM RICE PADDIES
(From Shearer and Khalil, 1993; reprinted by permission of the authors.)
Study Location Rice Cultivar
Cicerone etal., 1983 California, M101
1982 growing season United States
Cicerone etal., 1992 California, Not given
1983 growing season United States
1985 growing season Not given
(Only data from plots with
rice plants given here)
Seller et al., 1984 Andalusia, Bahfa
Spain
Schiitzet al., 1989a Vercelli, Italy Roma
Data reported from 1984 to
1986; includes data from
Holzapfel-Pschorn and
Seller, 1986.
Soil Type
Vertisol, Capay
Clay
Capay Silty Clay
Sacramento Clay
Not Given
Sandy Loam
(60% sand, 25%
silt, 12% clay,
2.5% organic C)
Treatment
Am. phosphate-Am, sulfate + Urea:
113kgN/ha
Preplan!: 36 kg N/ha + top dressing
after planting: 78 kg N/ha
No treatment
2.5 t/ha ground rice straw
5 t/ha ground rice straw
78 kg N/ha (Urea, preplan! fertilization)
78 kg N/ha + 2.5 t/ha rice straw
78 kg N/ha + 5 t/ha rice straw
Urea (before seeding): 160 kgN/ha;
Am. nitrate (at tillering): 40 kg N/ha
Unfertilized
Rice straw: average flux
5 t/ha
3 t/ha
6 t/ha
12 t/ha
24 t/ha
Compost: 60 t/ha
CaCN2: 200 kg N/ha
Urea: average flux
200 kg N/ha, raked
100 kg N/ha, raked
200 kg N/ha, incorporated
200 kg N/ha, surface
Flux
(mg/rtf/hr)
10.4
4.0
0.5
6.4
18.9
0.9
3.0
13.9
4.0
11.7
17.6
24.2
9.6
13.5
22.1
18.8
27.5
12.5
10.7
9.6
9.6
7.9
15.8
Emission
Season"
(days)
100
111
128
(approximate)
125
114
118
105
109
110
113
113
118
118
118
113
113
(continued)
-------
TABLE 3-1
METHANE FLUX MEASUREMENTS FROM RICE PADDIES (Continued)
Study .Location Rice Cultivar Soil Type
Schutz et al., 1989a Vercelli, Italy Roma Sandy Loam
(continued)
Yagi and Minami, 1990 Ibaraki Koshihikari Ryugasaki: Gley
Prefecture, Soil
Japan (Eutric Gleysols)
Kawachi: Peat Soil
(Dystric Histosols)
Mito: Humic
Andosol
Tsukuba: Light
Colored Andosol
(volcanic ash soil)
Treatment
Straw + Urea: average flux
6 t/ha + 200 kg N/ha
12 t/ha + 200 kg N/ha
Ammonium sulfate: average flux
200 kg N/ha, raked
200 kg N/ha, incorporated
100 kg N/ha, incorporated
50 kg N/ha, incorporated
200 kg N/ha, surface
Straw + CaCN2: average flux
2,5 t/ha + 37.5 kg N/ha
5 t/ha + 75 kg N/ha
12 t/ha + 200 kg N/ha
Non-nitrogen (0:60:30)b
Mineral (60+30:60:30)
Compost: 12 t/ha + Mineral (above)
Rice straw: 6 t/ha + Mineral (above)
Rice straw: 6 t/ha + Mineral
(60+25:60:60+25)
Non-nitrogen (0:120:80+30)
Mineral (50+30:120:80+30)
Compost: 12 t/ha
Rice straw: 6 t/ha
9 t/ha
Mineral (70+30:100:70+30)
Rice straw: 6 t/ha + Mineral (above)
Flux
(mg/m2/hr)
22.5
20.0
25.0
8.2
6.7
5.6
8.8
7.5
12.5
16.8
11.7
17.9
20.6
2.8
2.9
3.8
9.6
16.3
1.4
1.2
1.9
3.2
4.1
0.2
0.4
Emission
Season3
(days)
113
113
118
109
105
105
113
118
118
109
119
118
112
117
115
122
125
129
128
128
125
115
(continued)
-------
TABLE 3-1
METHANE FLUX MEASUREMENTS FROM RICE PADDIES (Continued)
Study
Dai, 1988; and Schutz et al.,
1990
Khalil et al., 1991
Sasset al., 1990, 1991 a
1989 and 1990 data
averaged for Lake Charles
and Beaumont fertilized
sites.
Sasset al., 1991b
Sasset al., 1992
t
Location Rice Cultivar
Hangzhou, Not given
China
Sichuan, Local and
China Hybrid
Texas, Jasmine 85
United States
Texas, Jasmine 85
United States
Texas, Jasmine 85
United States
Soil Type
Not given
"Purple Soil"
Lk. Charles clay
(Typic Pelludert)
Beaumont clay
(Entic Pelludert)
Verland silty clay
loam (Vertic
Ochraqualf)
Verland siity clay
loam
Treatment
Late rice, average flux
Early rice, average flux
Organic
Urea: 149 kg N/ha
Rice straw: 12 t/ha + Urea: 102 kg
N/ha
Urea: 190 kg N/ha
Rice straw: 8 t/ha + Urea: 102 kg N/ha
All sites, Urea: 190 kg N/ha
Planted April 9
Straw: 6 t/ha
w/o straw
Planted May 18
Straw: 6 t/ha
w/o straw
Planted June 15
Straw: 6 t/ha
w/o straw
All sites, Urea: 165 kg N/ha total
Normal flood irrigation (control)
Mid-season aeration
Multiple aeration (3 drained periods)
Late season flood irrigation
Flux
(mg/ma/hr)
26.4°
6.6C
36.6
8.7
15.2
2.5
5.6 •
23.5
18.3
18.3
11.8
12.7
12.0
4.4
2.3
0.6
6.3
Emission
Season8
(days)
62
70
120
85
86
85
86
85
81
76
87
87
87
99
(continued)
-------
TABLE 3-1
METHANE FLUX MEASUREMENTS FROM RICE PADDIES (Continued)
Study Location Rice Cultivar
Chen et al., 1993 Beijing, Huang jinguang
China
Nanjing, Shanyou 63
China
National Physical Laboratory, New Delhi P-615
New Delhi, India, 1992.
A.P. Mitra, ed.e Gargacha, Fine
Kharif season (Sept.-Dec.) Garia
Soil Type
Sandy Loam
1 .33% organic
matter
"yellow-brown
earth" 2.29%
organic matter
Alluvial Sandy
Loam
Not given
Treatment
610 kg/ha NH4HCO3 + 15 t/ha horse
manure: lld
900 kg/ha NH4HCO3: F
610 kg/ha NH4HCO3 + 15 t/ha horse
manure: F
710 kg/ha NH4HCO3 + 30 t/ha horse
manure: F
610 kg/ha NH4HCO3 + 15 t/ha horse
manure: D
190 kg/ha Urea + 15 t/ha barnyard
manure: L
45 t/ha barnyard manure: II
600 kg/ha Ammonium Sulfate: F
190 kg/ha Urea + 3 t/ha rapestraw: M
190 kg/ha Urea + 15 t/ha barnyard
manure: Semi-arid
Intermittent Irrigation'
Rainfed, flooded
Flux
(mg/m2/hr)
14.6
17.5
35.9
48.9
-0.0
10.8
9.5
2.6
14.3
6.6
0.2
11.8
Emission
Season3
(days)
79 .
101
106
110
(continued)
-------
TABLE 3-1
METHANE FLUX MEASUREMENTS FROM RICE PADDIES (Continued)
Study
Partial Studies: Studies where
Cicerone and Shelter (1981)
Khaliletal. (1990)
Parashar et al. (1991)
Wassman et al. (1993)
Location Rice Cultivar
measurements were not taken for a full
California, United States
Sichuan, China
New Delhi, Karnal,
Dehradun, and
Hyderabad, India
Zheijian, China
Soil Type Treatment
Flux
(mg/nvVhr)
Emission
Season"
(days)
season or are not yet reported in full. Ranges of emissions are shown.
1.3-7.8
1 -50
0.07 - 80.0
no fertilizer 7.4 - 47.0
694 kg/ha K,,SO4/KCI 7.7 - 35.4
694 kg/ha K2SO4/KCI + 1042 kg/ha
rapeseed cake 7.9 - 44.0
1042 kg/ha rapeseed cake 6.0 - 50.6
a. Number of days that CH4 was actually emitted from the rice paddies.
b. (NiPgOgiKgO) in kg/ha; rates of mineral fertilizer are expressed as basal + supplementary applications.
c. Values for Schiitz et al. (1990) were digitized from the figures.
d. F: flood irrigation; II: intermittent irrigation; D: dry culture.
e. Measurements were taken at 18 sites during the 1991 Kharif season in this study. Both systematic CH4 flux and water-level measurements for > 70%
of the growing season were published for the New Delhi and Gargacha sites only. Fertilization data are not available.
f. Irrigated field where water level was 0 at least once during the growing season.
-------
Methane is produced in saturated soils by anaerobic bacteria (methanogens), which
use compounds from the decay of organic matter as their food source and produce CH4 as a
by-product. This process takes place only where oxygen is not available, such as in flooded
paddy fields and wetlands (Bouwman, 1991; Boone, 1993; and Neue and Roger, 1993).
Temperature, water regime, nutrient availability, and soil factors (such as pH and redox
potential) may affect the growth and metabolic activity of the methanogen populations and
ultimately the CH4 flux from the paddy fields.
Experiments to measure the production and emission of CH4 from paddy soils have
shown that not all CH4 produced in the soil is emitted to the atmosphere (Holzapfel-Pschorn et
al., 1985; Schutz et al., 1989b; and Sass et al., 1992). Laboratory studies of paddy soil
indicate that 50-90% of the CH4 produced in the soil can be oxidized before it reaches the
surface (summarized in Neue and Roger, 1993). Strains of bacteria that consume CH4
(methanotrophs) are believed to live in both soil and water. They oxidize the carbon in CH4
and produce CO2 as a by-product. This reaction can take place even in the low-oxygen
environment of a paddy soil. Air transported to the roots of the rice plant creates an
oxygenated zone around the roots (Nouchi et al., 1990). There the methanotrophs can break
down CH4 before it is pumped out of the soil by the air circulation system of the plant.
In the same experiments, the pathways of CH4 emission to the atmosphere throughout
the growing season were also measured. Most of the CH4 (up to 90%) was found to be
transported through the intercellular spaces in the plant. About 10% of the CH4 was
transported in bubbles from the soil, while less than 1% of the CH4 diffused through the soil
and water to the atmosphere (see Figure 3-1). At the beginning of the growing season all the
CH4 was emitted through ebullition; as the rice plants developed, an increasing amount was
transported through the plant.
fields is not known.
Whether the numbers are representative of the world's rice
Internal Factors
The "internal factors" of soil microbiology, soil properties, and different rice cultivars are
at present the most difficult to incorporate into the global estimate of CH4 flux from rice
paddies. While the life cycle of methanogens is understood (see Boone, 1993, for a
summary), predicting the population fluctuations in rice fields is not yet possible.
Methanogens do not directly break down organic matter but require other bacteria to produce
the substrates they need for food. The population dynamics of the various interdependent
strains of bacteria may account for year-to-year variation found in field studies where all other
variables are nearly the same. Organic compounds exuded from the root system of the plants
may also influence the growth of methanogens or competing bacteria populations (Neue and
Roger, 1993).
The soil properties most important in affecting CH4 production are soil texture,
mineralogy, and Eh/pH buffer systems (Neue et al., 1990). Several experiments have directly
measured the effect of different soil textures during the same growing season (Yagi and
Minami, 1990; Sass et al., 1990 and 1991 a; and Chen et al., 1993). The differences in fluxes
between the soil types may reflect the ability of the soils to sustain a reducing environment.
In laboratory tests of paddy soils, a soil redox level of -150 mV was necessary for CH4
emission to begin (Mariko et al., 1991; and Masscheleyn et al., 1993).
Most of the CH4 emitted from rice paddies passes through the stems of the rice plant
(Schutz et al., 1989b). Different rice cultivars may have different capacities for gas transport,
Page 3-9
-------
FIGURE 3-1
SCHEME OF PRODUCTION, REOXIDATION, AND EMISSION OF CH4 IN A PADDY FIELD
Plant transport
90%
(Exudation, Decay)
anoxic
Source: Schtitz et al., 1989b; Reprinted by permission of the authors.
Page 3-10
-------
though this has not been established. The only published study using two different cultivars in
the same environment for flux measurements found no significant differences between the two
rice varieties (Khalil et al., 1991). The length of the growing seasons of different rice cultivars
also vary, requiring different periods of standing water in the paddies. While most studies
report the cultivar used, there is not enough information at present to relate cultivar type to
variations in CH4 flux.
External Factors
External factors affecting the flux of CH4 from rice paddies may be the result of the
weather, including solar radiation, which influences soil temperature, and rainfall; or
agricultural practices, such as irrigation and fertilization. The former are mainly factors that
change with season or stage of growth; the latter vary with geographic location, depending on
availability of water and mineral fertilizers, traditional agricultural methods, and local
economics.
There is a positive correlation between soil temperature and daily CH4 flux, which
changes slightly throughout the growing season with a Q10 of about 2 to 4 (see Figure 3-2)2
(Schiitz et al., 1989a; and Khalil et al., 1991). Temperature relationships may also vary from
season to season at the same site (Khalil et al., 1991). The relationship between CH4 flux
and temperature only holds within an optimum range of soil temperatures. Parashar et al.
(1993) tested the CH4 emission response to temperature by placing heating coils in a rice
field. They found that CH4 emissions increased rapidly at temperatures above 24°C, reaching
a maximum at 34.5°C. Higher temperatures inhibited production.
Changes in flux over the growing season have been found in every experiment (Schutz
et al., 1989a; Yagi and Minami, 1990; and Khalil et al., 1991). Planting date determines which
stage of the crop occurs during the longest and warmest days of the growing season. Schutz
et al. (1990) found significant differences between average fluxes for early and late rice crops
in Hangzhou, China. Sass et al. (1991b) compared the differences in flux over a planting
season by planting a crop at one-month intervals. The later plantings ripened more quickly,
and were irrigated for a shorter period, and differences in CH4 fluxes between the fertilizer
treatments decreased.
Rice cultivation may be divided into wetland and dryland (or upland) culture (Grist,
1986; and Neue et al., 1990). In wetland culture, the soil is prepared by puddling3 to reduce
water loss, and dikes or levees are built to contain the water. Wetland culture may be
separated into irrigated and rain-fed rice. While rain-fed rice is planted during the wet season,
all water is provided by rainfall, and dry spells decrease CH4 fluxes. Dryland rice culture is
generally a low-yield, subsistence agriculture, highly susceptible to drought, where no special
preparations are made to retain water in the fields (De Datta, 1975). Dryland rice may not be
a source of CH4 at all, as the soil is not saturated long enough for methanogen populations to
build up.
2 Q10 is an experimentally-derived temperature coefficient which is the.ratio of (1) the flux at a temperature 10°C
above the base temperature, to (2) the flux at the base temperature.
3 Puddling is a rice cultivation practice in which saturated and flooded soils are repeatedly plowed so that large soil
aggregates are broken down into finer fractions that are well mixed with the water. Puddled soil holds more water than
if it were in its natural state.
Page 3-11
-------
FIGURE 3-2
RELATIONSHIP BETWEEN THE FLUX AND SOIL TEMPERATURE IN A
PADDY FIELD IN SICHUAN, CHINA
100
80
60
40
20
0
16
20
24
28
32
Temperature (C)
• 1988 • 1989
Source: Khalil et al., 1991; Reprinted by permission of the authors.
Page 3-12
-------
Most of the studies in Table 3-1 were carried out where flood irrigation is used.
However, studies of water-management schemes where the soil surface is allowed to dry at
least once during the season show much lower average CH4 fluxes (Sass et al., 1992; and
Chen et al., 1993). Field measurements on wetlands and on tropical soils (Harriss et al.,
1982; and Keller et al., 1986) show that CH4 flux stops and all CH4 is oxidized once the soil
dries out. Fallow rice fields where the soils are still saturated continue to have low CH4
emissions, but emissions stop from paddy soils allowed to dry at harvest (Khalil et al., 1990).
Intermittent flooding may have undesirable side effects, however. Allowing the soils to dry
during some stages of plant growth, especially heading and flowering, leads to drastically
reduced yields (see summaries in Grist, 1986; and Bouwman, 1991). Sass et al. (1992)
compared normal flooding with both single- and multiple-aeration irrigation schemes; while the
multiple-aeration regime produced the least CH4, it used the most water.
Whether the crop is directly seeded or is transplanted may affect the length of time the
crop is kept flooded and may thus affect the total CH4 emissions over the growing season.
The Khalil et al. (1991) data were obtained from a region where the rice was transplanted to
the fields, then standing water was maintained in the fields until harvest about 120 days later.
In the study by Sass et al. (1991 a), the crop was seeded into nonflooded soil and was
permanently flooded after the young plants were established. Out of a growing season of 140
days, the crop was flooded for only 85 days.
Fertilizer treatments have also been found to affect CH4 flux. The addition of rice straw,
manure, compost, and other organic fertilizers appears to enhance CH4 production, probably
by providing a food supply for the methanogens (Schutz et al., 1989a and 1990; Yagi and
Minami, 1990; Sass et al., 1991 a; Bouwman, 1991; Cicerone et al., 1992; and Chen et al.,
1993). Nitrogen fertilizers, especially ammonium sulfate, may inhibit CH4 production, possibly
by chemical competition (reviewed in Bouwman, 1991). Use of mineral fertilizers in rice fields
has been increasing globally (Neue et al., 1990); however, there are still large reaions where
they are unavailable or not economic. In particular, organic manures are still widely used in
Asia and may result in higher fluxes for the largest rice-growing areas.
3.2.3 Regional Estimates of Methane Emissions
Four recent studies have included estimates of regional emissions from rice cultivation.
Two of these studies were undertaken to provide data for global transport-chemistry models
for CH4; they concentrated on different variables, and the authors made different types of
information available (Aselmann and Crutzen, 1989; and Matthews et al., 1991). Bachelet and
Neue (1993) estimated emissions from rice-growing countries in Asia, modifying the estimates
by soil and organic matter data. The study by Khalil and Shearer (1993) emphasized
emissions by water-management level.
Aselmann and Crutzen (1989) provided detailed tables of the percent of the global rice-
growing area in 2.5° latitude by 5° longitude boxes. The percentages reflect the area in rain-
fed and irrigated rice only for Asia (no dryland rice was included) but show total area for
Africa, Central America, and South America. Matthews et al. (1991) published detailed rice
crop calendars, indicating the months of possible cultivation of rice by country, each Indian
state, and each province of China. Area was allocated by 1° latitude by 1° longitude cells.
Khalil and Shearer (1993) developed an inventory of direct flux measurements and modified
the information from Matthews et al. on growing seasons to estimate global and regional
annual emission rates.
Page 3-13
-------
Estimates of global emissions of rice cultivation usually focus on regions in South and
East Asia, where 90% of the area planted to rice is located (FAO, 1991), and where
agricultural practices are most likely to favor CH4 production in rice paddies. Nine of the top
ten rice-growing countries are in Asia (by 1990 area: India, China, Bangladesh, Indonesia,
Thailand, Viet Nam, Myanmar, Brazil, Philippines, and Pakistan). Of the four non-Asian
countries in the top 20 rice-planting countries, three (Brazil, Madagascar, and Nigeria) have
60-80% of their total area planted in dryland rice (Grist, 1986); the fourth country (the United
States), uses cultivation practices different from most other large rice-growing countries.
The flux of CH4 from rice fields is usually calculated by estimating the length of the
season of CH4 emission, the area in rice (irrigated and rain-fed), and the flux of CH4
(mg/m2/hr) averaged over the entire growing season (see equation 3.1). Each of these main
variables is discussed separately.
Season Length
The estimate of the season length from the literature may be confused by whether
"season" refers to the season of CH4 emission, the rice-growing season (from seeding to
harvest), or the total growing season (the frost-free period). For example, in the study of Sass
et at. (1991 a), the total growing season was about 245 days, the rice-growing season was 140
days from planting to harvest, while CH4 was emitted during only 85 days (i.e., during the
flood-irrigated period). Because the number of areas where the CH4 emission season has
actually been measured is quite limited, some other variable must be used to approximate it.
Matthews et al. (1991) used crop calendars to estimate the months of the rice-growing
season by country; autumn, winter, and summer crops by Indian state; and early rice,
double/late crop rice, and single (mid-season) crop rice by Chinese province. Table 3-2
compares the growing season and season of CH4 emission reported in the literature with the
growing season estimated from crop calendars by Matthews et al. With the exception of Italy,
the difference between the growing season estimated using crop calendars and season of CH4
emission is about 40-50 days.
Methane Flux Factors
Ideally, each rice-growing region would have an average CH4 flux factor associated
with it that uniquely reflects the soil, climate, and cultivation practices of the area. Practically,
estimators must rely on the information that is available and their best judgment to apply
measured fluxes from one region to another where data are not available.
Aselmann and Crutzen (1989) calculated fluxes from the work of Holzapfel-Pschorn
and Seller (1986) in Italy, with a base rate of 300 mg/ma/day (12.5 mg/m2/hr) for average soil
temperatures of 20°C and below, and assumed a linear relationship between flux and
temperature up to 1,000 mg/m2/day (42 mg/m2/hr) for soil temperatures of 30°C. Matthews et
al. (1991) used a flux rate of 500 mg/m2/day (21 mg/m2/hr) for all areas, since the emphasis of
their study was to delineate the global distribution of rice cultivation rather than to estimate
fluxes.
Khalil and Shearer (1993) used averaged fluxes from the studies listed in Table 3-1 for
the major seasonal divisions of Chinese and Indian rice, for other countries where fluxes have
been measured, and to estimate fluxes for other rice-growing countries. Flux rates were
reduced by 40% for areas of rain-fed rice, approximately the reduction in flux found by Chen
Page 3-14
-------
TABLE 3-2
SEASONAL FACTORS FOR METHANE EMISSIONS FROM RICE FIELDS
References
Rice Growing
Season
(Days)
CH4 Emission
Period
(Days)
Crop Calendar
Season
(Days3)
Cicerone et al. (1983): United States
Holzapfel-Pschorn and Seiler (1986): Italy
Yagi and Minami (1990): Japan
Sass et al. (1991 a): United States
Khalil et al. (1991): Sichuan, China
Chen et al. (1993): Beijing, China
145
147
140
140
120
100
100
126
115
85
120
87
152
122
152
152
168
137
a. Estimated from Matthews et al. (1991). See text.
et al. (1993) for their intermittent irrigation regime. The flux from dryland rice was assumed to
be zero in this study.
Area Planted to Rice
Estimates of rice-growing area are usually taken from published agricultural statistics of
rice area harvested supplied by the United Nations (U.M.), the International Rice Research
Institute (IRRI), or the agricultural agencies of individual countries (see, for example: FAO
Production Yearbook; IRRI, 1990; China Agriculture Yearbook). For convenience, agricultural
statistics are reported by political or administrative subdivisions.
Aselmann and Crutzen (1989) and Matthews et al. (1991) did careful global rice area
allocations, described earlier. Khalil and Shearer (1993) reduced the total areas for each
country by the percentage of dryland rice grown. Estimates of the percentages were taken
from tables in De Datta (1975), Huke (1982), Morris et al. (1984), and Grist (1986).
Global Emissions
Three recent estimates of the global source of CH4 from rice cultivation are shown in
Table 3-3. (Matthews et al. (1991) assumed a global source of 100 Tg.)
The major difference between the first two estimates is probably in the way a weighted
average flux rate is calculated. The Khalil and Shearer estimate takes a seasonally averaged
flux, based on the field studies shown in Table 3-1, and applies it to the seasonal area in
wetland rice culture. The Aselmann and Crutzen estimate uses temperature-adjusted fluxes,
as described earlier. The IPCC estimate is essentially based on a review and assessment of
the scientific literature.
Page 3-15
-------
TABLE 3-3
ESTIMATES OF GLOBAL METHANE EMISSIONS FROM RICE AGRICULTURE
Khalil and Shearer
(1993)
Aselmann and Crutzen
(1989)
IPCC
(1992)
Tg/year
Year
65
1990
92 (53a)
1985
60
Not given
a. Number in parentheses assumes a constant flux of 13 mg/ma/hr.
Regional Emissions
Estimates of CH4 emissions by country are shown in Table 3-4, comparing the
countries of South and East Asia by three estimation techniques. These countries produce
about 90% of the estimated CH4 from rice cultivation. The countries with the largest estimated
CH4 source from wetland rice paddies are all in Asia and are shown in the table. The
estimates by Khalil and Shearer (1993) have been described earlier. Bachelet and Neue
(1993) calculated a range of 40-80 Tg/yr as the rice source for South and East Asia, making
different assumptions about the effects on CH4 emissions of soil type, rice yield, temperature,
and organic matter incorporated in the soil. Two of their estimates are presented in Table 3-4.
3.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The estimates of the global source discussed in section 3.2.3 range from 50 to 100 Tg
per year. Source estimates by country vary greatly due to assumptions made on the
importance of different factors that affect the CH4 flux, and the availability of information on
these factors for individual countries. A recommended methodology must focus on the factors
for which there is sufficient information on both the emission factors and the extrapolants.
Including the available information in the present estimates of country-by-country emissions
may improve the accuracy, but at present it is not certain which factors have the greatest
effect on emissions.
The factors clearly identified by field experiments are (1) water level and its history
during the growing season, (2) fertilizer applications and amendments of organic matter, (3)
soil type, (4) soil temperature, and (5) agricultural practices that affect the length of the CH4
emission season, such as direct seeding or transplanting. Different rice cultivars may
influence the amount of CH4 emitted from rice fields, but this is not yet documented. Data
show that higher temperatures, continuously flooded fields, and some types of organic
fertilizers lead to higher emissions compared to rice grown at lower temperatures, with
intermittent or managed irrigation in which the fields are not continuously inundated, and with
the use of chemical fertilizers.
Page 3-16
-------
TABLE 3-4
ESTIMATES OF METHANE EMISSIONS BY COUNTRY
FOR 1990 USING FOUR SETS OF ASSUMPTIONS
(Tg CH4/yr)
Country
Bangladesh
Cambodia
China
India
Indonesia
Japan
N. Korea
S. Korea
Laos
Malaysia
Myanmar0
Nepal
Pakistan
Philippines
Sri Lanka
Taiwan
Thailand
Viet Nam
Khalil and
Shearer
(1993)
3.6
0.9
21.4
15.8
5.0
0.2
0.3
0.8
0.2
0.3
1.3
0.2
0.7
1.3
0.4
0.4
4.9
3.2
Bachelet
and Neue:
Yield3
(1993)
4.1
0.4
21.3
18.5
4.5
1.0
0.5
0.8
0.2
0.3
2.0
0.3
0.8
1.2
0.4
0.4
3.8
2.7
Bachelet
and Neue:
Soil"
(1993)
4.0
0.4
14.7
14.5
3.5
0.8
0.4
0.6
0.1
0.2
1.4
0.2
0.4
0.8
0.3
0.4
2.2
1.8
Total
61
63
47
a. Estimates made by Bachelet and Neue (1993) using
IRRl rice yield data, and regional estimates of
incorporated organic matter.
b. The Bachelet and Neue (1993) estimate modified using
the FAO soil map, and authors' best judgement on the
CH4 potential of different soil types.
c. Formerly Burma.
At present there are insufficient data to incorporate most of these factors.
Nonetheless, there are sufficient data to allow for incorporation of current knowledge on the
effects of water levels and temperature. For some countries, the effects of organic and
mineral fertilizer can also be included. Inclusion of the remaining factors may become feasible
in the future.
3.3.1 Summary of Recommended Method
The recommended methodology follows the draft IPCC-OECD methodology (OECD,
1991), in which the area harvested is multiplied by a season length and an emission
coefficient (see equation 3.1, above). Three significant refinements to the draft IPCC-OECD
methodology, however, have been made here. First, the definition of "season length" has
Page 3-17
-------
been changed from length of cultivation season (or growing season) to length of emission
season (or flooding season). This change in meaning results in many cases in a shorter
season length and, therefore, a lower emission estimate. The second refinement is a
distinction between "continuously flooded" and "intermittently flooded" areas. In the
methodology described in this section, these two water-management regimes are assigned
different emission factors, and annual emission estimates for the two regimes are summed.
This also results in a lower emission estimate, as intermittently flooded areas are assigned
lower emission coefficients than continuously flooded areas. (Dryland, or upland, rice areas
are assumed to emit no CH4 in this methodology, as in OECD (1991).)
The last refinement to the draft IPCC-OECD methodology is the recommendation that
annual emissions be estimated for three contiguous years and then averaged, rather than
averaging three years of harvested area and multiplying the average area by a season length
and emission coefficient. The draft IPCC-OECD methodology only incorporated interannual
changes in area harvested. This change allows for incorporation of interannual changes in
season length and emission coefficients, where such data exist.
The steps in the recommended methodology are as follows:
(1) For each of three contiguous years, estimate annual emissions.
(i) Estimate annual emissions from flooded fields:
Flux(f loaded) =Y, ^(flooded) • As(flooded) • Ts(flooded)
s=1
where:
(j)(flooded)
A(flooded)
T(flooded)
emission factor for flooded conditions in Tg/m2/day;
area of rice agriculture under flooded regime in m2;
number of days under cultivation when flooded
(days/growing season);
number of growing season(s) per year when flooded
conditions are present in the rice field; and
year.
Emission factors should be based on local measurements or, if not available,
on measurements from a similar climatic zone. If measurements from a
different climatic zone must be used, correct flux for soil temperature effect:
multiplyby [O10](:
Q10 is the ratio of the flux at a soil temperature 10°C above the base
temperature to the flux at the base temperature, T is the soil temperature, and
Tb is the base temperature.
Page 3-18
-------
(ii) Estimate annual emissions from intermittently flooded fields:
n
Flux(intermittenf) = ^ $s(intermittent) • As(intermittent) T ^intermittent)
s=1
(iii) Sum the annual emission estimates to obtain the total annual flux:
Annual Flux = F = Flux(flooded) + Flux(intermittenf)
The procedure is described by the following general formula:
(3.3)
where i represents water-management regimes - flooded and intermittent.
(2) Average three annual emission estimates to obtain annual average:
where y represents one of three inventory years.
Other factors can be added where data are available. For example, where the effects of
fertilizer use on emission coefficients have been measured, and areas and season lengths
under different management regimes are known, fertilizer use can be incorporated into the
calculations as follows:
Flux(flooded) = Flux(flooded, organic) + Flux(flooded, chemical)
where Flux (flooded, organic) is calculated according to step (i), above, using the areas,
emission factors, length of flooding season, and temperatures applicable to the amount of rice
grown under flooded conditions using organic fertilizers. Flux(flooded, chemical) is calculated
analogously. Flux(intermittent) is also calculated as in (i), and again, the flooded and
intermittent fluxes are summed to yield a total annual flux.
Adding fertilizer effect:
(3.4)
where j represents different fertilizer types. Each component of equation 3.3 is calculated by
equation 3.4.
Page 3-19
-------
Additional factors can be added in the same way, for example:
Aijk Tijk
(3.5)
where k represents different soil types. Each component of equation 3.4 is calculated from
3.5 and then each component of equation 3.3 is calculated from equation 3.4. The process
may be continued for more factors.
3.4 RESULTS
Applying the methodology discussed in section 3.3 through equation 3.3 results in an
average global emission estimate of 65 Tg CH4/year. This estimate is based on three years
of data, centered on 1990, and incorporates the effects of water-management regime and
temperature on flux. Some of the daily fluxes implicitly incorporate information on organic vs.
mineral fertilizers, depending on the flux data available for a given region.
Together, China and India account for over half of the estimated global CH4 emissions
from rice cultivation. Six countries - China, India, Indonesia, Thailand, Bangladesh, and Viet
Nam -- account for over 80% of the estimated global emissions. Table 3-5 summarizes the
global distribution of emissions by country, as well as area, season length, and flux data.
3.5 TRENDS
Rice paddies played an important role in the growth of atmospheric CH4, particularly
from the middle of this century to the present. However, the rate of increase in atmospheric
CH4 has slowed in the last decade. Predicting future emissions from rice cultivation using
either the trend of past emissions or population-area relationships will probably lead to large
overestimates (Khalil and Rasmussen, 1990a; and Khalil et al., 1993). The increase in rice
cultivation was likely one of the main contributors to the increase of CH4 during the last
century (Figure 3-3). However, factors that were important causes of the increase in the past
are changing and probably will not be as important in the future. The time series of global
CH4 emissions from rice estimated by Khalil and Shearer (1993) is largely influenced by the
tremendous increase in area planted to rice in the last four decades. By the early 1980s, this
growth in area had slowed significantly.
Figure 3-4 shows a time series of per-capita rice field CH4 emissions for China (Khalil
et al., 1993). Except for a brief period in the 1950s, per-capita emissions have steadily
declined, despite the increase in area planted (nearly 10 million hectares). These data
illustrate the difficulties associated with using historical population-area relationships and
population projections to predict future CH4 emissions from rice-growing countries. Efforts to
increase rice yield without increasing the area planted, such as irrigation, fertilizers, and high-
yield rice cultivars, will affect the future CH4 emissions from rice paddies.
Page 3-20
-------
3.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
Uncertainty still exists in the estimate of global emissions, and there are substantial
uncertainties in estimates of regional fluxes of CH4 from rice cultivation. The principal
uncertainties still are in the association of a measured, seasonal flux with a specific region;
the length of the CH4 emission season in a specific region; and the effects of specific factors
of climate and management on the seasonal flux.
3.6.1 Specific Estimation Problems
Table 3-1 lists the areas where CH4 fluxes from rice cultivation have been measured.
The most comprehensive controlled studies are from the United States and Europe, which do
not represent the most important rice-planting areas. Systematic seasonal measurements
have been made in three locations in China but only preliminary data are available for India
(National Physical Laboratory, 1992), with the largest rice-growing area in the world. No
seasonal measurements have been made for the tropical rice-growing areas of Southeast
Asia or for the continents of Africa and South America. The uncertainties are particularly large
for Brazil, Nigeria, and Madagascar, countries that have large areas planted to rice but that
lack accurate data on the fractions of the area under different water-management regimes.
The assumption that dryland rice is not a significant source of CH4 uses the best
information currently available, but it has never been tested. The difficulties in estimating the
season of CH4 emission from the growing season for rice were demonstrated in Table 3-2.
Except for the limited number of areas where sampling has been done, the season of CH4
emission must be estimated from the growing season. Using the growing season without
adjustments will probably overestimate the CH4'emission season.
3.6.2 Verification of Country Inventories
In all cases, an emission inventory must be fully documented. The documentation has
two aspects. First, the method of calculation must be specified as in equation 3.5. Emission
factors and extrapolants, such as area or yield, must be delineated. Second, all data and
original sources must be referenced if not included explicitly as part of the inventory report. It
is desirable in all cases to rely on published information, whether from the country's
government, an international organization (such as the U.N. Food and Agriculture
Organization or the International Rice Research Institute), or the scientific literature. The final
data sets must be scrutinized by the scientific community and published in the open scientific
literature, according to long-established traditions of scientific evaluation and publication.
3.7 CONCLUSIONS
Rice cultivation is one of the most important anthropogenic sources of CH4. The main
factors that control emissions from rice fields are now fairly well understood, though the
effects of these factors are not well quantified.
Global estimates of CH4 emissions from rice fields are roughly constrained by the
lifetime of CH4 in the atmosphere, isotopic measurements, and measurements from ice cores.
Recent estimates of global emissions range from 50 to 100 Tg/year. There are no constraints
on emissions for smaller regions, geographical or political. Estimates for smaller regions tend
Page 3-21
-------
TABLE 3-5
RESULTS: AVERAGE ANNUAL (1990) METHANE EMISSION ESTIMATES AND AREA,
SEASON, AND FLUX DATA FOR MAJOR RICE-PRODUCING COUNTRIES
Country
India
China
Indonesia
Bangladesh
Thailand
Viet Nam
Myanmar
Brazil
Philippines
Pakistan
Japan
Cambodia
Nigeria
Nepal
S. Korea
Madagascar
United States
Sri Lanka
Taiwan
N. Korea
Malaysia
Laos
Rest of World
Global Total
Annual
Methane
Emissions3
(Tg CH4/yr)
15.8
21.4
5.0
3.6
4.8
3.2
1.3
0.5
1.3
0.7
0.2
0.9
0.2
0.2
0.8
0.1
0.3
0.4
0.4
0.3
0.3
0.2
3.0
65
Harvested
Area"
(103 ha)
42,321
33,265
10,403
10,303
9,878
6,069
4,774
4,450
3,413
2,093
2,073
1,800
1,567
1,440
1,237
1,135
1,114
793
700
673
640
625
Average
Season
(days)
106
117
110
123
123
119
139
101
98
103
123
134
103
90
103
167
123
122
119
103
109
123
Average
Flux
(mg/m2-hr)
19.1
25.1
22.3
17.4
26.7
22.4
11.7
18.1
22.1
13.4
4.0
27.3
11.7
8.9
26.7
11.7
9.2
21.4
22.4
21.6
19.6
36.3
Effective
Flux0
(mg/m2-hr)
15.8
22.9
18.4
11.8
16.4
18.3
7.9
4.2
16.4
13.4
3.8
15.6
4.4
7.2
25.6
2.9
9.2
17.6
21.8
17.0
15.7
12.6
Percent
Dry Land
15.3
1.6
15.2
14.3
12.3
7.3
14.9
77.0
11.8
0.0
4.0
27.3
62.0
4.0
1.0
75.0
0.0
6.8
3.0
13.3
12.9
49.2
Percent
Rain-
fed
32.5
5.2
7.2
71.5
65.7
27.8
' 43.1
N/A
34.3
0.0
0.0
39.0
N/A
67.2
8.0
N/A
0.0
27.6
0.0
20.0
23.5
39.9
a. These emission estimates are averages of estimates for 1989 through 1991. The number of significant digits
shown is not an indicator of precision; see discussion of uncertainties in the text.
b. Harvested area is annual average, using 1989 to 1991 data (FAO, 1992). This statistic includes double and
triple cropping.
c. Effective flux is the average flux adjusted for the percent area in dryland rice (flux = 0) and percent area in
shallow rain-fed rice (flux = 60% of seasonal average). Fluxes are weighted averages by season and area;
some rounding errors may result.
N/A = No data available.
Page 3-22
-------
JH
93
WD
H
80
70
60
50
40
30
20
FIGURE 3-3
TIME SERIES OF GLOBAL CH4 EMISSIONS FROM RICE AGRICULTURE
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990
Year
Source: Khalil and Shearer, 1993; Reprinted by permission of the authors.
-------
FIGURE 3-4
PQ
"SB
H
(M
0)
u
s
O
fl
O
Tf
=
u
40
35 -
30 -
25 -
20
PER-CAPITA EMISSIONS OF METHANE (Tg CH4 PER BILLION POPULATION (BP))
FROM RICE FIELDS IN CHINA DURING THE LAST CENTURY
(The Per-Capita Emissions Have Declined by 35% From the 1930s to the 1980s)
1900
1930
1960
1990
Time
Source: Khalil et al., 1993; Reprinted by permission of the authors.
-------
I
to be much more variable. For some countries, even the area planted to rice is not accurately
known, and there are no local flux measurements. Unless it is one of the few areas where
measurements have been made, the smaller the region, the more uncertain the estimate.
This report estimates 1990 emissions from rice cultivation at 65 Tg CH4.
Although the growth in area harvested during the last century is believed to have
caused a concurrent growth in CH4 emissions from rice cultivation, future emissions are likely
to be controlled in large part by other factors that are now changing. Chemical fertilizers are
replacing traditional organic fertilizers in many areas of the world, new high-yield varieties of
rice have been developed, and irrigation systems make possible multiple crops of rice. These
factors will influence CH4 emissions from rice cultivation in the next few decades, making it
difficult to make credible estimates of future global emissions based on current knowledge.
3.8 REFERENCES
Aselmann, I., and P.J. Crutzen. 1989. Global distribution of natural freshwater wetlands and
rice paddies, their net primary productivity, seasonality and possible wetland emissions. J. of
Atmos. Chem. 8:307-358.
Bachelet, D., and H.U. Neue. 1993. Methane emissions from wetland rice areas of Asia.
Chemosphere 26(1-4) :219-238.
Bansil, P.C. 1984. Agricultural Statistics in India: A Guide. 3rd Rev. Edition. Oxford & IBM
Publishing Co., New Delhi, India.
Bingemer, H.G., and P.J. Crutzen. 1987. The production of methane from solid wastes. J.
Geophys. Res. 92:2,181-2,187.
Blake, D.R. 1984. Increasing concentrations of atmospheric methane. Ph.D. dissertation,
University of California at Irvine, Irvine, California.
Blake, D.R., and F.S. Rowland. 1986. World-wide increase in tropospheric methane, 1978-
1983. J. Atmos. Chem. 4:43-62.
Bolle, H.-J., W. Seiler, and B. Bolin. 1986. Other greenhouse gases and aerosols. In Bolin,
B., B.R. Doos, J. Jager, and R.A. Warrick, .eds. The Greenhouse Effect, Climate Change and
Ecosystems (SCOPE 29). J. Wiley and Sons, New York, New York. 157-198.
Boone, D.R. 1993. Biological formation and consumption of methane. In Khalil, M.A.K., ed.
Atmospheric Methane: Sources, Sinks and Role in Global Change. Springer-Verlag, Berlin,
Germany. In press.
Bouwman, A.F. 1991. Agronomic aspects of wetland rice cultivation and associated methane
emissions. Biogeochemistry 15:65-88.
Chen, Z., L. Debo, K. Shao, and B. Wang. 1993. Features of CH4 emission from rice paddy
fields in Beijing and Nanjing, China. C/7emospftere26(1-4):239-246.
China Agriculture Yearbook 1985, 1986, 1987, 1989, 1990. Agribookstore, Hampton, Virginia.
Page 3-25
-------
Cicerone, R.J., and J.D. Shelter. 1981. Sources of atmospheric methane: Measurements in
rice paddies and a discussion. J. Geophys. Res. 86(C8):7,203-7,209.
Cicerone, R.J., J.D. Shelter, and C.C. Delwiche. 1983. Seasonal variation of methane flux
from a California rice paddy. J. Geophys. Res. 88(C15):11,022-11,024.
Cicerone, R.J., C.C. Delwiche, S.C. Tyler, and P.R. Zimmerman. 1992. Methane emissions
from California rice paddies with varied treatments. Global Biogeochemical Cycles 6(3):233-
248.
Cicerone, R.J., and R.S. Oremland. 1988. Biogeochemical aspects of atmospheric methane.
Global Biogeochemical Cycles 2:299-327.
Dai, Aiguo. 1988. CH4 Emission rates from rice paddies in Hangzhou China during fall of
1987. M.S. Thesis, Institute of Atmospheric Physics, Academia Sinica, Beijing, China.
De Datta, Surajit K. 1975. Upland rice around the world. In Major Research in Upland Rice.
International Rice Research Institute, Los Banos, Philippines.
Donahue, T.M. 1979. The atmospheric methane budget. In Aikin, A.C., ed. Proceedings of the
NATO Advanced Study Institute on Atmospheric Ozone: Its Variation and Human Influences.
U.S. Department of Transportation, Washington, D.C.
Ehhalt, D., and U. Schmidt. 1978. Sources and sinks of atmospheric methane. PAGEOPH
116:452-464.
FAO (Food and Agriculture Organization of the United Nations). 1976-1992. FAO Yearbook:
Production (year), vols. 30 through 45. FAO, Rome, Italy.
Fraser, P.J., M.A.K. Khalil, R.A. Rasmussen, and L.P. Steele. 1984. Tropospheric methane in
the mid-latitudes of the Southern Hemisphere. J. Atmos. Chem. 1:125-135.
Grist, D.H. 1986. Rice. 6th edition. Longman, Inc., New York, New York.
Harriss, R.C., D.I. Sebacher, and F.P. Day, Jr. 1982. Methane flux in the Great Dismal
Swamp. Nature 297:673-674.
Holzapfel-Pschorn, A., R. Conrad, and W. Seller. 1985. Production, oxidation and emission of
methane in rice paddies. FEMS Micro. Ecol. 31:343-351.
Holzapfel-Pschorn, A., and W. Seiler. 1986. Methane emission during a cultivation period from
an Italian rice paddy. J. Geophys. Res. 91 (D11):11,803-11,814.
Huke, R.E. 1982. Rice Area by Type of Culture: South, Southeast, and East Asia.
International Rice Research Institute (IRRI), Los Banos, Philippines.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
Page 3-26
-------
IRRI (The International Rice Research Institute). 1990. World Rice Statistics 1989. IRRI,
Philippines.
Keller, M., W.A. Kaplan, and S.C. Wofsy. 1986. Emissions of N2O, CH4 and CO2 from tropical
forest soils. J. Geophys. Res. 91(D11):11,791-11,802.
Khalil, M.A.K., and R.A. Rasmussen. 1983. Sources, sinks and seasonal cycles of
atmospheric methane. J. Geophys. Res. 88:5,131-5,144.
Khalil, M.A.K., and R.A. Rasmussen. 1990a. Atmospheric methane: Recent global trends.
Environ. Sci. Technol. 24:549-553.
Khalil, M.A.K., and R.A. Rasmussen. 1990b. Constraints on the global sources of methane
and an analysis of recent budgets. Tellus 420:229-236.
Khalil, M.A.K., and R.A. Rasmussen. 1991. The global methane cycle. In Proceedings of the
Indo-U.S. Workshop on Impact of Global Climatic Changes on Photosynthesis and Plant
Productivity. New Delhi, India. 8-12 January 1991.
Khalil, M.A.K., R.A. Rasmussen, M.-X. Wang, and L. Ren. 1990. Emissions of trace gases
from Chinese rice fields and biogas generators: CH4, N2O, CO, CO2, chlorocarbons, and
hydrocarbons. Chemosphere 20(1-2):207-206.
Khalil, M.A.K., R.A. Rasmussen, M.-X. Wang, and L. Ren. 1991. Methane emissions from rice
fields in China. Environ. Sci. Tech. 25:979-981.
Khalil, M.A.K., and MJ. Shearer. 1993. Methane emissions from rice agriculture.
Chemosphere. In press.
Khalil, M.A.K., MJ. Shearer, and R.A. Rasmussen. 1993. Methane sources in China:
Historical and current emissions. Chemosphere 26(1 -4):127-142.
Mariko, S., Y. Harazono, N. Owa, and I. Nouchi. 1991. Methane in flooded soil water and the
emission through rice plants to the atmosphere. Environ, and Exp. Bot. 31:343-350.
Masscheleyn, P.M., R.D. DeLaune, and W.H. Patrick, Jr. 1993. Methane and nitrous oxide
emissions from laboratory measurements of rice soil suspension: Effect of soil oxidation-
reduction status. Chemosphere 26(1 -4):251 -260.
Matthews, E., I. Fung, and J. Lerner. 1991. Methane emission from rice cultivation:
Geographic and seasonal distribution of cultivated areas and emissions. Global
Biogeochemical Cycles 5(1):3-24.
Mitchell, B.R. 1980. European Historical Statistics. 2d Rev. Ed. Facts on File, New York, New
York.
Mitchell, B.R. 1982. International Historical Statistics, Africa and Asia. New York University
Press, New York, New York.
Mitchell, B.R. 1983. International Historical Statistics: The Americas and Australasia. Gale
Research Co., Detroit, Michigan.
Page 3-27
-------
Morris, R.A., DJ. Greenland, and R.E. Huke. 1984. Remote sensing and rice research. In
Deepak, A., and K.R. Rao, eds. Applications of Remote Sensing for Rice Production. A.
Deepak Publishing, Hampton, Virginia. 33-54.
National Physical Laboratory. 1992. Greenhouse Gas Emission in India, 1991 Methane
Campaign. Mitra, A.P., ed. Scientific Report No. 2. New Delhi, India.
Neue, H.-U., P. Becker-Heidmann, and H.W. Scharpenseel. 1990. Organic matter dynamics,
soil properties, and cultural practices in rice lands and their relationship to methane
production. In Bouwman, A.F., ed. Soils and the Greenhouse Effect. J. Wiley and Sons, New
York, New York. 457-466.
Neue, H.-U., and P.A. Roger. 1993. Rice agriculture: Factors controlling emissions. In Khalil,
M.A.K., ed. Atmospheric Methane: Sources, Sinks and Role in Global Change. Springer-
Verlag, Berlin, Germany. In press.
Nouchi, I., S. Mariko, and K. Aoki. 1990. Mechanism of methane transport from the
rhizosphere to the atmosphere through rice plants. Plant Physiol. 94:59-66.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Parashar, D.C., P.K. Gupta, J. Rai, R.C. Sharma, and N. Singh. 1993. Effect of soil
temperature on methane emission from paddy fields. Chemosphere 26:247-250.
Parashar, D.C., J. Rai, P.K. Gupta, and N. Singh. 1991. Parameters affecting methane
emission from paddy fields. Indian Journal of Radio & Space Physics 20:12-17.
Rasmussen, R.A., and M.A.K. Khalil. 1981 a. Atmospheric methane (CH4): Trends and
seasonal cycles. J. Geophys. Res. 86:9,826-9,832.
Rasmussen, R.A., and M.A.K. Khalil. 1981b. Increase in the concentration of atmospheric
methane. Atmos. Env. 15:883-886.
Sass, R.L, P.M. Fisher, P.A. Harcombe, and FT. Turner. 1990. Methane production and
emission in a Texas rice field. Global Biogeochemical Cycles 4(1):47-68.
Sass, R.L, F.M. Fisher, P.A. Harcombe, and F.T. Turner. 1991 a. Mitigation of methane
emissions from rice fields: Possible adverse effects of incorporated rice straw. Global
Biogeochemical Cycles 5(3):275-287.
Sass, R.L, F.M. Fisher, F.T. Turner, and M.F. Jund. 1991b. Methane emission from rice fields
as influenced by solar radiation, temperature, and straw incorporation. Global Biogeochemical
Cycles 5(4):335-350.
Sass, R.L, F.M. Fisher, Y.B. Wang, F.T. Turner, and M.F. Jund. 1992. Methane emission
from rice fields: The effect of floodwater management. Global Biogeochemical Cycles
6(3):249-262.
Page 3-28
-------
SchCitz, H., A. Holzapfel-Pschorn, R. Conrad, H. Rennenberg, and W. Seller. 1989a. A 3-year
continuous record on the influence of daytime, season, and fertilizer treatment on methane
emission rates from an Italian rice paddy. J. Geophys. Res. 94(D13):16,405-16,416.
Schutz, H., W. Seller, and R. Conrad. 1989b. Processes involved in formation and emission of
methane in rice paddies. Biogeochemistry 7:33-53.
Schutz, H., W. Seiler, and H. Rennenberg. 1990. Soil arid land use related sources and sinks
of methane (CH4) in the context of the global methane budget. In Bouwman, A.F., ed. Soils
and the Greenhouse Effect. J. Wiley and Sons, New York, New York. 269-285.
Seiler, W., A. Holzapfel-Pschorn, R. Conrad, and D. Scharffe. 1984. Methane emission from
rice paddies. J. ofAtmos. Chem. 1:241-268.
Shearer, M.J., and M.A.K. Khalil. 1993. Rice agriculture: Emissions. In Khalil, M.A.K., ed.
Atmospheric Methane: Sources, Sinks and Role in Global Change. Springer-Verlag, Berlin,
Germany. In press.
Steele, L.P., P.J. Fraser, R.A. Rasmussen, M.A.K. Khalil, T.J. Conway, A.J. Crawford, R.H.
Gammon, K.A. Masarie, and K.W. Thoning. 1987. The global distribution of methane in the
troposphere. J. Atmos. Chem. 5:125-171.
U.S. Department of Agriculture. 1984. Agricultural Statistics of People's Republic of China,
1949-1982. Statistical Bulletin No. 714. U.S. Government Printing Office, International
Economics Division, Economic Research Service, Washington, D.C.
Wassman, R., H. Papen, and H. Rennenberg. 1993. Methane emission from rice paddies and
possible mitigation strategies. Chemosphere 26(1-4):201-218.
Yagi, K., and K. Minami. 1990. Effect of organic matter application on methane emission
from some Japanese paddy fields. Soil Sci. Plant Nutr. 36(4):599-610.
Page 3-29
-------
-------
CHAPTER 4
METHANE EMISSIONS FROM ANTHROPOGENIC BIOMASS BURNING
4.1 SUMMARY
Biomass burning has been recognized as a significant source of atmospheric methane
at least since the late 1970s. Like most other sources of methane, it has both natural and
anthropogenic causes, although anthropogenic causes now predominate. Most of the
estimates of methane emissions from biomass burning in the past have relied on a uniform
emission factor for all types of burning. This results in the share of trace-gas emissions for
different types of burning being the same as the amounts of biomass burned in those types.
In fact, the emission factors for various types of burning differ by a factor of 15. This chapter
estimates global emissions of methane from biomass burning based on an approach followed
for Africa by Delmas et al. (1991). Several recent papers also attempt the same estimation
(see e.g., Levine et al., 1993; Andreae and Warneck, 1993; and Hao and Ward, 1993).
This chapter estimates that anthropogenically caused biomass fires account for
approximately one-seventh of the anthropogenic methane from all sources. Excluding
emissions from prescribed forest burning and certain other minor categories for which data are
sparse, this number is estimated to be ~ 49 teragrams (Tg)1 CH4/yr, with rough confidence
limits of ± 40%. Confined burning of biofuels, such as firewood and charcoal, agricultural
residues, and cattle dung, accounts for 38% (of all) biomass burned and 43% (of
anthropogenic) methane emissions from biomass burning. Savanna burning accounts for 21%
of the biomass burned, but only for 12% of methane emissions because of the nature of the
biomass that burns. In contrast, deforestation for land-use change, although accounting for
11% of the biomass burned, contributes 17% of anthropogenic methane emissions. Sixteen
countries, all in the developing world, account for almost two-thirds of the total anthropogenic
methane emissions from biomass burning. By far, the largest emitters are China and Brazil,
followed by Indonesia, India, Zaire, Nigeria, and Mexico.
Much of the uncertainty in estimating emissions from biomass burning comes from the
uncertainty in the amounts burned. A large source of uncertainty is information about biofuel
use. A compilation of country-specific consumption based on household surveys would help
reduce this uncertainty.
When prescribed forest burning, wildfires and other sources that are not included in
this chapter for lack of published information are included, and when some of the uncertainties
are removed, the contribution of biomass burning could be close to that indicated by isotopic
studies—i.e., 58 Tg/yr.
It is difficult to forecast trends in methane emissions from biomass burning. Each type
of biomass burning is influenced by a different constellation of forces—among them, population
pressure, the need for agricultural land, economic development, and the availability of
alternatives. Future biomass use also will be determined by the shifts in the mix of energy
sources in response to the threat of global warming. Though biomass usage and burning are
1 Teragram = 108 metric tonnes = 1012 grams.
Page 4-1
-------
likely to increase in the future, it should be possible to hold down methane emissions as
newer technologies for burning biofuels will be more efficient than those in use today.
4.2 BACKGROUND
4.2.1 Introduction
Biomass burning must have existed on earth soon after land plants evolved 350 to 400
million years ago (Andreae, 1991). There is more certainty of large fires that occurred
approximately 66 million years ago (Anders et al., 1991). Natural fires can be triggered by
any one of several causes, such as lightning or meteorite strikes, volcanic eruptions, and even
constant rubbing of bamboo plants (Kaul, 1991). The capture of fire by early hominids 1.5
million years ago (Pyne, 1991) changed the etiology of fire on earth. Since then, most fires
have been of anthropogenic origin.
Burning of some type of biomass is connected intimately with human life in all
societies; the more traditional a society, the more culturally entrenched the practice is likely to
be. Fire is used for land management, for deriving energy, for waste disposal, and for
personal, aesthetic, or religious reasons. Often the motivation is a mixture of two or more of
the above reasons. Prescribed burning of savannas, agricultural residues, or forests can be
considered as examples of land or waste management. Open-dump burning of municipal
solid wastes in developing countries is purely a waste disposal activity, but the burning of the
same wastes in incinerators is associated sometimes with energy recovery. Cooking energy
is derived from biomass burned in stoves, which because of their copious production of
smoke, also have multiple uses in insect control and in thatch and food preservation. Space
heat is derived from wood stoves, steam and electricity from boilers, and gaseous fuels from
gasifiers. Wood burning in a well-designed fireplace provides both energy and a feeling of
warmth. In many cultures, fire is a witness at all religious ceremonies and a vehicle for the
disposal of the dead. Tobacco smoking, too, is a form of biomass burning. Some reasons
have become less important than they used to be, and some activities have disappeared from
some continents altogether, such as burning of grasslands to drive game for hunting.
There are more types of biomass burned than there are reasons for burning them.
Whole tree trunks are burned in boilers in electricity generating stations in Minnesota, and
bark chips in Vermont. In many areas of developing countries where the cooking energy
situation is bleak, any available combustible material is burned, including dried leaves, roots,
and cattle dung. Just as varied as the types of biomass consumed are the conditions under
which burning occurs. These range from completely unconfined burning to a rudimentary
three-stone fire to carefully optimized efficient modern boilers.
It is difficult to estimate how much biomass is burned worldwide annually, given the
multitudinous settings in which it is burned. Estimates range from 5 to 13 petagrams (Pg) of
dry matter per year (Crutzen and Andreae, 1990; and Andreae, 1991). As shown in Section
4.2.2 and in Table 4-1, an overwhelming fraction, close to 95%, of biomass burned is from
anthropogenic causes.
Since biomass is composed essentially of cellulose and lignin—which are
carbohydrates consisting of carbon, hydrogen, and oxygen—the products of its complete
Page 4-2
-------
TABLE 4-1
ESTIMATES OF BIOMASS BURNED GLOBALLY
(Pg dm/yr)
TYPE
Savannah-unintentional
Savannah-prescribed
Forests-unintentional
Forests-prescribed
Forests-shifting cultivation
Forests-land-use change
Urban park wastes
Urban MSW
Agricultural residues
Fuelwood
Animal dung
Charcoal
Other processed biofuels
TOTAL
NATURAL
Open
Low High
(small?)
X
0.27 0.47
X
0.3 0.5
ANTHROPOGENIC
Open
Low High
(small?)
0.72 3.71
?
0.06 0.20
1.12 2.22
0.442 1.562
Not estimated
Not estimated
0.472
X
2.8 8.1
Confined
Low High
X
Not estimated
0.64 1.55
1.32 2.72
0.093-0.404
0.021
Not estimated
2.1 4.4
TOTAL
(small?)
0.7-3.7
0.33-0.672
1.1-2.2
0.44-1.56
Not estimated
Not estimated
1.12-2.021
1.3-2.7
0.09-0.4
0.02
Not estimated
5-13
Sources: Andreae, 1991; Crutzen and Andreae, 1990; Meyers and Leach, 1989; and Armitage and Schramm, 1989.
-------
combustion are carbon dioxide (CO2) and water vapor, as the following simplified equation
shows:
25O = 2AnCO\
(3.1)
For a variety of reasons, an uncontrolled combustion process is never fully complete.
The insufficiency of oxygen or a drop in fire temperature could prevent some carbon in
biomass from being oxidized to CO2. Consequently, burning generates several carbonaceous
products of incomplete combustion. Some of these are particulates and some gaseous, as
Figure 4-1 shows schematically. Scientists have identified hundreds, if not thousands, of
chemical compounds in biomass smoke (Cooper, 1980). Methane (CH4) is one such gaseous
product of incomplete combustion of biomass and also the major hydrocarbon. While CO2 is
the largest constituent in all cases, the emissions of the products of incomplete combustion
vary by over an order of magnitude, depending upon the type of fire.
Several investigators have estimated amounts of CH4 emitted globally from biomass
burning. Figure 4-2 shows the most likely point estimates and ranges. While a downward
trend is indicated by this figure, this does not mean necessarily that CH4 emissions from
biomass burning are decreasing; on the contrary, they may be increasing. It is just that
estimates are improving with more information on the levels of burning activities and on
activity-specific emission factors. So while early central estimates2 were roughly 60 Tg CH4 /yr
(e.g., Crutzen et al., 1979), many current best central estimates are also between 50 and 60
Tg CH4/yr (e.g., Quay et al., 1988; and Andreae, 1991). The early ranges were wide, from 25
to 110 Tg CH4/yr. Now, while estimates of the lower limit (~25 Tg CH4/yr) have not changed
much, the upper-range limit is thought to be 80 Tg CH4/yr (IPCC, 1992).
Every biomass fire is comprised of four phases of combustion: flaming, pyrolysis,
smoldering, and glowing. These phases are chemically different processes and coexist in
different proportions during the fire. Fuel characteristics and combustion conditions decide
their relative durations and magnitudes. The differences in the relative durations and amounts
of gases and particulates emitted during these phases make for the enormous variety of
products from fires (Ward and Hardy, 1991).
Flaming dominates the start of a fire. Surface materials supply the volatile
hydrocarbons needed to sustain it. The heat from the flame initiates the thermal breakdown
(pyrolysis) of cellulosic- and lignin-containing materials and provides fuel gases, including CH4,
that sustain the visible flaming process.
In the smoldering phase, carbon begins to build on the solid-fuel surfaces, and the
pyrolytic reactions no longer produce sufficient fuel gases to maintain the flame envelope.
The process ultimately leads to the production of charcoal, where the only combustion
occurring is of the glowing type. This is a surface reaction of oxygen with carbon (Ward,
1990) and differs from smoldering in that it is not accompanied by appreciable emissions of
volatile compounds.
! All numbers for trace-gas weights in this chapter are in the full molecular weight basis.
Page 4-4
-------
FIGURE 4-1
CARBONACEOUS PRODUCTS OF BIOMASS BURNING
Methane
Non-methane
Volatile Organic]
Compounds
C Hydro-carbons ^
(Carbon Monoxide)
Q41MOUS
Products of
IncompUU
Combustion
Gases
Air-borne
Products
J
Note: Arrows are not drawn to scale
Page 4-5
-------
FIGURE 4-2
ESTIMATES OF METHANE EMISSIONS FROM GLOBAL BIOMASS BURNING
120
110
100
n
§"
D) TO
CZ^
g •
0
8 50
* «
30
20
10
n
-H
•
}
L
i1 n2
J LJ
I 1
I
T
L n8 .
L
JJ
1
• i
•
3«
i
•
jf
l"
1
1
•
. 5°D"
f" '
LJ
_1«
u
M
•P*
if
D
it
ft
T
>,..,
i
I
1 4
,»_
•
c
,»
J
• II
T
i
•
1
., u
I1
.1
1979 I960 1831 1982 1983 1984 1985 1988 1937 1968 1989 1990 1991 1992 1993
Year Estimate Published
Note: The superscripts are references to the following publications.
1. Crutzen et al., 1979
2. Sheppard et al, 1982
3. Crutzen, 1983
4. Cicerone, 1983
5. Khalil and Rasmussen, 1983
6. Seiler, 1984
7. Blake, 1984
8. Crutzen, 1987
9. Bolle et al., 1986
10. Bingemer and Crutzen, 1987
11. Cicerone and Oremland, 1988
12. Craig et al, 1988
13. Stevens and Engelkemeir, 1988
14. Quay et al, 1988
15. Wahlen et al, 1989
16. IPCC, 1991
17. IPCC, 1990
18. Crutzen and Andreae, 1990
19. Quay et al, 1991
20. Fung et al, 1991
21. Andreae, 1991
22. Crutzen, 1991
23. Lobert et al, 1991
24. IPCC, 1992
25. Levine et al, 1992
26. Andreae and Warneck, 1993.
27. Stevens, 1993
28. Delmas, 1993
-------
It has been reported that CH4 is produced in much larger quantity during the
smoldering phase than in the flaming phase, with emission factors that are 2-3 times greater
(Ward, 1990; and Lobert et al., 1991). It is possible that it may not be the production but the
release of CH4 that is greater during smoldering because it is oxidized to a smaller degree in
the flame. Since combustion efficiency is invariably higher during the flaming phase than it is
during the smoldering phase (Ward, 1990), the emissions of products of incomplete
combustion will be greater, and the emissions of CO2 smaller, during smoldering than during
flaming.
Any factor that influences pyrolysis and, therefore, the production of CH4 or its
subsequent combustion in the flame will affect the emissions of CH4 from biomass burning.
Thus, the type of biomass and its chemical composition, its moisture content, the fuel
geometry, size distribution, and combustion conditions-turbulence, availability of oxygen, etc.~
influence emissions. During the burning of dung, the smoldering phase dominates; therefore,
this type of burning is expected to have a high CH4 emission factor. Conversely, burning of
crop residues and grasses is mostly flaming combustion and can be expected to have lower
emission rates. Up to a point, turbulence promotes flaming and lowers emissions, but
excessive turbulence can cause the products of pyrolysis to escape before being consumed in
the flames. Finally, since charcoal burning is mostly glowing, it too should be associated with
lower emission factors.
The preceding discussion brings up the importance of considering the entire fuel cycle
when evaluating emissions from different technologies that use biomass as a fuel. For
example, although CH4 emissions from charcoal-burning stoves may be less than those from
wood-burning cook stoves, this does not imply necessarily that charcoal-burning stoves are
preferable from the perspective of greenhouse gas emissions. Substantial amounts of CH4
emissions associated with charcoal use occur during pyrolysis in charcoal making. Similarly, if
one were to compare emissions from biogas combustion with those from dung burning, one
must consider also the leakages from the digester dome and connecting pipes (Figure 4-3).
4.2.2 Varieties of Biomass Burning
Regional emissions of CH4 are estimated by multiplying the emission factors
associated with a burning activity by its extent and summing these subsectoral emissions over
all activities. Since the emission factors for burning a particular type of biomass under a
particular set of conditions are unlikely to vary geographically, the differences in CH4
emissions between regions depend on the relative mix of the types of burning activity.
Methane emissions from biomass burning in Africa, therefore, are determined largely by
savanna burning, while in South America, they are determined by deforestation, and in Asia,
by biofuel use.
This section describes briefly the different biomass burning activities. Figure 4-3
provides a synoptic view of the different activities. Wherever appropriate, distinctions are
made between natural and anthropogenic, open and contfined, and intentional and accidental
burning-trie first of these distinctions being required by the Framework Convention on Climate
Change.
Page 4-7
-------
FIGURE 4-3
TYPES OF BIOMASS BURNING
1
Natural
Grasslands
rhn
Biomass Burning
I
Unconfined
Forests
1
Anthropogenic
Croplands
Confined
ii
Transformation Devices
_i i
Notes: 1. Subcategories possible within shaded boxes (e.g., land-use change occurring from open vs. closed forests).
2. For each fuel category, information may be available for residential (cooking and space heating), industrial, and commercial uses.
-------
Open Burning
Savanna Burning
Savannas are tropical and subtropical ecosystems consisting mainly of grasses with
varying amounts of shrubs and trees. They have alternating wet and dry seasons and high
average annual temperatures. An annual rainfall of 700 mm divides arid savannas from
humid savannas (Lanly, 1982). Humid savannas have almost a continuous grass cover that
makes them susceptible to fire transmission during the dry season. Patchy covers
exacerbated by a high density of herbivores limit fire propagation in arid regions.
There is evidence of lightning-induced fires in West Africa from geologic times (Menaut
et al., 1991). Natural fires still occur, but they are rare and difficult to identify unequivocally.
There also have been some instances of spontaneous combustion occurring in the wet
season following fermentation in thick herbaceous litters.
Human beings have increased the frequency of savanna fires tremendously, especially
after the transition from hunting-gathering to farming 10,000 years ago (Menaut et al., 1991).
Wherever grazing of animals is an important activity, savanna burning is undertaken to
remove unpalatable stubble and to promote the regrowth of new grasses and succulent
shoots (Hao etal., 1990; and Menaut et al., 1991). While eliminating dead grasses and litter
accumulation, burning also gets rid of weeds, shrubs, arid tree seedlings and reduces the
population of pests, insects, and snakes. Setting fire also helps drive game out of hiding.
This may have been one of the earliest uses of fire by man (Andreae, 1991). Other reasons
for setting fire to savannas include clearing the ground for agriculture, establishing fire breaks
around settlements, and improving communications. While it is likely that burning recycles
phosphorus and some micronutrients, the belief that burning stimulates the recycling of
nitrogen, or that burning is desirable for maintaining the long-run productivity of grasslands,
might be fallacious (Andreae, 1991), since fixed nitrogen is lost through pyro-denitrification
when the burn temperature is high.
Most savanna fires are surface fires, rather than ground or crown fires (Menaut et al.,
1991). Recent tree litter and grasses burn almost completely. Savanna trees seldom burn,
as they have evolved to fire regimes by having a low terpene content, and so have a low
degree of flammability.
Savannas, grasslands, and shrublands are found on all continents, except for
Antarctica, and are still burned in Africa, Australia, South America, and Asia. Estimates of
areal extent range from 1300 (Hao et al., 1990) to 1900 million hectares (mha) (Bolin et al.,
1979)-60% of which is thought to be humid savannas subjected to periodic burning, and the
remainder is arid. The interval between burnings varies from place to place. In the Brazilian
cerrado, it is 1-2 years. Approximately 75% of the African humid savannas burns every year
(Hao et al., 1990). Though in localized instances savannas in Africa burn twice a year, almost
all savanna lands burn no more than every three years (Menaut et al., 1991). Burning the
bush is an integral part of land management in northern Australia, happening typically every
third year (Howden et al., 1992a). Burning of grazing lands is the predominant biomass
burning activity in Australia.
The global amount of dry matter (dm) biomass (i.e., corrected for moisture content)
that is burned annually in grassland fires has been estimated to be 0.7-3.6 Pg dm/yr (Hao et
al., 1990; and Crutzen and Andreae, 1990).
Page 4-9
-------
Because of the periodic nature of grassland burning, all the carbon that is emitted to
the atmosphere each year as carbon dioxide is reabsorbed by the new growth that occurs in
the wet season. However, all the emissions of CH4 are net additions to the atmosphere.3
Since much of the combustion that occurs in savanna burning is flaming, CH4 emissions per
unit amount of dry-matter biomass burned are believed to be low. But if strong winds were to
accompany burning, the emissions could increase, if capture in the flames is reduced.
There is a factor of 2-5 uncertainty in the published estimates of biomass burned
globally in the savannas and so at least that much also in the emissions of CH4 from this
activity. A part of the uncertainty comes from not all lands classified as savanna being used
in the way expected. Thus, some areas so classified may be under permanent agriculture or
under shifting agriculture. Another part stems from the paucity of data on densities of above-
ground biomass exposed to fires. The relative proportions of dead and living biomass (i.e.,
the distribution of moisture contents) at the time of the.fire vary from year to year but are
known only for a few regions. This, too, will have implications for uncertainties associated
with estimating emissions.
Shifting Cultivation
Shifting cultivation is a name given to a cyclical sequence of subsistence farming
events by which a mixture of crops is grown by rotation on different fields. The process
begins by slashing vegetation on a plot of land, allowing the vegetation to dry, isolating the
plot by cutting a fire line, burning the vegetation on the plot one or more times, cultivating the
land until its fertility declines, and finally abandoning the plot and leaving it fallow while moving
on to another plot. The cycle restarts when a cultivator returns to the first plot. In different
parts of the world, the cycle goes by different names: swidden agriculture, field-forest rotation,
bush fallowing, and, in India, as jhum agriculture. The duration of cultivation on a plot is
typically three to five years but could be as short as one year (Hao et al., 1990; and Kaul,
1991). Slash-and-burn agriculture differs from shifting cultivation in that the farmer does not
return to reuse the plot in the former case (Fearnside, 1993). The fallow period is 10-20 years
(Haoetal., 1990).
Shifting cultivation is a legacy from the neolithic times: the practice dates back 10,000
years. It is a part of the cultural ethos of many tribal communities (Kaul, 1991). Often,
although not always, it is practiced on hill slopes, which are often unsuitable for terraced
farming. If performed on sparsely distributed, small plots with long enough fallow periods, it is
an agronomically sustainable practice.
Estimates of the extent of shifting cultivation, according to Kaul (1991), are mostly
educated guesses. It is thought to be practiced by some 500 million people worldwide (as of
1980) (Lanly, 1982) on some 300-500 mha (million hectares) with an annual clearing of 20-60
mha (Seller and Crutzen, 1980; and Andreae, 1993). Not all standing biomass is burned. As
in savanna burning, larger trees survive many burn cycles. Besides, some useful biomass,
such as bamboo, may be extracted before the first burn. Seiler and Crutzen (1980) estimated
that 0.9 to 2.5 Pg dm are burned every year-75% in tropical secondary forests, and the
remaining in humid savannas. The more recent estimates narrow this range: Hao et al.
(1990) estimate 1.2-1.5 Pg dm are burned, and Crutzen and Andreae (1990) estimate this to
3 Also, there could be soil changes due to burning that could cause changes in soil methane uptake over some
post-bum duration.
Page 4-10
-------
be between 1.1 and 2.2 Pg dm/yr. Crutzen and Andreae (1990) also estimate that perhaps an
equal amount of standing biomass is exposed to the fire but escapes burning.
Because a greater mixture of various biomass types with a wider size distribution are
burned during shifting cultivation than when savannas are burned, higher CH4 emission factors
are believed to be associated with shifting cultivation. Some amount of smoldering must take
place as not everything burns on the first try. There is an uncertainty factor of at least 2 in the
amounts burned, and a comparable level of uncertainty in the emission factors.
Burning of Forests for Permanent Land-Use Change
The permanent conversion of forests to pasture or to cropland begins with the initial
cutting of undergrowth and felling of trees in a dry season. In some areas where biomass is
scarce, such as in India, and the land-use change is for agriculture, all usable biomass is
removed, and the remainder is set on fire after drying (Kaul, 1991). Where pasture is the end
use, tree trunks are left on the ground and often are not consumed in the first burn. Tree
trunks burn completely only after multiple burns. Ranchers in Brazil burn pastures at two- to
three-year intervals to combat the invasion of woody vegetation. Subsequent fires consume
progressively charred logs left over from previous burns, but do not affect charcoal
incorporated in the soil.
Conversion of forests to pasture and agricultural lands has been going on for centuries,
ever since agriculture became prominent. Europe was deforested extensively in the 13th
century to make way for agriculture, and more recently, North America was deforested in the
19th century. Current deforestation is confined mainly to tropical and subtropical regions in
Central and South America, Asia, and Africa.
Population pressure and related demand for agricultural and pastoral land have been
historically the most significant reason for deforestation. Although they are still the most
important reasons, many other reasons characterize modern times: submergence by hydro-
electric reservoirs; construction of electricity transmission lines, roads, industries, and
townships for human settlements; extraction of timber and other forest products for paper
manufacture; etc. For a long time, government incentives promoted deforestation in some
countries.
Because the estimate of amount of land area deforested is subject to much
uncertainty, the estimates of amounts burned are subject to even greater uncertainty. Recent
estimates (see Hao et al., 1990; and Andreae, 1991) put the current extent of forest
conversion at between 4 and 5 mha/yr in primary forests and another 10 mha in secondary
forests. Not all biomass is consumed in the fires. Fearnside (1991) estimates that 28% of the
initial above-ground biomass burns on the first burn in Amazonia, but with three burnings, this
rises to 40%. Crutzen and Andreae (1990) assume that only 40-50% in all is released
through combustion; the rest decays or is decomposed by termites. They estimate that about
1.1-3.1 Pg dm/yr of biomass are exposed to the fire, resulting in burning of 0.4-1.6 Pg dm and
immediate emissions of 0.7-2.6 Pg CO^yr. Results from a recent study coordinated by
Lawrence Berkeley Laboratory (Makundi and Sathaye, 1992) fall within this range. The study
estimates that in seven countries the emissions of carbon dioxide from deforestation burning
are 1.3 Pg CO2/yr. Together these countries account for approximately two-thirds of the
annual prompt emissions from deforestation in developing countries (Makundi and Sathaye,
1992).
Page 4-11
-------
Aircraft observations over Amazonia forest fires show that a substantial portion of
burning is in the smoldering phase (Andreae et al., 1988). The presence of prolonged
smoldering implies higher emissions of CH4 from burning of forests than from burning of
savanna. Logs consumed in reburns are burned primarily through smoldering rather than
flaming (Fearnside, 1992), though it is possible that some smaller charcoal pieces created in
earlier fires would burn in the glowing phase.
To the extent that deforestation rates remain constant, current CH4 emissions from
subsequent burns in areas deforested in previous years will be equal to future releases from
areas being cleared now (Fearnside, 1992). Burning brings about a whole host of other
changes as well. N2O emissions from soils, as well as CH4 uptake by soils, change in the
post-burn period (OECD, 1991). The sudden appearance of huge quantities of dead wood
also increases termite attacks on the material left after a burn, but if they are fungal termites,
they do not produce any more CH4 (OECD, 1991). Little is known about these processes to
allow their quantification in current national budgets.
Prescribed Open Burning in Forests
Prescribed open burning is a method of forest management by which combustible
debris in limited areas is burned intentionally to reduce biomass accumulation and thereby
reduce extensive natural fires. In the last century, many countries have gone from
indiscriminate burning, to aggressive fire suppression, to now a mixture of fire control and fire
use (Pyne, 1990). Generally, these ground fires consume mostly litter, grasses, shrubs, and
undergrowth. Often fire lanes also are burned periodically to keep them free of biomass.
Forestry practices vary from place to place. Besides its use in fire conservancy, open
burning is used to clear weeds and undergrowth before establishing new plantations and for
the propagation and regeneration of certain species, like teak and some varieties of pine
(Kaul, 1991). Besides natural fires and prescribed burns, many wildfires are caused
unintentionally and accidentally by human activity. In boreal forests, especially in Canada,
China, and Russia, emissions from these fires can be large.
Extensive prescribed burning is confined largely to North American temperate and
boreal forests. Undisturbed moist tropical forests do not burn easily, but prescribed burning
also occurs in the subtropical and dry tropical forests in India and Australia. Stocks (1990)
estimates that 8 mha are subjected to natural wildfires in temperate and boreal regions.
Crutzen and Andreae (1990) estimate that 2-3 mha are subjected to prescribed burning;
further, with wildfires, 0.33-0.67 Pg dm/yr are burned. If this amount is apportioned in the ratio
of areas involved, estimates of the amount of biomass burned as dry matter globally in
prescribed burns would lie in the range of 0.06-0.2 Pg dm/yr. Since prescribed fires should be
expected to consume less material than natural wildfires, estimates more close to the lower
end of this range would be attributable to prescribed burns.
Prescribed burning in boreal and temperate forests is an anthropogenic replacement
for what would occur naturally. Its objective is to reduce the total amount of combustible
material available to a future natural fire; therefore, emissions will reduce as well. Thus, over
a period (several decades to a century), CH4 emissions from prescribed burning do not add to-
-and perhaps subtract from-what would have occurred naturally. Therefore, the IPCC-OECD
(OECD, 1991) methodology does not include them in the national inventories, and they are
not discussed further, since the intent of this report is to estimate anthropogenic emissions.
Should one decide to include them, however, the methodology for estimating emissions would
Page 4-12
-------
be similar to that for estimating emissions from shifting .agriculture or deforestation for land-
use change, as described in section 4.3.2 of this chapter. What happens to soil fluxes (N2O
emissions, CH4 sinks) because of these prescribed burns is a more complex and unresolved
question (see Moore, 1992).
Open Burning of Agricultural Residues
Agricultural residues are by-products of agricultural production and associated
processing activities. Traditionally, after a harvest, the field is prepared for the next sowing by
burning the straw and the stubble remaining on the field. Similarly, outside rice mills, husk
may be disposed of by burning.
Agricultural residues have multiple uses and disposal routes. They are used for fuel,
fodder, mulch, fertilizer, building materials, and as raw materials for cottage industries and
handicrafts. The by-products are valued sometimes more than the crop itself (Barnard, 1990).
Traditional farming communities in areas of high population density are often very frugal in
their use of biomass resources. Open burning in the field is generally a last option. Only
when there are no alternate uses, or when it is too difficult to plough the residues back into
the soil, as on non-mechanized farms, is burning resorted to for disposal.4 With residues from
cotton and tobacco plants, there is an increased risk of pest infestation and plant diseases if
the residues are left in the field without being properly incorporated into the ground. In
wetland farming systems with multiple annual croppings (some fields grow three rice crops in
a year) because the next crop is to be sown immediately after the harvest of the previous one,
the advantage of being quick and easy favors burning as the preferred disposal option
(Barnard, 1990). In the case of rice cultivation, incorporation of residues in the soil may
enhance CH4 emissions from anaerobic bacterial decomposition.
Information on the types of residues that different countries burn is difficult to obtain.
We know that rice, sugarcane, cotton, tea, coffee, and tobacco residues are burned.
Sugarcane fields are burned before harvests (Howden et al., 1992a) and after harvests (Joshi,
1991). The use of agricultural residues, such as bagasse, for energy purposes is discussed in
the following section on confined combustion. Similarly, when rice husks are left to smolder
outside factory gates (Barnard, 1990), they should be accounted for here, but when they are
used in specially designed stoves and boilers, they should be accounted for in the section on
confined combustion.
The amount of agricultural residues produced is difficult to quantify because, unlike
crops, residues do not enter markets, and no records are collected. Besides, annual
production is large. Cereal crops produce 0.7 to 2.5 tonnes of straw per tonne of grain, and
approximately 3.1 Pg are produced each year from food crops alone (Barnard, 1990). Of this,
50-60% is produced in developing countries, and 40-50% in the developed countries (Barnard,
1990; and Crutzen and Andreae, 1990). One recent estimate puts the annual amounts
produced in both groups of countries at 1.7 Pg dm each (Crutzen and Andreae, 1990, quoting
Barnard, 1990).
The fraction of residues burned in the fields is also not known precisely. Rice straw
makes up 31% of agricultural residues produced in developing countries, and at least in
4 Even in developed countries, when plowing is not cost-effective, farmers burn fields every year, as in Corvallis
Valley, Oregon, in August-October (Andrasko, 1992).
Page 4-13
-------
Southeast Asia, burning is the preferred method of disposal. Another 11 % is sugarcane
residues (including bagasse), which are also burned. Crutzen and Andreae (1990) estimate
that about 25% of the residues produced in developing countries is burned in the fields,
amounting to 425 Tg dm/yr. About 10% is thought to be disposed off similarly in the
developed countries. Their estimate for total burning of agricultural residues is between 1.1
and 1.8 Pg dm/yr. Other estimates are 1.7-2.1 Pg dm/yr (Seller and Crutzen, 1980) and 2.02
Pg dm/yr (Andreae, 1991).
The uncertainty in estimating CH4 emissions from open burning of agricultural
residues stems from our ignorance of (1) the amounts produced and then disposed of in the
different routes and (2) emission factors. The amounts of residues produced in different
countries will vary because of the differences in crops grown and in management systems.
Agricultural residues being dry and thin can burn rapidly with predominantly flaming
combustion. Methane emission factors, therefore, are believed to be relatively low, just as for
savanna burning, unless strong winds accompany burning. Although no measurements exist,
open burning of rice husks is predominantly smoldering and, therefore, should be associated
with higher emission factors.
Other Miscellaneous Sources
There are several other instances of open burning in developing countries-burning
practices that vary from country to country. In garbage dumps around large cities in
developing countries, smoldering spontaneous combustion and intentional burning by
scavenger communities to recover metals are common sights (Smith and Thorneloe, 1992).
Park wastes (mostly fallen leaves) in some cities are often collected and set on fire. This
practice was also prevalent in the now developed countries until two decades ago, when it
was discontinued to improve local air quality. Very little information exists on the amounts of
park wastes or of urban solid wastes that are burned. Compared with the other sources,
these amounts can be expected to be small, but since the type of burning is predominantly
smoldering, emission factors can be high. Lack of sufficient information precludes their
estimation in this report, but as more information becomes available, these sources, if
significant, should be added to national inventories.
Confined Burning
Direct Burning of Biomass
Just as biofuels have supplied human energy needs for tens of thousands of years
(Smil, 1983), they continue to supply the energy needs of over two billion people even today.
Biofuels-wood, charcoal, agricultural residues, dung, and biogas-are primarily used for
cooking (50%); space heating (30%) is the second most common use. Industrial uses in
boilers, bakeries, brick kilns, foundries, and steel making are common in many countries and
account for the remainder. Biomass-processing industries often use biomass residues (bark,
chips, sawdust, rice husk, bagasse, etc.) for their internal energy needs, and in areas with
abundant biomass resources, electricity also may be produced. Tobacco and coffee curing,
tea drying, beer brewing, fish smoking, pottery, and ceramics are other uses. In Malawi, 23%
of fuelwood consumption goes to cure tobacco, where 17 kg of fuelwood are required to
produce 1 kg of tobacco (Kaale, 1990).
Globally, biomass provides one-seventh of the energy supply (Hall, 1991). In
developing countries, it accounts for over a third, and in the rural areas of developing
Page 4-14
-------
countries, even more. There is some use even in developed countries: 14% in Sweden, and
about 4% in the United States, where biomass-based electric power plant capacity is 9,000
megawatts (Hall, 1991). Because biofuels generally are not traded commercially, published
production statistics underestimate their use. Global estimates based on surveys of average
per-capita biofuel consumption (-1-5 kg/cap/day--RosilIo-Calle and Hall, 1992) range between
1.3 and 2.7 Pg dm/yr (Crutzen and Andreae, 1990). Generally, the consumption in rural areas
is more than that in urban areas, and within rural areas, consumption is greater where
biomass resources are plentiful. The scarcity of higher-quality fuels forces people to burn
lower-quality fuels; in Asia and Africa, some 400 million tonnes of dung are estimated to be
burned annually as fuel (Armitage and Schramm, 1989).
The uncertainty in CH4 emissions from biofuel burning under confined conditions
comes from the uncertainty in per-capita consumption figures and in emission factors for
different biofuel-device combinations. Average consumption varies enormously, depending
upon local scarcity of biofuel and the adaptation of fire-tending practices by cooks to this
scarcity. The fuels vary as well as the devices in which they are burned. In principle,
because of the differences in fuel geometry (thickness, etc.) and the conditions under which
burning takes place, each device-biofuel combination would be associated with a unique
emission factor for CH4 and other products of incomplete combustion. But measurements
exist for only a few instances upon which we must rely for preparing inventories. Surely as
the data on both amounts burned and emission factors improve, emission estimates will
improve. •
Incineration of Municipal Solid Wastes: With the shortage of landfill space close to
major urban centers, many communities have been considering and installing incinerators to
dispose of municipal solid wastes. Often these incinerators are equipped with energy-recovery
units to help defray the costs of operating these facilities. While information on amounts
incinerated may be collected and known, there is no information available on emission factors
from these sources. As a start, default values may be assumed, such as from wood-fired
utility or industrial boilers. As the data base on emission factors improves, this source can be
added to national inventories.
Burning of Processed Biomass
Biomass is not only burned directly, frequently, it is upgraded before being burned,
which results usually in an enhancement of calorific value or of combustion properties, or
both. Figure 4-3 shows that the upgraded biomass fuels can be solid (charcoal, briquettes) or
liquid (methanol, ethanol) or gaseous (producer gas, synthesis gas, or biogas). A systematic
accounting of emissions must estimate CH4 emissions not only during the burning of these
upgraded fuels but also during their conversion processes. Information, albeit scanty,
currently exists on only two of these routes-those of charcoal manufacture and conversion of
biomass to biogas in anaerobic digesters. Estimation of emissions from the charcoal cycle is
discussed in the methodology in the following section. Flecent measurements from one type
of biogas digester in China led Khalil et al. (1990) to conclude that biogas digesters were an
insignificant source of CH4 globally (5-30 Gg/yr). Even if other types of biogas domes are
leakier (Gunnerson and Stuckey, 1986; and Kishore et al., 1987), their numbers and total CH4
production make their contribution to global emissions small. Contributions from other cycles
can be added to the national accounts as more information becomes available and if
emissions are appreciable.
Page 4-15
-------
Charcoal Manufacture and Use: Making charcoal out of wood is a way of increasing
its calorific value (which makes its long-distance transport more economic) and of improving
combustion properties at the point of use. Charcoal burning is not a significant source of CH4
globally because the emission factor is small (Delmas et al., 1991; and Smith et al., 1993).
Charcoal manufacturing, however, carried out in somewhat inefficient kilns in developing
countries has emission factors that are 70 times greater (Delmas et al., 1991) than those from
charcoal burning. With the amount of fuelwood that is converted to charcoal being roughly 0.1
Pg/yr, this source yields significant amounts of CH4 that should be accounted for in national
inventories.
4.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The basic procedure for estimating emissions from biomass burning is similar to the
two-part procedures already discussed in previous chapters for estimating emissions from rice
cultivation or livestock. In each case, one starts with an estimate of amount of the "activity"
(rice acreage, number of animals, or amount of biomass) and multiplies this by an emission
"factor" (emissions per unit acreage, animal, or amount of biomass). To estimate the total
emissions from each of these sectors, one sums over all the cases for which the emission
factors are known to differ~for example, by type of irrigation in rice, by species of animal for
livestock, and by the type of burning activity for biomass. Thus,
Emissions =
x
EF,
(3.2)
where A is the amount of biomass burned, EF is the emission factor, and i denotes the
different categories for which emissions need to be summed. The units used for emission
factors can vary. Besides, on a g/kg basis, there are at least three other common ways of
expressing emission factors: as a ratio of the trace gas to CO2 emitted (see Figure 4-1), as a
relative enhancement of the trace gas in the plume over the ambient atmosphere, and on a
g/MJ basis, which relates emissions to the energy content of the biomass that is burned.
Although this two-part procedure summarizes the basic methodology, several steps and
different routes are involved in estimating each part. As Figure 4-4 shows, there are different
preferred ways of estimating the amounts of biomass burned for different types of burning.
4.3.1 Review of IPCC-OECD Methodology
The most comprehensive methodology for estimating CH4 emissions from biomass
burning is the draft IPCC-OECD methodology (OECD, 1991). While the methodology is simple
and straightforward, the following provides a brief review. There are at least four criteria by
which any methodology can be judged: completeness, consistency, and correctness and
transparency of its assumptions.
As it now stands, the sections on biomass burning are strewn in many different places
in the OECD document. Thus, we find biofuels in the energy chapter, dung burning under
animal wastes, and savanna and agriculture residues treated in a separate section in the
chapter on land-use change. This is explained partly by the primary focus of the IPCC-OECD
methodology being on estimating CO2 emissions, and partly by the existence of many
legitimate alternate ways of organizing and aggregating emissions from different subsectors.
But the unintended consequences of this method of organization are that it is easy to miss
Page 4-16
-------
FIGURE 4-4
ESTIMATING METHANE EMISSIONS FROM BIOMASS BURNING
Savannahs,
Shifting Cultivation
Agricultural Residues
Non-commercial
Confined Biofuel Use
Deforestation
Commercial
Confined Biofuel Use
Biomass Exposed
to Rre Annually
Dry Matter Fraction
• (1- Moisture Fraction)
Note: Italics shows boxes where one starts for different types ofbiomass
-------
some emissions and to double count others, and it is in general difficult to maintain
consistency across different estimating procedures.
Thus, for example, the burning of municipal solid wastes in incinerators or their
smoldering in open dumps in developing countries is accounted for in the chapters on
stationary sources. Excluded are emissions during all intermediate processing of biofuels to
other forms, such as during charcoal manufacture or anaerobic digestion into biogas. Also,
sometimes there is a possibility of double counting, such as fuelwood use for energy purposes
being counted under deforestation in the land-use section, and agricultural residues being
counted once under field burning and again for energy extraction.
The current IPCC-OECD methodology does not account explicitly for all the different
routes by which carbon is transformed during biomass burning. There is incomplete
accounting of carbon in ash, in aerosol particulates, or in the nonmethane volatile organic
compounds (see Figure 4-1). Similarly, as discussed in OECD (1991), there is a possible
double counting of carbon as CO and CH4 as carbon in CO2.
Because the methodology is scattered, amounts left as charcoal are accounted for in
some types of burning (e.g., agricultural residues), acknowledged in some types but ignored in
others (e.g., savanna burning, p. 6-32). Moisture content, too, is recognized explicitly as a
factor for some types of burning (e.g., agricultural residues), and is unacknowledged but
possibly embedded in others (e.g., biofuels, p. 2-47). Some assumptions are implicit in the
methods but are not stated explicitly.
Finally, values recommended for certain default assumptions, to be used in the
absence of country-specific data, are dated. Values for the percent of agricultural residues
burned in developed countries are overestimates, and so are those for combustion
efficiencies. The method also makes no distinction between burning fraction and combustion
efficiencies, between open and confined burning, or between open burning of different types
and uses of the same range of emission factors for all types of burning. It relies on fuelwood
production data for estimating biofuel consumption, whereas at least for the noncommercial
consumption (which constitutes the largest fraction), surveys of per-capita consumption will
yield more reliable estimates (Scurlock and Hall, 1989; and Crutzen and Andreae, 1990).
Some of these problems are recognized and acknowledged in the report but are
relegated to footnotes in the interest of keeping the methodology simple. These problems are
easy to fix (more types of burning can be added, default assumptions can be changed), and
none will make a large difference in estimates compared with the large uncertainties that
currently exist in the amounts burned and in the emission rates. Making default assumptions
explicit, no matter how conjectural, is preferable, however, to ignoring or submerging them.
By forcing research into areas where gaps exist, estimates can be improved over time. The
methodology recommended here (see Figure 4-4) is similar to the one proposed by IPCC-
OECD (OECD, 1991). All types of biomass burning are presented together to force
consistency across methods, so that the same factors are accounted for in all.
4.3.2 Methodology Used in this Chapter
Figure 4-4 summarizes the methodology for estimating CH4 emissions from different
types of biomass burning when the emission factor of CH4 for that particular type of burning is
known as a ratio (to CO2 emitted). The methodology is shown as a series of (multiplication)
Page 4-18
-------
r
products, and the essential difference between the various types of biomass burning is the
way in which the amount of biomass exposed to fire is estimated.
Statistics for amounts of biomass used for large-scale electricity generation are
published and available (middle left, Figure 4-4). For alt other types of biomass burning, the
amounts that are exposed to the fire must be estimated. Different ways have been devised to
make reasonable estimates. Thus, the areas burned annually in savannas and in shifting
cultivation (top left, Figure 4-4) are estimated from a knowledge of the total areas (ha)
involved and the average frequency of burning (#/yr; often <1). For forest areas being burned
permanently for agricultural or pastoral use, repeated burnings, as described by Fearnside
(1992), can be treated in a similar fashion to savanna burning. Areas affected annually
multiplied by the total above-ground biomass density (Mg/ha) yield the amount of biomass
exposed annually to fires.
The estimate of crop residues produced annually in a country is the product of annual
crop production and the average residue to crop ratio (top middle, Figure 4-4). To estimate
what fraction of this is burned in-situ in the fields, amounts used in each of the different
nonburn routes (fodder, fuel, fiber, construction material, raw material for handicrafts, residues
ploughed in or left to decay in the fields, etc.) must be subtracted from the amounts produced.
This calculation is needed for each crop grown in the country.
Estimates of amounts of biofuels-fuelwood, charcoal, crop residues, and dung directly
burned-that do not enter commercial market channels (as in rural areas of developing
countries) are made best by a knowledge of the average consumption per person (top right,
Figure 4-4). This, coupled with a knowledge of the population with that consumption, enables
an estimate to be made of the amounts burned annually. This method is often useful for
estimating consumption in urban areas as well, where although biofuels may be traded,
records are not kept, as most of this trade is in the informal sector of the economy.
The next series of steps after the calculation of biomass exposed to fires aims at the
accounting of carbon that is present in the biomass. The biomass that is exposed to (open)
fires is generally a mixture of living biomass (which can be up to 50% water) and dead
biomass in varying degrees of desiccation (moisture content as low as 10%). The next step
estimates the amount of dry matter that is exposed to the fire (Figure 4-4). Not all biomass
exposed to fire burns. The fraction burned (often termed as burning efficiency) accounts for
the portion of dry biomass exposed to the fire that actually burns. Carbon content is
expressed generally as a fraction (or a percent) of dried biomass. Multiplying this by the
amount of dry matter gives us the amount of carbon that is involved in the fire. While burning
efficiency is defined as the portion of biomass that was exposed that burns, combustion
efficiency accounts for the fraction of carbon in the fire that is oxidized completely to CO2.
The product of this multiplication yields CO2 emitted in g-C, multiplying it by the emissions
ratio gives us an estimate of CH4 emitted in g-C. The final multiplication by the ratio of
molecular weight of CH4 to the atomic weight of carbon gives the estimate of the amount of
CH4 emissions expressed as g-CH4.
As mentioned earlier, emission factors instead of being expressed as ratios are also
sometimes expressed in g CH4/kg dm or g CH4/MJ. It is a simple matter to convert an
emission factor from g CH4/MJ to g CH4/kg dm, with a knowledge of the calorific value of the
biomass that is involved in burning. Multiplying amounts of dry matter burned by emission
factors on a mass basis will yield directly estimates of CH4 emissions. Table 4-2 summarizes
information on emission ratios, arranged in an increasing order by type of burning, that were
Page 4-19
-------
TABLE 4-2
BIOMASS-BURNING METHANE EMISSION FACTORS
TYPE
Wood boilers
Wood stoves
Charcoal burning
Savannas
Agricultural residues
Fuelwood--noncommercial
Dung burning
Shifting cultivation
Deforestation
Charcoal manufacture
Emission Factor Definitions
ESTIMATES OF EMISSION FACTORS (LOW-HIGH)
R=ACH4 /AC02 (%)
-
-
0.14-0.85
0.28-0.66
0.5-0.8
1.40-1.79
-
-
1.10-1.23
6.3-12.06
Emission ratio R is the ratio
of relative enhancement of
CH4 concentration in the
plume of the fire over the
background before the fire
as compared to the
enhancement of CO2.
g CH4/kg dm
-
-
-
0.6-2.2
2.7
4.8-13.9
-
-
5.2-10.8
13.0-21.0
Emission factor is
defined as the
emissions of
grams of CH4 per
kilogram of dry
matter of biomass
burned.
CH4 /C0a (%)
0.02
0.05
-
0.12
-
0.57
-
-
0.73
3.28?
Emission factor is
also sometimes
expressed as the
ratio (or percent)
of the mass of
CH4 emitted in a
fire to that of CO2
emitted.
g CH4-C/g-C (%)
0.04
0.13
0.14
0.33-0.40
0.3?
1.56
1.7
1.2-1.8
1.99
6.2-9.0
This is same as in the
previous column
except that the mass
of gases emitted is
expressed in grams of
carbon instead of the
full molecular weight
basis.
-------
TABLE 4-3
DEFAULT VALUES OF PARAMETERS FOR USE IN CALCULATING METHANE EMISSIONS FROM BIOMASS BURNING
(See Table 4-2 for definition of emission factors.)
Type
Savanna
Australia
Africa
Tropical America
Asia
Shifting Cultivation
India
Rest (Open fallow)
(Closed fallow)
Deforestation
F-7
Rest
Agricultural
Residues
Developing Countries
Developed Countries
Confined Burns
Fuelwood (noncom.)
Agr. Resid. (noncom.)
Dung (noncom.)
Charcoal
Commercial Biomass
Commercial Fuelwood
Large-Scale Biomass
Charcoal Manufacture
Basic data
unit
Burning
Interval
(Years)
Biomass
Density
(T dm/ha)
Areas
Cheng
Delmas
Hao
Hao
2-3
1-3
1-2
2
5.0
6.6?
6.6
4.9
Moisture
Content
(%)
Do not
need, as
biomass
densities
are
reported
in dry-
matter
units
Fraction
Burned
(%)
Carbon
Content
(%)
Combustion
Efficiency
(%)
Range of Emission Factors
R=ACH4/ACO2 (%)
g CH4 /kg dm
80-85
40-46
83-85
0.4
(0.2-0.6)
2.2
(1.1-3.2)
Areas
4.9
Hao
Hao
4-5
Annual
Annual
23-33
38
126
X
40-50
45
80?
'
'
Areas
F-7 project
FAO-1988
Crop
Production
FAO
FAO
Population
X
Res/ Crop
Ratio
0.7-4
0.7-4
p.cap. use
Country or region-
specific data
Amount
FAO
FAO
FAO
114-354
Burned in-
situ (%)
25
2.5?
X
30-40
45
80?
1.2
(0.9-1.5)
6.5
(4.9-9.3)
-10
-10
90-95
90-95
40-48
85-90
85-90
0.5
(0.3-0.7)
2.7
(1.6-3.8)
X
10-20
-10
-10
9
100
45-50
40-48
36-42
85
87
88
85
88
1.2(0.9-1.5)
.
.
(0.14-0.85)
6.5 (4.9-13.9)
-
.
.
X
-10
-10
-
-
45-50
45-50
-
87
90
-
Same as above for fuelwood
Technology-dependent
6.3 (4.0-9.0)
13.0(10.0-20.0)
-------
used in this chapter to estimate emissions from individual countries. In the absence of actual
data, default values for other variables may be picked from the ranges given in Table 4-3 and
used to estimate CH4 emissions from biomass burning. Where direct measurements of
emissions have not been made~e.g., open field burning of agricultural residues-values must
be interpolated from those measured in other settings--e.g., savanna burning.
4.4 RESULTS
Table 4-1 presents recent published estimates of the ranges of the amounts of
biomass burned globally each year. Some categories, such as municipal solid wastes, have
not been estimated, but their inclusion is unlikely to broaden significantly the current range of
5-13 Pg dm/yr. The table also indicates that anthropogenically caused fires account for
approximately 95% of the biomass burned. Similarly, open fires dominate over confined
burns, which account for 35-40% of the biomass burned. If average values in the ranges are
assumed to be true, then confined biofuels use is the single largest category, accounting for
38% of all biomass burned. Natural, prescribed, and other anthropogenic burning of forests
account for about 35%, savanna burning accounts for 21 %, and open field burning of
agricultural residues the remainder.
If the same emission factor were to be used for all types of burning, CH4 emissions
would be roughly proportional to the amount of dry biomass consumed in the fires. However,
emission factors between different types of burning vary by a factor of 15 (Table 4-2).
Therefore, the relative contribution to CH4 emissions from various types of biomass burning is
appreciably different from the relative contribution of amounts burned. Although emission
factor data are not good enough to permit differentiation among countries for any one type of
burning, they are sufficiently reasonable for differentiating among different types of burning.
Table 4-4 shows CH4 emissions from major countries and continents by type of burning. This
table is based on anthropogenic sources only, but excludes emissions from prescribed forest
burning for reasons already mentioned.
Methane emissions from anthropogenic biomass burning are roughly 49 Tg/yr.
Traditional biofuel combustion accounts for 43% of anthropogenic CH4 emissions from
biomass burning. Shifting cultivation in primary and secondary forests accounts for 21 %,
deforestation 17%, and savanna burning only 12%. This result is significantly different from
some earlier studies, which held savanna burning responsible for a majority of trace-gas
emissions.
The same table also shows the breakdown by countries. Most of the anthropogenic
burning is on three continents: Asia (-40%), Africa (-30%), and Latin America (-25%). China
and Brazil are by far the largest emitters. Together with Indonesia, they account for over one-
third of CH4 emissions from biomass burning. Seven countries—including India, Zaire, Nigeria,
and Mexico-account for about half the emissions. Approximately two-thirds of the emissions
are from 16 countries, each of which emits at least 0.7 Tg/yr, as shown in Table 4-4. Ivory
Coast, Mozambique, Tanzania, Zambia, Malaysia, and Venezuela are the remaining other
countries with emissions in excess of 0.5 Tg/yr.
Page 4-22
-------
TABLE 4-4
ESTIMATES OF METHANE EMISSIONS FROM ANTHROPOGENIC BIOMASS BURNING
(Gg CH4/year)
Country/
Region
Zaire
Nigeria
Sudan
Ethiopia
Angola
Other Africa
China
Indonesia
India
Myanmar
Thailand
Viet Nam
Bangladesh
Pakistan
Other Asia
Brazil
Mexico
Colombia
Other L. A.
U.S..
Other N. A.
Australia
Other Ocn.
E. Europe
W. Europe
Total (Tg)
Note: The numbers
Open Burning
Savanna Shifting Permanent
Burning Cultivation Deforestation
569.0
94.6
400.3
226.4
401.0
2447.7
58.8
23.6
20.7
23.2
0.0
4.6
0.0
1.0
45.9
971.8
9.7
26.2
218.7
0.0
0.0
321.0
14.5
0.0
0.0
5.9
in bold indicate
395.0
316.1
168.5
145.3
256.6
2282.7
0.0
486.5
112.0
576.4
25.8
346.3
10.1
0.0
482.0
2230.5
837.6
279.2
1102.3
0.0
0.0
0.0
47.2
0.0
0.0
10.1
the country with
471.0
282.1
124.3
11.7
98.0
914.3
537.5
1290.0
219.0
101.0
216.6
66.4
5.7
5.9
590.0
1425.6
255.4
607.6
1132.3
0.0
0.0
0.0
0.0
0.0
0.0
8.4
the maximum
Agricultural Confined
Residues Bio-fuels
3.3
22.2
5.8
5.7
0.5
68.6
226.7
32.3
189.9
7.1
21.4
10.4
14.3
20.2
43.6
101.9
35.8
10.7
76.1
81.2
10.7
8.4
4.0
72.5
46.5
1.1
emissions
193.5
717.5
179.3
478.6
58.4
2322.9
6437.0
1064.9
1822.6
247.8
343.0
403.6
720.8
702.7
1710.0
684.3 '
391.9
151.5
813.4
401.1
32.0
22.7
31.7
437.0
206.7
20.6
for that burning
Charcoal
Manu-
facture
87.3
143.7
240.0
23.1
17.2
550.9
0.0
12.3
188.4
0.0
272.7
0.0
0.0
0.0
75.8
616.4
12.7
51.3
124.2
48.0
0.0
3.1
3.3
16.5
20.2
2.5
category.
Total
Tg/Yr
1.7
1.6
1.1
0.9
0.8
8.6
7.3
2.9
2.6
1.0
0.9
0.8
0.8
0.7
2.9
6.0
1.5
1.1
3.5
0.5
0.0
0.3
0.1
0.5
0.2
48.5
Page 4-23
-------
4.5 TRENDS
Biomass burning from both natural and anthropogenic sources evidently has affected
the global atmosphere from well before the industrial era. Kammen and Marino (1993) have
estimated that the global source term from biomass burning rose steadily from 4 Tg CH4 /yr in
800 A.D. to 15 Tg CH4/yr in 1800. Since they did not account for all the types of biomass
burning (e.g., burning to drive game from hiding) their numbers could be underestimates.
For more recent times, Bolle et al. (1986) have estimated that CH4 emissions from
biomass burning increased from 49 Tg CH4 /yr in 1940 to 79 Tg CH4 /yr in 1980. Figure 4-2
and more recent evidence from isotopic studies suggest that while gauging the trend, these
numbers might overestimate the true emissions. The IPCC (1992) estimates that CH4
emissions from biomass burning are currently 40 Tg CH4.
Nonetheless, the increase in biomass burning in the last few decades has contributed
significantly to the increase in the burden of atmospheric CH4. Though one paper suggests
that CH4 emissions from biomass burning from the Southern Hemisphere alone are still
increasing at an annual rate of 3.5 Tg CH4 (Stevens, 1993), the recent decline in the rate of
increase of CH4 concentrations (Steele et al., 1992) makes it unlikely that today global
biomass-burning emissions are increasing significantly (unless some other biogenic source is
decreasing even more). Following is a discussion of the current changes in each major
biomass burning subsector.
Savannas
The frequency of savanna burning may be increasing because of growing population
pressure and more intensive use of rangeland (Crutzen and Andreae, 1990). A legitimate
question to ask is, If humans stopped burning savannas, would nature burn as much from
accumulated biomass? It is difficult to give an unambiguous answer now because over the
long term a transition may occur from this inherently fire-prone ecosystem to a different
ecosystem that may have different fire-resistance properties and size distribution.
Shifting Cultivation
Traditionally, the fallow period ranged from 10 to 50 years, but this period has been
declining of late because of population pressures. For example, half a century ago, the
average fallow period between cycles in India was 20-40 years; now the range is 4+ 2 years
(Kaul, 1991). The area under shifting cultivation is not expected to increase much beyond its
present extent because virgin or secondary forests that could be brought under this practice
are becoming scarce (Andreae, 1991). In India, the Forest Act of 1980 forbids any further
conversions of existing forest lands, even to shifting cultivation. Instead, the trend is to
convert existing shifting cultivation lands to permanent agriculture (Andreae, 1991).
Entrenched tribal customs, poor soil conditions, and difficulties of proper soil management in
remote and difficult terrains all make this conversion a slow process.
Deforestation
There has been much concern about the rate of destruction of the world's remaining
forests. Although the rates in such countries as Brazil and India seem to have peaked
recently (Makundi and Sathaye, 1992) due to economic recession and government policies,
the global rates are still higher than what they were in the early 1970s and before.
Page 4-24
-------
Agricultural Residues
OECD countries discourage their farmers from burning crop stubble because of the air
pollution that results. In Australia, this practice is becoming less frequent as low-tillage
farming practices become more common and as residue retention replaces pre-harvest
burning as a management practice (Howden et al., 1992b). Thus, while earlier estimates
using data from the 1960s (OECD, 1991) put the fraction of agricultural residues disposed of
by burning in developed countries at 50% (Seiler and Crutzen, 1980), more recent estimates
put this figure at 10% (Andreae, 1993).
In some parts of developing countries, agricultural residues used to be common
property, available to anyone who had a use for them. With the increase in population and
the reduced availability of alternate energy sources, however, residues are becoming more
privatized (Barnard, 1990). According to one estimate 800 million people, mostly in Asia, rely
on agricultural residues as their principal fuel (Barnard, 1990). So while open field burning of
residues might decrease in developing countries as well, their use as a fuel is likely to
increase. People normally shift down to residues and dung as fuelwood, and other more
desirable fuels become scarcer or less affordable. An early estimate of the fraction of (all)
residues burned was 0.8 (Seiler and Crutzen, 1980); the more recent estimates of 0.6-0.9
encompass this value (Crutzen and Andreae, 1990).
Biofuel Use: On the one hand, economic development in this century has been
accompanied by a movement away from biofuels to higher-quality fuels: from fuelwood, to
charcoal, to kerosene, to natural gas (or LPG), and finally to electricity. But the increases in
oil prices in the 1970s and balance of payment constraints in the 1980s made remote the
hope of a transition to these fuels for a substantial portion of the world's population. In fact, in
many areas, increasing population pressure and decreasing fuelwood supplies have hastened
a regression to even poorer-quality fuels, like agricultural residues, twigs, cattle dung, roots,
and leaves (Smith, 1987). On the other hand, in all the options considered for stabilization of
atmospheric concentrations of greenhouse gases, processed and upgraded biofuels are seen
to play the major role (Lashof and Tirpak, 1990).
Other Changes
The frequency and the intensity of naturally or accidentally caused wildfires in the next
20-30 years could increase significantly. The amount of combustible material is affected
synergistically by a combination of anthropogenic and climatic factors, which include
succession of severe drought years, disease, and pest infestations. Human management
practices include aggressive fire suppression and closely packed plantation densities. Of
course, it is possible that the frequency of droughts could increase with climate change
caused by anthropogenic emissions of greenhouse gases, as well as with changes in rainfall
regimes caused by deforestation. Both the frequency and the intensity of wildfires have been
increasing in California, Canada, and Russia. (In Russia the increase has been due to a
transition from aggressive (fire) suppression to laissez faire policies (GECR, 1992).)
4.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
Most studies of trace-gas emissions from biomass combustion rely on measurements
of gas concentrations in the exhaust plume of the fire and compare these concentrations to
Page 4-25
-------
the background concentrations at that place before the fire. Thus, the relative enhancement
R, often used as an emission ratio for CH4, is defined by:
R = AC#4/ACO2
(3.3)
where ACH4 is the increase in concentration of CH4 in the plume over the background, and
ACO2 is the same for carbon dioxide. This method has been used for forest fires, and in one
Instance to estimate trace-gas emissions from cook stoves in Manila (Smith et al., 1993) (see
Table 4-3). The basic uncertainty in this procedure stems not from the measurements of
gases, which are standard procedures, but from knowing (or accounting for) the differential
dilution of the gases in the plume by the background gases. Also embedded is an
assumption, perhaps a reasonable one, that the plume gases participate in the combustion
process only once. Different sampling protocols also would make measurements done by
different teams incomparable. These comments do not apply to the chamber method of
measuring emission factors (e.g., Lobert et al., 1991), but it is difficult to simulate in the
chamber method the conditions that obtain in open burning in the wild. Clearly, additional
work is required to sort out these matters.
In spite of the formidable difficulties, estimating CH4 emissions from forest burning is
slightly easier than estimating net carbon emissions, because only prompt emissions are of
concern, and because the latter involves the additional uncertain step of estimating CO2
uptake by vegetation and soils. However, there are a few second-order effects that fires might
set in motion that are highly uncertain. It is known that methanotrophic bacteria in soils act as
a CH4 sink. One would expect burning to modify this sink, but neither the magnitude nor the
direction is known. Besides releasing greenhouse gases, open biomass combustion may
perturb soils sufficiently to release additional N2O (Anderson et al., 1988), but it is not clear if
this exceeds the background emissions of N2O from the forests (Goreau and de Mello, 1988).
Also, according to Fearnside (1992), relative to the uncut forest, termite attacks on the
biomass left to decay on the ground increase following the first burn, converting between 0.2
and 0.8% of the carbon to CH4 and the rest to CO2. Not enough is known yet about the long-
term changes in these fluxes to warrant their inclusion in national inventories at this stage
(Moore, 1992).
Two types of methods have been used by scientists thus far to estimate global
emissions from biomass burning. Based on measurements of 813C, an isotope of carbon,
pyrogenic emissions of CH4 have been estimated to be ~11 % of the global source, or about 59
Tg CH4 /yr (Quay et al., 1988). The other type of studies is based on estimates of amounts of
biomass burned in different settings multiplied by emission factors measured in actual,
experimental, and laboratory fires. Estimates from these types of studies fall into two clusters:
one close to the estimate obtained from isotopic studies (e.g., 51 Tg CH4/yr, or -10% of the
global source (Andreae, 1991)) and others significantly lower than this (e.g., 30 ± 15 Tg-CH4
/yr in Crutzen (1991), quoting Crutzen and Andreae (1990)).
Reconciliation among these three sets of numbers will be possible when more detailed
studies are undertaken at the national level. There must be differences in burning practices in
different ecosystems that only detailed local investigations will uncover. With one exception
(Delmas et al., 1991), the studies that estimate CH4 emissions from biomass burning tend to
Page 4-26
-------
use the same emission factor for all types of burning.5 When emissions from charcoal
manufacture are included, however, emission factors can vary by two orders of magnitude.
As Menaut et al. (1991) conclude:
A substantial improvement will be made only when estimations are performed,
for a given year, at a regional scale and then summed up at a larger one. It
seems unrealistic to found such estimates on studies done at a scale larger
than the basic bioclimatic and phytogeographic units of a region.
Beginnings in this direction are being made. Several localized studies have appeared
in the last two years (e.g., Joshi, 1991; Kaul, 1991; Fearnside, 1992; and Menaut et al., 1991).
While more accurate national inventories would help in making accurate estimates of the
global emission source, the task of independently verifying national inventories with the
current technology is very daunting. The propagation of uncertainty in a chain of
multiplications might yield estimates so broad so as to be of little use in national comparisons
(Robinson, 1989). Satellite measurements have been proposed to measure areas burned, but
these have their uncertainties (Robinson, 1991). Since emission factors from confined
biomass burning could be verified in laboratories around the world, they are likely to prove
less contentious than those from rice cultivation. Standing biomass densities may be another
matter.
Chapter 11 of this report discusses technology-based verification schemes. Until such
schemes yield unequivocal results, peer-reviewed and consensus-based default values that
countries feel comfortable with should be used, rather than unenforceable and disputable
mechanisms. More research is required on the extent of the various types of burnings in
different countries, especially in the few largest emitters, and on the ensemble of emissions
from them.
4.7 CONCLUSIONS
Methane emissions from biomass burning are estimated at roughly 49 Tg CH4 /yr.
Even though prescribed forest fires are also anthropogenic, they are excluded from this
estimate for reasons explained in that section. Also excluded for reasons of data
unavailability are emissions from several other categories, such as burning of urban park
wastes, municipal solid wastes, and processed biofuels, except charcoal. Naturally caused
wild forest fires are excluded as well. If these sources were included, the estimate of CH4
from biomass burning would become close to those obtained from isotopic studies~i.e., 55-60
Tg/yr. While no formal uncertainty analyses were performed, other studies indicate that the
variation around this central estimate would be wide, perhaps ± 40%.
Confined burning of biofuels is the largest single category of emissions, amounting to
-20 Tg/yr. It is likely that emissions from this category are overestimated because of reliance
on a uniform default value of per-capita biofuel consumption for all but a few countries. Use
of country-specific consumption based on household surveys (many carried out in the late
1970s and early 1980s) would help reduce this uncertainty. For this category, technology-
specific emission factors for large-scale biomass combustors also have not been considered.
5 This deficiency has been corrected in the more recent studies that have appeared. See, for example, Levine
et al., 1993; Andreae and Warneck, 1993; Hao and Ward, 1993.
Page 4-27
-------
Similarly, emissions from charcoal manufacture may have been underestimated, since they
are sensitive to the assumptions made about the efficiency of the charcoal manufacture
process.
Forty percent of the anthropogenic burning is in Asia. Africa (-30%) and Latin America
(-25%) are also significant emitters. Methane emissions from biomass burning in Asia are
caused largely by biofuel use; in Africa, by savanna burning; and in Latin America, by
deforestation. China and Brazil are the two largest emitting countries. With Indonesia, they
account for over one-third of CH4 emissions from biomass burning. Seven countries, including
India, Zaire, Nigeria, and Mexico, account for about half the emissions.
Forecasting trends in CH4 emissions from biomass burning is a difficult exercise. Each
type of biomass burning is influenced by different factors—among them, population pressure,
the need for agricultural land, economic development, and the availability of alternatives. The
share of biomass in the mix of global energy sources in response to the threat of global
warming also will affect emissions in the future. Though biomass use (mostly of upgraded
fuels) and burning are likely to increase, it should be possible to hold down CH4 emissions as
newer uses become more efficient.
4.8 REFERENCES
Anders, E., W.S. Wolbach, and I. Gilmour. 1991. Major wildfires at the Cretaceous-Tertiary
boundary. In Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic, and
Biospheric Implications. MIT Press, Cambridge, Massachusetts. 485-492.
Anderson, I.C., J.S. Levine, M.A. Poth, and PJ. Riggan. 1988. Enhanced biogenic emissions
of nitric oxide and nitrous oxide following surface biomass burning. J. Geophys. Res.
93(D4):3,893-3,898.
Andrasko, K. 1992. Personal communication.
Andreae, M.O. 1991. Biomass burning: Its history, use, and distribution and its impact on
environmental quality and global climate. In Levine, J.S., ed. Global Biomass Burning:
Atmospheric, Climatic, and Biospheric Implications. MIT Press, Cambridge, Massachusetts. 3-
21.
Andreae, M.O. 1993. The influence of tropical biomass burning on climate and the
atmospheric environment. In Oremland, R.S., ed. The Biogeochemistry of Global Change:
Radiatively Active Trace Gases. Chapman and Hall, New York, New York. In press. Page 5.
Andreae, M.O. 1993. Personal communication.
Andreae, M.O., E.V. Browell, M. Garstang, G.L. Gregory, R.C. Harriss, G.F. Hill, D.J. Jacob,
M.C. Pereira, G.W. Sachse, A.W. Setzer, P.L. Silva Dias, R.W. Talbot, A.L. Torres, and S.C.
Wofsy. 1988; Biomass-burning emissions and associated haze layers over amazonia. J.
Geophys. Res. 93(D2):1,509-1,527.
Andreae, M.O., and P. Warneck. 1993. Global methane emissions from biomass burning
and comparison with other sources. Pure and Applied Chemistry. In press.
Page 4-28
-------
Armitage, J., and G. Schramm. 1989. Managing the supply of and demand for fuelwood in
Africa. In Schramm G., and J.J. Warford, eds. Environmental Management and Economic
Development. The Johns Hopkins Press, Baltimore, Maryland. 139-171.
Barnard, G.W. 1990. Use of agricultural residues as fuel. In Pasztor, J., and L.A.
Kristoferson, eds. Bioenergy and the Environment. Westview Press, Inc., Boulder, Colorado.
85-112.
Bingemer, H.G., and P.J. Crutzen. 1987. The production of methane from solid waste. J.
Geophys. Res. 92:2,181-2,187.
Blake, D.R. 1984. Increasing concentrations of atmospheric methane. Ph. D. dissertation.
University of California at Irvine, Irvine, California.
Bolin, B., E.T. Degens, P. Duvigneaud, and S. Kempe. 1979. The global biogeochemical
carbon cycle. In Bolin, B., E.T. Degens, P. Duvigneaud, and S. Kempe, eds. The Global
Carbon Cycle. SCOPE 13, John Wiley & Sons, Inc., Chichester, United Kingdom. 1-56.
Bolle, H.-J., W. Seller, and B. Bolin. 1986. Other greenhouse gases and aerosols. In Bolin,
B., B.R. Doos, J. Jager, and B. Warrick, eds. The Greenhouse Effect and Climatic Change
and Ecosystems. John Wiley & Sons, Inc., Chichester, United Kingdom. 157-203.
Cicerone, R.J. 1983. Methane in the atmosphere. Presented at The Twelfth International
Conference on the Unity of the Sciences. November 24-27. Chicago, Illinois. 1 -28.
Cicerone, R.J., and R.S. Oremland. 1988. Biogeochemical aspects of atmospheric methane.
Global Biogeochemical Cycles 2(4):229-327.
Cooper, J.A. 1980. Environmental impact of residential wood combustion emissions and its
implications. J. Air Pollution Control Association 30:885-861.
Craig, H., C.C. Chou, J.A. Welhan, C.M. Stevens, and A. Engelkemeir. 1988. The isotopic
composition of methane in polar ice cores. Science 242:1,535-1,538.
Crutzen, P.J. 1983. Atmospheric interactions - homogenous gas reactions of C, N, and S
containing compounds. In Bolin, B., and R.B. Cooke, eds. The Major Biogeochemical Cycles
and Their Interactions. John Wiley & Sons, Inc., New York, New York. 67-112.
Crutzen, P.J. 1987. The role of the tropics in atmospheric chemistry. In Dickinson, R.
Geophysiology of Amazonia. John Wiley & Sons, Inc., New York, New York. 107-130.
Crutzen, P.J. 1991. Methane's sinks and sources. Nature 350:380-381.
Crutzen, P.J., and M.O. Andreae. 1990. Biomass burning in the tropics: Impact on
atmospheric chemistry and biogeochemical cycles. Science 250:1,669-1,678.
Crutzen, P.J., LE. Heidt, J.P. Kransec, W.H. Pollock, and W. Seller. 1979. Biomass burning
as a source of atmospheric gases CO, H2, N2O, NO, CH3, and COS. Nature 282:253-256.
Delmas, R.A. 1993. An overview of present knowledge on methane emissions from biomass
burning. In van Amstel, A.R., ed. Methane and Nitrous Oxide, Methods in National Emissions
Page 4-29
-------
Inventories and Options for Control. Proceedings of an International IPCC Workshop, 3-5
February 1993, Amersfoort, the Netherlands. RIVM, Bilthoven, Netherlands.
Delmas, R.A., A. Marenco, J.P. Tathy, B. Cros, and J.G.R. Baudet. 1991. Sources and sinks
of methane in the African savanna: CH4 emissions from biomass burning. J. Geophys. Res.
96(D4):7,287-7,299.
Ehhalt, D.H., and U. Schimdt. 1978. Sources and sinks of atmospheric methane. Pure &
Applied Geophysics 116:452-464.
Fearnside, P.M. 1991. Greenhouse gas contributions from deforestation in Brazilian
Amazonia. In Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic, and
Biospher/c Implications. MIT Press, Cambridge, Massachusetts. 92-105.
Fearnside, P.M. 1992. Greenhouse gas emissions from deforestation in the Brazilian
Amazon. Review Draft 1.
Fearnside, P.M. 1993. Personal communication.
Fung, I., J. John, J. Lerner, E. Mathews, M. Prather, L.P. Steele, and P.J. Fraser. 1991.
Three-dimensional model synthesis of the global methane cycle. J. Geophys. Res.
96(D7):13,033-13,065.
\
GECR (Global Environment Change Report). 1992. Burning the carbon store: Will wildfire
steal the offsets? IV(16): Arlington, Massachusetts. 28 August 1992.
Goreau, T.J., and W.Z. de Mello. 1988. Tropical deforestation: Some effects on atmospheric
chemistry. Ambio 17(4):275-281.
Gunnerson C.G., and D.C. Stuckey. 1986. Anaerobic Digestion: Principles and Practices for
Biogas Systems. World Bank Technical Paper Number 49, Washington, D.C.
Hall, D.O. 1991. Biomass energy. Energy Policy ~\9(8):7-\ 1-737.
Hao, W.M., M.-H. Liu, and P.J. Crutzen. 1990. Estimates of annual and regional releases of
CO2 and other trace gases in the atmosphere from fires in the tropics, based on the FAO
statistics for the period 1975-1980. In Goldammer, J.G., ed. Fire in the Tropical Biota:
Ecosystem Processes and Global Challenges. Ecological Studies 84. Springer-Verlag, Berlin,
Germany. 440-462.
Hao, W.M., and D. Ward. 1993. Methane production from global biomass burning. Journal
of Geophysical Research. In press.
Howden, S.M., G.M. McKeon, and I.E. Galbally. 1992a. Methane: Other Agricultural
Sources. In Proceedings: AIAS Symposium on Australian Agriculture and the Greenhouse
Effect
Howden, S.M., G.M. McKeon, J.C. Scanlan, J.O. Carter, D.H. White, and I.E. Galbally.
1992b. Managing Pastures in Northern Australia to Minimize Greenhouse Gas Emissions.
Presented at The IPCC/AFOS Workshop, Canberra, Australia.
Page 4-30
-------
IPCC (Intergovernmental Panel on Climate Change). 1990. Climate Change: The IPCC
Scientific Assessment. Cambridge University Press, Cambridge, United Kingdom.
IPCC (Intergovernmental Panel on Climate Change). 1991. Climate Change: The IPCC
Response Strategies. Island Press, Washington, D.C.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
Joshi, V. 1991. Biomass burning in India. In Levine, J.S., ed. Global Biomass Burning:
Atmospheric, Climatic, and Biospheric Implications. MIT Press, Cambridge, Massachusetts.
185-193.
Kaale, B.K. 1990. Traditional fuels. In Pasztor, J., and L.A. Kristoferson, eds. Bioenergy and
the Environment. Westview Press, Inc., Boulder, Colorado. 85-112.
Kammen, D.M., and B.D. Marino. 1993. On the origin and magnitude of pre-industrial
anthropogenic CO2 and CH4 emissions. Chemosphere26(1-4):69-86.
Kaul, O.N. 1991. Forest biomass burning in India. TISGLOW (TERI Information Service on
Global Warming) 2(2):2-17.
Khalil, M.A.K., and R.A. Rasmussen. 1983. Sources, sinks, and seasonal cycles of
atmospheric methane. J. Geophys. Res. 88:5,131-5,144.
Khalil, M.A.K., R.A. Rasmussen, M-X. Wang, and L. Ren. 1990. Emissions of trace gases
from Chinese rice fields and biogas digesters: CH4, N2O, CO, CO2 , chlorocarbons, and
hydrocarbons. Chemosphere 20(1 -2):207-226.
Kishore, V.V.N., P. Raman., and V.V. Ranga Rao. 1987. Fixed Dome Biogas Plants. Tata
Energy Research Institute, New Delhi, India.
Lanly, J.P. 1982. Tropical Forest Resources. FAO (United Nations Food and Agriculture
Organization) Forestry" Paper 30. FAO, Rome, Italy.
Lashof, D.A., and D.A. Tirpak, eds. 1990. Policy Options for Stabilizing Global Climate:
Report to Congress. U.S. Environmental Protection Agency, Washington, D.C.
Levine, J.S., W.R. Gofer III, and J.P. Pinto. 1993. Biomass burning and the production of
methane. In Khalil, M.A.K., ed. Atmospheric Methane. Springer-Verlag, New York, New York.
In press.
Lobert, J.M., D.H. Scharffe, W.H. Hao, T.A. Kuhlbusch, R. Seuwen, P. Warneck, and P.J.
Crutzen. 1991. Experimental evaluation of biomass burning emissions: Nitrogen and carbon-
containing compounds. In Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic,
and Biospheric Implications. MIT Press, Cambridge, Massachusetts. 289-304.
Page 4-31
-------
Makundi, W.R., and J.A. Sathaye. 1992. Carbon Emissions and Sequestration in Tropical
Forests: A Comparative Analysis Using Country Studies. Draft Summary Report of the
Research Network on Tropical Forestry and Global Climate Change. Lawrence Berkeley
Laboratories, Berkeley, California.
Menaut, J.C., L. Abbadie, F. Lavenu, P. Loudjani, and A. Podaire. 1991. Biomass burning in
West African savannas. In Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic,
and Biospheric Implications. MIT Press, Cambridge, Massachusetts. 133-142.
Meyers, S., and G. Leach. 1989. Biomass Fuels in Developing Countries: An Overview.
Lawrence Berkeley Laboratory, Berkeley, California. LBL-27222. July 1989.
Moore, B. III. 1992. Draft, Chapter 6: Emissions from land-use and biomass-burning for
OECD Final Report Estimation of Greenhouse Gas Emissions. 2-43.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France. 256 pp.
Pyne, S.J. 1990. Fire conservancy: The origins of wildland fire protection in British India,
America, and Australia. In Goldammer, J.G., ed. Fire in the Tropical Biota: Ecosystem
Processes and Global Challenges. Ecological Studies 84. Springer-Verlag, Berlin, Germany.
319-336.
Pyne, S.J. 1991. Sky of ash, earth of ash: A brief history of fire in the United States. In
Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic, and Biospheric
Implications. MIT Press, Cambridge, Massachusetts. 504-511.
Quay, P., S.L. King, J.M. Lansdown, and D.O. Wilbur. 1988. Isotopic composition of methane
released from wetlands: Implications for the increase in atmospheric methane. Global
Biogeochemical Cycles 2:385-397.
Quay, P.D., S.L. King, J. Stutsman, D.O. Wilbur, L.P. Steele, I. Fung, R.H. Gammon, T.A.
Brown, G.W. Farwell, P.M. Grootes, and F.H. Schmidt. 1991. Carbon isotopic composition of
atmospheric CH4: Fossil and biomass burning source strengths. Global Biogeochemical
Cyc/es5(1):25-47.
Robinson, J.M. 1989. On uncertainty in the computation of global emissions from biomass
burning. Climatic Change 14:243-262.
Robinson, J.M. 1991. Problems in global fire evaluation: Is remote sensing the solution? In
Levine, J.S., ed. Global Biomass Burning: Atmospheric, Climatic, and Biospheric
Implications. MIT Press, Cambridge, Massachusetts. 67-73.
Rosillo-Calle, F., and D.O. Hall. 1992. Biomass energy, forests and global warming. Energy
Po//cy20(2):124-136.
Scurlock, J.M.O., and D.O. Hall. 1989. The contribution of biomass to global energy use.
Biomass 20:1-7.
Page 4-32
-------
Seller, W.R. 1984. Contribution of biological processes to the global budget of CH4 in the
atmosphere. In Klug, M.J., and C.A. Reddy, eds. Current Perspectives in Microbial Ecology.
American Society of Microbiology. Washington, DC. 468-477.
Seller, W.R., and P.J. Crutzen. 1980. Estimates of gross and net fluxes of carbon between
the biosphere and the atmosphere from biomass burning. Climatic Change 2:207-247.
Seller, W.R., R. Conrad, and D. Scharffe. 1984. Field studies of methane emission from the
termite nests into the atmosphere and measurements of methane uptake by tropical soils. J.
Atmos. Chem. 1:171-186.
Sheppard, J.C., H. Westberg, J.F. Hopper, K. Ganesan, and P. Zimmerman. 1982. Inventory
of global methane sources and their production rates. J. Geophys. Res. 87(C2):1,305-1,312.
Smil, V. 1983. Deforestation in China. Arnbio 12(5):226-231.
Smith, K.R. 1987. Biofuels, Air Pollution and Health: A Global Review. Plenum Publishing
Co. New York, New York.
Smith, K.R., M.A.K. Khalil, R.A. Rasmussen, S.A. Thorneloe, F. Manegdeg, and M. Apte.
1993. Greenhouse gases from biomass and fossil fuel stoves in developing countries: A
Manila pilot study. Chemosphere 26(1-4):479-505.
Smith, K.R., and S. Thorneloe. 1992. Household fuels in developing countries: Global
warming, health, and energy implications. Presented at the EPA Symposium on Greenhouse
Gas Emissions and Mitigation Research. Washington, D.C. 20 August 1992.
Steele, L.P., E.J. Dlugokencky, P.M. Lang, P.P. Tans, R.C. Martin, and K.A. Masarie. 1992.
Slowing down of the global accumulation of atmospheric methane during the 1980s. Nature
358:313-316.
Stevens, C.M. 1993. Record of isotopic abundances in the atmosphere and sources. The
Global Cycle of Atmospheric Methane. NATO ASI Book. Springer-Verlag, Heidelberg,
Germany. In press.
Stevens, C., and A. Engelkemeir. 1988. A stable carbon isotopic composition of methane
from some natural and anthropogenic sources. J. Geophys. Res. 93(D1):725-733.
Stocks, B.J. Quoted in Crutzen and Andreae, 1990.
Wahlen, M., N. Tanaka, R. Henry, B. Deck, J. Zeglen, J.S. Vogel, J. Southon, A. Shemesh, R.
Fairbanks, and W. Broecker. 1989. Carbon-14 in methane sources and in atmospheric
methane: The contribution from fossil carbon. Science 245:286-290.
Ward, D.E. 1990. Factors influencing the emissions of gases and particulate matter from
biomass burning. In Goldammer, J.G., ed. Fire in the Tropical Biota: Ecosystem Processes
and Global Challenges. Ecological Studies 84. Springer-Verlag, Berlin, Germany. 418-436.
Ward, D.E., and C.C. Hardy. 1991. Smoke emissions from wildland fires. Environment
International 17:117-134.
Page 4-33
-------
-------
CHAPTER 5
METHANE EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS
5.1 SUMMARY
Oil and gas systems are an important source of methane emissions, accounting for
about 33 to 68 teragrams (Tg)1 of methane in 1990, or almost 15% of total anthropogenic
methane. Methane is emitted during oil and gas production, processing, transportation, and
consumption. Sources of emissions within oil and gas systems include: emissions during
normal operation, such as chronic leaks or discharges from process vents; emissions during
routine maintenance, such as pipeline repair; emissions during system upsets and accidents;
emissions associated with venting and flaring gas during oil and gas production; and
emissions from incomplete combustion during the use of the oil and gas.
The basis for estimating emissions is weak for most regions at this time. Only a few
detailed studies of emission rates have been performed. Better emission data that take into
account region- and country-specific factors are needed. Currently available information
indicates that gas production and transportation in the former Soviet Union and Eastern
Europe are by far the most important sources, accounting for about 50% of emissions. This is
a region for which relatively little data are available, however, so these emission estimates
must be viewed with caution. The data also indicate that emissions from oil and gas
combustion and crude oil transportation and refining are small, accounting for about 5% of
total emissions from this source. Overall, additional primary data are needed to improve the
estimates of emissions from oil and natural gas systems.
Future trends of emissions from oil and gas systems are ambiguous. Oil and gas
production and consumption are expected to increase, which could lead to increased
emissions. However, emissions from venting and flaring activities (particularly among the
OPEC countries) and emissions from the oil and gas system in the former Soviet Union are
expected to decline. Given the uncertainty in the current emission estimates, the trend in
future emissions also remains uncertain at this time.
5.2 BACKGROUND
5.2.1 Overview of Oil and Natural Gas Systems
Oil and gas systems can be divided into four main parts: oil and gas production; crude
oil transportation, storage, and refining; natural gas processing, transportation, and
distribution; and fuel combustion (Figure 5-1).
Oil and Gas Production
Oil and gas are withdrawn from underground formations using on-shore and off-shore
wells. Oil and gas are frequently withdrawn simultaneously from the same geologic formation,
and then separated. Gathering lines are generally used to bring the crude oil and
1 Teragram = 106 metric tonnes = 1012 grams.
Page 5-1
-------
FIGURE 5-1
STAGES IN THE OIL & NATURAL GAS SYSTEM
Oil and Natural Gas Production
Crude Oil Transportation
Storage and Ref i n i ng
Natural Gas Process ing., Storage
Transmission and Distribution
I
Fuel Combustion
JU
a
Page 5-2
-------
raw gas streams to one or more collection points within a production field. Because natural
gas is about 90% methane (CH4), leaks or venting from these systems result in CH4
emissions. Oil and/or gas are produced in approximately 186 countries worldwide. In 1990,
about 133,000 petajoules (PJ), or 20.2 billion barrels, of oil were produced, and 73,000 PJ, or
71.2 trillion cubic feet (Tcf) of gas were produced globally.2 Most of the world's oil and gas
production is concentrated in several countries (Table 5-1). The top 10 countries account for
69% of oil production and 85% of gas production. These top 10 countries consume 53% of
the global oil production and 72% of the global gas production.
TABLE 5-1
MAJOR OIL- AND GAS-PRODUCING AND OIL-REFINING COUNTRIES
Oil Production: 1990
Country
Former Soviet
Union
United States
Saudi Arabia
Iran
Mexico
China
Venezuela
U.A.E.
Iraq
United Kingdom
Other Countries
Global Total
Production (PJ)
23,722
17,758
14,191
6,672
6,191
5,782
4,840
4,495
4,277
3,848
41,610
133,386
Gas Production: 1990
Country Production (PJ)
Former Soviet
Union
United States
Canada
Netherlands
Algeria
United Kingdom
Indonesia
Norway
Saudi Arabia
Mexico
Other Countries
Global Total
29,071
17,542
3,992
2,541
1,975
1,904
1,579
1,131
1,101
1,011
11,188
73,035
Oil-Refining Capacity
as of Jan. 1, 1990
Country
United States
Former Soviet
Union
Japan
Italy
China
Canada
United Kingdom
France
Mexico
Germany
Other Countries
Global Total
Capacity (PJ) a
39,131
29,631
10,114
6,754
5,300
4,461
4,411
4,382
3,647
3,631
66,930
178,392
a. Estimated by converting from barrels per day to petajoules per year.
Sources: Oil and gas production data from UN (1992). Oil-refining capacity from OGJ (1989).
2 1 PJ (1015 joules) is the energy equivalent of 975 million cubic feet (28 million cubic meters) of natural gas, or
0.019 Tg of natural gas. 1 PJ is the energy equivalent of 151,515 barrels of oil (6 million gallons), or 0.023 Tg of oil.
Page 5-3
-------
Crude Oil Transportation, Storage, and Refining
Crude oil is transported by pipelines and tankers to refineries. Often, the crude oil is
stored in tanks for a period of time. Methane is usually found in the crude oil stream, and
leaks or venting of vapors from these facilities result in methane emissions.
Refineries process crude oil into a variety of hydrocarbon products, such as gasoline
and kerosine. During the refining process, methane is separated from the other hydrocarbons
and may be leaked or vented. Refinery outputs, referred to as "refined products," generally
contain negligible amounts of methane. Consequently, methane emissions are not estimated
for transporting and distributing refined products. Refineries are operated in 102 countries of
the world, with the top 10 countries accounting for 62% of global capacity (Table 5-1).
Natural Gas Processing. Storage, Transportation, and Distribution
Natural gas is processed to recover liquid hydrocarbons, such as butane and propane,
and to prepare the dried gas for transporting to consumers. Most gas is transported through
transmission and distribution pipelines. During periods of low gas demand, the excess natural
gas is injected into underground storage reservoirs before being distributed for consumption.
A small amount of gas is shipped by tanker as liquefied natural gas (LNG). Because only a
small portion of gas is transported as LNG, emissions from LNG facilities are not estimated.
The following are the main processing, transportation, and distribution activities for
natural gas:
Gas-processing plants. Natural gas is usually processed in gas plants to
produce products with specific characteristics. Depending on the composition
of the unprocessed gas, it is dried, and a variety of processes may be used to
remove hydrogen sulfide and most of the heavier hydrocarbons, or condensate,
from the gas. The processed gas is then injected into the natural gas
transmission system, and the heavier hydrocarbons are marketed separately.
Storage and injection/withdrawal facilities. During periods of low gas demand,
natural gas is injected into underground reservoirs for storage. During periods
of high demand, the gas is withdrawn from the reservoir, processed if needed,
and sent into the distribution network. The storage and injection/withdrawal
facilities include a variety of processes and equipment, including compressors,
injection/withdrawal wells, and separators and dehydrators.
Transmission pipelines. Transmission facilities include high-pressure lines that
transport gas from production fields, processing plants, storage facilities, and
other sources of supply over long distances to distribution centers or large-
volume customers. Although transmission lines are usually buried, a variety of
above-ground facilities supports the overall system. These facilities include
metering stations, maintenance facilities, and compressor stations located along
the pipeline routes. Compressor stations, which maintain the pressure in the
pipeline, generally include upstream scrubbers, where the incoming gas is
cleaned of particles and liquids before entering the compressors. Reciprocating
engines and turbines are used to drive the compressors. Compressor stations
normally use pipeline gas to fuel the compressors. They also use the gas to
fuel electric power generators to meet the station's electricity requirements.
Page 5-4
-------
Distribution systems. Distribution pipelines are extensive networks of generally
small-diameter, low-pressure pipelines. Gas enters distribution networks from
transmission systems at "gate stations," where the pressure is reduced for
distribution within cities or towns.
Fuel Combustion
Natural gas and refined oil products are used in internal and external combustion
engines and turbines. During the combustion of refined products, methane is produced and
emitted. Because natural gas is primarily composed of methane, incomplete combustion also
results in some methane emissions. The rate of methane emissions from fuel combustion
varies significantly with engine type, operating conditions, and emission control technologies
used.
5.2.2 Sources of Methane Emissions in Oil and Natural Gas Systems
Emissions from oil and gas systems can be divided into five main types: (1) venting
and flaring, (2) normal operations, (3) routine maintenance, (4) system upsets, and (5)
combustion emissions.
Venting and Flaring
Venting and flaring refers to the disposition of gas that cannot be contained or
otherwise handled.3 Venting and flaring activities release methane because the vented gas
typically has a high methane content and because flares do not always destroy 100% of the
methane in the gas. Usually, these activities are associated with oil- and gas-production
activities. For example, an oil well may produce an amount of gas that is too small to collect
and market. This gas may be re-injected into the underground formation or may be flared if
re-injection is not feasible or economical. If flaring is not feasible, this gas may be vented,
although venting is prohibited in most areas of the world and is undesirable for safety reasons.
Gas-production facilities are generally unlikely to vent or flare gas, although in some cases
flaring or venting may be necessary during periods of equipment maintenance or as the result
of other factors, such as equipment failure.
The combined quantity of gas vented and flared is reported by countries that produce
oil and gas (Barns and Edmonds, 1990). The reliability of the data is questionable in many
cases because vented and flared amounts are not metered and are often an "accounting
balance," whereby total gas use (sales, re-injection, on-site use) is set equal to the total gas
withdrawn, by putting any differences between gas withdrawn and gas used in the category of
vented and flared.
3 Recently, Radian (1992) expanded the definition of "venting and flaring" to include gas releases from any
equipment that is designed to release gas, such as a pneumatically operated device that is powered by a pressurized
gas stream. Similarly, gas released during turbine engine start-up would also be considered a venting emission. For
purposes of this study, planned incidental releases associated with the operation of equipment in the production,
processing, and distribution of natural gas are not considered part of venting and flaring. Using the traditional concept
of venting and flaring, this study considers emissions from venting and flaring associated with the production of oil and
natural gas separately from the emissions from routine equipment venting.
Page 5-5
-------
Normal Operations
Normal operations are the day-to-day operations of a facility absent the occurrence of
abnormal conditions. Facilities emit methane during normal operations due to a wide variety
of operating practices and factors, including:
Emissions from pneumatic devices. Pneumatic devices are commonly used to
regulate and control gas pressures and flows throughout the natural gas
system. These devices rely on pressurized gas as an energy source for their
operation. In most cases, the natural gas stream itself is a suitable supply of
pressurized gas. Most pneumatic devices are designed to release the
pressurized gas they use. These emissions depend on the size, type, and age
of the devices, the frequency of their operation, and the quality of their
maintenance.
"Fugitive" emissions from system components. These emissions are
unintentional and usually continuous releases associated with leaks from the
failure of a seal or the development of a flaw, crack, or hole in a component
designed to contain or convey oil or gas. Connections, valves, flanges,
instruments, and compressor shafts can develop leaks from flawed or worn
seals, while pipelines and storage tanks can develop leaks from cracks or from
corrosion.
Emissions from starting and stopping reciprocating engines and turbines at
compressor stations.
Emissions from process vents, such as still vents on glvcol dehydrators4 and
vents on crude oil tankers and storage tanks. Vapors, including methane, are
emitted from the vents as part of the normal operation of the facilities.
Routine Maintenance
Routine maintenance includes regular and periodic activities performed in the operation
of the facility. These activities may be conducted frequently, such as launching and receiving
scrapers (pigs)5 in a pipeline, or infrequently, such as evacuation of pipes ("blowdown") for
periodic testing or repair. In each case, the required procedures release gas from the affected
equipment. Releases also occur during maintenance of wells ("well workovers") and during
replacement or maintenance of fittings.
System Upsets
System upsets are unplanned events in the system, the most common of which is a
sudden pressure surge resulting from the failure of a pressure regulator. The potential for
unplanned pressure surges is considered during facility design, and facilities are provided with
pressure-relief systems to protect the equipment from damage due to the increased pressure.
* Glycol dehydrators are used to remove water from natural gas through continuous glycol absorption.
8 Scrapers, or "pigs," are devices used for the routine maintenance of pipe, including cleaning, dewaxing, and
monitoring the condition of the pipeline.
Page 5-6
-------
Relief systems vary in design. In some cases, gases released through relief valves
may be collected and transported to a flare for combustion or re-compressed and re-injected
into the system. In these cases, methane emissions associated with pressure-relief events
will be relatively small. In older facilities, relief systems may vent gases directly into the
atmosphere or may send gases to flare systems where complete combustion may not be
achieved.
The frequency of system upsets varies with the facility design and operating practices.
In particular, facilities operating well below capacity are less likely to experience system
upsets and related emissions.
Emissions associated with accidents are also included under the category of upsets.
Occasionally, gas transmission and distribution pipelines are accidentally ruptured by
construction equipment or other activities. These ruptures not only result in methane
emissions; they can be extremely hazardous as well.
Combustion Emissions
Methane emissions related to combustion of oil and gas result, for the most part, from
the incomplete combustion of fuel. In the case of natural gas, methane is a major component
of the uncombusted fuel and is thus emitted to the atmosphere. In the cases in which
methane is not a component of the fuel, methane may be created in the combustion process.
In general, the methane emissions resulting from fuel combustion are much less than those
associated with the production, processing, or transportation of the fuels.
The amount of methane produced per unit of energy delivered from gaseous and liquid
fuels is very sensitive to the combustion technology employed, including the use of emission
control equipment (OECD, 1991). For purposes of this analysis, fuel consumption has been
divided into two source groups:
Stationary Sources. These sources include gas-fired boilers; oil- and gas-fired
electric utility turbines; gas-, gasoline-, and diesel-powered industrial equipment;
large-bore diesel and dual-fuel engines; and gas- and oil-fired residential
heating sources.
Mobile Sources. These sources contribute less methane than stationary
sources. They include highway and off-highway vehicles, aircraft, railway
transportation, and agricultural, industrial, and construction machinery. Nearly
all of this energy use is in liquid fuels, with gaseous fuels playing a very minor
role.
Table 5-2 lists those emission types that are the most important sources within each
segment of the oil and gas industry. Based on available information, the sources listed as
Page 5-7
-------
TABLE 5-2
EMISSIONS FROM OIL AND NATURAL GAS SYSTEMS
Segment
Major Emission Sources
Other Potential Emission Sources
Oil and Gas Production
Oil and Gas Wells
Gathering Lines
Treatment Facilities
Venting and Flaring
Normal Operations: fugitive emissions;
deliberate releases from pneumatic devices and
process vents
Routine Maintenance
System Upsets and Accidents
Crude Oil Transportation, Storage, and Refining
Pipelines
Tankers
Storage Tanks
Refineries
Normal Operations: fugitive emissions;
deliberate releases from process vents at
refineries, during loading and unloading of
tankers and storage tanks
Routine Maintenance
System Upsets and Accidents
Natural Gas Processing, Transportation, and
Distribution
Gas Plants
Underground Storage Reservoirs
Transmission Pipelines
Distribution Pipelines
Normal Operations: fugitive emissions;
deliberate releases from pneumatic devices and
process vents
Routine Maintenance
System Upsets and Accidents
Fuel Combustion
Stationary Sources
Oil: Utility, Industrial, Residential Boilers;
Engines
Gas: Utility, Industrial, Residential Boilers;
Engines
Mobile Sources: Liquid Fuelsa
Light-Duty Passenger Vehicles, Trucks, Buses,
Railways, Ships, Aviation
Combustion Exhaust Emissions
Combustion Exhaust Emissions
Exhaust Emissions
a. A small amount of gaseous fuel is also used in mobile sources.
-------
"major" account for the majority of emissions from each segment. Because data are limited
and there is considerable diversity among oil and gas systems throughout the world, other
potential sources are also listed, some of which may be important contributors to emissions.
5.2.3 Emission Data
Only very limited data are available that describe methane emissions from natural gas
and oil systems. Estimating the emission types defined earlier is complicated by the fact that
emission rates from similar systems vary among regions and countries due to differences in
supporting infrastructure, operating and maintenance practices, and level of technology used.
Because natural gas and oil systems are composed of a complex set of facilities, simple
relationships between emissions and gross descriptors of the systems are not easily
ascertained.
The available published data reviewed to identify emission estimates include: a
detailed consideration of the physical attributes of oil and gas systems; the operation and
maintenance characteristics of key facilities; and country- or region-specific factors that may
influence emission rates. Three major types of data were identified: (1) survey data, (2)
estimates based on reported, unaccounted-for gas, and (3) data from engineering studies and
measurements.
Surveys
Several surveys of system operators have estimated emissions as a portion of
production or throughput. These studies include Alphatania (1989), AGA (1989), and INGAA
(1989). While these surveys provide a basis for identifying the portions of the systems that
operators believe are likely to be major sources of emissions, they are not based on detailed
assessments of emission rates. Consequently, these surveys do not provide an adequate
quantitative basis for estimating methane emissions from oil and natural gas systems.
Estimates Based on Reported Unaccounted-For Gas
Several studies, such as Hitchcock and Wechsler (1972), Abrahamson (1989), and
Cicerone and Oremland (1988), assumed that emissions can be approximated by reported
amounts of "unaccounted-for" gas. Unaccounted-for gas is defined as the difference between
gas production and gas consumption on an annual basis. Like estimates of venting and
flaring, unaccounted-for gas often is used as an accounting convenience to balance company
or national production and consumption records.
The applicability of unaccounted-for gas estimates is very limited because factors other
than emissions account for the majority of the gas listed as unaccounted for, including: meter
inaccuracies, use of gas within the system itself, theft of gas (PG&E, 1990), variations in
temperature and pressure, and differences in billing cycles and accounting procedures
between companies receiving and delivering the gas (INGAA, 1989). Furthermore, because
known releases of gas are not reflected in unaccounted-for gas estimates, such as emissions
from compressor exhaust, the unaccounted-for gas estimates cannot unambiguously be
considered an upper or lower bound on emissions.
Page 5-9
-------
Engineering Studies and Measurements
A few studies are based on detailed engineering and/or field measurement analyses.
Several engineering analyses have considered how actual or model facilities are built and
operated, and extrapolate facility emissions to a system-wide basis. Several measurement
studies have measured emissions from operating facilities or have identified actual leaks and
have extrapolated these measurements to estimate system-wide emissions.
For purposes of this analysis, data from engineering studies and measurements are
the preferred basis for making estimates. Unfortunately, only a few of these types of studies
have been performed, which limits the ability to estimate emissions regionally and globally
from oil and gas systems. Table 5-3 lists the studies identified and the information they
contain. The emission estimates from the studies in the table have been converted to
common units of kilograms of methane emissions per petajoule of energy (kg/PJ). Seven
studies are listed, with emission estimates for portions of the United States. (U.S. EPA, 1993),
Eastern Europe (Rabchuk et al., 1991), and Western Europe (Schneider-Fresenius et al.,
1989; and Norwegian SPCA, 1992). Additionally, Barns and Edmonds (1990) present
estimates based on a global assessment. Representative uncontrolled emission factors for
fuel combustion from OECD (1991) were adopted for developing the fuel-combustion
emissions for non-U.S. regions for which region-specific emission factors were not easily
available. API (1987) estimates for emissions from marine-vessel loading and ballasting
operations were used to estimate global crude oil transportation emissions. Additional studies
of this type are needed to improve the basis for making emission estimates. At this time,
these data are used as the best available for estimating emissions.
5.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
Emissions were estimated as follows:
1. The global oil and gas systems were divided into five regions. Within each
region, the characteristics of the oil and gas system were relatively
homogeneous.
2. For each region, representative emission factors for each emission type within
each segment were selected, with the objective of taking into account the
various system designs and operating practices found in each region.
3. Region-specific activity levels were estimated for use in conjunction with the
emission factors to estimate emissions from each segment in each region.
Total emissions were estimated as the sum of the region-specific estimates. This
method follows the approach discussed in OECD (1991). As recommended, the oil and gas
systems are divided into their major segments, and emissions are estimated for each of the
maih emission types. Although OECD recommends that emissions from oil and gas
combustion be estimated separately, these emissions are combined here. Each step of the
method used is discussed in turn.
Page 5-10
-------
TABLE 5-3
SUMMARY OF EMISSION FACTORS
Data Source
Study Methodology
Emission Factors and Applicability
U.S. EPA (1993)
Compilation of estimates from
detailed engineering analyses and
field measurement studies
Oil and Gas Production:
300 - 5,000 kg/PJ of oil produced
45,900 - 84,200 kg/PJ of gas
produced
3,000 -13,900 kg/PJ of total oil and
gas produced
Emissions from nongas-producing oil wells,
including fugitive emissions and routine
maintenance emissions in the United States.
Emissions from gas production, including fugitive
emissions, dehydrator venting, bleeding from
pneumatic devices, routine maintenance, and
system upsets in the United States.
Venting and flaring emissions from oil and gas
production and fugitive emissions from gas-
producing oil wells in the United States.
Crude Oil Transportation and Refining:
105 - 1,650 kg/PJ of oil refined
Emissions from oil refining and related oil storage
tanks in the United States.
Natural Gas Processing, Transmission, and Distribution:
56,600 -117,700 kg/PJ of gas
consumed
Emissions from gas processing, transmission, and
distribution, including fugitive emissions, dehydrator
venting, bleeding from pneumatic devices, routine
maintenance, and system upsets in the United
States.
Fuel Combustion in Stationary and Mobile Sources
70-130 kg/PJ of fuel oil used for
stationary combustion
900 - 3,500 kg/PJ of gas used for
stationary combustion
17,400 - 31,500 kg/PJ of gas used
for stationary combustion
5,600 -16,900 kg/PJ of oil used for
mobile combustion
Emissions from oil-fired equipment used in the
utility, industrial, residential, commercial, agricultural,
and military sectors in the United States.
Emissions from gas-fired equipment used in the
utility, industrial, residential, commercial, agricultural,
and military sectors in the United States.
Emissions from compressor engine exhaust used in
gas production in the United States.
Emissions from highway vehicles and other mobile
sources in the United States.
(continued)
-------
TABLE 5-3
SUMMARY OF EMISSION FACTORS (Continued)
Data Source
Study Methodology
Emission Factors and Applicability
Rabchuketal. (1991)
Compilation of estimates from
previous measurement studies
and from official data for 1989
Oil and Gas Production:
139,700 - 314,200 kg/PJ of gas produced Emissions from leakages at gas wells,
including routine equipment venting in the
former Soviet Union
Natural Gas Processing, Transmission, and Distribution:
288,100 - 628,500 kg/PJ of gas produced Emissions from leakages at collection
networks, underground storage facilities,
compressor stations, linear part of main
pipelines, and distribution networks in the
former Soviet Union
174,600 - 384,100 kg/PJ of nonresidential Emissions from leakages at industrial plants
gas consumption and power stations
87,300 -192,000 kg/PJ of residential gas Emissions from leakages at residential and
consumption commercial sectors
Schneider-Fresenius et
al. (1989)
Compilation of results from the
Battelle study's 1988 literature
survey
Oil and Gas Production:
14,800 - 27,500 kg/PJ of gas produced
Emissions from gas production and
treatment facilities in Germany
Natural Gas Processing, Transmission, and Distribution:
71,700 -133,100 kg/PJ of gas consumed Emissions from transportation, distribution,
and storage of gas in Germany
Norwegian SPCA (1992) Summary emission estimates for
1989, based on information and
measurements collected from oil
companies and industry
associations
Oil and Gas Production:
1,100 - 3,400 kg/PJ of oil produced
Emissions from cold venting and flaring from
oil activity in Norway
(continued)
-------
TABLE 5-3
SUMMARY OF EMISSION FACTORS (Continued)
Data Source
Study Methodology
Emission Factors and Applicability
Barns and Edmonds
(1990)
Compilation of official reports and
projections on international
emissions
Oil and Gas Production:
96,000 kg/PJ of natural gas production
Eastern Europe and Former Soviet Union
6,300 - 29,700 kg/PJ of gas produced
Other Oil-Exporting Countries
757,600 -1,046,400 kg/PJ of gas produced
Rest of the World
174,600 - 209,500 kg/PJ of gas produced
Emissions from gas production and
separation facilities in the world
Emissions from venting and flaring activity
Emissions from venting and flaring activity
Emissions from venting and flaring activity
OECD (1991)
Compilation of representative
uncontrolled emission factors
drawn from other sources
Fuel Combustion:
Mobile Sources:
31,400 kg/PJ of oil consumed
560,000 kg/PJ of gas consumed
Stationary Sources:
Eastern Europe and Former Soviet Union
10,200 kg/PJ of oil consumed
850 kg/PJ of gas consumed
Western Europe
4,000 kg/PJ of oil consumed
1,100 kg/PJ of gas consumed
Other Oil-Exporting Countries
5,000 kg/PJ of oil consumed
1,100 kg/PJ of gas consumed
Rest of the World
5,200 kg/PJ of oil consumed
700 kg/PJ of gas consumed
Emissions from transport fuel combustion
Emissions from stationary fuel combustion-
electric utility, industry, residential/
commercial, and other (agriculture and
military)
API (1987)
Estimates of hydrocarbon
emissions and evaporative cargo
losses, based on recent ship and
barge emission tests performed
during typical operations
Crude Oil Transportation, Storage, and Refining
745 kg/Pj of oil tankered
Emissions from loading and unloading of oil
tankers globally.
-------
5.3.1 Definition of Regions
Regions must be defined considering the limitations in data on emission factors and
activity levels, but also recognizing the key differences in oil and gas systems that are found
globally. The following five regions were adopted:
United States. The United States is a large producer and importer of oil and is
a large producer of gas. Detailed emission estimates are available for the
United States.
Eastern Europe and Former Soviet Union. Emission rates for this region are
probably much higher than those for other regions, in particular for the gas
system. This region includes the former Soviet Union (which is by far the
largest oil and gas producer in the region), Albania, Bulgaria, the former
Czechoslovakia, Hungary, Poland, Romania, and the former Yugoslavia.
Western Europe. This region is a net importer of oil and gas, and mainly
produces oil and gas offshore. This region includes Austria, Belgium, Denmark,
Faroe Islands, Finland, France, Germany, Gibraltar, Greece, Iceland, Ireland,
Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Sweden,
Switzerland, and the United Kingdom.
Other Oil-Exporting Countries. This region includes the world's other major oil-
producing countries: the 13 OPEC members (Algeria, Gabon, Libya, Nigeria,
Ecuador, Venezuela, Indonesia, Iran, Iraq, Kuwait, Qatar, Saudi Arabia, and the
United Arab Emirates) and Mexico. Generally, these countries produce large
quantities of oil and have limited markets for gas.
Rest of the World. This region includes the remaining countries of Asia, Africa,
the Middle East, Oceania, Latin America, and Canada.
For purposes of defining these regions, countries were aggregated with relatively similar oil
and gas systems. Additional investigation would most likely improve the definition of the'
regions.
5.3.2 Emission Factors
As discussed above, the basis for selecting emission factors is weak because very few
detailed studies of emissions have been performed. Using the information summarized in
Table 5-3, emission factors were selected by industry segment and emission type for each of
the regions. In some cases, data from the United States were used when region-specific
information was not available.
Tables 5-4 through 5-8 list the emission factors selected. Emission factors from U.S.
EPA (1993) were used for the United States (Table 5-4). Key emission factors for Eastern
Europe and the former Soviet Union were taken from Rabchuk et al. (1991) (Table 5-5).
Barns and Edmonds' (1990) estimates were used for emission factors for venting and flaring
for the several regions, including Eastern Europe.
Studies by Schneider-Fresenius et al. (1989) and Norwegian SPCA (1992) were
adopted as representative of emission factors for Western European gas production and
Page 5-14
-------
TABLE 5-4
U.S. EMISSION FACTORS
Emission Types
Emission Factor
(kg/petajoiile)
Source
Oil and Gas Production
Oil
Gas
Oil & Gas
Crude Oil Transportation, Storage, and
Refining
Transportation
Refining
Storage Tanks
Natural Gas Processing, Transport,
and Distribution
Fuel Combustion
Stationary Sources
Oil
Gas: Compressor Exhaust
Gas: Other stationary sources
Mobile Sources
Oil
Gas
300 - 5,000 of Oil Produced
45,900 - 84,200 of Gas Produced
3,000 - 13,900 of Oil & Gas Prod.
745 of Oil Tankered
90-1,400 of Oil Refined
20 - 250 of Oil Refined
56,600 - 117,700 of Gas Consumed
70-130 of Oil Consumed
17,400 - 31,500 of Gas Consumed
900 - 3,500 of Gas Consumed
5,600 - 16,900 of Oil Consumed
Not Estimated
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
API (1987)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
TABLE 5-5
EASTERN EUROPE AND THE FORMER SOVIET UNION - EMISSION FACTORS
Emission Types
Emission Factor
(kg/Petajoule)
Source
Oil and Gas Production
Oil
Gas
Oil & Gas
Crude Oil Transportation, Storage, and
Refining
Transportation
Refining
Storage Tanks
Natural Gas Processing, Transport,
and Distribution
Process., Transport, and Dist.
Nonresidential Consumption
Residential Consumption
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
300 - 5,000 of Oil Produced
139,700 - 314,200 of Gas Produced
6,300 - 29,700 of Gas Produced
745 of Oil Tankered
90- 1,400 of Oil Refined
15-250 of Oil Refined
288,000 - 628,500 of Gas Produced
174,600 - 384,100 of Gas Consumed
87,300 - 192,000 of Gas Consumed
10,200 of Oil Consumed
850 of Gas Consumed
31,400 of Oil Consumed
560,000 of Gas Consumed
U.S. EPA (1993)
Rabchuketal. (1991)
Barns and Edmonds
(1990)
API (1987)
U.S. EPA (1993)
U.S. EPA (1993)
Rabchuketal. (1991)
Rabchuketal. (1991)
Rabchuketal. (1991)
OECD (1991)
OECD (1991)
OECD (1991)
OECD (1991)
Page 5-15
-------
TABLE 5-6
WESTERN EUROPE - EMISSION FACTORS
Emission Types
Emission Factor
(kg/petajoule)
Source
OH and Gas Production
Oil
Gas
Oil & Gas
Crude OH Transportation,
Storage, and Refining
Transportation
Refining
Storage Tanks
Natural Gas Processing,
Transport, and Distribution
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
300 - 5,000 of Oil Produced
14,800 - 27,500 of Gas Produced
1,100 - 3,400 of Oil Produced
745 of Oil Tankered
90-1,400 of Oil Refined
15-250 of Oil Refined
71,700 - 133,100 of Gas Consumption
4,000 of Oil Consumed
1,100 of Gas Consumed
31,400 of Oil Consumed
560,000 of Gas Consumed
U.S. EPA (1993)
Schneider-Fresenius et al.
(1989)
Norwegian SPCA (1992)
API (1987)
U.S. EPA (1993)
U.S. EPA (1993)
Schneider-Fresenius et al.
(1989)
OECD (1991)
OECD (1991)
OECD (1991)
OECD (1991)
Page 5-16
-------
TABLE 5-7
OTHER OIL-EXPORTING COUNTRIES - EMISSION FACTORS
Emission Types
Emission Factor
(kg/petajoule)
Source
Oil and Gas Production
Oil
Gas
Oil & Gas
Crude Oil Transportation, Storage,
and Refining
Transportation
Refining
Storage Tanks
Natural Gas Processing, Transport,
and Distribution
Process., Transport, and Dist.
Nonresidential Consumption
Residential Consumption
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
300 - 5,000 of Oil Produced
45,900 - 96,000 of Gas Produced
757,600 - 1,046,400 of Gas
Produced
745 of Oil Tankered
90- 1,400 of Oil Refined
15-250 of Oil Refined
117,700 of Gas Consumed - 288,100
of Gas Produced
174,600 of Gas Consumed3
87,300 of Gas Consumed3
5,000 of Oil Consumed
1,100 of Gas Consumed
31,400 of Oil Consumed
Not Estimated
U.S. EPA (1993)
U.S. EPA (1993) and Barns
and Edmonds (1990)
Barns and Edmonds (1990)
API (1987)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993) and
Rabchuketal. (1991)
Rabchuketal. (1991)
Rabchuketal. (1991)
OECD (1991)
OECD(1991)
OECD (1991)
a. Used for the high estimate of Natural Gas Processing, Transport, and Distribution emissions.
Page 5-17
-------
TABLE 5-8
REST OF THE WORLD - EMISSION FACTORS
Emission Types
Emission Factor
(kg/petajoule)
Source
Oil and Gas Production
Oil
Gas
Oil & Gas
Crude Oil Transportation, Storage,
and Refining
Transportation
Refining
Storage Tanks
Natural Gas Processing, Transport,
and Distribution
Process., Transport, and Dist.
Nonresidential Consumption
Residential Consumption
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
300 - 5,000 of Oil Produced
45,900 - 96,000 of Gas Produced
174,600 - 209,500 of Gas Produced
745 of Oil Tankered
90-1,400 of Oil Refined
15-250 of Oil Refined
117,700 of Gas Consumed -
288,100 of Gas Produced
174,600 of Gas Consumeda
87,300 of Gas Consumed3
5,200 of Oil Consumed
700 of Gas Consumed
31,400 of Oil Consumed
560,000 of Gas Consumed
U.S. EPA (1993)
U.S. EPA (1993) and Barns
and Edmonds (1990)
Barns and Edmonds (1990)
API (1987)
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993) and Rabchuk
etal. (1991)
Rabchuk et al. (1991)
Rabchuk etal. (1991)
OECD(1991)
OECD (1991)
OECD(1991)
OECD (1991)
a. Used for the high estimate of Natural Gas Processing, Transport, and Distribution emissions.
venting and flaring (Table 5-6). In particular, the venting and flaring estimate is relatively low
for the North Sea production facilities examined. No region-specific data were available for
the Other Oil Exporting Countries (Table 5-7) and the Rest of the World (Table 5-8). Emission
factors in these regions are expected to fall between the relatively low rates found in the
United States and Western Europe (Tables 5-4 and 5-6), where emission controls are the
highest, and the relatively high rates found in Eastern Europe and the former Soviet Union
(Table 5-5), where emission controls are the least stringent. Consequently, a range of
emission factors is applied to these regions.
Emission factors for stationary and mobile combustion for the United States, shown in
Table 5-4, were taken from U.S. EPA (1993). The non-U.S. emission factors for stationary
and mobile combustion shown in Tables 5-5 through 5-8 were developed using data from
OECD (1991) and OECD-IEA (1991). The derivation of these emission factors is presented in
Appendix A.
5.3.3 Activity Levels
Data on the quantity of oil and gas produced, refined, and consumed were collected for
each region for 1990 (Table 5-9). Most of the data were obtained from UN (1992). Data
Page 5-18
-------
TABLE 5-9
ACTIVITY LEVELS FOR OIL AND NATURAL GAS SYSTEMS IN 1990 (petajoules/yr)
Factor Type
Oil Production"
Gas Production"
Oil Consumption
Stationary0
Mobile0
Totaf
Gas Consumption
Stationary0
Mobile0
Non-
Residential"
Residential"
Totaf
Oil Refined
(Throughput)6
United
States
17,758
17,542
9,283
21,460
30,743
18,466
0
11,601
6,865
18,466
30,064
E. Europe
& Former
Soviet
Union
24,454
29,071
11,640
6,795
18,435
26,574
262
21,803
5,033
26,536
24,591
Western
Europe
8,319
7,188
9,927
12,477
22,404
10;317
11
5,397
4,931
10,328
24,525
Other Oil-
Exporting
Countries3
57,410
9,240
6,156
4,198
10,354
6,665
0
6,109
556
6,665
14,511
Rest of the
World
25,445
9,994
19,835
13,846
33,681
9,796
7
8,415
1,389
9,803
38,486
World Total
133,386
73,035
56,841
58,776
115,617
71,818
280
53,325
18,774
72,098
132,176
a. Includes all the OPEC countries and Mexico.
b. Source: UN, 1992.
c. Total UN (1992) energy estimates divided between stationary and mobile sources, based on sectoral
distribution of OECD-IEA (1991).
d. Total UN (1992) energy estimates divided between Nonresidential and Residential, based on sectoral
distribution of OECD-IEA (1991).
e. Using McCaslin (1992) estimates of throughput of oil refined in 1990 adjusted, to be consistent with the UN
(1992) oil production data for 1990.
on oil-refining capacity were used to approximate oil refined in that year. Data on oil tankered
were not available by region. Total oil exported in 1990 was about 31 million barrels per day
(BP Statistical Review, 1992). Since some oil is loaded and unloaded more than once before
reaching its final destination, and given the occurrence of intrastate tankering, this estimate of
total oil exported was adopted as the lower bound for crude oil tankering globally. An upper
bound of total oil production (55.4 million barrels per day, based on the UN (1992) estimate for
1990) was used to reflect the fact that some oil is loaded and unloaded many times before it
is refined. A regional breakout of this emission factor was not attempted.
Page 5-19
-------
5.4 RESULTS
Table 5-10 summarizes the emission estimates by region and industry segment.
These emission estimates are the product of the emission factors and activity levels presentee!
earlier. The total global emissions shown in Table 5-10 range from about 33 to 68 Tg for
1990. This range is comparable to the IPCC (1992) estimate of 30 to 70 Tg of methane
emissions from oil and gas systems, which represents almost 15% of total methane emissions
from anthropogenic sources. Appendix B presents a more detailed estimation of these
emissions for each of the different regions.
The largest major category of emissions is Natural Gas Processing, Transportation,
and Distribution (16 to 39 Tg). Eastern Europe and the former Soviet Union contribute most
to these emissions — about 13 to 28 Tg. The next largest source of emissions is estimated to
be Oil and Gas Production (15 to 27 Tg). The Other Oil Exporting Countries and Eastern
Europe and the former Soviet Union, which have similar emission estimates, are the major
contributors to emissions (7 to 11 Tg and 4 to 10 Tg, respectively). The Other Oil Exporting
Countries are estimated to contribute most to the Venting and Flaring emissions in this
category (7 to 10 Tg of the 9 to 13 Tg total).
Very little information was available for estimating emission factors for Crude Oil
Transportation, Storage, and Refining and for Fuel Combustion. The estimates of emissions
from these sources indicate that they are very minor contributors to the total emission
estimates.
Table 5-11 summarizes the emissions for the key countries estimated to have the
largest emissions. Emissions by country were estimated using the region-specific emission
factors with the individual country's set of activity levels. The top 10 countries shown in Table
5-11 are estimated to account for about 83% of the total global emissions.
The former Soviet Union, the world's largest producer of oil and natural gas, is
estimated to have the highest emissions, ranging from 16 to 36 Tg in 1990, accounting for
about 50% of the total global emissions. Emissions from the United States are estimated to
be the second highest and account for about 8% of the global emissions. Emission estimates
for each of the remaining eight countries are all less than 3 Tg.
5.5 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
Because relatively few detailed emission studies have been conducted, the emission
estimates must be considered very uncertain. The overall magnitude of the emissions is
driven by two key studies:
Rabchuk et al. (1991) report that emissions from gas production and
transportation in the former Soviet Union are high, about 2-4% of total gas
production. Their estimates are based partly on previous measurement studies
and partly on official data on gas losses. Recent visits to this region indicate
that system construction, maintenance, and operations may be consistent with
high emission rates (Craig, 1992). However, a better quantitative evaluation is
needed to validate the current emission estimates.
Page 5-20
-------
TABLE 5-10
GLOBAL EMISSIONS BY REGION AND INDUSTRY SEGMENT (Tg of CH4 in 1990)
Industry Segment
Oil and Gas
Production
Crude Oil
Transportation,
Storage, and
Refining
Natural Gas
Processing,
Transport, and
Distribution
Fuel Combustion
Total
United
States
0.9 - 2.1
<0.05
1.0-2.2
0.5- 1.0
2.4 - 5.3
Former
Soviet
Union and
Eastern
Europe
4.3- 10.1
<0.05
12.6 - 27.6
0.5
17.4 - 38.3
Western
Europe
0.1 - 0.3
<0.05
0.7-1.4
0.4
1.3-2.1
Other Oil-
Exporting
Countries a
7.4- 10.8
<0.03
0.8 - 3.8
0.2
8.4- 14.8
Rest of the World
World Total
2.2 - 3.2 14.9 - 26.5'
<0.1 <0.5
1.2-4.5 16.3-39.4
0.5 1.8-2.1
3.9 - 8.3 33.2 - 68.3
TABLE 5-11
GLOBAL EMISSIONS BY KEY COUNTRIES (Tg of CH4 in 1990)
Country
Former Soviet Union
United States
Algeria
Canada
Indonesia
Saudi Arabia
Mexico
Iran
Romania
Venezuela
All Other Countries
Total Global Emissions
Region
Former Soviet Union and Eastern EEurope
United States
Other Oil-Exporting Countries
Rest of the World
Other Oil-Exporting Countries
Other Oil-Exporting Countries
Other Oil-Exporting Countries
Other Oil-Exporting Countries
Former Soviet Union and Eastern Europe
Other Oil-Exporting Countries
Emissions
16.3-36.0
2.4 - 5.3
1.7-3.0
1.2-2.8
1.3-2.4
1.0-1.8
1.0-1.7
0.9- 1.5
0.6-1.4
0.7- 1.3
5.9- 11.1
33.2 - 68.3
Page 5-21
-------
Barns and Edmonds (1990) report emissions from venting and flaring by region
based on official reports and projections. The emission estimates for the OPEC
countries are relatively high and account for most of the emissions from this
category. The safety concerns associated with venting, and the value of re-
injecting gas into oil reservoirs to maintain reservoir pressures, would tend to
question the high emission estimates. Improved data are needed to resolve
this question.
The adoption of emission factor estimates from U.S. EPA (1993) for various non-U.S. regions
also adds uncertainty to the overall estimates. U.S. oil- and gas-production facilities and
refineries are subject to emission control requirements. The use of U.S. emission factors,
particularly for refining, may underestimate emissions in other regions. Nevertheless, if the
emission factors for oil production and oil refining were increased by a factor of 10 for the
entire world, the estimate of total global emissions would only increase by about 1 -9 Tg for
1990.
Emissions from both stationary and mobile fuel combustion are also very uncertain
because the implications of emission control technologies were not considered. The
estimates, which are based on emission factors from uncontrolled sources, indicate that this
source is unlikely to contribute significantly to total emissions.
5.6 TRENDS
Overall, future emissions could follow expected trends in oil and gas production and
consumption. Pepper et al. (1992) indicate that under current policies, oil and gas production
and consumption are expected to increase by about 0.7-1.7% per year globally over the next
10 to 20 years. If production practices remain unchanged during this period, emissions would
be expected to increase at a similar rate.
Emission factors will most likely decline in the future, however, for several reasons.
Rrst, Barns and Edmonds (1990) report that venting and flaring is declining globally, and in
particular has been declining in the OPEC countries, which are the major source of these
emissions. By marketing or re-injecting associated gas, methane emissions from this source
may decline substantially in the future. Additionally, steps are underway to modernize the oil
and gas infrastructure in the former Soviet Union. By improving the efficiency of this system,
emissions will decline substantially.
Given the potential for emissions to be reduced from two of the major sources, there is
the possibility that emissions will decline over the next 10 to 20 years, even as oil and gas
production and consumption increase. Given the paucity of data on current emissions, the
trend in future emissions must also remain uncertain at this time.
5.7 CONCLUSIONS
• The methane emitted from oil and gas systems in different parts of the world depends
on the industry's size, supporting infrastructure, operating and maintenance practices, and
level of technology. The method used to estimate the global emissions of 33-68 Tg in this
study is consistent with the recommendations in OECD (1991) and is based on developing
emission factors for each of the major segments of the oil and gas systems for distinct regions
Page 5-22
-------
of the world and using published statistics on the activity levels associated with each of these
segments. Lack of sufficient measurement studies makes the emission factors very uncertain.
In particular, improved data are needed for venting and flaring emission factors and emission
rates for the former Soviet Union.
5.8 REFERENCES
Abrahamson, D. 1989. Relative greenhouse effect of fossil fuels and the critical contribution
of methane. Presented to the Oil Heat Task Force, Washington, D.C.
AGA (American Gas Association). 1989. Natural gas transmission and distribution methane
emissions. AGA Engineering Technical Note. AGA, Arlington, Virginia.
Alphatania (The Alphatania Group). 1989. Methane Leakage from Natural Gas Operations.
The Alphatania Group, London, United Kingdom.
API (American Petroleum Institute). 1987. Atmospheric hydrocarbon emissions from marine
vessel transfer operations. API publication 2514A. API, Washington, D.C.
Barns, D.W., and J.A. Edmonds. 1990. An evaluation of the relationship between the
production and use of energy and atmospheric methane emissions. Prepared for the Office of
Energy Research, U.S. Department of Energy, Washington, D.C.
BP (British Petroleum) Statistical Review. 1992. Statistical Review of World Energy. BP,
London, United Kingdom.
Cicerone, R.J., and R.S. Oremland. 1988. Biogeochemical aspects of atmospheric methane.
Global Biogeochemical Cycles 2(4):299-327.
Craig, B. 1992 (December). Personal communication with Bruce Craig, Oil and Gas
Systems, Global Change Division, U.S. Environmental Protection Agency, Washington, D.C.
Hitchcock, D.R., and A.E. Wechsler. 1972 (March). Biological Cycling of Atmospheric Trace
Gases. Prepared for the National Aeronautics and Space Administration, Washington, D.C.
INGAA (Interstate Natural Gas Association of America). 1989. Global Warming and Methane
Loss from Interstate Natural Gas Pipeline. INGAA, Rate and Policy Analysis Department,
Washington, D.C.
IPCC (Intergovernmental Panel on Climatic Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom. 200 pp.
McCaslin, J.C., ed. 1992. International Petroleum Encyclopedia. 1992. vol. 25. PennWell
Publishing Co., Tulsa, Oklahoma.
Norwegian SPCA (State Pollution Control Authority). 1992. Letter on "Methane Emissions
from Oil Activities" from Audun Rosland, SPCA, Oslo, Norway, to Craig D. Ebert, ICF
Incorporated, Washington, D.C.
Page 5-23
-------
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change,
OECD, Paris, France.
OECD-IEA (Organization for Economic Cooperation and Development - International Energy
Agency). 1991. World Energy Statistics and Balances. Paris, France.
OGJ (Oil and Gas Journal). 1989. Worldwide report. Oil and Gas Journal 87(52):41-43.
Pepper, W., J. Leggett, R. Swart, J. Wasson, J. Edmonds, and I. Mintzer. 1992 (May).
Emissions Scenarios for the IPCC: An Update. Prepared for the Intergovernmental Panel on
Climate Change Work Group 1.
PG&E (Pacific Gas & Electric Company). 1990. Unaccounted for Gas Project Summary
Volume. GRI-90/0067.1, PG&E Research & Development, San Ramon, California.
Rabchuk, V.I., N.I. Ilkevich, and Y.D. Kononov. 1991 (November). A Study of Methane
Leakage in the Soviet Natural Gas Supply System. Prepared for the Battelle Pacific Northwest
Laboratory, Siberian Energy Institute, Irkutsk, U.S.S.R.
Radian Corporation. 1992. Venting and Flaring Emissions from Production, Processing, and
Storage in the U.S. Natural Gas Industry. Updated draft report prepared for the U.S.
Environmental Protection Agency and the Gas Research Institute. Austin, Texas.
Schneider-Fresenius, W., R.A. Hintz, U. Hoffmann-Meienbrock, W. Klopffer, and J. Wittekind.
1989. Determination of Methane Emission into the Atmosphere due to Losses in the Natural
Gas Supply System of the Federal Republic of Germany -- Contribution of Methane to the
Global Greenhouse Effect. Battelle-lnstitut, Frankfurt, Germany.
UN (United Nations). 1992. UN Energy Statistics Yearbook 1990. UN, New York, New York.
U.S. EPA (U.S. Environmental Protection Agency). 1993. Anthropogenic Methane Emissions
in the United States - Report to Congress. Global Change Division, Office of Air and
Radiation, Washington, D.C.
Page 5-24
-------
APPENDIX A
DERIVATION OF NON-U.S. EMISSION FACTORS FOR FUEL COMBUSTION
The non-U.S. emission factors for stationary and mobile combustion were developed
using data from OECD (1991) and OECD-IEA (1991). OECD (1991) presents emission
factors by type of combustion process, while OECD-IEA (1991) reports estimates of energy
combustion by five sectors: electric utility, industry, residential/commercial, transport, and
other (agriculture and military).
Emission factors for stationary combustion of oil and gas were developed for each
region using an average of the OECD emission factors that corresponded to the electric utility,
industry, residential/commercial, and other (agriculture arid military sectors). This average
was weighted in terms of the amount of fuel combusted by the different stationary sectors
reported by OECD-IEA. Tables A-1 and A-2 list the data used to estimate the stationary-
source-combustion emission factors. For each region, emissions from the individual stationary
sectors are estimated as the product of the OECD emission factors, with the corresponding
OECD-IEA estimates of fuel combustion. The average stationary emission factor for a region
is estimated by dividing the total emissions from all stationary sectors by the total stationary
fuel combustion within the region.
The OECD emission factors for transport were used as the emission factors for mobile
sources: 560,000 kg of CH4 per PJ of gas consumed and 31,400 kg of CH4 per PJ of oil
consumed in the transport sector.
These emission factors are based on the assumption that emission sources remain
uncontrolled. The combustion emission factors should consider that emissions are influenced
by the level of emission control technology employed. Detailed data on emission controls,
however, were not available. This assumption leads to an overestimation of emissions in
those countries that have implemented emission control requirements.
Page 5-25
-------
TABLE A-1
DERIVATION OF STATIONARY-SOURCE, GAS-COMBUSTION EMISSION FACTORS
Region
Eastern Europe & Former Soviet Union
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Western Europe
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Other Oil-Exporting Countries
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Rest of the World
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Utilities
10,563
100
1,056
1,581
100
158
1,522
100
152
5,054
100
505
Industry
11,240
1,400
15,737
3,816
1,400
5,342
4,587
1,400
6,422
3,361
1,400
4,706
Residential/
Commercial
4,000
1,200
4,800
4,337
1,200
5,204
515
1,200
618
1,366
1,200
1,639
Other (Agriculture &
Military)
771
1,300
1,002
583
1,300
758
41
1,300
54
16
1,300
. 21
Total
26,574
--
22,595
850
10,317
--
11,462
1,100
6,665
--
7,246
1,100
9,797
--
6,871
700
a. OECD-IEA (1991) figures scaled down to UN (1992) consumption estimates.
b. Source: OECD, 1991.
-------
TABLE A-2
DERIVATION OF STATIONARY-SOURCE, OIL-COMBUSTION EMISSION FACTORS
Region
Eastern Europe & Former Soviet Union
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Western Europe
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
i
Average Emission Factor (kg/PJ)
Other Oil-Exporting Countries
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Rest of the World
Consumption (PJ)a
Emission Factor (kg/PJ)b
Emissions (1000 kg)
Average Emission Factor (kg/PJ)
Utilities
3,493
700
2,445
1,787
700
1,251
1,559
700
1,092
4,708
700
3,296
Industry
4,248
2,900
12,321
3,664
2,900
10,626
'
2,681
2,900
7,775
9,132
2,900
26,483
Residential/
Commercial
618
1,600
989
3,791
1,600
6,066
1,284
1,600
2,055
3,863
1,600
6,180
Other (Agriculture
& Military)
3,280
31,400
102,989
685
31,400
21,510
631
31,400
19,825
2,132
31,400
66,947
Total
11,639
-
118,744
10,200
9,927
--
39,453
4,000
6,155
--
30,747
5,000
19,835
--
102,906
5,200
a. OECD-IEA (1991) figures scaled down to UN (1992) consumption estimates.
b. Source: OECD, 1991.
-------
APPENDIX B
EMISSION ESTIMATES BY REGION
The emission estimates for each of the regions were estimated as the product of the
emission factors and activity levels presented in Tables 5-4 to 5-9. The emission estimates
for each individual region by emission type are shown in Tables B-1 to B-5. The global
estimate is shown in Table B-6.
Page 5-28
-------
TABLE B-1
U.S EMISSIONS
Emission Type
Emissions
(million kg)
Description
Oil and Gas Production
Oil
Gas
Oil & Gas
Total
6-89 Emissions from nongas-producing oil wells, including fugitive emissions and routine
maintenance emissions.
805 -1,477 Emissions from gas production, including fugitive emissions, dehydrator venting,
bleeding from pneumatic devices, routine maintenance, and system upsets.
108 - 490 Venting and flaring emissions from oil and gas production and fugitive emissions
from gas-producing oil wells.
918-2,056
Crude Oil Transportation, Storage, and Refining
Transportation
Refining 3 - 42
Storage Tanks 0-8
Total 3 - 49
Natural Gas Processing, Transport, and 1,046 -
Distribution 2,174
Fuel Combustion
Stationary Sources
Oil
Gas
Compressor Exhaust
Other Stationary Sources
Mobile Sources
Oil 121 - 362
Gas 0
Total , 459 - 5,290
Estimated collectively for the world; see text.
Emissions from gas processing, transmission, and distribution, including fugitive
emissions, dehydrator venting, bleeding from pneumatic devices, routine
maintenance, and system upsets.
0.6 -1.2 Emissions from all stationary oil combustion.
321 - 582 Emissions from compressor engine exhaust.
16 - 65 . Emissions from all other stationary sources.
Emissions from all mobile oil combustion.
Emissions from all mobile oil combustion.
Total
2,426-5,290
-------
TABLE B-2
EMISSIONS FROM EASTERN EUROPE AND FORMER SOVIET UNION
(INCLUDES FORMER SOVIET UNION, ALBANIA, BULGARIA, FORMER CZECHOSLOVAKIA, HUNGARY,
POLAND, ROMANIA, AND FORMER YUGOSLAVIA)
Emission Type
Oil and Gas Production
Oil
Gas
Oil & Gas
Total
Crude Oil Transportation, Storage, and Refining
Transportation
Refining
Storage Tanks
Total
Natural Gas Processing, Transport, and Distribution
Processing, Transportation, and Distribution
Nonresidential Consumption
Residential Consumption
Total
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
Total
Total
Emissions
(million kg)
8-123
4,060 - 9,135
183-863
4,251 - 10,121
2-34
0 -6
3-40
8,374- 18,271
3,806 - 8374
439 - 966
12,620-27,611
119
23
213
147
501
17,374 - 38,274
Description
Emissions from nongas-producing oil wells, including fugitive emissions and
routine maintenance emissions.
Emissions from leakages at gas wells, including routine equipment venting.
Emissions from venting and flaring.
Estimated collectively for the world; see text.
Emissions from leakages at underground storage facilities, compressor
stations, linear part of main pipelines, and distribution networks.
Emissions from leakages at industrial plants and power stations.
Emissions from leakages at residential and commercial sectors.
Emissions from all stationary oil combustion.
Emissions from all stationary gas combustion.
Emissions from all mobile oil combustion, assuming no controls.
Emissions from all mobile gas combustion, assuming no controls.
-------
TABLE B-3
EMISSIONS FROM WESTERN EUROPE
(INCLUDES AUSTRIA, BELGIUM, DENMARK, FAROE ISLANDS, FINLAND, FRANCE, GERMANY, GIBRALTAR, GREECE,
ICELAND, IRELAND, ITALY, LUXEMBOURG, MALTA, NETHERLANDS, NORWAY, PORTUGAL, SPAIN, SWEDEN,
SWITZERLAND, AND UNITED KINGDOM)
Emission Type
Oil. and Gas Production
Oil
Gas
Oil & Gas
Total
Crude Oil Transportation, Storage, and Refining
Transportation
Refining
Storage Tanks
Total
Natural Gas Processing, Transport, and
Distribution
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
Total
Total
Emissions
(million kg)
3 -42
107-198
10- 29
119-268
2-34
0- 6
3-40
740- 1,374
39
11
392
6
449
1,310-2,132
Description
Emissions from nongas-producing oil wells, including fugitive emissions and
routine maintenance emissions.
Emissions from gas production and treatment facilities.
Emissions from venting and flaring.
Estimated collectively for the world; see text.
Emissions from transportation, distribution, and storage of gas.
Emissions from all stationary oil combustion.
Emissions from all stationary gas combustion.
Emissions from all mobile oil combustion, assuming no controls.
Emissions from all mobile gas combustion, assuming no controls.
-------
TABLE B-4
EMISSIONS FROM OTHER OIL EXPORTING COUNTRIES
(INCLUDES ALL OPEC COUNTRIES AND MEXICO)
Emission Type
Oil and Gas Production
Oil
Gas
Oil & Gas
Total
Crude Oil Transportation, Storage, and Refining
Transportation
Refining
Storage Tanks
Total
Natural Gas Processing, Transport, and Distribution
Processing, Transportation, and Distribution
Nonresidential Consumption
Residential Consumption
Total
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
Total
Total
Emissions
(million kg)
18-289
424 - 887
7,001 - 9,669
7,443-10,845
1 -20
0-4
1 -24
785 - 2,662
0-1,066
0-49
785 - 3,777
31
7
132
0
170
8,399 - 14,815
Description -
Emissions from nongas-producing oil wells, including fugitive emissions and
routine maintenance emissions.
Emissions from gas production and treatment facilities.
Emissions from venting and flaring.
Estimated collectively for the world; see text.
Emissions from processing, transportation, distribution, and storage of gas.
Emissions from leakages at industrial plants and power stations.
Emissions from leakages at residential and commercial sectors.
Emissions from all stationary oil combustion.
Emissions from all stationary gas combustion.
Emissions from all mobile oil combustion, assuming no controls.
Emissions from all mobile gas combustion, assuming no controls.
-------
TABLE B-5
EMISSIONS FROM THE REST OF THE WORLD
(INCLUDES ALL OTHER COUNTRIES OF ASIA, AFRICA, MIDDLE EAST, OCEANIA, AND LATIN AMERICA)
Emission Type
Oil and Gas Production
Oil
Gas
Oil & Gas
Total
Crude Oil Transportation, Storage, and Refining
Transportation
Refining
Storage Tanks
Total
Natural Gas Processing, Transport, and
Distribution
Processing, Transportation, and Distribution
Nonresidential Consumption
Residential Consumption
Total
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
Total
Total
Emissions
(million kg)
8-128
459 - 956
1,745-2,094
2,211 -3,181
3-53
1-10
4-63
1,154-2,879
0- 1,469
0-121
1,154-4,468
103
7
435
4
548
3,918 - 8,262
Description
Emissions from nongas-producing oil wells, including fugitive emissions and
routine maintenance emissions.
Emissions from gas production and treatment facilities.
Emissions from venting and flaring.
Estimated collectively for the world; see text.
Emissions from processing, transportation, distribution, and storage of gas.
Emissions from leakages at industrial plants and power stations.
Emissions from leakages at residential and commercial sectors.
Emissions from all stationary oil combustion.
Emissions from all stationary gas combustion.
Emissions from all mobile oil combustion, assuming no controls.
Emissions from all mobile gas combustion, assuming no controls.
-------
TABLE B-6
WORLD EMISSIONS
Emission Type
Emissions
(million kg)
Description
Oil and Gas Production
Oil
Gas
Oil & Gas
Total
Crude Oil Transportation, Storage, and Refining
Transportation
Refining
Storage Tanks
Total
Natural Gas Processing, Transport, and
Distribution
Fuel Combustion
Stationary Sources
Oil
Gas
Mobile Sources
Oil
Gas
Total
42 - 671
5,854 -12,657
9.045-13.144
14,941 - 26,472
56- 99
11 -183
2-33
70-316
16,345-39,405
292 - 293
65-114
1,293-1,534
157
1,806-2,097
Emissions from nongas-producing oil wells, including fugitive emissions
and routine maintenance emissions.
Emissions from gas production and treatment facilities.
Emissions from venting and flaring.
Emissions from processing, transportation, distribution, and storage of
gas.
Emissions from all stationary oil combustion.
Emissions from all stationary gas combustion.
Emissions from all mobile oil combustion, assuming no controls (except
for the United States).
Emissions from all mobile gas combustion, assuming no controls
(except for the United States).
Total
33,162 - 68,290
-------
CHAPTER 6
METHANE EMISSIONS FROM THE COAL FUEL CYCLE
6.1 SUMMARY
In 1990, the coal fuel cycle emitted an estimated 24-40 teragrams (Tg)1 of methane to
the atmosphere globally, which represented approximately 10% of total anthropogenic
methane emissions.2 An additional 1.3 Tg of methane, or approximately 3-5% of total
emissions, was recovered by coal mining operations and used as fuel instead of being vented
to the atmosphere.
Underground coal mining was the primary source of these emissions, accounting for
70-85% of total emissions (20-28 Tg). Surface mines and post-mining coal transportation and
handling operations were estimated to contribute 10-20% (3-8 Tg). Coal combustion was
estimated to contribute the remaining 5-10% of emissions (1-4 Tg). Abandoned mines and
coal waste piles are also believed to release small quantities of methane, although estimates
of these sources are not possible due to the lack of information on their numbers and
emission levels.
Almost three-quarters of global methane emissions from the coal fuel cycle was
emitted by three countries: the People's Republic of China, the former Soviet Union, and the
United States. These countries are also the world's largest coal-producing and -consuming
countries. Lesser amounts of methane were emitted by Australia, the former Czechoslovakia,
Germany, India, Poland, South Africa, and the United Kingdom. Together, the world's 10
largest coal producers were responsible for 90% of global emissions associated with the coal
cycle.
A variety of methods can be used to estimate methane emissions from the coal fuel
cycle, depending upon the availability of data. For mining, the simplest (but most uncertain)
approach is to estimate emissions by multiplying global average methane emission factors for
underground and surface mining and post-mining activities by the relevant coal production
levels. For coal combustion, emissions can be estimated using emission factors for various
coal end uses and multiplying by coal consumption. The ranges resulting from these methods
are large, due to uncertainties about the appropriate emission factors. In those countries
where more specific data are available, however, these emission factors have been refined to
prepare the U.S. EPA's "Best Estimates" of emissions.
For higher certainty, emissions can be estimated for specific coal basins or coal-
consumption sectors using more refined emission factors. If information is available on coal
and mining characteristics, it is also possible to develop models to estimate mining emissions
by relating various coal characteristics to methane emissions. Such models have been
developed based on analysis of coal data from the United States, but cannot be applied with
1 Teragram = 106 metric tonnes = 1012 grams.
2 Coal is mined for use as an energy source and for metallurgical purposes. Although this chapter is entitled
"Methane Emissions from the Coal Fuel Cycle" (to indicate that it encompasses emissions associated with mining,
as well as transportation and combustion), emissions due to mining of coal for metallurgical purposes are also
included.
Page 6-1
-------
any accuracy to other countries. Where such data are available, however, such models could
be developed for other countries.
Ultimately, the most detailed and reliable method available for estimating mining
emissions is to collect and analyze actual emission data from coal mines. These data are
currently only available for underground coal mines, where measurements have been made
for safety reasons. In some countries, however, these data may be unavailable or incomplete,
and the simpler (but more uncertain) estimation methods will be the most appropriate.
Measurement programs are underway in the United States and Australia (and potentially other
countries) to quantify emissions from surface mines, and it may be possible to develop
additional data of this type in the future for the preparation of more detailed emission
estimates.
Significant uncertainties are associated with some of the simpler estimation
methodologies because of the absence of measured emission data for many parts of the coal
fuel cycle and the lack of internationally available data on coal characteristics and emission
levels for key countries. In general, underground mining emissions are the best understood
and have the greatest data availability because of the necessity to monitor methane levels in
mines for safety reasons. In many countries, however, data on methane emissions from
underground mine ventilation and degasification systems may not be collected or published.
Methane emissions from surface mines and post-mining activities (i.e., coal transportation,
storage, and handling) are currently highly uncertain because of the lack of emission
measurements. Detailed emission factors for coal combustion have been developed for the
major coal end uses, particularly for developed countries. The applicability of these factors to
other countries is uncertain, particularly in the developing world, where combustion efficiencies
are much lower. No emission estimates have been prepared for abandoned mines or coal
waste piles due to the lack of necessary information. Abandoned mines and coal waste piles,
however, are not considered to be a significant source of emissions.
In the future, methane emissions associated with the coal fuel cycle are expected to
increase, due to higher levels of coal production and consumption and the use of coal from
more gassy mines in many countries. The rate of increase in methane emissions from the
coal cycle is uncertain due to questions about future coal-production and -consumption levels
and difficulty predicting future coal characteristics, including mine gassiness. In 1990, world
coal production was over 4.7 billion tonnes; future coal production could be significantly
higher, perhaps reaching 5.3-5.9 billion tonnes in 2000 and 5.9-6.8 billion tonnes in 2010 (U.S.
DOE/EIA, 1992b). In addition, the proportion of underground mining could increase as
surface-mined coal seams are depleted. At forecasted production levels, emissions could
range from 27 to 50 Tg in 2000 and 30 to 58 Tg in 2010, assuming methane emission factors
remain constant. If mine gassiness increases as a result of the mining of deeper, gassier coal
seams, or if the proportion of underground mining increases relative to surface mining,
however, emissions could be even higher than these projections.
Increased methane emissions are particularly likely in countries like China, where coal
production is forecast to grow by more than 2% annually between 1988 and 2000, and growth
is expected to continue beyond 2000 as well. China may also need to mine deeper and
gassier coal seams to maintain coal production, and may thus experience an increase in
emissions per tonne of coal mined.
In the future, the use of mine degasification systems may increase to cope with
increasing mine gassiness. To the extent that the additional recovered methane is used, the
Page 6-2
-------
growth in emissions would be moderated somewhat. In many countries, some of the methane
released by underground mines is currently collected by mine degasification systems instead
of being diluted in mine ventilation air. This methane typically has a concentration of at least
35% methane in air and can be used as fuel for many purposes. Some countries do not use
any of this fuel, while others use most of the methane they recover. In 1990, for example, a
reported 1.3 Tg of coalbed methane was used globally from the countries of Australia, China,
the former Czechoslovakia, Germany, Poland, the former Soviet Union, the United Kingdom,
and the United States. Some methane is also recovered and used in Japan, and perhaps in
other countries, but the amounts are unknown. The amount of methane likely to be recovered
by mine degasification systems and the extent to which methane utilization could increase in
the future are uncertain. There is, however, the potential for significant reductions in
emissions from coal mines.
6.2 BACKGROUND
This section briefly describes the processes of methane (CH4) generation and emission
from coal, the key components of the coal fuel cycle, arid the results of some previous studies
to quantify CH4 emissions from this source.
6.2.1 Processes of Methane Generation and Emission in Coal
Methane, which is the principal constituent of natural gas, is contained in coal seams
and is released to the atmosphere by coal mining and other parts of the coal fuel cycle.
Methane is formed during the process of coalification, where organic matter is converted into
coal over millions of years. Some CH4 remains trapped in coal and surrounding strata until a
reduction in pressure (caused by mining or natural faulting) enables it to escape into mine
workings or to the atmosphere.
The amount of CH4 contained in a coal seam is heavily dependent on the geological
characteristics and depositional history of the coalbed. Such factors as coal rank, depth, and
permeability affect the amount and distribution of CH4 in the coalbed and surrounding strata,
which in turn determine the quantity and rate of CH4 release during mining and at other parts
of the coal cycle.3 In general, higher-ranked coals and coals buried at greater depths tend to
contain more CH4 than shallow or low-ranked coals, up to a point of saturation. The unique
features of each coal basin (such as its geological history) can make generalizations difficult,
however.
The amount of CH4 emitted during coal mining is primarily a function of coal
characteristics (i.e., rank, gas content, and depth) and mining method. One of the most
important features of CH4 emissions from coal mines is that emissions tend to exceed the
amount of CH4 that would be predicted based solely on the gas content of the mined coal.
This is because CH4 can be emitted not only from the mined coal, but also from coal left
behind in the mine, by surrounding coal strata, and, in some cases, by other surrounding gas-
bearing strata. Based on research by the U.S. Bureau of Mines, it appears that the emission
3 Coal rank is a measure of the degree of coalification; as coalification proceeds, coals are transformed from
lignite, through the ranks of subbituminous to bituminous and, under certain conditions, to anthracite and graphite.
Depth is related to the depth of burial of the coal. Permeability represents the ability of methane to flow through the
coal matrix.
Page 6-3
-------
factors could be six to nine times the CH4 content of the underground-mined coal seams
(Kissell et al., 1973). Numerous other studies have observed similar relationships in coals
mined throughout the world (BCTSRE, 1992; Pilcher et al., 1991; and Bibler et al., 1992).
Similarly, preliminary research on surface mines indicates that emissions could exceed the
CH4 content of the mined coal by as much as five times (Kirchgessner et al., 1993).
6.2.2 Key Emission Processes of the Coal Fuel Cycle
Underground mining is the principal source of CH4 emissions associated with the coal
fuel cycle. In addition to underground mining, CH4 is emitted by surface mines, post-mining
activities, and coal combustion. In general, underground mines release more CH4 than
surface mines because the coals tend to be of higher rank and are more deeply buried, which
means that more of the CH4 generated during coalification remains trapped in the coal seam
and surrounding strata (Kim, 1977). Post-mining activities include emissions during coal
transportation, storage, and handling, covering all activities from the time the coal leaves the
mine until combustion. Emissions from coal combustion occur as a result of incomplete or
inefficient coal combustion at the point of end use. A variety of other potential CH4 emission
sources are associated with the coal cycle, including abandoned mines and coal waste piles;
however, little is known about emissions from these sources.
Underground Mining
In underground mines, CH4 can be liberated from two sources: (1) ventilation shafts
and (2) degasification systems. Ventilation is used in all underground mines to maintain safe
mining conditions because CH4 is explosive in concentrations of 5-15% in air. In most
countries, ventilation air must contain CH4 concentrations of less than 1% in air. Although its
concentration is very low, the CH4 in this air constitutes most of global underground-mining
emissions because of the enormous volumes used to ventilate mines.
Degasification systems are typically employed only at those mines where additional
CH4 recovery is needed to supplement ventilation for safety reasons. These systems can be
drilled in advance of mining or into fractured gob areas, which are created after mining occurs
and the roof collapses. Degasification systems remove CH4 that would otherwise have to be
diluted in the ventilation air, and they produce CH4 in concentrations that can be used for fuel.
Certain degasification systems, for example, can produce CH4 in concentrations of more than
95%, which can be injected directly into conventional natural gas pipelines. At gassy coal
mines, degasification systems can account for a significant fraction of total CH4 emissions
(U.S. EPA, 1993; Pilcher et al., 1991; Bibler et al., 1992; and Marshall et al., 1993).
Surface Mining
In surface mines, exposed coal faces and surfaces, as well as areas of coal rubble
created by blasting operations, are believed to be the major sources of CH4. As in
underground mines, however, emissions may come from underlying strata, which may be
fractured and distressed due to removal of the overburden, or in some cases from gas-bearing
strata in the overburden, which is fragmented during the mining process. Because surface-
mined coals are less deeply buried and are often of lower rank, they do not tend to contain as
much CH4 as underground-mined coals. Thus, emissions per tonne of coal mined are
generally much lower for surface mines. Due to the large amount of coal produced from
surface mines, however, emissions from this source are significant.
Page 6-4
-------
Post-Mining Activities
Emissions from post-mining coal storage, transportation, and handling operations result
from the natural desorption of CH4 from the coal after it has left the mine and before
combustion. The amount of CH4 emitted depends upon specific coal characteristics, such as
CH4 content, permeability, and desorption rates. It is believed that underground-mined coals
emit more CH4 during post-mining activities because they tend to have higher gas contents
than surface-mined coals. In addition, these coals tend to have lower permeability, which
means that CH4 is emitted from the coal at a slower rate so that a larger proportion remains in
the coal and can subsequently be emitted after the coal has left the mine.
Coal Combustion
Some CH4 is most likely released during coal utilization as a result of incomplete or
inefficient combustion. The amount of CH4 emitted will depend on the combustion unit. In
general, well-maintained utility boilers are believed to release small amounts of CH4, whereas
poorly maintained or aging industrial or residential boilers and stoves are believed to have
higher emission rates.
Coal Mine Waste Piles
Methane may be emitted from the piles of waste rock and coal that are produced
during mining. There are currently no emission measurements for this source. Emissions are
believed to be low, however, because much of the CH4 would most likely be emitted during
mining, and the waste rock would have a low gas content compared with the coal being
mined.
Abandoned Mines
Emissions from abandoned mines may come from unsealed shafts and from vents
installed to prevent the buildup of CH4 in mines. Except in limited circumstances, CH4
emissions from abandoned mines are believed to be small. There is very little information on
the number of abandoned mines, however, and comprehensive data are currently unavailable
on emissions from these mines.
6.2.3 Previous Methane Emission Studies
As shown in Table 6-1, numerous estimates of CH4 emissions from coal mining have
been prepared. Koyama (1964) is credited with the first coal mining emissions study, in which
he estimated that 20 Tg of CH4 were released annually from coal mining. In preparing this
estimate, Koyama multiplied an assumed emission factor of 17.7 m3 per tonne of coal mined
by world hard coal production. His estimate used 1960 coal production data; since
Page 6-5
-------
TABLE 6-1
EMISSION ESTIMATES FROM SELECTED STUDIES
Study Author
Emission
Estimate
(Tg)
Year of
Estimate
Methodological
Issues
Koyama(1963) 20
Hitchcock & Wechsier (1972) 8 - 28
Seller (1984) 30
Crutzen (1987) 34
Okken & Kram (1989) 15 - 45
Zimmermeyer (1991) 24
Selzer&Zittel(1990) 23
Bams & Edmonds for 25
U.S. DOE (1990)
Boyer et al. for 33 - 64
U.S. EPA (1990)
Kirchgessner(1993) 47
CIAB (1992) 24
Present study 24 - 40
1960 Hard coal only; no emission estimates for surface
mining or post-mining activities. 1960 coal
production data.
1967 Post-mining not included. Source of emission
factors, particularly low end, unspecified. 1967
coal production data.
1975 Based on Koyama, with 1975 coal production
data.
n/a Source of emission factors unclear. Hard coal
only; no emission estimates for surface mining or
post-mining activities.
n/a Source of emission factors unclear. Hard coal
only; no emission estimates for surface mining or
post-mining activities.
n/a Only underground mining considered; no
emissions estimates for surface mines or post-
mining activities.
n/a Adjusted Zimmermeyer by: (1) including surface
mines; (2) assuming that 15% of underground
mining emissions (3.6 Tg) not emitted to the
atmosphere due to CH4 utilization.
1986 Assumed mathematical relationship between coal
rank and depth and that in-situ CH4 content was
equal to the mining emission factor.
1988 Statistical approach related CH4 emissions to in-
situ CH4 content. Correlation based on U.S. data
only. Large uncertainty in application of results
for global estimates.
1988 Statistical approach related coal characteristics to
in-situ gas content and in-situ gas content to
emissions. Correlation based on U.S. data only.
No uncertainty analysis conducted.
1990 Country-specific data used where available.
Emission factor assumptions tend to be
conservative, compared to present study. No
uncertainty analysis.
1990 Country-specific data used where available to
prepare best estimates. Global average emission
factors used for rest of countries. Only study to
include emission estimate for coal combustion.
Page 6-6
-------
1960, coal production has increased by more than 75%. In addition, it appears that his study
considered only emissions from hard coal production.4
Several studies followed Koyama's, and many used similar methodologies with more
recent coal production data (Hitchcock and Wechsler, 1972; Seller, 1984; Crutzen, 1987;
Okken and Krarri, 1989; Zimmermeyer, 1991; Selzer and Zittel, 1990; and Barns and
Edmonds, 1990). These studies resulted in emission estimates ranging from 8 Tg to 45 Tg,
depending on the emission factors assumed, the year for which coal production data were
obtained, and whether the estimate was prepared for hard coal alone or hard and brown coal.
For the most part, the studies estimated emissions from hard coal only and did not include
surface mining, post-mining activities, or coal combustion.
Unlike most previous studies, the Boyer et al. (1990) study estimated emissions from
underground and surface mines and post-mining activities and provided country-specific
emission estimates. The estimates were developed using available data on CH4 contents and
emissions from 59 U.S. underground mines to develop a regression equation. First, CH4
emissions from underground mining, surface mining, and post-mining activities for the United
States were estimated using the available data, which were obtained from a relatively small
number of mines. Methane emission factors for underground mining and surface mining were
then developed by dividing estimated CH4 emissions by U.S. coal production. The resulting
emission factors ~ 27.1 m3/tonne of coal mined in underground mines and 2.5 m3/tonne for
surface mines ~ were multiplied by coal production for other countries to estimate their
emissions.
Using this methodology, Boyer et al. estimated global CH4 emissions to be 47 Tg in
1987. The uncertainty range on this estimate was 33-64 Tg, reflecting the uncejjainty of the
regression with which the U.S. emissions were estimated and the uncertainty associated with
applying U.S. factors internationally.5
In 1991, the emission factors developed by Boyer et al. were adopted by the
Organization of Economic Cooperation and Development (OECD) for use as a simple
estimation method for global CH4 emissions from coal mining (OECD, 1991). Since
publication of the Boyer et al. study, however, a number of country-specific studies have been
published, and a great deal of additional data have become available. As a result, the OECD
is revising its methodology, which will be published in 1993. The latest recommendations by
the OECD regarding methods of estimating CH4 emissions from coal mining are presented in
the next section.
The Boyer et al. approach has also been refined as part of a project undertaken by
U.S. EPA's Office of Research and Development (Kirchgessner et al., 1993). Like Boyer et
al., Kirchgessner et al. used regression analysis to develop statistical relationships between
4 For general emission studies, coal is sometimes classified as one of two types: hard coal or brown coal.
Hard coal includes anthracite, bituminous, and some categories of subbituminous coal. It is frequently assumed
that hard coal is produced in underground mines, although there are exceptions. Brown coal is generally assumed
to be produced in surface mines and includes the lower-ranked subbituminous coals and lignite.
6 The regression equation developed in Boyer et al. (1990) had an r2 value of 0.35. The uncertainty associated
with applying this equation to estimate U.S. emissions was thus determined to be ± 23%. An additional factor of +.
10% was used to reflect the uncertainty associated with applying the U.S. data to estimate global emissions. In the
absence of detailed emission data from other countries, this factor was chosen arbitrarily.
Page 6-7
-------
selected coal characteristics and CH4 content and emissions. The Kirchgessner et al.
equations, while also developed using U.S. data, are more detailed than Boyer et al.'s and
require data on depth, moisture content, heating value, and fuel ratio (fixed carbon to volatile
matter). The equations were developed based on U.S. data. Kirchgessner et al. then
collected data as available for various countries and used it in the equations to develop a
global emission estimate of 46 Tg in 1989. No uncertainty analysis has been conducted,
however, and the validity of the data used in estimating emissions for the other countries has
not been assessed.
A final recent CH4 emissions study has been prepared by the Coal Industry Advisory
Board (ClAB, 1992), for the Global Climate Committee. This study presents a CH4 emission
estimate of 24 Tg in 1990, and includes estimates from underground and surface mines and
post-mining activities. It does not include an estimate of CH4 emissions from coal combustion.
Country-specific emission data were used where available for underground mining, and global
emission factors were developed to estimate emissions from surface mines and post-mining
activities. In general, the emission factors used in this study, as well as some of the country-
specific data used, are considered conservative. Thus, the ClAB estimate is consistent with
the lower emissions bound developed in the present study.
6.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
Methane emission estimates should be developed for the four principal sources of CH4
emissions: underground mines, surface mines, post-mining activities, and coal combustion, as
shown in Figure 6-1. Because different types and levels of data are likely to be available in
different countries, this study recommends use of a "tiered" approach for estimating
emissions.6 The first tier requires basic and readily available data and can be used to
estimate emissions for all coal-producing countries within a relatively large uncertainty band.
Higher tiers require additional data and can only be applied if the data are available from the
countries in question. The selection among the tiers will depend upon the quality of the data
available in the countries for which the estimates are being prepared.
In this study, emission estimates are prepared for most of the world's coal-producing
countries using the most basic, "first-tier," approach. In addition, more detailed, higher-tier,
estimates have been prepared wherever possible. The coal production and consumption data
used in preparing these estimates are shown in Table 6-2.
6.3.1 Underground Mining
Methane emissions from underground mines should include estimated emissions from
ventilation systems and degasification systems (if any of a country's mines uses degasification
systems to supplement ventilation). This section presents methods of estimating emissions
from both of these sources.
Three tiers are available for estimating emissions from underground mining, with the
choice among them depending upon the availability of data and the degree to which coal
mining is considered a significant source of emissions by particular countries. The first-tier
8 This approach will also be adopted by the OECD in its revised estimation methodology, due to be published
in early 1994.
Page 6-8
-------
FIGURE 6-1
FLOW CHART FOR ESTIMATING METHANE EMISSIONS FROM THE COAL CYCLE
I. Estimate
Emissions from
Underground
Mines
Correct for Any
Methane Used
Instead of
Emitted
Estimated \.
I Emissions for
I Underground J
Nv Mines ./
II. Estimate
Emissions from
Surface
Mines
,/ Estimated
^ I Emissions from
~~ I Surface
N. Mines
III. Estimate
Emissions from
Post-Mining
Activities
Estimated
Emissions from
Post-Mining
Activities
IV. Estimate
Emissions from
CqaJ
Combustion
s^ Estimatedl\.
^ [ Emissions from 1
^"1 Coal J
^v Combustion/'^
Total Emissions
for the
Coal Cycle >^
Page 6-9
-------
TABLE 6-2
1990 COAL PRODUCTION AND CONSUMPTION DATA
(million tonnes)
Country
China
United States
Former Soviet Union
Germany
India
Poland
Australia
South Africa
Former
Czechoslovakia
United Kingdom
Canada
Greece
Turkey
Spain
Colombia
France
New Zealand
Austria
Belgium/Luxembourg
Italy
Norway
Ireland
Subtotal
World Total
Sources: IEA, 1990,
Coal
Underground
1,023.6
385.1
392.8
76.6
109.4
153.5
51.8
111.7
22.0
75.3
3.9
0
7.5
15.2
0
11.4
0.6
0.6
1.0
0
0.3
0
2,442.5
1992a, and 1992b.
Production
Surface
42.7
548.4
308.6
358.9
128.6
58.1
153.9
62.8
84.7
14.0
64.4
51.9
39.1
20.6
20.5
1.4
2.0
1.9
0
1.0
0
0
1963.5
Coal Consumption
Total
1 ,066.3
933.6
701.4
435.5
238.0
211.6
205.7
174.5
106.7
89.3
68.3
51.9
46.6
35.8
20.5
12.8
2.6
2.4
1.0
1.0
0.3
0
4,406.0
4,740.0
Utility
272.0
700.1
364.4
146.7
122.3
114.5
81.3
76.2
71.6
84.4
42.2
50.5
30.4
39.5
1.4
13.2
0.2
3.5
4.5
11.8
0
2.0
2,232.7
2,467.5
Industry
575.6
104.5
325.1
44.7
95.8
72.8
3.5
56.6
33.8
17.9
6.8
2.3
11.4
7.0
3.8
15.8
1.8
2.9
11.7
10.5
0.8
05
1,405.6
1,666.6
Residential/
Commercial
167.0
6.1
0
0.4
1.2
0
0.2
1.8
1.4
5.7
0.1
0
10.4
0.6
0.3
1.6
0.2
0.3
0.7
0.1
0
0.8
198.9
222.0
Total
1,014.6
810.7
671.5
191.8
219.3
187.3
84.9
134.7
106.8
108.0
49.1
52.9
52.2
47.1
5.5
30.7
2.2
6.7
16.8
22.5
0.8
3.3
3,837.2
4,356.1
Page 6-10
-------
approach -called the "Global-Average Method" - uses a predetermined range of emission
factors (based on experience in a number of countries) to estimate emissions. The most
complex, third-tier, approach -- called the "Mine-Specific Method" -- develops emission
estimates using detailed emission data for most, if not all, of a country's underground coal
mines. In between these two methods is an intermediate, second-tier, approach - called the
"Basin- or Country-Average Method" - in which more limited information, including either
measurements from a subset of mines or statistical or other types of models, can be used to
refine the range of possible emission factors presented in the Global Average Method. Each
of these approaches is described in more detail below.
Tier 1: Global-Average Method
The simplest method for estimating CH4 emissions is to multiply underground coal
production by a factor or range of factors representing global-average emissions from
underground mining, including emissions from both ventilation and degasification systems.
This method is useful for those countries where total coal production from underground mines
is available but more detailed data on mining emissions, geological conditions, coal
characteristics, and the like are not. The emission estimates generated using this method
should be presented as a range to reflect the high degree of uncertainty associated with it.
The Tier 1 method is presented as equation 6.1.
In the draft IPCC/OECD (OECD, 1991) methodology, a single emission factor of 27.1
m3 of CH4 per tonne of coal mined was recommended for all underground mining, based on
the Boyer et al. (1990) study. This factor included emissions from both mining and post-
mining activities associated with underground coal production.7 Based on more recent studies
and additional country-specific emission data, the OECD is revising this emission factor to
reflect some additional issues. First, use of a range of emission factors is suggested to reflect
the large variation possible in CH4 emissions from underground mines in different coal basins
and countries. Second, this emission factor should represent only those emissions associated
with underground mining (ventilation and degasification systems); post-mining emissions
should be handled separately.
In this study, revised global average emission factors of 10-25 m3/tonne of coal mined
are recommended (not including post-mining activities). This range reflects the findings of
various country studies, as shown in Table 6-3, and also has been adopted by the OECD for
its revised emission methodology (OECD, 1993). As more detailed emission data are
published by various countries, the factors can be further revised, if necessary.
7 Due to a mistake in the draft OECD report, this factor appeared to exclude emissions from mine
degasification systems and post-mining activities. However, this emission factor did include these two additional
emission sources.
Page 6-11
-------
r
• « """ " • • """""• • -MI « EQUATION'S 1 ' '' '
TIER 1: GLOBAL AVERAGE METHbD~ UtfOERGRQUND'MiNES"
Low CH4 Emissions = Low CH4 Emission Factor "'
(tonnes) (m3 CH4/tonne of coal mined)
^. Underground Coal Production
(tonnes)
x Conversion Factor ' '
High CH4 Emissions = High CH4 Emission Factor
(tonnes) " (m3 CH4/tonne of coal mined)
x Underground Coal Production'
(tonnes) ''
; x Conversion Factor
> J " O *
Where; '
Low CH4 Emission Factor =10 ms/tonne,
High CH4 Emission Factor = 25 nrYtonne.
Conversion Factor converts the volume of CH4 to a weight measure,
. density of methane at 20°C and ^ atm, which is: '
0.671 tonnes/103m3.
Tier 2: Country- or Basin-Average Method
The suggested Tier 2 approach ~ called the "Country- or Basin-Average Method" -
can be used to refine the range of emission factors used for underground mining by
incorporating additional country- or basin-specific information. Basically, this method can be
used where limited amounts of additional data are available to develop more appropriate, and
most likely narrower, ranges of emission factors for underground mining. For many countries,
it is expected that this range will fall within the global average emission factor range of 1025
m3/tonne. The range of possible emission factors is not constrained under the Tier 2
approach, however, and in some countries the underground mining emission factors may lie
outside the global average emission factor range.
Page 6-12
-------
TABLE 6-3
ESTIMATED UNDERGROUND MINING EMISSION FACTORS
FOR SELECTED COUNTRIES
Country
Emission Factor
(m3/tonne)
Source
Former Soviet Union 17.8-21.7
United States 11.0-15.3
Germany (East and West) 22.4
United Kingdom 15.3
Poland 5.9 - 10.5
Former Czechoslovakia 23.9
Australia 15.6 - 22.3
Canada 23.1
Marshall et al., 1993;
and Airuni, 1992
U.S. EPA, 1993
Zimmermeyer, 1991
BCTSRE, 1992
Pilcher et al., 1991
Bibleretal. 1992
Lama, 1991; Saghafi
and Williams, 1992
Jaques, 1992
The best means of developing country- or basin -specif ic emission factors is to examine
measurement data from a limited number of underground coal mines to approximate what a
reasonable narrower range of emission factors might be.8 These measurement data can be
used to develop simple emission models based on physical principles or to select a narrower
range based on judgment. Among the key types of data that should be considered in model
development are:
the gas content of the coal, which contributes to the total amount of CH4
available for emission during mining;
other potentially relevant coal characteristic data, such as depth, moisture,
heating value, and fuel ratio;
the frequency of coal within the strata above and below the mined coal seam,
which also contributes to the total amount of CH4 available for emission during
mining; and
the method of mining, which determines the amount of ground that is disturbed
by mining the coal and the extent to which the CH4 contained in the mined coal
seam and in the coal seams in the surrounding strata is liberated during mining.
This approach was used by Boyer et al. (1990) and, more recently, by Kirchgessner et
al. (1993) to estimate CH4 emissions from coal mining. As mentioned previously, Boyer et al.
used data from 59 U.S. mines to relate the gas content of the mined coal to CH4 emissions
from mining. Kirchgessner et al.'s approach used more detailed coal characteristics data to
estimate emissions.
8 If measurement data are available for most or all of a country's underground coal mines, the Tier 3 approach
called the "Mine-Specific Method" - should be used to estimate emissions.
Page 6-13
-------
Models based on coal characteristics and other data will be most accurate if applied on
the coal-basin level. Because coal characteristics, mining methods, and CH4 emissions can
vary significantly within coal basins and particularly within different regions of a country,
developing general national models will significantly increase the uncertainty of the emission
estimates.
Tier 3: Mine-Specific Method
Because CH4 is a serious safety hazard in underground mines, many countries have
collected data on CH4 emissions from mine-ventilation systems, and some also collect data on
CH4 emissions from mine-degasification systems. Where such data are available, the Tier 3
approach ~ called the "Mine-Specific Method" ~ should provide the most accurate estimate of
CH4 emissions from underground mines. Since these data have been collected for safety, not
environmental, reasons, hov/ever, it is necessary to ensure that they account for total
emissions from coal mines. The key issues that should be considered when using mine
safety data, as well as the recommendations for resolving them, are summarized as follows:
Where and how are ventilation system emissions monitored?
When used to develop overall CH4 emission estimates, ventilation air monitors
should be placed at the point where ventilation air exhausts to the atmosphere.
If ventilation emissions are not monitored at the point of exhaust, emission data
should be corrected based on estimated additional CH4 emissions between the
point of measurement and the point of exhaust to the atmosphere.
Are ventilation system emissions monitored and/or reported for all mines?
In some countries, emissions are only reported for "gassy mines." Estimates
should be developed for nongassy mines as well. Estimates can be prepared
using information about the definitions of gassy and nongassy mines and data
on the total number of mines and the coal production at these mines.
Are CH4 emissions from degasification systems reported?
Some countries collect and report CH4 emissions from ventilation and
degasification systems, while others only report ventilation system emissions.
Both emission sources must be included in emission estimates. If
degasification system emissions are not included in reported data, those mines
with degasification systems should be identified, and estimates should be
prepared on emissions from their degasification systems. Emission estimates
can be based on knowledge about the efficiency of the degasification system in
use at the mine or the average efficiency of degasification in the country.
6.3.2 Treatment of Methane Utilization
All of the methods described above, with the possible exception of the Mine-Specific
Method, assume that all of the CH4 liberated by mining will be emitted to the atmosphere. In
many countries, however, some of the CH4 recovered by mine degasification systems is used
as fuel instead of being emitted. Wherever possible, the emission estimates should be
corrected for the amount of CH4 that is used as fuel by subtracting this amount from total
estimated emissions.
Page 6-14
-------
f
In several countries, data on the disposition of CH4 recovered by degasification
systems (i.e., whether it is used or emitted to the atmosphere) have been obtained from the
coal industry or energy ministries. Poland, for example, reports that its mine degasification
systems recovered 286 million m3 of CH4 in 1989, of which 201 million m3 was used and the
remaining 85 million m3 was emitted to the atmosphere (Polish Central Mining Institute, 1990).
Regardless of the method used to develop the emission estimates, the Polish emission
estimate should be adjusted to reflect the use of CH4 by subtracting 201 m3 from total
emissions.
6.3.3 Surface Mining
Two possible approaches for estimating CH4 emissions from surface mining are
recommended, depending on data availability. These approaches are similar to those
developed for underground mining, but the results will be much more uncertain due to the
absence of emission data. If emission measurements are developed in the future, it should
be possible to refine these estimation methodologies.
Tier 1: Global-Average Method
As in the methodology for underground mining, the simplest (Tier 1) approach for
surface mines -- called the "Global Average Method" -- is to multiply surface coal production
by a range of emission factors representing global average emissions, as shown in equation
6.2.
In the draft IPCC/OECD methodology, an average emission factor of 2.5 m3/tonne was
recommended (OECD, 1991), based on the results of Boyer et al. (1990).9 Based on more
recent analyses and additional studies, a revised emission factor range of 0.3 to 2.0 m3/tonne
has been developed. This revised range has been adopted by the OECD (1993). The low
end of the range is consistent with the average in-situ CH4 content for surface-mined coals
reported by various recent studies (Saghafi and Williams, 1992; CIAB, 1992; and Jaques,
1992), and it assumes that no CH4 is emitted from the strata surrounding the coal. The
assumption that surrounding strata do not contribute to emissions from surface mines has
been made by some recent studies, including the BCTSRE: (1992) study and the CIAB (1992)
study. The high end of the range is based on an average in-situ gas content of 0.4 m3/tonne
(which is the U.S. average) and assumes that emissions are increased by a factor of five due
to the contribution of surrounding strata. The factor of five is consistent with the results of
Kirchgessner et al. (1993) at a U.S. surface mine. Other studies have similarly assumed that
surrounding strata contribute to emissions, including Jaques (1992), which assumed a factor
of three for Canada.
Given the lack of information and measurements on CH4 emissions from surface
mines, this range must be considered extremely uncertain, and it should be refined in the
future as more data become available.
9 In the draft OECD report, surface mining emission factor appears to exclude emissions from post-mining
activities. In fact, the factor of 2.5 m3/tonne includes both direct emissions from surface mining and those from
post-mining activities.
Page 6-15
-------
EQUATION 6,2
TIER 1: GLOBAL AVERAGE*METHOD - SURFACE bUNES
Low CH4 Emissions = Low CH4 Emission Factor ' ,
(tonnes) (m3 CH4/tonne of coal mined)
x Surface Coal Production
(tonnes)
x Conversion Factor
fc High CH4 Emissions = High CH4 Emission Factor
*' (tonnes) (ma CH4/tonne of coal mined)
i x Surface Coal Production
: (tonnes)
; x Conversion Factor
Where:
Low CH4 Emission Factor = 0.3 nrf/tonne.
High CH4 Emission Factor » 2.0 nfVtonne. < „ t
Conyersipn Factor converts the volume of CH4 to a weight measure based on the
density of methane at 20°C and ^ atm, which isi • -
0.671 tonnes/103m3. , ;
Tier 2: Country- or Basin-Specific Method
A second-tier estimation of CH4 emissions - called the "Country or Basin Specific
Method" -- can be used if additional information is available on in-situ CH4 content and other
characteristics of a country's surface-mined coals. Depending on the degree of detail desired,
emissions can be estimated for specific coal basins or countries, using equation 6.3.
In equation 6.3, "In-Situ Gas Content" represents the CH4 actually contained in the coal
being mined, as determined by measuring the gas content of coal samples. Average
values for a coal mine, coal basin, or country may be available, depending on the
country. For surface mines, unlike underground mines, it is frequently assumed that all
of the CH4 contained in the coal is released during mining and that post-mining
emissions from surface-mined coals are effectively zero (BCTSRE, 1992; and ClAB,
1992). This assumption could be modified for some countries, however, based on their
particular coal characteristics.
Page 6-16
-------
Cqal Pro'durctfoh \ -Y'• • •>
., , ,- < - • . ,
I n;Situ ^as -'pgpfenj^rjcj ^s^urjied Brilssion iFa,otor for Sufrdurfdlng Strata are
'-ple^crfbed jn^tffetext ;;; ; "r\ \ "- ^ . f, -;, ,1 -v-;. .;" . V'XC'-V r- ^."-x
Covrsion Factor ddneshe.yoiarn'e 'of ^CH to,-a weiht-measure based on
of ^CH4 to,-a weight-measure based on
6arjci,I atmi 'whichlsr '' . - °' ' :-•",-
"
"Assumed Emission Factor for Surrounding Strata" represents the possibility that more
CH4 will be emitted during surface mining than is contained in the coal itself because of
emissions from the strata below (or in limited cases, above) the coal seam. Some
country-specific studies have assumed that there are no emissions from surrounding
strata associated with surface-mined coals (BCTSRE, 1992; and CIAB, 1992). If,
however, available information indicates that there are gas-bearing strata surrounding
the mined coal seam and that these strata are emitting their gas in conjunction with the
mining, then these emissions should also be included in the estimates. Other studies,
including Jaques (1992) and Kirchgessner et al. (1993), have assumed or observed
through measurements that surrounding strata contribute to emissions.
Emission factors for the surrounding strata can be developed using one of two
approaches. Ideally, the assumed emission factor should be based on an evaluation
of the gas content of the surrounding strata and verified by measurements. If such
data are unavailable, an alternative method of developing an emission factor is to
assume that some multiple of the gas content of the mined coal is emitted by the
surrounding strata. The Canadian study by Jaques (1992), for example, assumes a
factor of three, and Kirchgessner et al. (1993) report that emissions at a U.S. surface
mine exceeded the in-situ gas content by a factor of five. It should be noted that the
Page 6-17
-------
alternative approach is highly speculative, given the lack of data upon which to base
such assumptions.
6.3.4 Post-Mining Activities
Like surface mining emissions, there are currently few measurements of CH4 emissions
from post-mining activities. In fact, many past studies have overlooked this emission source,
while others have developed only rudimentary estimation methodologies. Two possible
approaches for estimating emissions from post-mining activities are recommended.
Tier 1: Global-Average Method
For the simplest estimates, a global-average emission factor range can be multiplied
by coal production for underground mining and surface mining, as shown in equation 6.4. It is
important to distinguish between underground- and surface-mined coals because the gas
contents and potentially the post-mining emission processes are likely to be very different;
hence, emissions could vary significantly.
For underground mining, emission factors of 0.9-4 rrvVtonne are recommended, based
on recent studies. The low end of the range was developed by the CIAB (1992) by assuming
that underground-mined coal has an average gas content of 1.5 m3/tonne and that 60% of this
gas is emitted after the coal leaves the mine. The high-end emission factor assumes that the
underground-mined coal has an average gas content of 10 m3/tonne. It also assumes that
40% of the CH4 is emitted after mining.
For surface mining, emission factors of 0-0.2 nrvVtonne are recommended. The low
end of the range assumes that there are no post-mining emissions associated with surface
mining, based on assumptions made by BCTSRE (1992) and the CIAB (1992). The high end
of the range assumes an average gas content of 0.5 m3/tonne, which is consistent with the
average gas content in the United Kingdom (BCTSRE, 1992). It also assumes that 40% of
the in-situ CH4 content is released after mining.
Tier 2: Country- or Basin-Specific Method
Emission estimates can be refined if additional data are available on coal
characteristics. This method may be preferable if higher-tier methods have been used to
estimate emissions from underground and surface mines. Equation 6.5 summarizes the
approach for preparing refined emission estimates.
"In-Situ Gas Content" represents the CH4 actually contained in the coal being mined,
as determined by measuring gas contents in coal samples. Average values for a coal
mine, coal basin, or country can be used, depending upon the availability of data in a
particular country.
The "Fraction of Gas Released During Post-Mining Activities" represents the
percentage of the in-situ gas content that is assumed to be emitted during post-mining
activities. There are three key issues related to the development of this fraction:
Page 6-18
-------
:> ; -^. « - .- >-,^ ,,»v •>>•
r ? vi'. rf x" <>^' »„ • > ^ %V . '» v^'^v '"/• , <".',»••'> ,»" V«"""i * :^''t, % *.%v, *» '" > "> , ; / , ,t *,^ ^> % <, ^ . • •
> -v.^v>>Vl x»; 5"^tf ^/'-.r-V <• ".v-;X '.%:'•: '«;v js-^^ V<,T---I ^-fc> -- -C';''^ '-^">'< ^"'v °." ^
>'' ' '
;:w(tc»i>rte$}"r'>^ 'X*.''>.,v"; ;©Wtame,ol Gdal.mwb'd} ;-• , rv .....*. • ,;;:>; '. ;-
I/?"-" 'sVv'.N'V"/*-^ -"-i-x doal.P^^uctton lOrjderoroucid and Surface)* , J •, °- c-- --\ \
>S> ' - ",™ { .-> * ^**t( »'»* ,««,«- . ^ , x ,' ' , ^ «.'«'..'- >
5 ^ * ' " •
s_HigH",OH4Bw^^ s'" " "*' "'' ' * ' '"^
J*J ° < o ' * " sf •" i J- ^> ,,* s C * * ^ ,°0 ""**".* ^ <" "£•-•" X V- t ^v ' O " 3-i \ < -, "* ," ' <*• -1* * ^ ^ > '
o" „ } < ^ ^S"'ri^Jjv ^ < ^ >.\. „ -;s "*,•.-
'^5'JkX-x >J%"*°M/'V- •".; s-'•"%-^'''? -^V' ' -•*' ' '* r ""''' '"'ss- y- ' H-"S"<:%- '"*"' S"X\° '/•>^-'"*>
Where:;;^ * /„',_' o/rv-^ *""" ">^;f"" '/'**<• > <^<% '••..<'.'- ,X? . ,v'^. •'-*>> ,„« '
v^-x ^.j^Ur^ergro^ ', ,'k \'":' "•'"• ' •' ,-s
•v".v 'XJ^o^^^nd'JH^h'CH^EmisskHtii^dbr-^^Om^hne. / „' ,-',-' V'1.'-'
und°High"CH
CH4 .Emission
"\V<' >;«Vi &'•'," '• '-' ,'° •' 'X''"i >'" ">.''„* ^ v'"-^'.'">" '"''" * ; V'^V" ,\-V'" °,' ' >• '= ''"' >- • i •" " '~'! '"'"'
' "^ i° ?> /Em|sSfOJhs;ar,e.dalc,u|atedK^^ ^ '% ,:
'f7-y^/vu^s1tigtr1elrr,especliye*e^ * V ^\,_,/' /'"'',;
'','?"?''•>'* ^'^/''' '-r'^'^"'^^;-^^'^"'"'?-'-^"" I-"'' v %:!4 " '" :"'s^ '^-\i-?*--Sr ,-• ••:
; -"11 ^ >^i. Sx ^-< i-*>v>,x •«,y :^:V?-> ^- V -...:.?..^y.:iv' '• -4 ' ~r' -I °;.'.. ^.:.-.' ':'-'•..".'! "' '.L.
(a) For Surface-Mined Coal: In most cases, if the Tier 2 approach (equation 6.3) is
used to estimate CH4 emissions from surface mines, post-mining emissions
from surface-mined coals are assumed to be zero.10 If it has been assumed
that some of the CH4 contained in surface-mined coal remains after mining,
however, equation 6.5(b) should be used to estimate post-mining emissions,
and the value selected for "Fraction of Gas Released During Post-Mining
Activities" should be consistent with the assumption used in equation 6.3.
10 This assumption should be made if the full in-situ gas content is multiplied by surface-mined coal production
to estimate emissions. Several studies have made this assumption, including BCTSRE (1992) and CIAB (1992).
Page 6-19
-------
Ill Ill III III 111 I I 111 III I Ilk J ,1 ll
1 TIER 2: COUNTRY OR BASIN SPECIFIC ..„
. P0ST4/riNING ACTIVITIES
"' ' 1^1 •' I, A _ . i< »,? "'« > «.«•».
a. Underground CH4 = In-Situ Gas Content (m3 CH^tonnej
Emissions (tonnes) ' f r
, x Fraction "of Gas Released During Post-Mining Activities
f_ " (percent) ' " * ' -'-
r x Underground Coal Production (tonnes)
i *
- x Conversion Factor
i- When Necessary:
. b. Surface CH4
I Emissions (tonnes)
In-Situ Gas Content (m3 CH/tonne)
x Fraction of Gas Released During Post-Mining Activities.
(percent)
x Surface Coal Production (tonnes)
x Conversion Factor
* Where:
• In-Situ Gas Content and Fraction of Gas Released During Mining are described in the
text. ' ' -
• Conversion Factor converts the volume of CH4 to a weight measure based on the .
density of methane at 20°C and 1 atm, which Is:
0.671 tonnes/103m3.
(b) For Underground-Mined Coal: The assumed fractions for underground mining
should be developed based on information about coal permeability, desorption
rates, mining methods, and other factors. Recent studies have assumed that
25-60% of the in-situ CH4 content of underground-mined coal is emitted during
post-mining activities (U.S. EPA, 1993; BCTSRE, 1992; and CIAB, 1992).
(c) Potential Fraction of CH4 Not Emitted: It is currently assumed that all of the
CH4 contained in mined coal will be emitted to the atmosphere, although it is
possible that a fraction could remain in the coal until the point of combustion
and could be burned instead of emitted. At this time, estimates of the extent to
which this may be the case have not been developed. If such information
becomes available in the future, however, it should be incorporated into
equation 6.5.
Page 6-20
-------
6.3.5 Coal Combustion
Because data on coal combustion by end-use sector are available from the
International Energy Agency (IEA, 1992a and 1992b), one method of estimating emissions
from coal combustion is recommended. This method relies on available data on CH4 emission
factors for various types of coal-fired combustion units (Radian, 1990; Selzer and Zittel, 1990;
Watt, 1993; Smith and Sloss, 1992; and OECD, 1991). As shown in Table 6-4, the measured
emission factors vary substantially, however, and there is substantial uncertainty related to the
potential emission levels associated with certain types of coal-fired equipment. Emission
levels from utility power plants in Western countries are generally the best understood and are
believed to be very low. It is possible, however, that the emissions associated with power
plants, industrial boilers, and domestic stoves in non-Western countries could be significantly
higher; to date few emission measurements have been reported. The recommended emission
factors are shown in the last column of Table 6-4, and the estimation methodology is shown in
equation 6.6.
6.3.6 Treatment of Other Coal Fuel Cycle Emissions
Due to lack of data, no emission estimates have been prepared for other fuel cycle
emissions, principally from abandoned mines and coal waste piles. Some national experts
have speculated that for their countries these sources are negligible (BCTSRE, 1992).
However, additional research is warranted before concluding that these emission sources are
negligible on a global basis.
Abandoned coal mines are of particular concern because there are many of them, and
some could emit significant quantities of CH4. Most available evidence indicates that CH4 flow
rates decay rapidly once deep-mine coal production ceases (Williams and Mitchell, 1992; and
Greedy, 1991). In some abandoned mines, however, CH4 can continue to be released from
surrounding strata for many years. In Belgium, France, and Germany, for example, several
abandoned mines are currently being used as sources of CH4 that can be added to the gas
system (Smith and Sloss, 1992; and KfA, 1993). Some U.S. abandoned mines are also
producing marketable amounts of gas (Hite, 1992).
TABLE 6-4
SELECTED COAL COMBUSTION EMISSION FACTORS, BY SOURCE
(m3/tonne of coal consumed)
Source
Utility
Industry
Boilers
Kilns/Dryers
Watt
1993
0.019-0.022
0.033-0.079
-
-
Smith &
Sloss
1992
0.04
0.04-0.48
-
•
OECD
1991
0.024-0.028
-
0.096
0.04
Selzer &
Zittel
1990
0.04
0.14-0.5
-
-
Radian
1990
0.025-0.026
-
0.092
0.012-0.026
Present
Study
0.02-0.04
0.03-0.5
-
Residential/Commercial
Boilers
Residential Stoves
0.033
0.4
9.2-30
0.93-2.3
12.4-31
0.33
0.33-2.3
9.2-30
Page 6-21
-------
EQUATION 6.6 ' "" ' :
TIER 1: COAL COMBUSTION EMISSIONS, BY SECTOR
Coal Combustion Emissions =
(tonnes)
Sum, over aH sectors* of: ,
Coal Consumption^ by sector (tonnes) ,
x Sectoral Coal Combustion Emission Factor '
,(m3/tonrte)
' ^ > /
»„ / ^ i. *-
x Conversion Factor ' '
Where: " ' . ' •" '
Recommended Coal Combustion Emission Factors are determined'for each sector
(power, industry, residential/commercial), as shown in Table 6-6, '' ,
Conversion Factor converts the volume of CH4 to a weight measure, based on the
density of methane at 20°C and 1 atm, which is: „'
0.671 tonnes/103m3.
U.S. EPA is currently developing a data base of U.S. abandoned and inactive mines in
an effort to produce emission estimates. Based on the results of this work, it should be
possible to make rough U.S. emission estimates for this class of mines in the future. The
number of abandoned mines in other countries is unavailable, but is expected to be large.
6.4 RESULTS
A number of methods of estimating CH4 emissions have been described in the
previous section. One of these methods, the Global-Average Method, can currently be
applied in all countries to estimate emissions because it requires only coal-production data.
Other methods require more detailed data, and so can be applied only in certain countries.
For several countries, however, such data are available and have been used in preparing
estimates.
6.4.1 General Findings
Table 6-5 presents emission estimates developed using Global-Average Emission
Factors' and, where possible, U.S. EPA's Best Estimates, which are based on more
information. For comparison, the emission estimates prepared by the CIAB (1992), which also
provide recent specific estimates for key countries, are also included. As the table shows, the
Global-Average Emission method has been used to develop emission estimates for most of
Page 6-22
-------
the world's coal-producing countries.11 Total emissions in 1990 were estimated to range from
19 to 57 Tg using this method.
For several of the major coal-producing countries, more detailed estimates have been
developed. For the most part, these estimates use more detailed emission data on
underground mining and more refined assumptions about likely emission levels from surface
mines and post-mining activities. In many cases, these estimates have been developed by or
in conjunction with in-country personnel well versed in coal mining emission issues. These
emission estimates are also shown in Table 6-5, in the column "U.S. EPA Best Estimates,"
and the methods of preparing these estimates for each country are described in the next
section.
When the more detailed data are used, the resulting global emission estimates for coal
mining are 24-40 Tg. This estimate is the preferred range of likely emissions from coal
mining, because it incorporates the most recent and detailed information from a number of the
key CH4-emitting countries.
Most of the CH4 emitted by the coal fuel cycle is emitted by mining in the major coal-
producing countries (Table 6-6). In fact, China, the United States, and the former Soviet
Union are estimated to emit almost three-quarters of the world's emissions from this source.
The top-ten emitting countries, which also include Australia, the former Czechoslovakia,
Germany, India, Poland, South Africa, and the United Kingdom, together accounted for
approximately 90% of both global emissions and global coal production.
11 For some countries, such as North Korea, data on coal production are not available from international
sources. The lack of estimates for these countries is not expected to significantly affect the accuracy of global
emissions estimates, however, because coal production data are available for more than 90% of the world's coal
production.
Page 6-23
-------
TABLE 6-5
ESTIMATED 1990 METHANE EMISSIONS FROM THE COAL FUEL CYCLE
(Tg)
Global Average Method
Country
China
United States
Former Soviet Union
Germany
India
Poland
Australia
South Africa
Former Czechoslovakia
United Kingdom
Canada
Greece
Turkey
Spain
Colombia
France
New Zealand
Austria
Belgium/Luxembourg
Italy
Norway
Ireland
Subtotal
World Total
Low
8.4
2.7
2.8
0.4
0.8
1.0
0.3
0.8
0.1
0.5
<0.1
<0.1
0.1
0.1
0
0.1
0
0
<0.1
0
0
0
18.1
19.4
High
23.4
8.1
8.0
1.8
2.4
3.0
1.2
2.3
0.5
1.4
0.2
0.1
0.2
0.3
<0.1
0.2
<0.1
<0.1
<0.1
0
<0.1
0
53.1
57.0
U.S.
EPA
Best Estimate
Low High
9.5
3.6
4.8
1.0
0.4
0.6
0.5
0.8
0.3
0.6
0.1
<0.1
0.1
0.1
0
0.1
0
0
<0.1
0
0
0
22.6
24.4
16.6
5.7
6.0
1.2
0.4
1.5
0.8
2.3
0.5
0.9
0.2
0.1
0.2
0.3
<0.1
0.2
<0.1
<0.1
<0.1
0
<0.1
0
37. 1
39.6
CIAB
7.6
3.3
4.8
1.0
0.5
1.2
0.5
0.9
0.3
0.8
--
-
-
--
--
-
--
--
--
--
--
--
20.8
25.0
Note: CIAB estimates do not include emissions associated with coal combustion.
Page 6-24
-------
TABLE 6-6
1990 ESTIMATED METHANE EMISSIONS FROM THE COAL FUEL CYCLE
FOR THE TOP-10 EMITTING COUNTRIES
Coal Production (million tonnes)
Estimated CHA Emissions (Tg)
Country
China
United States
Former Soviet
Union
Germany
India
Poland
Australia
South Africa
Former
Czechoslovakia
United Kingdom
Subtotal - Top 10
World Total
Underground
1,024
385
393
77
109
154
52
112
22
75
2,043
Surface
43
548
309
359
129
58
154
63
85
U
1,762
Total
1,066
934
701
436
238
212
206
175
107
89
4,164
4,740
Low
9.5
3.6
4.8
1.0
0.4
0.6
0.5
0.8
0.3
0.6
22.1
24.4
High
16.6
5.7
6.0
1.2
0.4
1.5
0.8
2.3
0.5
OL9
35.9
39.6
The estimates also indicate that underground mining is responsible for the bulk of the
emissions from the coal cycle, with its share ranging from 70% to 85% (Figure 6-2). Surface
mines and post-mining emissions account for another 10-20%. Coal combustion is estimated
to emit the remaining 5-10%. About 90% of the coal combustion emissions are estimated to
come from China, where large amounts of coal are burned in small, inefficient domestic
stoves.
Wherever possible, the estimated CH4 emissions presented in Table 6-5 have been
adjusted to reflect CH4 utilization in countries with ongoing CH4-recovery programs. The
amounts of CH4 currently being utilized in those countries for which data are available are
shown in Table 6-7, and total about 1.3 Tg (1.9 billion" m3), which represents 3-5% of total
emissions. In addition, another 1.5-2.6 Tg (2.2-3.9 billion m3) are being vented by mine-
degasification systems. It is probable that the amounts of CH4 being recovered, used, and
vented by mine-degasification systems are underestimated because no information is
available for several countries that may have such systems.
6.4.2 Country-Specific Information
For several of the world's key coal-producing countries, more detailed information has
been used to develop the U.S. EPA Best Estimate of CH4 emissions from coal mining. The
information used in developing these estimates is discussed in this section, and the results
are compared with those of the Global-Average Method and other recent studies.
Page 6-25
-------
FIGURE 6-2
1990 METHANE EMISSIONS FROM THE COAL FUEL CYCLE
Underground
Mines
Surface
Mines
Post-
Mining
Coal
Combustion
Total
Emissions Source
-------
TABLE 6-7
METHANE RECOVERY BY DEGASIFICATION SYSTEMS
FOR KEY COUNTRIES
(Tg)
Degasification Emissions
Country
Recovered
Used
Emitted
Year
Australia
China
Former Czechoslovakia
Germany
Poland
Former Soviet Union
United Kingdom
United States
0.10
0.29
0.09
0.35
0.19
0.83
0.20
0.7-1.8
0.05-0.08
0.18
0.08
0.25
0.14
0.19
0.14
0.25
0.02-0.05
0.11
0.01
0.10
0.05
0.64
0.06
0.45-1.55
1989
1990
1990
1990
1989
1990
1990
1988
Sources: Lama, 1991; Saghafi & Williams, 1992; MOE, 1991; Bibler et al., 1992;
Zimmermeyer, 1991; Pilcher et al., 1991; Marshall et al., 1993; Airuni,
1992; Greedy, 1992; and U.S. EPA, 1993.
People's Republic of China
In 1990, the People's Republic of China produced 1,024 million tonnes of coal in
underground mines and 46 million tonnes of coal in surface mines (IEA, 1990), making it the
world's largest coal producer. Of the underground coal, 44% was produced in large state-
owned mines, 20% in large provincial mines, and 36% in small local mines.
Based on data provided by the Chinese Ministry of Energy, it was assumed that
underground state mines emitted 10,000 million m3 of CH4 in ventilation air in 1990 and an
additional 434 million m3 in degasification systems (MOE, 1991). Thus, the resulting average
emission factor for the state-owned underground mines was 23.1 m3/tonne. Of the total
emissions from these mines, moreover, a reported 270 million m3 of CH4 were used in 1990
(MOE, 1991). No emission data are currently available for emissions from other underground
mines, so a range of possible emissions was estimated based on the state mines. It was
assumed that the large nonstate mine would have average emission factors of 25-100% of the
state-owned mines (5.8 to 23.1 m3/tonne), yielding possible emissions of 1,179-4,715 million
m3 (0.79-3.16 Tg). For small local mines, an average emission factor of 1 m3/tonne was
assumed, given the low level of mechanization. This assumption yielded additional emissions
of 369 million m3 (0.25 Tg). The Global-Average Method was used to estimate emissions
from surface mines and post-mining activities. Surface mines emitted an estimated 13-86
million m3 (0.01-0.06 Tg) in 1990, and post-mining activities an additional 922-4,105 million m3
(0.62-2.76 Tg). China is estimated to have the largest emissions from coal combustion,
Page 6-27
-------
constituting 90% of total emissions from this source. This is due to the large amount of coal
consumed by industry and in residential stoves. The estimated emissions from 1990 range
from 1.05 to 3.56 Tg. These estimates should be considered highly uncertain, however, due
to the lack of available data on CH4 emissions from Chinese combustion sources.
Based on these assumptions, the U.S. EPA Best Estimate for Chinese CH4 emissions
from coal mining in 1990 is 9.5-16.6 Tg. This estimate is within the range developed using
the Global-Average Method. The high-end level is significantly lower than that developed
using the Global-Average Method, however, due to the likelihood that the nonstate mines will
have significantly lower emissions than those represented by the high-global average emission
factor. This estimate is higher than the estimate of 7.6 Tg presented in the CIAB (1992)
study. There are two main differences between the U.S. EPA and CIAB estimates. First, the
CIAB assumed that all coal production in nonstate mines emitted CH4 at a level of 0.9
ma/tonne. While emissions are likely to be low for the many small local underground mines,
U.S. EPA considers this assumption unduly conservative for the larger nonstate mines,
particularly given the reportedly high incidence of CH4 accidents and explosions in these
mines. In addition, the CIAB estimate did not include CH4 emissions from coal combustion,
which are believed to be significant in China.
United States
The United States produced 385 million tonnes of coal from underground mines and
548 million tonnes of coal from surface mines in 1990 (U.S. DOE/EIA, 1992a). U.S. EPA
(1993) has prepared a detailed estimate of CH4 emissions from coal mines in 1988, which
totals 3.3-5.2 Tg.12 Another 0.25 Tg of CH4 were recovered and used by mine degasification
systems. Correcting the 1988 estimate for 1990 coal production (assuming similar emission
factors) and including coal combustion, which was not estimated in U.S. EPA (1993), results in
a 1990 emission estimate of 3.6-5.7 Tg.
This estimate is within the range of 2.7-7.9 Tg estimated using the Global-Average
Method. The CIAB (1992) estimated total emissions of 3.3 Tg in 1990, which is somewhat
lower than the U.S. EPA Best Estimate. The key differences between the CIAB study and the
U.S. EPA Best Estimate (based on U.S. EPA, 1993) are:
Treatment of Surface Mines and Post-Mining Activities: The CIAB used conservative
emission factors for these sources, and estimated emissions were 462 million m3 (0.31
Tg). U.S. EPA 1993 developed a wider range of emission levels of 1,043-2,235 million
m3(0.7-1.5Tg);
Amount of Utilization: The CIAB reported utilization of 462 million m3 (0.31 Tg), while
U.S. EPA reported utilization of 373 million m3 (0.25 Tg) associated with projects at six
coal mines;
18 Estimates were prepared for 1988 because this is the most recent year for which the U.S. Bureau of Mines
has published mine-by-mine ventilation emission data.
Page 6-28
-------
Level of Deqasification System Emissions: The CIAB reported degasification system
emissions of 1,471 million m3, based on personal contacts. U.S. coal companies are
not required to report these emissions, and limited data are publicly available on actual
emissions. U.S. EPA estimated emissions of 745-2,384 million m3 (0.5-1.6 Tg), based
on an identification of mines reported or believed to have degasification systems in
place and on the assumed recovery efficiencies of these systems.
Former Soviet Union
In 1990, the former Soviet Union produced 393 million tonnes of coal in underground
mines and 309 million tonnes of coal in surface mines (IEEA, 1990), which represents a drop of
almost 5% from 1989 (from 740 million tonnes to 709 million tonnes). The Skochinsky Mining
Institute reports that CH4 emissions in 1990 were 7,005 million m3 (4.7 Tg), of which 280
million m3 (0.19 Tg) were used (Marshall et al., 1993). This estimate was used to develop the
low-end emission estimate, with additional emissions estimated using the Global-Average
Method. Surface mining and post-mining activities added an estimated 447 million m3 (0.3
Tg). Another recent study by Airuni (1992) reports that total CH4 emissions in the former
Soviet Union in 1990 were 8,730 million m3 (5.86 Tg), not including another 432 million m3
(0.29 Tg), which was reportedly used. Methane emissions from coal combustion in the former
Soviet Union were estimated to be larger than for many countries, because of the amount of
coal consumption by industrial and residential customers. These emissions could range from
15 to 179 million m3 (0.01 to 0.12 Tg) in 1990.
The Skochinsky data were used for developing the low end of the U.S. EPA Best
Estimate, and the Airuni estimate for the high end of the range. This yields a Best Estimate of
emissions from coal mines in the former Soviet Union in 1990 of 4.82-5.98 Tg. This estimate
falls within the range of 2.7-8.1 Tg developed using the Global-Average Method. In addition,
the low end of the range is consistent with the estimate developed by the CIAB (1992), which
used the Skochinsky data and conservative assumptions of emissions from surface mines and
post-mining activities.
Germany
In 1990, Germany produced 77 million tonnes of coal in underground mines and 359
million tonnes of coal in surface mines (IEA, 1990). Reported CH4 emissions from
underground mining, including ventilation and degasification, were 1,724 million m3 (1.16 Tg)
in 1990, of which 371 million m3 (0.25 Tg) were utilized (Zimmermeyer, 1991). Surface-mining
emissions were estimated to be only 5.4 million m3. An emission factor of 0.015 rrvVtonne was
used for this estimate, instead of the global-average emission factor range, because Germany
reports that its surface-mined coals are of extremely low rank and low gas contents. Based
on the Global-Average Method, post-mining emissions were estimated to be 70-308 million m3
(0.05-0.26 Tg), with only emissions from underground mining estimated. Methane emissions
from coal combustion were estimated to range from negligible to 30 million m3 (0.02 Tg).
Total CH4 emissions in 1990 for the U.S. EPA Best Estimate were thus estimated to be
0.96-1.19 Tg for Germany. This estimate is consistent with the range of 0.4-1.8 Tg developed
in the Global-Average Method. It is also consistent with the CIAB (1992) study, which used
similarly conservative assumptions.
Page 6-29
-------
India
India's reported coal production in 1990 was 109 million tonnes in underground mines
and 129 million tonnes in surface mines (IEA, 1990). A recent study prepared by the Indian
government (Mitra, 1992) was used to prepare the U.S. EPA Best Estimate. This study
estimated emissions for the period April 1, 1990, to March 31, 1991; the estimates for this
period were assumed to correspond to those for the 1990 calendar year.
Mitra (1992) reported that only a small portion (3.5 million tonnes) of India's
uriderground-mined coal was produced in gassy mines, and that the average emission factor
for these mines was 20 m3/tonne. An additional 27 million tonnes of coal were produced in
underground mines, with an average emission factor of 10 m3/tonne, and the remainder of the
underground-mined coal, as well as the surface-mined coal, was assumed to emit 1 m3/tonne
during mining. Post-mining emissions were estimated on the basis of measurements taken to
determine the amount of gas remaining in various types of coal after mining. The amount of
residual gas ranged from a low of 0.09 m3/tonne for nongassy underground- and surface-
mined coal to a high of 3.7 m3/tonne for the coal produced in the gassiest mines. Methane
emissions from coal combustion are estimated to range from negligible to 60 million m3 (0.04
Tg).
Using these assumptions results in a Best Estimate of Indian coal mining emissions of
0.4 Tg in 1990. This estimate is significantly lower than that estimated by the Global-Average
Method, which ranges from 0.8 to 2.3 Tg. This is because the available data indicate that the
underground coal mined in India is much less gassy than that mined in other countries. Over
half of India's underground coal production reportedly emits CH4 at levels comparable to its
surface mines, for example, and only 3 percent of the underground coal is mined in gassy
mines. The Best Estimate is also slightly lower than the estimate prepared by the CIAB
(1992), which,estimated an emission level of 0.45 Tg using a statistical model based on
available data on coal rank and depth and emission data from other countries.
Poland
In 1990, Poland produced 154 million tonnes of coal in underground mines and 58
million tonnes in surface mines (IEA, 1990). The most recent estimates of Polish CH4
emissions from underground mines were for 1989, when underground coal production was
somewhat higher, at 176 million tonnes. The 1989 emission estimates ranged from 1,045
million m3 to 1,844 million m3 (0.70 to 1.24 Tg), with the lower estimate including emissions
from only those mines classified as "gassy" according to the Polish system (Pilcher et al.,
1991). Emissions from underground mines in 1990 were estimated by calculating the
emission factors associated with underground mining in 1989 and assuming that they would
be unchanged. Based on this approach, Polish underground-mining emissions in 1990 were
estimated to be 909-1,617 million m3 (0.61-1.09 Tg). In 1989, 201 million m3 (0.14 Tg) were
utilized; and it was assumed that the same amount was used in 1990. Poland has not
developed estimates of emissions from surface-mining or post-mining activities. For U.S.
EPA's Best Estimate, the Global-Average Method was used to estimate emissions from these
sources, with a resulting estimate of 156-744 million m3 (0.11-0.5 Tg). In the low case, CH4
emissions from coal combustion were estimated to be negligible, while in the high case they
were estimated at 45 million m3 (0.03 Tg).
Page 6-30
-------
Based on these assumptions, U.S. EPA's Best [Estimate of 1990 CH4 emissions from
coal mining in Poland is 864-2,150 million m3 (0.58-1.45 Tg). The U.S. EPA Best Estimates
are lower than those estimated using the Global-Average Method, which yields a range of
1.0-3.0 Tg. This is because the average emission factor for underground mines in Poland is
at the low end of the global-average emission factor range and, in the case of the estimate for
gassy mines only, is below the bottom end of the range. The CIAB (1992) estimated
emissions of 1.21 Tg in 1990. It used the high-end 1989 data for underground mines (without
adjusting it for lower 1990 coal production) and low-end global average emission factors for
surface mines and post-mining activities. A Polish study estimated emissions of 912 million
m3 from underground mining in 1990 (Surowka and Pasierb, 1992), which results in estimates
that are consistent with the low end of the U.S. EPA Best Estimate range.
Australia
In 1990, Australia's coal production was reported to be 52 million tonnes from
underground mines and 154 million tonnes from surface mines (IEA, 1990). Two recent
Australian studies were used to prepare U.S. EPA's Best Estimate of Australia's coal-mining
emissions. These studies reported that 1990 CH4 emissions from underground mining,
including both ventilation and degasification systems, ranged from 809 million,m3 (0.54 Tg) in
Lama (1991) to 1,162 million m3 (0.78 Tg) in Saghafi and Williams (1992). Of this, between
70 and 122 million m3 (0.05 and 0.08 Tg) were reportedly used. Surface-mining emissions
were estimated to range from 46 to 124 million m3 (0.03 to 0.08 Tg), with the lower end of the
range associated with an average gas content of 0.3 m3/tonne and the higher end based on
some field measurements at major surface mines (Saghafi and Williams, 1992). No post-
mining emissions were assumed to be associated with surface mining, and post-mining
emissions from underground mines were estimated to range from 17 to 34 million m3 (0.01 to
0.02 Tg), assuming an average residual gas content of 0.5-1 m3/tonne in underground-mined
coal (Saghafi and Williams, 1992). Methane emissions from coal combustion are estimated to
be negligible.
Total emissions for Australia's coal mines were thus estimated at 750-1,250 million m3
(0.50-0.84 Tg) in 1990. This estimate is consistent with those developed using other
methods. It lies within the range of 0.3-1.2 Tg, developed using the Global-Average Method.
The CIAB (1,992) estimated emissions of 0.52 Tg, which is consistent with the low end of the
range, as would be expected since the CIAB study uses conservative assumptions for
surface-mining and post-mining activities and uses the low emission estimate for underground
mining, as presented by Lama (1991). The U.S. EPA Best Estimate is also consistent with
Lama's study (1991), which presented a best estimate of 0.5-0.6 Tg, and the Saghafi and
Williams (1992) study, which estimated emissions from coal mining at 0.76 Tg.
The Former Czechoslovakia
In 1990, the former Czechoslovakia produced 22 million tonnes of coal in underground
mines and 85 million tonnes of coal in surface mines (IEA, 1990). Total CH4 emissions from
underground mines in 1990 were reported to be 525 million m3 (0.35 Tg) (Bibler et al., 1992).
Of this, 124 million m3 (0.08 Tg) were used instead of being emitted. Estimates of CH4
emissions from surface-mining or post-mining activities are not available. Therefore, the
Global-Average Method was used to prepare U.S. EPA's Best Estimate of emissions from
these sources. These assumptions resulted in estimated additional emissions of 46-275
million m3 (0.03-0.19 Tg) of CH4. Methane emissions from coal combustion are estimated to
range from negligible to 30 million m3 (0.02 Tg).
Page 6-31
-------
U.S. EPA's Best Estimate of CH4 emissions in 1990 is 446-676 million m3 (0.3-0.47
Tg). This estimate is within the range of 0.1-0.5 Tg, developed using the Global-Average
Method for all sources. It tends toward the high end of the range because underground coal
mines in the former Czechoslovakia are very gassy and have an average estimated emission
factor of more than 25 m3/tonne. The CIAB (1992) study reported estimated emissions of 0.3
Tg, which is consistent with the low end of U.S. EPA's Best Estimate. The CIAB used the
same estimates for underground mining and conservative assumptions to estimate emissions
from surface mines and post-mining activities.
United Kingdom
In 1990, the United Kingdom's coal production was reported to be 75 million tonnes
from underground mining and 14 million tonnes from surface mining (IEA, 1990). The British
Coal Technical Services and Research Executive has done extensive work estimating CH4
emissions, and it reports that its average CH4 emission factor from underground mines,
including both ventilation and degasification systems, was 15.3 m3/tonne of coal mined in
1990 (BCTSRE, 1992). At its reported underground coal-production level, this emission factor
results in annual CH4 emissions of 1,148 million m3 (0.77 Tg). A reported 209 million m3 (0.14
Tg) of this gas were utilized (Greedy, 1992). British Coal reports that the average gas content
of its surface-mined coal is 0.5 + 0.3 m3/tonne. Using this range results in estimated
emissions from surface mining of 7-11 million m3. No post-mining emissions were assumed
for surface-mined coal. Underground coal was assumed by British Coal to emit 2 rrvVtonne
during post-mining activities, and this assumption results in estimated emissions of 150 million
m3 (0.1 Tg). Methane emissions from coal combustion are estimated to be very low, ranging
from negligible to 30 million m3.
These assumptions result in total estimated emissions of 0.75 Tg in 1990, which is
consistent with British Coal's estimate of 0.75 Tg + 0.1 Tg, based on its assessment of the
uncertainties associated with various assumptions (BCTSRE, 1992). Thus, U.S. EPA's Best
Estimate uses a range of 0.65-0.85 Tg. U.S. EPA's Best Estimate is within the range
generated using the Global-Average Method and is consistent with the CIAB (1992) study.
The Global-Average Method yields an emissions estimate of 0.5-1.4 Tg. The Best Estimate is
toward the low end of this range because all of the emission factors reported by British Coal
are in the lower end of the global-average emission factor ranges. The CIAB study estimates
emissions of 0.76 Tg in 1990, and its estimate, like the U.S. EPA Best Estimate, was based
directly on the work of British Coal.
Canada
In 1990, Canada produced 3.9 million tonnes of coal in underground mines and 64
million tonnes in surface mines (IEA, 1990). Environment Canada has prepared a recent
estimate of CH4 emissions from coal mining for 1990 (Jaques, 1992). This study reports that
underground coal mining emissions were 90 million m3 (0.06 Tg), based on the work of
Hollingshead (1990) and Stewart (1990). Although some gas utilization is reportedly
underway at Canadian underground coal mines, no utilization estimates for these mines are
available. Surface-mined coal is estimated by Hollingshead (1990) to contain 46 million m3 of
CH4, of which only 54% was assumed to desorb during mining based on measured residual
gas contents. Thus, emissions from surface-mined coal were estimated to be 25 million m3
(0.02 Tg). In addition, emissions from the overlying and underlying strata were assumed to be
Page 6-32
-------
three times the mined-coal releases, or 75 million m3 (0.05 Tg). Estimates of post-mining
emissions were not provided by Jaques (1992). The Global-Average Method was used to
estimate emissions from this source, which represented an additional 4-27 million m3.
Methane emissions from coal combustion are estimated to be negligible in Canada.
Using these assumptions, the U.S. EPA Best Estimate for Canadian CH4 emissions
from coal mining in 1990 is 0.13-0.15 Tg. The U.S. EPA Best Estimate is within the range
estimated by the Global-Average Method of 0-0.2 Tg. The CIAB (1992) study did not develop
an estimate of CH4 emissions from Canada.
6.5 TRENDS
This section discusses general and country-specific trends that may influence future
CH4 emissions from the coal cycle. In many countries, overall CH4 emissions are expected to
increase in the future if coal production and mine gassiness grow. It is possible that
emissions could fall in some Eastern European countries and the former Soviet Union,
however, due to the restructuring of the coal industry, the likely closure of some deep and
gassy mines for economic reasons, and possible declines in coal production. This section
discusses the key factors that will most likely influence future global trends in emissions and
potential developments in key countries.
6.5.1 General Factors Influencing Emissions
Three principal factors will affect future CH4 emissions from coal mining. Two of these
factors - higher coal production and increased mining in deep, gassy coal mines ~ will
contribute to higher emissions, while the third factor - CH4 utilization — could moderate the
emission increases.
Increasing Coal Production
In 1990, global coal production was 4.7 billion tonnes, of which approximately 55% was
mined in underground mines, and the remainder was mined using surface methods. Future
coal production is projected to be significantly higher, perhaps reaching 5.3-5.9 billion tonnes
in 2000 and 5.9-6.8 billion tonnes in 2010 (U.S. DOE/EIA, 1992b).
Several factors will affect future coal-production levels. Population and economic
growth will most likely increase energy demand and, hence, coal production. At the same
time, however, it is possible that more stringent environmental regulations of local and global
air pollutants, and the relative economics of coal production and use as compared to other
fuels, could reduce coal's share of future energy supply. Similarly, improvements in energy
efficiency, particularly in some of the most energy-intensive countries, could reduce the need
for additional energy supplies and limit the growth in CH4 emissions associated with increases
in coal production. One key uncertainty regarding future coal production involves the potential
impacts of carbon taxes or other policy instruments that could be enacted to reduce coal
consumption because of concerns about global warming. Several European countries have
proposed such policies, although none has enacted them to date.
Page 6-33
-------
If forecasted production levels are achieved and CH4 emission factors remain constant,
emissions could increase significantly, ranging from 27 to 50 Tg in 2000 and 30 to 58 Tg in
2010. To the extent that environmental or economic issues or improved energy efficiency
reduce coal production, however, emissions could be lower than these projections.
Increased Mining of Deep. Gassy Coal Seams
It is possible that underground coal mines will become gassier over time as shallower
coal seams are depleted and mining is initiated in deeper and potentially gassier coal seams.
In addition, as the proportion of underground versus surface mining increases, emissions
could also increase significantly. Quantifying a possible associated rate of increase in coal
mining emission factors is difficult, however, due to an absence of historical data, the difficulty
in generalizing about emissions in different coal basins, and the possibility that the most gassy
and unsafe mines in some countries may close. If a shift toward gassier coal seams occurs,
CH4 emissions could be higher than the levels projected above.
Increasing Methane Utilization
One moderating factor with respect to increasing CH4 emissions is the potential for
widespread use of the CH4 recovered by mine-degasification systems. As mines become
deeper and gassier, it is likely that they will use additional means of degasification to
supplement ventilation systems. As a result, they will recover larger quantities of CH4 that can
be used as fuel. Many countries are becoming increasingly aware of this fuel source, and in
the future could develop policies to encourage coal mines to increase their production and use
of the CH4 contained in coal seams.
In 1990, annual CH4 utilization at coal mines was estimated to be at least 1.3 Tg (1.9
billion m3). This represented an estimated 3-5% of total emissions from the coal fuel cycle.
With increasing recovery efficiencies of degasification systems and additional interest in
reducing the waste of a clean-burning fuel, it is likely that utilization rates will increase in the
future. If all of the CH4 that was reportedly recovered by degasification systems had been
used in 1990, for example, emissions could have been reduced by an additional 1.5-2.6 Tg
(2.2-3.9 billion m3).
6.5.2 National Trends
In any particular country, trends in CH4 emissions from the coal fuel cycle will depend
on developments in the rest of the energy sector, the particular distribution and accessibility of
coal basins, and the markets for recovered CH4, among other factors. For several of the key
CH4-emitting countries, some of the principal factors that could affect future CH4 emissions are
discussed below.
People's Republic of China
Based on government projections of future coal production and likely mining
conditions, the growth in CH4 emissions from coal mining in China is expected to be
substantial. In 1990, China was the world's largest coal producer, with production exceeding
1 billion tonnes (U.S. DOE/EIA, 1992b). Over 97% of this coal was produced in underground
mines. Many of China's principal coal basins have very gassy mines, with deeply buried,
high-rank coals in complex geologic formations.
Page 6-34
-------
In China's 8th Five-Year Plan (1991-95), the State Planning Commission approved
plans to increase coal production to 1.2 billion tonnes in 1995 and potentially to 1.4 billion
tonnes in 2000. Unless China makes a significant shift toward developing alternative'fuels,
such as natural gas or renewables, it is likely that this dramatic push to increase coal
production will continue.
Given China's current plans, CH4 emissions from coal mining are expected to increase
in the future. It is also likely, however, that China will increase mine degasification and QH4
utilization. A project is currently underway to demonstrate new CH4-recovery technologies at
mines, and this project should enable China to more effectively produce and use the CH4
contained in its coal seams.
United States
In 1990, CH4 emissions from the coal cycle in the United States were estimated to be
3.6-5.7 Tg, associated with the production of almost 1 billion tonnes of coal (U.S. EPA, 1993).
Approximately 0.25 Tg of CH4 was recovered and used instead of being vented. Methane
emissions are projected to increase between 1990 and 2010, due to increases in coal
production. In 2000, emissions could range from 4.0 to 6.7 Tg, and emissions in 2010 could
range from 5.2 to 8.9 Tg (U.S. EPA, 1993). These estimates do not account for the potential
increase in coal mine gassiness associated with the mining of deeper and gassier coal seams,
however, due to difficulties in quantifying this factor. If deeper and gassier coal seams are
mined, therefore, these estimates could be lower than actual emissions.
^As the range indicates, there is substantial uncertainty in future emission levels. The
principal uncertainties relate to the impact of new environmental regulations, particularly the
reduction of acid rain precursors, on future coal production. Two coal-production forecasts
were used in preparing these estimates, assuming different levels of coal production and
different mixes of underground- and surface-mined coals. To the extent that acid rain
legislation encourages additional use of low-sulfur coal, which is generally produced in surface
mines in the western United States, the growth in emissions will be moderated.
Emission growth may also be moderated by the potential to increase CH4 utilization.
Currently, CH4 utilization projects are constrained at many mines due to a variety of legal and
institutional barriers. The 1992 Energy Policy Act contains provisions to remove some of
these barriers, however, and more favorable state policies may also be enacted over the next
several years.
Former Soviet Union
In 1990, coal production in the former Soviet Union was about 700 million tonnes, third
in the world after China and the United States (U.S. DOE/EIA, 1992b). By 1991, however,
after the dissolution of the Soviet Union, coal production had fallen to only 630 million tonnes.
Assuming similar emission factors, this drop in coal production would correspond to emissions
of 3.9-5.2 Tg in 1991.
Future CH4 emissions in the former Soviet Union are difficult to predict because there
is substantial uncertainty about future coal production. In the past, many of the gassiest coal
mines in the former Soviet Union were subsidized heavily by the government, and they may
be unable to survive the transition to a market economy. Thus, it is possible that coal
production will not return to its historic levels and that the energy sector instead will
Page 6-35
-------
production will not return to its historic levels and that the energy sector instead will
concentrate on improving energy efficiency and shifting to cleaner-burning fuels, such as
natural gas. These trends would tend to reduce coal production and lower emissions.
Poland
Like the former Soviet Union, Poland's energy sector has suffered significant disruption
as a result of the transition to a market economy. In 1988, for example, Poland produced
almost 200 million tonnes of hard coal in underground mines and an additional 75 million
tonnes of lignite in surface mines. In 1990, hard-coal production was reported to have fallen
to only 150 million tonnes, and lignite production to only 60 million tonnes (U.S. DOE/EIA,
1992b).
It is not expected that coal production will continue to decline. The Polish government
is reportedly planning to maintain annual hard-coal production at around 140-150 million
tonnes and to use additional natural gas to meet its energy needs (FBIS, 1992). The
government is also strongly encouraging coalbed CH4 production, both in coal reserves and in
mining areas. Thus, it is likely that emissions in Poland will remain relatively constant or
perhaps decline in the future.
The Former Czechoslovakia
Like Poland, the states of the former Czechoslovakia are undergoing a transition to a
market economy and are also planning to close several coal mines. In 1980, Czechoslovakia
produced almost 30 million tonnes of hard coal and 100 million tonnes of lignite and brown
coal. By 1990, however, production had fallen to 22 million tonnes of hard coal and 84 million
tonnes of lignite. Production may continue to drop in the future, moreover, due to plans to
close at least four hard-coal mines by 1995.
Lower coal production, coupled with an increasing emphasis on coalbed CH4 utilization,
should result in lower CH4 emissions in the former Czechoslovakia. Currently, this region
uses 0.1 Tg of CH4 recovered from its coal mines (Bibler et al., 1992). By 2000, however,
utilization could be much higher due to an aggressive government program to increase
coalbed CH4 production in both nonmining and mining areas.
United Kingdom
In 1990, approximately 75 million tonnes of deep coal and 18 million tonnes of surface-
mined coal were produced in the United Kingdom. Over the last 22 years, coal production
has fallen significantly; it was 177 million tonnes in 1966. With the reduction in coal
production, there has been also a significant decrease in CH4 emissions associated with the
coal cycle, from an estimated 1.4 Tg in 1966 to 0.8 Tg in 1990 (BCTSRE, 1992). The United
Kingdom is planning to close several coal mines over the next few years, which is expected to
contribute significantly to lower CH4 emissions.
Page 6-36
-------
6.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
The estimates in this report contain two types of uncertainty. First, in many cases,
inadequate or missing data result in large uncertainties. Second, there are uncertainties
inherent in the methodologies for preparing estimates, which result from generalizing
complicated emission processes into global, national, basin-specific, or sectoral emission
factors. It is difficult to quantify rigorously the amount of uncertainty, and in this study large
ranges have been presented to indicate that the estimates are uncertain. This section
discusses some of the specific uncertainties related to particular data types and describes the
major sources of uncertainty related to various methodologies.
6.6.1 Data Uncertainties
The methods presented in this report require different types of data upon which to
develop emission estimates. All methods use coal-production data, for example, and many
require data on coal characteristics and geological information. The availability of such data
can greatly influence the selection of an appropriate estimation methodology, and the
accuracy of the data determines the reliability of the resulting emission estimates. The key
types of data required and some of the major uncertainties related to them are summarized
below.
Coal Production
Depending upon the method used, coal production can be required on a national or
basin level. In all cases, it is necessary to distinguish between underground- and surface-
mined coals, and this will likely be the largest source of uncertainty with respect to available
coal-production data. Where coal production is not identified by mining type, it may be
necessary to make assumptions about mining methods based on coal types. It is frequently
assumed, for example, that hard coal is mined underground and brown coal and lignite are
produced in surface mines. While this may be generally true, however, there are some
exceptions that can influence emission estimates. The misallocation of coal production among
these two categories can have a significant impact on emission estimates because of the
large differences in likely emissions from underground and surface mines.
Coal Characteristics
Some estimation methods require data on coal characteristics, including information on
coal gas contents, permeability, and other characteristics. The principal source of uncertainty
will relate to ensuring that the available data are representative of national- or basin-level
conditions. The degree of uncertainty caused by nonrepresentative data will depend on the
likely degree of variation in different parameters within the country or basin. Quantifying this
uncertainty may require conducting uncertainty analyses on a national or basin-specific level
by varying key parameters and observing the resulting changes in emission estimates.
Underground Mine Emission Data
In some countries, data may be available on underground mining emissions from
ventilation and, perhaps, degasification systems. Access to such data can enable countries to
prepare detailed and relatively reliable emission estimates. The adequacy of these data
should be assessed, however, by examining how and where measurements were taken,
whether all mines were included, and how emissions from degasification systems were
Page 6-37
-------
treated. Where appropriate, estimates of uncertainty in various parameters should be
determined in consultation with technical mining experts. The wide ranges used in this
chapter for global-average emission factors illustrate the uncertainties associated with these
factors (Table 6-8).
6.6.2 Methodological Uncertainties
The methods presented in this chapter are uncertain because they attempt to describe
in a simple and consistent manner very complicated emission processes. Methane emissions
can vary significantly within different parts of the same coal mine, for example; thus,
developing national or global emission factors will inevitably result in oversimplification and
inaccuracy. Of all of the components of the coal cycle, however, underground mining
emissions are the best understood and have the most available data.
TABLE 6-8
SUMMARY OF GLOBAL AVERAGE EMISSION FACTORS
FOR THE COAL FUEL CYCLE
Source
Emission Factor
(m3/tonne)
Underground Mines
Surface Mines
Post-Mining Activities
For Underground-Mined Coal
For Surface-Mined Coal
Coal Combustion
Power Generation
Industry
Residential/Commercial Boilers
Residential Stoves
10.0-25.0
0.3 - 2.0
0.9 - 4.0
0 - 0.02
0.02 - 0.04
0.03 - 0.5
0.33 - 2.3
9.2 - 30.0
Underground Mining
Three estimation methods have been presented, and the choice among them will
depend on data availability and the degree of certainty and detail required in the estimate. In
general, those methods that are the easiest to use and rely on the most general information
will have the largest associated uncertainties.
Global Average: The Global-Average Method only requires information on
underground coal production for each country, and applies a wide emission
factor range to that production. Thus, this method does not take into
consideration any country-specific factors that could affect CH4 emissions, such
as mining method or depth, coal characteristics, or geology. The range of
suggested emission factors varies by more than a factor of two, which reflects
Page 6-38
-------
the uncertainty in this approach. In addition, in some countries, emissions
associated with underground mining could lie outside the range if its mines are
significantly more or less gassy than coal mines in an "average" basin.
Country- or Basin-Specific: The Country- or Basin-Specific Method may be
somewhat more accurate than the Global-Average Method because it
incorporates additional information, potentially including actual CH4 emission
measurements from some mines or data on coal characteristics that might
provide insight into the gassiness of the coal. This information will not be
comprehensive, however, and there will be uncertainties related to its
representativeness and applicability in different parts of a country or coal basin.
Mine-bv-Mine: This method is likely to be the most reliable because it uses
emission data from individual mines. Even mine-specific emission data are
uncertain, however. The U.S. Mine Health and Safety Administration, for
example, estimates that its ventilation air estimates are accurate to +20%. In
addition, in some cases, available data may not accurately reflect total
emissions to the atmosphere, because of how the measurements were taken,
the lack of information on degasification systems, and other factors.
Surface Mining
Information on CH4 emissions from surface mines is generally unavailable. Two
methods can currently be applied -- the Global-Average and Basin-Specific methods ~ and
both are highly uncertain. U.S. EPA currently has a program underway to measure emissions
from surface mines, which should reduce some of the uncertainties associated with estimating
these emissions. Other countries, such as Australia, have also measured CH4 emissions from
surface mines.
Global Average: The key methodological uncertainty in this approach concerns
the use of a global emission factor range. As discussed with respect to
underground mining, global emission factors do not account for country-specific
factors that could affect CH4 emissions. To reflect this, the low and high ends
of the emission-factor range used vary by more than an order of magnitude.
Thus, while this method should provide a reasonable approximation of global
emissions, the results could vary substantially from national estimates.
Basin-Specific: Where the data are available to apply it, this method can
reduce the uncertainty by using basin-average gas contents. These gas
contents are multiplied by coal production and, if appropriate, then by a factor
representing the contribution of surrounding strata to emissions. Key
uncertainties are related to the reliability of average gas content values to entire
countries or coal basins and the difficulty of determining a reasonable
approximation of CH4 emissions from surrounding strata.
Page 6-39
-------
Post-Mining Operations
Like surface mines, CH4 emissions from post-mining coal transportation, storage, and
handling operations are uncertain because there are no available emission data. The Global-
Average Method uses highly speculative emission factor ranges for underground and surface
mines, and without initiation of a measurement program, the accuracy of these estimates
cannot be determined.
Combustion Processes
The principal uncertainty in the method of estimating combustion emissions is related
to the emission factors selected. Emission factors can vary significantly, depending on the
type of combustion unit and its history (i.e., operating and maintenance practices).
Furthermore, the range of measured emission factors is large, and for many coal-consumption
sectors, limited data are available. As a result, the emission estimates prepared to date for
coal consumption range from insignificant levels of less than 0.1 Tg (Barns and Edmonds,
1990) to levels of more than 10 Tg, which would constitute more than 25% of coal-cycle
emissions (Selzer and Zittel, 1990).
6.6.3 Research Needs
As the discussion of uncertainties has indicated, research in several areas could
significantly improve estimated CH4 emissions from the coal cycle. Some of the key areas for
future emphasis are:
Underground Mines: Accurate, mine-specific data, including emissions from both
ventilation and degasification systems, are needed. Because underground mining is
the largest emission source, efforts to improve the certainty of these emissions will be
very important to the development of reliable emission estimates.
Surface Mines: Expanded measurement programs at surface mines would significantly
reduce the uncertainty in emission estimates. To date, only a few countries have
undertaken CH4 measurement programs at surface mines, and information on key
issues, such as the degree to which surrounding strata contribute to emissions, is
sparse.
Post-Mining Activities: Expanded measurement programs during post-mining activities
could also significantly reduce uncertainty. Key questions to resolve involve the
amount of residual CH4 contained in coals after mining and the degree to which this
CH4 is emitted during post-mining activities, as opposed to being burned during
combustion.
Coal Combustion: Improved measurements of emissions from coal combustion are
critical, particularly related to industrial and domestic combustion units, which could be
significant sources of CH4.
Abandoned Mines and Other Sources: Emissions from other points in the coal cycle
currently cannot be estimated due to the lack of information. Measurements should be
undertaken at abandoned mines, coal waste piles, and other possible sources to
determine whether emissions from these sources are significant.
Page 6-40
-------
6.7 CONCLUSIONS
In 1990, CH4 emissions from the coal fuel cycle were an estimated 24-40 Tg, not
including the 1.3 Tg of CH4 that was recovered and used by coal mines. These emissions
represented approximately 10% of total anthropogenic CH4 emissions. Underground mining
was the largest source of CH4 emissions from the coal fuel cycle, accounting for an estimated
70-85% of the total emissions. Surface mining and post-mining activities (coal transportation,
storage, and handling) contributed 10-20% of emissions, and coal combustion contributed the
remaining 5-10%.
Methane emissions from the coal fuel cycle coulcl increase in the future if coal
production increases. Current projections are for coal production to increase by 10-25% by
2000 and perhaps by 25-50% by 2010. It is likely that this production will be associated with
much higher CH4 emissions from the coal fuel cycle, particularly if the proportion of
underground mining increases and if deeper, gassier seams are mined. The future will also
present additional opportunities to recover and use CH4 from coal mining, however, which
could serve to moderate the growth in emissions.
The emissions associated with many components of the coal fuel cycle are uncertain,
and additional research is necessary to better understand and quantify these sources.
Underground mining is the best understood of the sources, but additional measurements,
particularly for degasification system emissions, would improve estimates. The data on which
surface mining and post-mining emission estimates have been based are very sparse and
uncertain. Some measurements have been undertaken in the United States, and other
countries may also benefit from undertaking such research efforts. There are several sources,
moreover, for which emission estimates cannot currently be prepared due to lack of data,
including abandoned mines and coal waste piles. Additional research is warranted to better
understand these sources.
6.8 REFERENCES
Airuni, A.t. 1992 (March). Personal communication with AT. Airuni, Russian Institute of
Comprehensive Mineral Resources Exploitation, Moscow, Russia.
Barns, D.W., and J.A. Edmonds. 1990. An Evaluation of the Relationship Between the
Production and Use of Energy and Atmospheric Methane Emissions. TR047,
USDOE/NBB0088P, U.S. Department of Energy, Washington, D.C.
BCTSRE (British Coal Technical Service Research Executive). 1992. Quantification of
Methane Emissions From British Coal Mine Sources. Prepared for the Working Group on
Methane Emissions, The Watt Committee on Energy, United Kingdom.
Bibler, C.J., U.S. Marshall, and R.C. Pilcher. 1992. Assessment of the Potential for Economic
Development and Utilization of Coalbed Methane in Czechoslovakia. EPA/430/R-92/1008,
prepared for the U.S. Environmental Protection Agency, Office of Air and Radiation,
Washington, D.C.
Boyer, C.M. II, J.R. Kelafant, V.A. Kuuskraa, and K.C. Manger. 1990. Methane Emissions
From Coal Mining: Issues and Opportunities for Reduction. EPA-400/9-90/008, U.S.
Environmental Protection Agency, Office of Air and Radiation, Washington, D.C.
Page 6-41
-------
CIAB (Coal Industry Advisory Board). 1992. Global Methane Emissions in the Coal Industry.
Prepared for the Global Climate Committee, International Energy Agency, Paris, France.
Greedy, D.P. 1991. An introduction to geological aspects of methane occurrence and control
in British deep coal mines. Quarterly Journal of Engineering Geology 24:209-220.
Greedy, DP. 1992. Personal communication with D.P. Greedy. British Coal Corp., Coal
Research Establishment Technical Services, Gloucestershire, United Kingdom.
Crutzen, PJ. 1987. Geophysiology of Amazonia. In Dickenson, R., ed. Role of the Tropics in
Atmospheric Chemistry. John Wiley, New York, New York. 107-132.
FBIS (Foreign Broadcast Information Service). 1992. Joint Publications Research Service
(JPRS) Report: East Europe. 8 September 1992. Quoting from Wielkopolska, Anna. "Coal
Industry Production Reviewed; Forecast Made."
Hitchcock, D.R., and A.E. Wechsler. 1972. Biological Cycling of Atmospheric Trace Gases.
Final Report, NASW-2128. 177-154.
Hite, T. 1992. Personal communication with T. Hite, president, Hite Operating Co.,
Indianapolis, Indiana.
Hollingshead, B. 1990. Methane Emissions from Canadian Coal Operations: A Quantitative
Estimate. Coal Mining Research Company, Devon, Alberta.
IEA (International Energy Agency). 1990. Coal Information 1990. Paris, France.
IEA (International Energy Agency). 1992a. Energy Statistics and Balances of Non-OECD
Countries: 1989 - 1990. Paris, France.
IEA (International Energy Agency). 1992b. Energy Statistics of OECD Countries: 1989-1990.
Paris, France.
Jaques, A.P. 1992. Canada's Greenhouse Gas Emissions: Estimates for 1990. EPS 5/AP/4,
prepared by Environment Canada.
KfA. 1993. Personal communication with W. Katscher, KfA, Julich, Germany.
Kim, A.G. 1977. Estimating Methane Content of Bituminous Coalbeds from Adsorption Data.
RI 8245, U.S. Department of the Interior, Bureau of Mines, Pittsburgh, Pennsylvania.
Kirchgessner, D.A., S.D. Piccot, and J.D. Winkler. 1993. Estimate of global methane
emissions from coal mines. Chemosphere26('\-4YA53-472.
Klssell, F.N., C.M. McCulloch, and C.H. Elder. 1973. The Direct Method of Determining
Methane Content of Coalbeds for Ventilation Design. RI 7767, U.S. Department of the Interior,
Bureau of Mines, Pittsburgh, Pennsylvania.
Koyama, T. 1963. Gaseous metabolism in lake sediment and paddy soils and the production
of atmospheric methane and hydrogen. Journal of Geophysical Research 68(13):3,971.
Page 6-42
-------
Koyama, T. 1964. Biogeochemical studies on lake sediments and paddy soils and the
production of hydrogen and methane. In Miyake, Y., and T. Koyama, eds. Recent Researches
in the Fields of Hydrosphere, Atmosphere, and Nuclear Geochemistry. Muruzen, Tokyo,
Japan. 143-177.
Lama, R.D. 1991. Methane gas emissions from coal mining in Australia: Estimates and
control strategies. In Proceedings of the IEA/OECD Conference on Coal, the Environment and
Development: Technologies to Reduce Greenhouse Gas Emissions. IEA/OECD (International
Energy Agency/Organization of Economic Cooperation and Development), Paris, France. 255-
266.
Marshall, J.S., R.C. Pilcher, and D.W. Kruger. 1993. Assessment of the Potential for Economic
Development and Utilization of Coalbed Methane in Russia and Ukraine. Draft report prepared
for the U.S. Environmental Protection Agency, Office of Air and Radiation, Washington, D.C.
MOE (Ministry of Energy, Department of Safety and Environmental Protection). 1991. Personal
communication with Li Xuecheng, Director of Safety and Environmental Protection, Beijing,
People's Republic of China.
Mitra, A.P. 1992 (August). Greenhouse Gas Emission in India: 1992 Update. Scientific Report
No. 4, prepared under the auspices of the Council of Scientific and Industrial Research and
Ministry of Environment and Forests.
OECD (Organization of Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France.
OECD (Organization of Economic Cooperation and Development). 1993. Proposed revised
draft chapter for OECD report ~ Methane emissions from coal mining. Prepared by expert
group at the Methane and Nitrous Oxide Workshop, 3-5 February 1993, Amersfoort, the
Netherlands.
Okken, P.A., and T. Kram. 1989. CH4/CO - Emission from fossil fuels global warming
potential. ESC-WR-89-12. Paper presented at the IEA/ETSAP - Workshop, June 1989, Paris,
France. Energy Study Centre: Netherlands Energy Research Foundation, Petten, the
Netherlands.
Pilcher, R.C., CJ. Bibler, R. Glickert, L. Machesky, and J.W. Williams. 1991. Assessment of
the Potential for Economic Development and Utilization of Coalbed Methane in Poland.
EPA/400/1-91/032, U.S. Environmental Protection Agency, Washington, D.C.
Polish Central Mining Institute. 1990. Official Polish Methane Emissions Data for 1989.
Provided to Raven Ridge Resources, Grand Junction, Colorado.
Radian. 1990. Emissions and Cost Estimates for Globally Significant Anthropogenic
Combustion Sources of NOX, N2O, CH4, CO, and CO2. Prepared for the U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Page 6-43
-------
Saghafi, A., and D. Williams. 1992. Estimates of methane emissions from Australian coal
mines. Paper presented at symposium sponsored by Institute of Minerals, Energy, and
Construction, Division of Coal and Energy Technology, on Coalbed Methane Research and
Development, November 1992, Townsville, Queensland, Australia.
Seller, W. 1984. Contribution of biological processes to the global budget of CH4 in the
atmosphere. In Klug, M., and C. Reddy, eds. Current Perspectives in Microbial Ecology.
American Society for Microbiology, Washington, D.C. 468-477.
Selzer, H., and W. Zittel. 1990. Emissions of Methane Gas Affecting the Climate: An
Investigation of Global Methane Emissions from the Use of Fossil Fuels, Waste Incineration,
Agriculture, and the Earth. LBST-Report No. 10/90, Ludwig-Bolkow-Systemtechnik GmbH,
Ottobrunn, Germany.
Smith, I.M., and L.L. Sloss. 1992. Methane Emissions from Coal: IEA Perspectives.
IEAPER/04, International Energy Agency, London, United Kingdom.
Stewart, B. 1990. Personal communication with A. Jaques, Pollution Data Analysis Division,
Environment Canada, Ottawa, Ontario.
Surowka, J., and S. Pasierb. 1992. Environmental Impact of Larger Utilization of Coalbed
Methane in the Upper Silesia Region. FEWE/K/25/92, Polish Foundation for Energy Efficiency,
Katowice, Poland.
U.S. DOE/EIA (Department of Energy/Energy Information Administration). 1992a. Energy
Consumption and Conservation Potential: Supporting Analysis for the National Energy
Strategy. DOE/EIA UC-98, EIA, Office of Energy Markets and End Use, Washington, D.C.
U.S. DOE/EIA (Department of Energy/Energy Information Administration). 1992b. International
Energy Outlook. DOE/EIA-0484(92), EIA, Washington, D.C.
U.S. EPA (Environmental Protection Agency). 1993. Anthropogenic Methane Emissions in the
United States: Report to Congress. Prepared by Office of Air and Radiation, Washington,
D.C.
Watt. 1993. Methane Emissions. Papers presented at the 29th Consultative Conference of
The Watt Committee on Energy, 28 January 1993, Royal Geographical Society, London,
United Kingdom.
Williams, A., and C. Mitchell. 1992. Methane Emissions From Coal Mining. Department of Fuel
and Energy, The University of Leeds, Leeds, United Kingdom. In preparation.
Zimmermeyer. 1991. Recovery and Use of Coalbed Methane (Replies to the Questionnaire
Transmitted by the Federal Republic of Germany). United Nations Economic Commission for
Europe.
Page 6-44
-------
CHAPTER 7
MINOR INDUSTRIAL SOURCES OF METHANE
7.1 SUMMARY
Over the past several years research efforts on the global methane (CH4) budget have
focused on the largest anthropogenic and natural sources and sinks of methane. Recent
research indicates there is a large and diverse group of minor industrial sources that also emit
methane into the atmosphere (Lacroix, 1993; Piccot et al, 1990b; and Saeger et al., 1989).
Individually, these minor sources emit small quantities of methane, but their collective
contribution to the global budget may be significant. Global-scale emission estimates have
recently been developed for some minor source categories, but additional research will be
needed to identify and characterize all source categories which may exist. As a result, there
is a high degree of uncertainty associated with current global estimates.
There are hundreds of individual industrial sources that emit some quantity of methane
into the atmosphere. In this chapter, emissions are estimated for six minor source categories:
peat mining; coke, iron, and steel production processes; miscellaneous industrial processes;
paper and printing processes; and chemical manufacturing (excluding ethanol and charcoal
production). Peat mining and coke, iron, and steel production processes are the two most
significant minor sources discussed in this chapter. Together, these two categories are
estimated to account for about 3.3 Teragrams (Tg)1 per year of methane or approximately
87% of all the emissions from the minor sources examined here. Other potentially significant
minor sources include miscellaneous industrial processes, paper and printing processes, and
chemical manufacturing. All five source categories discussed in this chapter are estimated to
account for about 3.6 Tg/year of methane.
Changes in the technology, economic activity, environmental regulations, and other
factors can significantly influence the rate of emissions from many of the industrial sources
examined here. Thus, some understanding of country-specific manufacturing, refining, and
other processes must be developed before any assessment of how technological and other
processes variations may influence current and future emissions from specific source
categories. It is clear that research is needed to: (1) identify all significant minor industrial and
other sources of methane emissions, (2) develop generic methane emission factors for those
sources, (3) develop country-specific methane emission factors for those sources that are
representative of the process variations and emissions control differences that exist among
countries, and (4) develop representative country- and source-specific production and other
activity data needed to represent the activity levels of the individual sources.
7.2 BACKGROUND
Over the past several years researchers have conducted numerous studies to identify
and characterize the principal sources and sinks of atmospheric methane (CH4). Logically,
these studies have focused on the largest anthropogenic and natural sources and sinks of
1Teragram = 106 metric tonnes = 1O12 grams.
Page 7-1
-------
CH4. Research on anthropogenic sources has concentrated on ruminants, rice paddy fields,
natural gas production and distribution, coal mining, biomass burning, and landfills, as
described in other chapters of this document. However, there is a large and diverse group of
minor industrial sources that also emits CH4 into the atmosphere. These minor sources have
not been included in most global CH4 budget estimates to date. Individually, they emit small
quantities of CH4, but collectively their contribution to the global budget may be significant.
These sources include activities such as coke production and peat mining.
Recently, global-scale emission estimates have been developed for some of the minor
source categories (Piccot and Beck, 1993; Berdowski et al., 1993; Lacroix, 1993; Piccot et al.,
1990b; Blake, 1984; and Darmstadter et al., 1984). Since very little research has been done
to identify and characterize these source categories, this chapter presents preliminary data
that are based on limited measurements and may contain large uncertainties. In addition, it is
likely that not all minor sources of CH4 have been identified.
This chapter identifies several minor source categories that have not been included in
previous chapters and discusses the research that has been done to estimate emissions from
them. Specifically, this chapter includes the following minor source categories: peat mining,
coke production and iron and steel production processes, miscellaneous industrial processes,
paper and printing processes, and chemical manufacturing. The chapter discusses and
recommends emission estimation methodologies and presents global and country-specific
emission estimates.
7.2.1 Recent Research
Several researchers have recently performed work to identify, characterize, and
estimate emissions from minor sources of CH4. However, since there has been no targeted
effort to identify and characterize all minor sources of CH4 emissions, it is likely that the
sources identified so far represent an incomplete listing. This section provides a brief
overview of this body of research and identifies source categories and emissions estimates.
National Acid Precipitation Assessment Program (NAPAP)
During the 1980s, the United States and Canada completed a joint research effort to
develop representative emission inventories for the United States and Canada for
anthropogenic and natural sources of acid rain precursors as a component of the National
Acid Precipitation Assessment Program (NAPAP). Although the primary focus of this effort
was on more conventional pollutants, such as sulfur oxides, nitrogen oxides, and volatile
organic compounds, preliminary data on CH4 emissions from specific industrial and other
activities were reported as well (Saeger et al., 1989).
The NAPAP emission inventory contains detailed information on hundreds of individual
sources that may emit some quantity of CH4 to the atmosphere. Table 7-1 presents a ranked
listing of the most significant CH4 source categories in the United States (as estimated by
NAPAP) and identifies the primary emission sources associated with each category. Although
NAPAP clearly ignores the more traditional sources of CH4, such as ruminants, coal mines,
and rice production, it does identify and characterize several nontraditional CH4 sources. As
Table 7-1 shows, CH4 can be emitted from a wide range of sources, including various
residential wood- and waste-combustion activities, mobile sources, fuel production and refining
operations, industrial fossil fuel-fired equipment, and various industrial processes, such as
Page 7-2
-------
TABLE 7-1
RANKED LISTING OF SIGNIFICANT INDUSTRIAL SOURCES OF
METHANE EMISSIONS IN THE UNITED STATES3
Source Category
Significant Emission Sources
Residential Fuelwood Combustion
Mobile Sources
Residential On-Site Waste Burning
Mobile Sources
Petroleum Refining
Oil- and Gas-Production Processes
Internal Industrial and Other Gas
Combustion
Industrial Waste Treatment, Storage, and
Disposal Facilities
Miscellaneous Industrial Manufacturing
Processes
Other External Industrial Combustion
Woodstoves and Fireplaces
Light Duty Vehicles
Household Trash Burning
Medium and Heavy Duty Vehicles
Catalytic Cracker and Process Heater Exhausts
Gas Flares and Incinerators
Equipment Leaks and Waste Water Fugitives
Process Vents
Crude Oil and Product Loading and Storage
Compressors
Wellheads and General Fugitive Emissions
Stack Gases from Gas-Fired Turbines
Stack Gases from Gas-Fired I.C. Engines
Stack Gases from Utility, Industrial, and Other Boilers
Waste Combustion Facilities
Industrial Waste Ponds
Miscellaneous Other Waste Treatment/Disposal Sites
Organic Chemical Manufacturing Operations
Coke and Commercial Charcoal Manufacturing Operations
Industrial Waste Incineration
Various Manufacturing Operations
Stack Gases from Wood-Fired Boilers
a Ranked by significance based on U.S. data (Saeger et al., 1989).
chemical manufacturing and industrial waste disposal operations. Based on the NAPAP
emission inventory, the most significant minor sources of CH4 in the United States not
included in other chapters of this document are the miscellaneous industrial manufacturing
processes, such as organic chemical manufacturing operations, coke manufacturing
operations, and other various manufacturing operations. The most significant chemical
manufacturing operations in the United States appear to be processes involved in the
production of carbon black, charcoal, various alcohols (e.g., ethanol and isopropanol), and
various feedstock chemicals (e.g., ethylene and ethylene oxide).
Page 7-3
-------
Summary of Reported Global Emissions Estimates
Only a few attempts have been made to develop emissions estimates for selected
minor industrial sources of CH4. An early attempt to estimate CH4 emissions from sources
located in urban areas was conducted by Blake (1984). In Blake's analysis, a limited number
of ambient air samples collected in several cities were observed to routinely exhibit elevated
levels of CH4. Using these measurements, an emission flux rate for the world's cities was
calculated (0.06/m2/day) and then multiplied by an estimate of the land-surface area covered
by the world's cities. Based on these rather crude calculations, Blake estimated that non-
automobile-related emissions from the world's urban areas were about 12 Tg/year in the early
1980s. Urban CH4 sources can include a variety of industrial and other activities examined in
this chapter.
In a study by Piccot et al. (1990b), research was conducted to identify and quantify the
CH4 emissions from a wide range of previously uncharacterized sources. Most of the source
types examined in this study included industrial processes, such as coke production facilities,
petroleum refineries, paper- and metal-coating operations, printing operations, gasoline
storage and marketing operations, organic-chemical manufacturing operations, residential
wood and waste burning, and fossil fuel combustion. Global emissions for these sources
were estimated using source-specific emission factors (i.e., CH4 emissions per unit of source-
specific activity) that were developed from U.S. emission data contained in the NAPAP
inventory discussed earlier. For example, country-specific emissions from petroleum refineries
were estimated by multiplying an individual country's refinery crude oil throughput by a U.S.-
based emission factor that quantifies the CH4 emission rate per quantity of crude oil
throughput at U.S. refineries. Because the CH4 emission factors were developed based on
the emissions from industrial and other operations in the United States, differences in the
types of industrial processes, product characteristics, and pollution control equipment used in
other countries are not represented in these estimates. The effects of these differences on
emission estimates are unknown, but they could be significant for some source categories. In
addition, it should be emphasized that data quality ratings for the emission factors used were
reported by Piccot et al. to be very low (i.e., highly uncertain), emphasizing the need for
additional research on emission factor development.
Global estimates of CH4 emissions from several minor industrial sources have also
been developed by Lacroix (1993). Source-category emissions estimated by Lacroix include
petrochemicals (2.0 Tg/year ± 50%) and peat mining (2.0 Tg/year + 50%). Total emissions
from these two sources are 4.0 Tg per year. To generate the global estimate of CH4 from
peat mining, Lacroix multiplied an estimate of world peat production by the CH4 content of
peat, as reported by Glotov et al. (1985). Lacroix's estimate of CH4 emissions from peat
mining includes only the mining process itself. As with coal mining, methane may be emitted
from the peat mining process as the peat is disturbed and the methane migrates from the peat
into the atmosphere. As identified by Lacroix, the complete cycle of peat production includes
dewatering, harvesting, and briquetting, all of which contribute to CH4 emissions. Specific
details of the processes producing emissions from these activities are not yet well-understood
and well-quantified. As such, Lacroix suggests that his emission estimate for peat mining is
only a crude approximation and may represent a minimum range.
At a conference on global climate issues held in the Netherlands in February 1993,
Piccot and Beck (1993) and Berdowski et al. (1993) presented the results of their recent
studies on minor sources of CH4. In their study, Piccot and Beck used 1990 activity data to
develop CH4 emission estimates for a 1990 base year and organized the source categories
Page 7-4
-------
into logical groups. Estimates of the global CH4 emissions from selected minor source
categories examined by Piccot and Beck are summarized in Table 7-2, along with the type of
activity data used to estimate emissions. Additional categories studied by Piccot and Beck
have been discussed in other chapters of this document. The most globally significant minor
source categories include peat mining, and coke, iron, and steel production processes.
(Emissions estimates for peat mining were based on Lacroix (1993).) Other significant
TABLE 7-2
1990 GLOBAL CH4 EMISSIONS ESTIMATED FOFt SELECTED MINOR SOURCES
Source Description
Peat Mining
Coke Production and Iron
Source-Specific Activity
Parameter Used
Peat production
Coke production
Global Emissions
(Tg/year)
2.0
0.3
and Steel Production
Processes
Miscellaneous Industrial
Processes
Paper and Printing Processes
Chemical Manufacturing
Processes
Value added in manufacturing 0.2
Kraft paper, newsprint, and 0.1
other paper production
Ethylene and propylene production 0.1
Total Emissions
2.7
Source: Piccot and Beck, 1993.
source groupings in Table 7-2 include miscellaneous industrial processes, paper and printing
processes, and chemical-manufacturing processes (including synthetic organic chemical
manufacturing and carbon black operations). In the paper and printing category, kraft-paper
manufacturing processes represent approximately 20% of the category's emissions, and
printing processes represent approximately 20% of the category's emissions. In the chemical
manufacturing category, organic-chemical manufacturing represents over 50% of the
category's emissions, and carbon-black production processes represent approximately 25% of
the category's emissions. The group of sources found to emit negligible amounts of CH4
include diesel vehicles, paper-coating operations, solvent use, and asphalt paving. Together,
the five source categories included in Table 7-2 were estimated by Piccot and Beck to
contribute 2.7 Tg per year to the global budget in 1990.
Berdowski et al. estimated methane emissions from fuel combustion, including non-
commercial fuels, and industrial noncombustion sources (the "industrial processes"). Industrial
noncombustion sources and processes included production of iron and steel (coke), oil
refining, and production of carbon black, ethylene, dichloroethene, styrene, and methanol.
Table 7-3 compares the global emission estimates from Piccot and Beck (1993) and
Berdowski et al. (1993) for the minor industrial source categories. Since fossil fuel combustion,
Page 7-5
-------
residential on-site incineration, and oil refining are included elsewhere in this document, their
emission estimates are not presented in Table 7-3.
TABLE 7-3
SELECTED MINOR SOURCE CATEGORIES AND EMISSIONS ESTIMATES
Piccot and Beck
Berdowski et alT
Source Category
Emissions
(Tg/yr)
Source Category
Emissions
(Tg/yr)
Peat Mining
2.0
na
Coke, Iron, and Steel Production 0.3
Processes
Iron and Steel Production,
Including Coke
Miscellaneous Industrial
Processes
0.2
na
Paper and Printing Processes
Chemical Manufacturing
Total Emissions
0.1
0.1
2.7
na
Chemical Manufacturing
Total Emissions
0.2
2.2
Sources: Piccot and Beck, 1993; and Berdowski et al., 1993.
7.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The approach recommended for estimating country-specific emissions from individual
source categories is the same as the general method adopted by the Intergovernmental Panel
on Climate Change (IPCC) (OECD, 1991). In this method, the first step is to multiply source-
specific emission factors by some activity data that represent a surrogate for the level of
activity associated with the emission source. However, several important factors must be
considered before using this general emission estimation method. First, emission factors for
sources can vary significantly among countries as a result of differences in processes,
technology vintages, fuel compositions, operational practices, and environmental policies.
Therefore, some effort should be made to use emission factors that are representative of the
individual countries to which they are applied. This is of particular concern with the types of
industrial process and other sources examined here because there is a great potential for
variations to exist in the manufacturing process and emission controls associated with these
sources. For example, different coke production processes can yield significantly different
coke-oven gas compositions. The low-temperature process favored in Europe produces coke-
oven gas containing a concentration of CH4 which is two times higher than that produced by
the high-temperature process favored in the United States (Shreve and Brink, 1977). Since
coke-oven gas is a principal source of CH4 emissions at coke production facilities,
assumptions made about the types of processes used in a country can significantly affect
emission estimates.
Page 7-6
-------
Since very little research has been devoted to examining the CH4 emissions from
minor industrial sources, very limited data are available for developing emission estimates,
and country-specific estimates have not been reported in the literature. In a recent study
conducted for the IPCC, it was concluded that industrial-process-level data for greenhouse
gases are so limited that a proposed emission estimation methodology could not be
developed for most industrial processes (OECD, 1991). Process-specific CH4 emission
factors for industrial processes either have not been developed or are highly suspect. For
example, in recent studies, CH4 emission factors for a limited number of industrial process and
other minor CH4 emission sources were presented but were characterized as highly uncertain
(OECD, 1991; Piccot et al., 1990a).
Preliminary country-specific estimates of CH4 emissions were developed for selected
minor source categories for this chapter using the global emission estimates reported by
Piccot and Beck (1993) and Berdowski et al. (1993). These five minor source categories are
peat mining, coke-production and iron-and-steel production processes, miscellaneous
industrial processes, paper and printing processes, and chemical manufacturing. Country-
specific emission estimates were not developed for categories discussed elsewhere in this
document. The estimation and allocation procedures used to generate county-specific
emissions for each category are discussed in the following paragraphs.
7.3.1 Peat Mining
Total global emission estimates for peat mining developed by Lacroix (1993) were
apportioned to individual countries, based on estimates of peat production for each country.
Data on peat mining activity were obtained from the U.N. Energy Statistics Yearbook (UN,
1992a).
7.3.2 Coke, Iron, and Steel Production Processes
This category includes emissions from all production of metallurgical coke, including
coke-oven charging and pushing operations, coke-quenching operations, coke-oven door
leaks, coke-oven gas combustion, and other significant sources. Emissions associated with
all significant iron-and-steel production processes, such as sintering and blast furnace
operations, are also included. Finally, emissions from fluid coking processes at refineries,
although very minor in extent, are also included.
An overall emission factor for this category (3.2 tonnes/106 tonnes coke produced) was
calculated by averaging the emission factor developed using the NAPAP data and the
emission factor presented by Berdowski et al. The NAPAP-based emission factor was
developed for coke production and iron-and-steel production processes by dividing the total
1985 NAPAP CH4 emissions from these activities by 1985 coke production in the United
States. Darmstadter et al. (1984) estimated emissions from coke-manufacturing processes to
be 6 Tg. The corresponding emission factor was not included in the averaging here for
several reasons: (1) the age of the data used for the 6 Tg estimate; (2) the estimate
represented only coke production and did not include iron-and-steel production processes; and
(3) the estimate was extremely high when compared with other emission estimates for this
category. Coke production was used as the activity statistic for this category (UN, 1988 and
1992b).
Page 7-7
-------
7.3.3 Miscellaneous Industrial Processes
The overall emission factor for this category (0.0676 tonnes/109 dollars manufacturing
value added) was calculated using the NAPAP data approach described for coke production.
Value added in manufacturing for each country was used as the activity statistic for this
category (UN, 1988 and 1992b). The NAPAP data do not allow for a specific identification of
the industrial processes included under this category. However, the majority of emissions
from this category are process-related emissions.
Clearly, there is a significant potential for variations to occur among countries in the
emissions from industrial processes. However, because of the wide range of different
processes included in this category and the limited country-specific data available on CH4
emissions from industrial processes, it was concluded that such a refined estimation approach
was unjustified.
7.3.4 Paper and Printing Processes
This category primarily includes kraft-paper manufacturing processes and printing and
publishing operations. The overall emission factor for this category (0.787 tonnes/106 tonnes
paper produced) was calculated using the NAPAP data approach described for coke
production. The activity statistic for this category was production of kraft paper and
paperboard, other machine-made paper, other printing and writing paper, and newsprint.
As with the miscellaneous industrial processes category, there is a significant potential
for country-specific variations in the emissions from industrial processes. Because of the
range of different processes included in this category and the limited country-specific data
available on CH4 emissions from industrial processes, it was concluded that a more refined
estimation approach was unjustified.
7.3.5 Chemical Manufacturing Processes
The dominant sources of CH4 in the chemical manufacturing processes category are
fugitive leaks from manufacturing operations, carbon-black manufacturing, and the production
of various alcohols, ethylene oxide, ethylene, ammonia and methanol. The overall emission
factor for this category (1.846 tonnes/106 tonnes feedstock chemical produced) was calculated
using the NAPAP data approach described for coke production. The activity statistic for this
category was production of basic feedstock chemicals (i.e., ethylene and propylene). (This
category specifically does not include ethanol production or charcoal production.)
As with the miscellaneous industrial processes category, there is a significant potential
for country-specific variations in the emissions from industrial processes. Because of the
range of different processes included in this category and the limited country-specific data
available on CH4 emissions from industrial processes, it was concluded that a more refined
estimation approach was unjustified.
7.4 RESULTS
Country-specific CH4 emission estimates developed as described in this chapter are
presented in Table 7-4 for the five minor source categories. The balance of the global
emissions for each source category are reported as the "rest of world" at the bottom of each
Page 7-8
-------
TABLE 7-4
METHANE EMISSIONS FROM SELECTED MINOR SOURCES (1,000 tonnes/yr)
Country
Former Soviet
Union
Finland
Ireland
United States
Japan
China
(mainland)
Former
West Germany
Poland
France
Former
Czechoslovakia
United Kingdom
Italy
India
Brazil
Canada
South Korea
Belgium
Spain
Australia
Romania
North Korea
Turkey
Netherlands
Sweden
Mexico
Austria
Former
Yugoslavia
South Africa
Customs Union
Peat
Mining
968.7
555.6
432.5
0.0
0.0
0.0
32.3
0.0
0.0
8.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Coke
Production
257.8
1.6
0.0
84.6
153.3
196.6
58.0
54.8
23.5
34.0
26.8
20.5
31.4
24.6
11.9
27.1
17.5
10.3
14.6
14.5
11.6
10.4
8.8
3.5
6.4
5.5
7.3
6.3
Misc.
Industrial
Processes
0.0
0.9
0.5
63.0
33.7
0.0
16.1
0.0
10.0
0.0
7.6
7.5
2.9
3.9
4.6
3.0
1.2
3.1
2.1
0.0
0.0
1.3
1.7
1.5
1.5
1.5
1.0
0.7
Paper and
Printing
Processes
0.0
6.4
0.0
49.5
19.2
3.2
8.0
0.5
5.0
0.7
3.3
4.0
1.4
3.0
12.6
0.0
0.8
1.6
0.9
0.6
3.1
0.3
2.0
6.2
1.7
2.2
0.5
1.4
Chemical
Manuf.
8.1
0.0
0.0
48.8
18.5
2.6
9.0
0.9
6.8
1.7
4.2
4.2
0.4
4.3
5.9
3.1
0.0
3.0
0.0
0.8
0.0
1.2
0.0
0.8
1.8
0.9
0.7
0.0
Total
Emissions
1,234.7
564.5
433.0
246.0
224.7
202.4
123.4
56.2
45.3
45.0
41.9
36.2
36.1
35.8
35.1
33.2
19.5
18.0
17.6
15.8
14.7
13.2
12.7
12.0
11.4
10.1
9.5
8.4
[continued]
Page 7-9
-------
TABLE 7-4
METHANE EMISSIONS FROM SELECTED MINOR SOURCES (1,000 tonnes/yr)
(Continued)
Country
Former
East Germany
Argentina
Rest of the
World
Total
Peat
Mining
0.0
0.0
2.2
2,000.0
Coke
Production
3.5
3.1
29.7
1,159.5
Misc.
Industrial
Processes
0.0
1.1
17.3
187.7
Paper and
Printing
Processes
0.3
0.6
10.8
150.0
Chemical
Manuf.
1.2
0.0
9.7
138.6
Total
Emissions
5.0
4.8
69.8
3,635.8
column. From the data in Table 7-4, it is estimated that total global emissions from these five
minor source categories were approximately 3.6 Tg in 1990. Although this estimate should be
considered highly uncertain, it is not possible to develop a representative uncertainty range
until some additional basic research is conducted.
The most significant minor global sources appear to be peat mining (2.0 Tg) and coke,
Iron, and steel production processes (approximately 1.2 Tg). Together, these two source
categories account for approximately 87% of all the CH4 emissions from the minor source
categories. Emissions from the remaining categories account for about 0.5 Tg.
On a country-specific basis, the former Soviet Union contributes over 1.2 Tg of the
total 3.6 Tg of CH4 from these five source categories. Emissions from the next three highest
contributing countries are 0.6 Tg from Finland, 0.4 Tg from Ireland and 0.2 Tg from the United
States. Together, these four countries contribute approximately 68% of total global emissions
from the five minor source categories. The 30 countries listed in Table 7-4 contribute over
98% of total global emissions from the five minor source categories.
7.5 TRENDS
Without a sound technical understanding of current CH4 emissions and emission
factors for specific processes and countries, it is difficult to examine how future changes in
technology, economic activity, environmental regulations, or other factors may affect future
emissions from the sources examined here. Therefore, this section presents a qualitative
discussion of technological and other factors which could affect future emissions from the
major sources examined.
i
Emissions trends for peat production will likely depend on the energy market.
Changing from conventional energy sources to alternative energy sources, such as peat, is
unlikely when the price of conventional fuels is low.
Coke production processes were also identified as a significant minor industrial source
of CH4 emissions. For this category, emissions from the coking process and the combustion
of coke-oven gas are the most significant contributors to emissions. This is not surprising,
Page 7-10
-------
since the major constituents of coke-oven gas are CH4 (22-32% in the United States) and
hydrogen (U.S. EPA, 1987). Coke is manufactured by carbonizing coal (i.e., heating coal in
an oxygen depleted atmosphere) and is one of the basic materials used in the conversion of
iron ore into iron and steel. In the coking process, emissions of CH4 and various other
organic compounds escape into the atmosphere when the coke ovens are opened and
charged with fresh coal and when inadequate seals exist and emissions escape from
around leaking coke-oven doors and other openings. Since leaking doors are the most
significant emission source of coke-oven gas, the age, design, and operation of the process
can have a significant impact on the emissions released (U.S. EPA, 1987).
Future emissions from coke production processes can be significantly altered,
depending on the operation and maintenance procedures used, the age and state of repair of
the equipment, and new technology developments. The U.S. Environmental Protection
Agency (U.S. EPA) conducted a study that estimated that if a comprehensive leak-control
program were initiated at U.S. coke facilities, and if coke-oven seals were modified and
charging operations were staged, emissions in the United States could be reduced by about
40% (U.S. EPA, 1987). The study also estimated that if existing U.S. coke-oven batteries
were rebuilt, emissions would be reduced by 86%. Clearly, there is a potential for existing
coke-producing facilities to reduce emissions in the future, but it is likely that the steps just
mentioned would only be taken in response to pending environmental regulation. Current
regulations for this industry are a result of concern over the carcinogenic effects associated
with coke-oven gas emissions. Without some understanding of the current vintages of coke
production facilities in specific countries and the environmental regulations which apply to
those facilities, it is not possible to estimate accurately how the types of technological and
operational influences discussed here could affect future emissions.
Except as noted earlier, it is unlikely that CH4 emissions associated with the production
of coke and the combustion of coke-oven gas will be reduced in the future, unless the actual
amount of coke required is reduced and/or if technological changes in the coking process
occur. It is possible to adopt a process that reduces coke consumption by using high-grade
iron ores and supplemental fuels in the ore reduction process. For example, in the United
States, the amount of coke produced per megagram of pig iron produced was reduced by
about 40% between 1956 and 1983 as a result of these factors. There is also an innovative
process modification that is being developed, whereby coke quenching occurs within the oven
rather than outside in a coke-quench car. This modification could reduce the amount of hot-
quench gases which escape into the atmosphere during the coke production process.
Miscellaneous industrial processes and various chemical manufacturing processes
were also identified as significant minor industrial sources of CH4 emissions. The substantial
number of different processes associated with these two emission categories is too broad to
discuss here. However, there is an enormous potential for changes in many of these
processes to occur in the future, which could significantly affect CH4 emissions. For example,
two different chemical manufacturing processes could be used to produce the intermediate
chemical cumene, which is used in the production of phenol and acetone. One process uses
a phosphoric acid catalyst, and CH4 is emitted from the distillation towers. The emissions of
all hydrocarbons from a "typical" plant are 14.5 kg/hour. In another process, an aluminum
chloride catalyst is used and the hydrocarbon emissions increase to 50 kg/hour, but the
emissions of CH4 from the distillation towers are eliminated (U.S. EPA, 1980). As was the
case with coke manufacturing processes, some understanding of country-specific
manufacturing and refinery processes must be developed before any assessment of how
technological and processes variations may influence future emissions.
Page 7-11
-------
7.6 CONCLUSIONS
In 1990, the total global emissions from the five minor source categories examined in
this chapter are estimated to be 3.6 Tg. Although the data used to develop this estimate
contain significant uncertainty, they suggest that the combined emissions from these minor
sources may account for a portion of the "missing methane" identified in recent global budget
studies.
It is clear from the results presented here that basic research is needed to develop an
improved understanding of the contribution of minor anthropogenic sources to the global CH4
budget. The first, and perhaps most important, step in this research should be to develop a
complete list of process and other minor sources of CH4 since this has not yet been done
comprehensively. This list should be highly process-specific to facilitate the development of
the most specific and appropriate emission factors and activity data possible. The next step
should be to develop "generic" emission factors for all sources and to estimate total global
emissions using these factors. This preliminary global estimate should only be used to
identify the most significant sources upon which future research should be focused. For the
most significant sources, research should then be conducted to develop country-specific
emission factors that are representative of the individual countries to which they are applied.
This is needed because the types of industrial processes and other sources examined here
have a great potential to exhibit significantly different emission characteristics as a result of
differences in manufacturing processes and emissions controls used in different countries.
The preliminary results presented here suggest that research on country-specific emission
factors should be focused on coke production processes, various organic-chemical
manufacturing operations, and petroleum-refinery operations.
7.7 REFERENCES
Berdowski, J., J. Olivier, and C. Veldt. 1993. Methane from fuel combustion and industrial
processes. In van Amstel, A.R., ed., Methane and Nitrous Oxide, Methods in National
Emissions Inventories and Options for Control, Proceedings of an International IPCC
Workshop, 3-5 February 1993, Amersfoort, the Netherlands. RIVM, Bilthoven, Netherlands.
131-141.
Blake, D.R. 1984. Increasing Concentrations of Atmospheric Methane, 1979-1983. Ph.D.
Thesis, University of California, Irvine, California.
Darmstadter J., L. Ayres, R.U. Ayres, W.C. Clark, P. Crosson, P. Crutzen, T.E. Graedel, R.
McGill, R.F. Richards, and J.A. Tarr. 1984. Impacts of World Development on Selected
Characteristics of the Atmosphere: An Integrative Approach. Volume 2 - Appendices. Oak
Ridge National Laboratory Report ORNL/Sub/86-22033/1/V2, Oak Ridge National Laboratory,
Oak Ridge, Tennessee.
Glotov, V., V.V. Ivanov, and N.A. Shilo. 1985. Migration of hydrocarbons through permafrost
rock. Transactions (Doklady) of the U.S.S.R. Academy of Science (Earth Science Sections)
285:192-194.
lannacchione, AT., R.H. Grau, A. Sainato, T.M. Kohler, and S.J. Schatzel. 1984.
Assessment of Methane Hazards in an Anomalous Zone of a Gulf Coast Salt Dome. Report
No. RI-8861, U.S. Bureau of Mines, Pittsburgh Research Center, Pittsburgh, Pennsylvania.
Page 7-12
-------
Lacroix, A.V. 1993. Unaccounted-for sources of fossil and isotopically-enriched methane and
their contribution to the emissions inventory: A review and synthesis. Chemosphere 26(1-
4):507-557.
OECD (Organization For Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from the OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Piccot, S.D., and L. Beck. 1993. Development of a Screening CH4 Emissions Inventory for
Minor Industrial Sources. Paper presented to the working group on methane emissions from
fuel combustion and industrial processes at the International IPCC Workshop on Methane and
Nitrous Oxide, 3-5 February 1993, Amersfoort, Netherlands.
Piccot, S.D., J.A. Buzun, and H.C. Frey. 1990a. Emissions and Cost Estimates for Globally
Significant Anthropogenic Combustion Sources of NOX, A/,,0, CH4, and CO2. EPA-600/7-90-
010, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina.
Piccot, S.D., A. Chadha, J. DeWaters, T. Lynch, P. Marsosudiro, W. Tax, S. Walata, and J.D.
Winkler. 1990b. Evaluation of Significant Anthropogenic Sources of Radiatively Important
Trace Gases. EPA-600/8-90-079 (NTIS PB91-127753), U.S Environmental Protection Agency,
Research Triangle Park, North Carolina.
Saeger, M.S., J. Langstaff, R. Walters, L. Modica, D. Zimmerman, D. Fratt, D. Dulleba, R.
Ryan, J. Demmy, W. Tax, D. Sprague, D. Mudget, and A. Werner. 1989. The 1985 NAPAP
Emissions Inventory (Version 2): Development of the Annual Data and Modelers' Tapes. EPA-
600/7-89-012a, U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina.
Shreve, R.N., and J.A. Brink. 1977. Chemical Process Industries. McGraw-Hill Publishing,
New York, New York.
UN (United Nations). 1988. 1986 Industrial Statistics Yearbook. Volumes 1 and 2. United
Nations, New York, New York.
UN (United Nations). 1992a. 1990 Energy Statistics Yearbook. United Nations, New York,
New York.
UN (United Nations). 1992b. 1990 Industrial Statistics Yearbook. Volumes 1 and 2. United
Nations, New York, New York.
U.S. EPA (U.S. Environmental Protection Agency). 1980. Organic Chemical Manufacturing,
Volume 7: Selected Processes. EPA-450/3-80-028b, Emission Standards and Engineering
Division, Research Triangle Park, North Carolina.
U.S. EPA (U.S. Environmental Protection Agency). 1987. Coke-oven Emissions from Wet-
Coal Charged By-product Coke-oven Batteries - Background Information for Proposed
Standards. EPA-450/3-85-028a, Emission Standards and Engineering Division, Research
Triangle Park, North Carolina.
Page 7-13
-------
-------
CHAPTER 8
METHANE EMISSIONS FROM LANDFILLS AND OPEN DUMPS
8.1 SUMMARY
Methane produced by decomposing wastes buried in landfills and open dumps is a
significant contributor to global methane emissions, with estimates ranging from 10 to 70
teragrams (Tg)1 per year (see Table 8-1). Global anthropogenic methane sources emit 360
Tg/yr (IPCC, 1992), which suggests that landfills may account for 3-19% of the total. Methods
of managing municipal solid waste (MSW) vary widely, ranging from open dumps and open
burning to sanitary landfills with leachate collection systems and landfill gas control. Different
management methods affect the rate of methane emissions.
This chapter presents two methods for estimating methane emissions from this source.
Application of the IPCC/OECD methodology (OECD, 1991) results in an estimate of 57 Tg/yr.
A second, empirical model, developed by the U.S. Environmental Protection Agency's Air &
Energy Engineering Research Laboratory (EPA/AEERL), is based on data from landfill gas-
recovery projects. EPA/AEERL global methane estimates range from 19 to 39 Tg/yr.
The United States is by far the biggest contributor, with estimates ranging from 8 to 16
Tg/yr. This is a somewhat larger range than reported in EPA's Report to Congress:
Anthropogenic Methane Emissions in the United States: Estimates for 1990 (1993b). This is
due to still unresolved variations in the methods used to estimate waste-in-place figures for
U.S. landfills. The U.S. domestic report to Congress estimates a range of 8 to 12 Tg per year
and is considered the current official figure for U.S. landfill methane emissions.
In the future, some OECD countries plan to place less waste in landfills in favor of
recycling and incineration. Also, controls for landfill air emissions are being considered by
some OECD countries. The United States plans to issue rules for municipal landfills that will
result in a methane emission reduction of about 7 Tg/yr by 2000. With regard to developing
countries, no conclusions about the future amount of waste that will be landfilled or dumped
can be drawn, due to lack of data. Substantial uncertainty in these estimates results from a
lack of data characterizing (1) country-specific waste generation or waste-in-place, (2) waste
management practices, (3) methane potential of the waste, and (4) methane oxidation in
landfill covers.
8.2 BACKGROUND
8.2.1 Methane Production from the Anaerobic Decomposition of Waste
The anaerobic decomposition of organic matter, as it occurs in a landfill, is a complex
process that requires that several groups of microorganisms act synergistically under
favorable environmental conditions. Three trophic groups of anaerobic bacteria must be
present to produce methane (CH4) from biological polymers, such as cellulose, hemicellulose,
1 Teragram = 106 metric tonnes = 1012 grams.
Page 8-1
-------
and protein: (1) the hydrolytic and fermentative microorganisms, (2) obligate proton-reducing
acetogens, and (3) methanogens (Wolfe, 1979; and Zehnder et al., 1982).
The hydrolytic and fermentative group is responsible for the hydrolysis of biological
polymers. The initial products of polymer hydrolysis are soluble sugars, amino acids, long-
chain carboxylic acids, and glycerol. Following polymer hydrolysis, the hydrolytic and
fermentative microorganisms ferment the initial products of decomposition into short-chain
carboxylic acids, alcohols, carbon dioxide (CO^), and hydrogen. Acetate, a direct precursor of
CH4, is also formed.
TABLE 8-1
TOTAL GLOBAL AND U.S. ESTIMATES OF METHANE EMISSIONS FROM LANDFILLS
AND OPEN DUMPS, ACCORDING TO VARIOUS METHODS
Global
Average
50
62
na
15
na
39
(To/Vrt
Range
30-70
na
na
10-20
na
25-52
Reference
Bingemer & Crutzen, 1987a
OECD, 1991"
na
Richards, 1989a
na
Thorneloe, 1993
United States (Tg/yr)
Average Range
na
23
6
na
10
12
na
na
3-8
na
8-12
8-18
Reference
na
OECD, 1991"
Augenstein, 1990a'°
na
EPA, 1993d
Doom et al., 1994
a. Potential emissions, not corrected for the amount that is flared or utilized.
b. Uses IPCC/OECD methodology (OECD, 1991) with updated country-specific data on MSW generation rates.
c. Uses estimated annual placement rates from 1950 to 1990.
d. U.S. EPA, 1993b.
The second group of bacteria — the obligate proton-reducing acetogens ~ convert the
fermentation products of the hydrolytic and fermentative microorganisms to CO2, hydrogen,
and acetic acid. The conversion of fermentation intermediates, such as butyrate, propionate,
and ethanol, is thermodynamically favorable only at very low hydrogen concentrations. Thus,
these substrates are used only when the obligate proton-reducing acetogenic bacteria can
function in syntrophic association with hydrogen scavengers, such as CH4-producing or
sulfate-reducing organisms.
The third group of bacteria necessary for the production of CH4 are the methanogens.
Major substrates used by methanogens for producing CH4 are acetate, formate, methanol,
methylamines, and hydrogen plus CO2 (Wolin and Miller, 1985).
While CH4 and CO2 are the terminal products of anaerobic decomposition, CO2 and
water are the terminal products of aerobic decomposition. Aerobic decomposition occurs in
Page 8-2
-------
management facilities where waste is exposed to air, such as when compost is turned for
aerating, and in uncontrolled dumps, such as when refuse is spread in thin layers or otherwise
exposed to oxygen (e.g., by scavenging). However, when refuse is buried in large piles,
whether at an open dump or in a sanitary landfill, the oxygen entrained at burial is consumed
rapidly, and substantial quantities of CH4 may be produced (Bhide et al., 1990).
8.2.2 Factors Affecting CH4 Potential from Landfilled Waste
Methane formation does not occur immediately after refuse is placed in a landfill or
dump. It can take months or years for the proper environmental conditions and the required
microbiological populations to become established. Numerous factors control decomposition,
including moisture content, nutrient concentrations, presence and distribution of
microorganisms, particle size, water flux, pH level, and temperature. Reviews of the effects of
each of these factors on CH4 production are provided in Barlaz et al. (1990), Pohland and
Harper (1987), and Halvadakis (1983).
The two factors that appear to have the most impact on CH4 production are moisture
content and pH. The effect of refuse moisture content has been summarized by Halvadakis
(1983), although some of the data in the summary relate to manure and not municipal waste.
The broadest data sets are those of Emberton (1986) arid Jenkins and Pettus (1985).
Emberton measured CH4 production rates in excavated landfill samples under laboratory
conditions. Jenkins and Pettus sampled refuse from landfills and tested how CH4 production
was affected by the moisture content of refuse. In both studies, the CH4 production rate
exhibited an upward trend with increasing moisture content, despite differences in refuse
density, age, and composition. It is difficult to translate the results of these laboratory studies
to actual landfills. A statistically significant correlation between landfill gas recovery and
precipitation could not be found, although an inferred correlation was demonstrated by
differentiating between landfills in arid and non-arid regions. (Peer et al., 1993; and U.S. EPA,
1993b).
A second key factor influencing the rate and onset of CH4 production is pH. The
optimum pH level for the activity by methanogenic bacteria is between 6.8 and 7.4. CH4
production rates decrease sharply with pH values below about 6.5 (Zehnder, 1982). When
refuse is buried in landfills, there is often a rapid accumulation of carboxylic acids; this results
in a temporary pH decrease and a long time lapse between refuse burial and the onset of CH4
production.
Neutralizing leachate and recycling it back through refuse has been shown to enhance
the onset and rate of CH4 production in laboratory studies (Pohland, 1975; Buivid, 1981; and
Barlaz et al., 1987). Given that moisture and pH have been reported as the two most
significant factors limiting CH4 production, the stimulatory effect of leachate neutralization and
recycling is logical. Neutralization of leachate provides a means of externally raising the pH of
the refuse ecosystem. Recycling neutralized leachate back through a landfill increases and
stabilizes both the moisture content and substrate availability of the refuse. It also enhances
mixing in what would otherwise be an immobilized batch reactor. Field experience with
leachate recycling systems is limited, and more information is needed to fully document its
value. In the next few years, new information should become available.
The lapsed time preceding the onset of CH4 production in landfills is an important
aspect when considering the management of individual landfills for biogas recovery or
emission mitigation. For evaluating global CH4 emissions from solid waste management
Page 8-3
-------
systems, the age at which landfills and uncontrolled dumps begin to produce CH4 is less
important than the total CH4 production potential of landfilled refuse.
8.2.3 Determination of the CH4 Potential of Solid Waste
The CH4 potential of landfilled refuse can be determined in three basic ways. The
theoretical CH4 potential of the main chemical constituents may be calculated, or laboratory or
field tests may be conducted. All three methods have in common the question of whether the
data are representative. Even in field tests, waste composition and other parameters that
affect CH4 generation may show unpredictable variety from one location to the next.
Not all CH4 produced in a landfill would be emitted to the air. Some CH4 may be
converted to CO2 in the presence of oxygen by aerobic methanotrophic bacteria, as it passes
through the cover soil. CH4 oxidation has been documented in landfill cover soil studied under
laboratory conditions (Whalen et al., 1990). However, there are no data on the quantitative
significance of CH4 oxidation above landfills. CH4 escaping through cracks in a landfill cover
most likely will not reside in the cover for a period sufficient to undergo significant oxidation.
Theoretical Approach
Knowledge of the chemical composition and the mass of refuse buried in a landfill
makes it possible to estimate the volume or mass of CH4 that may be produced. The amount
of CH4 that would be produced if all of a given constituent were converted to CH4, CO2, and
ammonia may be calculated from Equation 8.1 (Parkin and Owen, 1986):
CnHaObNc
(8.1)
Using the above stoichiometry, the potential of, for instance, cellulose and hemicellulose is
415 and 424 liters (I) CH4 at standard temperature and pressure per dry kilogram (kg),
respectively. (Cellulose (C6H10O5) and hemicellulose (C5H8O4) are major constituents of MSW
in the United States (Barlaz, 1988; and Barlaz et al. 1989).) These CH4 potentials represent
maximum CH4 production if 100% of the cellulose and hemicellulose were converted to CH4.
However, decomposition of these constituents in landfills is well below 100%, mainly because
(1) some cellulose and hemicellulose are surrounded by lignin or other recalcitrant materials
(such as plastic) and, therefore, are not biologically available; and (2) without active
intervention, buried refuse is not evenly exposed to moisture, microorganisms, and nutrients.
Instead of relating CH4 potential to individual constituents, it can also be expressed as a
function of the degradable organic carbon (DOC) content of the waste. The higher the DOC
percentage, the more CH4 could be formed, provided the waste degrades under anaerobic
conditions.
Laboratory Tests
CH4 yields of 42-120 liters CH4/dry kg refuse have been reported in laboratory tests
conducted with leachate recycling and neutralization (Barlaz et al., 1987; Barlaz 1988; Kinman
et al., 1987; and Buivid, 1981). The laboratory data are not perfectly comparable in that
experimental conditions (e.g., moisture, particle size, temperature) are not uniform among
studies. Sampling size is limited to volumes that can be reasonably handled and reduced by
proven techniques. However, it is possible to obtain multiple samples at presumably
representative locations within a landfill to get an estimate of the range and extent of
Page 8-4
-------
decomposition. Most laboratory experiments were conducted to explore techniques for
enhancing CH4 production. The enhanced CH4 production rates could not be expected at
field-scale landfills, unless certain enhancement techniques were employed in the field.
A technique for assessing the CH4 potential of refuse is the biochemical CH4 potential
(BOP) test (Shelton and Tiedje, 1984; and Bogner, 1990). In the BOP test, the anaerobic
biodegradability of a small sample of refuse (5-10 grams) is measured in a small batch reactor
(100-200 milliliters). While the BOP represents an upper bound of CH4 potential from refuse,
it will be lower than the stoichiometric estimate described in the previous section. BOPs also
require representative sampling in landfills.
Field Tests
In field-scale test cells, CH4 yields were measured as part of the Controlled Landfill
Project in Mountain View, California (Pacey, 1989). Yields of 38.6-92.2 liters CH4 /dry kg of
refuse were measured after 1,597 days. However, mass balance data suggested that in
certain test cells, significant volumes of CH4 were not recorded. A number often used by the
landfill gas industry as an estimate of CH4 production in field-scale landfills is 0.1 cubic foot
CH4 per wet pound per year (ft3/wet Ib/yr). Assuming refuse buried at 20% moisture, this
converts to 7.8 liters CH4/dry kg/yr, a number comparable to some of the lower values
reported in the literature. Comparison of CH4 production data between field-scale landfills and
laboratory experiments is difficult because there are essentially no data in the open literature
on CH4 production rates in field-scale facilities. Interpretation of data from field-scale landfills
or test cells is complicated by questions concerning the mass of refuse responsible for
production of a measured volume of gas ancl the efficiency of gas collection. More scientific
CH4 production data are collected under laboratory conditions than under field conditions
(Thorneloe, 1991; and Barlaz, 1991).
8.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
Existing emission methodologies for this source tend to assume that optimal conditions
for anaerobic decomposition exist within a landfill. However, this is rarely the case as the
information in the previous section and an article by Rathje (1991) indicate. To address this
concern, AEERL has used data on landfill gas recovery to develop an empirical model.
References used to develop inputs for the estimates are identified in Table 8-2. The
methodology developed by AEERL adjusts for gas-recovery efficiency and CH4 oxidation.
Two techniques for estimating emissions from landfills are presented here: the AEERL
Regression Model and the draft IPCC/OECD methodology. Because waste-in-place data exist
for very few countries, the models use waste-generation rates.
8.3.1 Estimation of Waste-Generation Rates
Municipal solid waste (MSW) consisting primarily of household and commercial refuse
is the largest contributor to CH4 from landfills and dumps. Often MSW is co-disposed with
other types of waste, i.e., demolition debris, fly ash, industrial waste, etc. Estimates of waste
generation are presented in Table 8-3. Country-specific data were available to determine
MSW generation and land-disposal values, especially for developed countries. Information on
MSW generation in developing countries is more difficult to obtain and is often anecdotal.
Page 8-5
-------
TABLE 8-2
REFERENCES USED
Geographic Region
References*
Africa
Asia and the Middle East
Europe
Latin America
Australia and Oceania
North America
World
1-12
13-21
22-36
37-38
39
40-43
44-51
1. El Halwagi et al.,
1988.
2. El Halwagi et al.,
1986.
3. Kallwasser, 1986.
4. UNDP etal., 1987.
5. Oluwande, 1984.
6. Monney, 1986.
7. Cointreau, 1982.
8. Rettenberger and
Weiner, 1986.
9. The World Bank,
1985.
10. Mwiraria et al., 1991.
11. United Republic of
Tanzania, 1989.
12. Verrier, 1990.
13. Maniatis and Vanhille,
1987.
*Reference Key
14. Lohani and Tanh, 27. Ernst, 1990.
1980.
15. Ahmed, 1986.
28. Cossu and Urbini,
1990.
16. Pairoj-Boriboon, 29. Beker, 1990.
1986.
17. Gadi, 1986.
18. Mei-Chan, 1986.
19. Bhideand
Sundaresan, 1990.
20. Cossu, 1990a.
21. Hayakawa, 1990.
30. Gandolla, 1990.
31. Bartone, 1990.
32. Cossu, 1990b.
33. Frantzis, 1988
34. Scheepers, 1990.
22. Christensen, 1990. 35. Bartone and Haley,
1990.
23. Carra and Cossu, 36. Barres et al., 1990.
1990.
24. Ettala, 1990.
37. Kessler, 1990.
25. Stegmann, 1990. 38. Yepes and Campbell,
1990.
40. U.S. EPA, 1988.
41. U.S. EPA, 1992b.
42. Thorneloe, 1993.
43. El Rayes and
Edwards, 1990.
44. Bartone et al., 1991.
45. Holmes, 1984.
46. Cointreau, 1984.
47. Bingemer and
Crutzen, 1987.
48. UN, 1989.
49. Richards, 1989.
50. World Resources
Institute, 1990.
51. Diaz and Golueke,
1987.
26. Lechner, 1990
39. Bateman, 1988.
Page 8-6
-------
TABLE 8-3
WASTE GENERATION RATES AND METHANE EMISSION ESTIMATES
USING IPCC/OECD AND AEERL METHODS
Country
Congo
Egypt
Gambia
Ghana
Kenya
Liberia
Morocco
Nigeria
South Africa
Sudan
Tanzania
Uganda
Zimbabwe
Other Africa
Total Africa
Bangladesh
China
India
Iran
Iraq
Israel
Japan
Kuwait
Malaysia
Mongolia
Myanmar
Waste
Generated
(Tg/yr)
0.24
6.99
0.08
2.35
2.28
0.32
3.12
10.61
11.17
2.79
2.29
1.47
1.90
31.46
77
7.99
134.50
66.79
10.76
4.21
1.20
41.00
0.59
2.01
0.18
3.11
IPCC/
AEERL's
OECD
Method Lower Bound
(Tg/yr) (Tg/yr)
Africa
0.01
0.32
0.01
0.05
0.09
0.01
0.12
0.48
0.43
0.09
0.09
0.06
0.08
1.27
3.1
Asia
0.42
3.87
0.80
0.31
0.12
0.05
1.04
0.14
0.64
0.02
0.08
0.00
0.08
0.00
0.03
0.04
0.01
0.05
0.18
0.11
0.03
0.03
0.02
0.02
0.48
1.1
0.08
0.64
0.74
0.16
0.06
0.01
0.24
0.01
0.03
0.00
0.03
Regression
Mid-Point
(Tg/yr)
0.01
0.13
0.00
0.05
0.06
0.01
0.08
0.28
0.18
0.05
0.04
0.03
0.03
0.75
1.7
0.13
0.99
1.15
0.25
0.10
0.02
0.38
0.01
0.05
0.00
0.05
Model
Upper Bound
(Tg/yr)
0.01
0.17
0.00
0.06
0.08
0.01
0.11
0.38
0.24
0.07
0.06
0.04
0.04
1.02
2.3
0.17
1.35
1.56
0.34
0.13
0.03
0.51
0.02
0.07
0.01
0.07
(continued)
Page 8-7
-------
TABLE 8-3
WASTE GENERATION RATES AND METHANE EMISSION ESTIMATES
USING IPCC/OECD AND AEERL METHODS (Continued)
Country
Waste
Generated
(Tg/yr)
IPCC/ AEERL's Regression Model
OECD
Method Lower Bound Mid-Point Upper Bound
(Tg/yr) (Tg/yr) (Tg/yr) (Tg/yr)
Asia (continued)
North Korea
Pakistan
Philippines
Saudi Arabia
South Korea
Sri Lanka
Thailand
Turkey
United Arab Emirates
Viet Nam
Other Asia
Total Asia
Albania
Austria
Belgium
Bulgaria
Former Czechoslovakia
Denmark
Finland
France
Germany
Greece
Hungary
Ireland
Italy
Netherlands
3.74
10.34
7.90
3.54
28.11
2.39
7.04
9.58
0.41
6.29
34.46
400
0.37
2.60
3.10
2.20
2.83
2.35
2.50
34.00
33.94
1.78
3.20
1.10
17.30
8.50
0.01
0.16
0.75
0.42
0.10
0.10
0.28
0.20
0.01
0.22
2.04
12
Europe
0.01
0.17
0.14
0.06
0.15
0.07
0.24
0.88
1.97
0.48
0.24
0.11
1.45
0.44
0.06
0.11
0.08
0.05
0.04
0.02
0.09
0.18
0.01
0.09
0.60
3.3
0.01
0.05
0.04
0.02
0.05
0.02
0.09
0.41
0.48
0.05
0.06
0.03
0.34
0.12
0.09
0.17
0.13
0.08
0.07
0.04
0.15
0.28
0.01
0.14
0.94
5.2
0.01
0.08
0.06
0.03
0.09
0.03
0.13
0.64
0.75
0.08
0.09
0.05
0.53
0.19
0.12
0.22
0.17
0.11
0.09
0.05
0.20
0.38
0.01
, 0.20
1.29
7.1
0.02
0.11
0.08
0.04
0.12
0.04
0.18
0.87
1.02
0.10
0.12
0.06
0.72
0.26
(continued)
Page 8-8
-------
TABLE 8-3
WASTE GENERATION RATES AND METHANE EMISSION ESTIMATES
USING IPCC/OECD AND AEERL METHODS (Continued)
Country
Waste
Generated
(Tg/yr)
IPCC/
OECD
Method
(Tg/yr)
AEERL's Regression Model
Lower Bound Mid-Point Upper Bound
(Tg/yr) (Tg/yr) (Tg/yr)
Europe (continued)
Norway
Poland
Romania
Spain
Sweden
Switzerland/
Liechtenstein
Former Soviet Union
United Kingdom
Former Yugoslavia
Other Europe
Total Europe
Canada
United States (1990)
Argentina
Brazil
Colombia
Venezuela
Others
Total America
Australia
New Zealand
Other Oceania
Total Oceania
Total Global
2.00
7.90
4.50
11.00
2.30
5.80
40.84
32.00
3.26
3.20
229
North and
21.00
263
5.67
31.00
6.80
5.23
'38.35
371
11.00
2.10
0.54
14
1097
0.11
0.37
0.13
0.85
0.08
0.12
2.49
2.85
0.17
0.08
14
Latin America
2.02
20.00
0.28
2.23
0.44
0.05
1.48
26.5
Australia and
1.16
0.15
0.03
1.3
57
0.03
0.11
0.04
0.22
0.03
0.03
0.83
0.75
0.06
0.06
3.9
and the Caribbean
0.57
7.50
0.10
0.66
0.15
0.08
0.53
9.6
Oceania
0.23
0.05
0.01
0.3
19
0.05
0.17
0.06
0.35
0.04
0.05
1.29
1.18
0.10
0.10
6.2
0.89
10.00
0.15
1.03
0.24
0.12
0.83
13
0.37
0.08
0.01
0.5
27
0.06
0.23
0.08
0.48
0.05
0.07
1.76
1.60
0.13
0.13
8.3
1.21
16.00
0.20
1.40
0.33
0.17
1.13
21
0.50
0.11
0.02
0.6
39
Note: Decimals in country-specific estimates are not meant to indicate precision. Estimates are considered
precise to within two significant figures. Totals may not equal sum of individual numbers due to rounding.
Page 8-9
-------
Most of the available data for developing countries are provided on a per-capita basis for only
the larger cities. This information was combined with population (UN, 1990) and urbanization
data (Population Reference Bureau, Inc., 1989) to determine the amount of waste generated
in the urban centers of these countries. Two estimates for a rural per-capita refuse generation
rate (Kessler, 1990 and United Republic of Tanzania, 1989) were then combined with the rural
population value to determine rural waste generation. In cases where no data were available,
data from comparable countries were used.
Per-capita MSW generation was in the range of 1.7-1.8 kg/person/day for both the
United States and Canada (U.S. EPA, 1992b; and El Rayes and Edwards, 1991). For the
United States, these numbers were not used, since estimates of total waste-in-place are
available (U.S. EPA, 1988; and Doom et al., 1994). MSW generation rates in other OECD
countries average out to be 1.1 kg/person/day. Information on the amount of MSW generated
and landfilled in the European countries that are not OECD members and the former Soviet
Union is limited. Average MSW generation for Greece, the former Soviet Union, and Eastern
Europe is approximately 0.6 kg/person/day. (References are listed in Table 8-2.)
For most Asian countries, estimates of MSW generation were at best identified for one
or two major cities, but not for the entire country. National per-capita MSW generation
estimates were found for Indonesia, Sri Lanka, the Philippines, Singapore, Taiwan, and
Pakistan. These estimates range from 0.4 kg/person/day for the Philippines to 1.0
kg/person/day for Singapore. The average per-capita MSW generation for these countries is
estimated to be 0.6 kg/person/day. (References are listed in Table 8-2.)
Few data are available on MSW production and management in Latin America and the
Caribbean Islands. Most of the available information is only for the larger cities. The average
per-capita MSW generation rate in Costa Rica, Mexico, Brazil, Colombia, Chile, Paraguay,
Peru, and Venezuela is estimated to be 0.8 kg/person/day. (References are listed in Table
8-2.)
Information on MSW generation and disposal for African and Middle Eastern countries
is very limited. Some information pertaining to generation rates for African countries was
located, but information for only two Middle Eastern countries -- Israel and Yemen — was
obtained. Based on the very limited information for African and Middle Eastern countries, it is
estimated that per-capita generation rates range from 0.3 to 1.1 kg/person/day. (References
are listed in Table 8-2.)
Not all waste that is generated will actually be landfilled. It may also be incinerated or
recycled. For most OECD countries, ample data are available to determine the percentage of
landfilled waste. For other countries, a default had to be defined. The default for each
country is based on expert judgment and relies on information that is primarily anecdotal. In
many countries, much of the garbage is scavenged before it is collected, especially paper,
textiles, and metal products. It may also be (1) burned for heating or cooking purposes,
(2) fed to domestic animals, (3) dumped in rivers or other bodies of water, or (4) swept out
onto the street or buried. In addition, garbage is often burned at the dump to reduce the
volume. This practice decreases the amount of material available for anaerobic
decomposition. Finally, the open dumps are often scavenged again by humans and animals.
Page 8-10
-------
8.3.2. AEERL Regression Model Methodology
The AEERL Regression Model Methodology is detailed in a report entitled "Global
Methane Emissions from Landfills and Open Dumps," which AEERL will publish in early 1994.
The AEERL model is empirical, based on the assumption that CH4 flow rates from landfills
with landfill gas (LFG) recovery systems can be used as surrogates for CH4 generation and
successively for CH4 emissions. To investigate the validity of this assumption, AEERL
initiated a program to collect data from 30 landfills with LFG recovery systems. The objective
of this program was to develop a statistical model of annual landfill CH4 emissions as a
function of climate, refuse mass, age, waste acceptance rate, composition, and compaction,
as well as to obtain an emission factor, which could be used to estimate both U.S. and global
CH4 emissions from landfills. Sites were chosen to represent a wide range of climate
conditions, as they occur in the United States. The research concluded that the mass of
waste-in-place showed a significant correlation with CH4 flow rates. None of the climate
variables - precipitation, average temperature, and dewpoint -- proved to have significant
correlations with CH4 flow rates. The effect of refuse age on gas production was also
analyzed. Gas flow rates correlated most strongly with refuse age for 10- to 20-year-old
refuse. Although these results were not conclusive, they suggest that the generation time for
gas production is 20-30 years, with an average of 25 years (U.S. EPA, 1991 b; and U.S. EPA
1992b). This generation time is within the range of generation times assumed in other landfill
gas recovery models (EMCON, 1982; and Augenstein and Pacey, 1990).
To relate CH4 flow rates from recovery projects to CH4 generation rates, two
assumptions needed to be made. It was assumed that the average recovery efficiency of a
gas-collection system is 75% (adapted from Augenstein and Pacey (1990)). Furthermore, it
was assumed that 10% of nonrecovered CH4 is oxidized (adapted from Whalen et al. (1990)).
Both assumptions are subject to discussion, as very limited data exist.
The model is detailed in Doom et al. (1994). As mentioned before, for most countries
total waste-in-place data are not available and have to be developed from annual waste
generation rates. As the variability of country-specific waste generation rates is unknown, it is
assumed that waste generation is steady state. Annual waste generation rates are multiplied
by the U.S. generation time (25 years) to obtain a surrogate for waste-in-place, which is in
turn multiplied by the U.S. emission factor and two country-specific parameters. (As
mentioned in section 8.3.2., CH4 flow rates are not significantly correlated with climate.) The
first parameter accounts for differences in chemical composition, expressed in CH4 potential.
The U.S. average CH4 potential is corrected for other countries by using country-specific
waste composition data. The second factor expresses which fraction of total waste-in-place is
decaying under anaerobic conditions.
The U.S. values for generation time and emission factor may be used since country-
specific variations in generation times and emissions per waste mass (emission factor) are
eliminated in the above described multiplication. For instance, if in a country the generation
time would be 12.5 years instead of 25, then the emissions per waste mass (the emission
factor) would have to be twice as high, hence the product stays unaltered.
AEERL's global model can be expressed mathematically by:
Page 8-11
-------
'CH4POTUS
EGM-Yn
in which:
MSW generated (tg/yr)
MSW (%) to landfill or dump
Degree to which dump or landfill is anaerobic
Av. CH4 pot. of landfilled MSW country M (I/kg)
Av. CH4 potential of landfilled waste US (I/kg)
Emission factor (g/g.yr)
Time waste produces CH4 (years)
Methane recovered or flared (tg/yr)
Methane emissions (tg/yr)
M
CH4POTM
CH4POTUS
E
G
= 80
= 0.00187
= 25
8.3.3 Draft IPCC/OECD Methodology
The draft IPCC/OECD methodology (OECD, 1991) was adapted from a model
developed by Bingemer and Crutzen (1987). The IPCC/OECD model uses theoretical
calculations based on a mass-balance approach, where an instantaneous release of CH4 is
assumed to enter the atmosphere during the same year that refuse is placed in a landfill. It
also assumes that all of the CH4 that is produced escapes to the atmosphere (i.e., none is
oxidized on route to the atmosphere). A third assumption of this method is that all developing
nations generate and dispose of municipal solid waste at the same per-capita rate. The
model also considers just urban waste, assuming that the large majority of landfilled waste is
produced in cities. Of course, country-specific data on MSW generation rates and percentage
of waste landfilled are preferred, recognizing that these data, too, have uncertainties.
To calculate the annual CH4 emissions from MSW, IPCC/OECD used the following
equation:
CHA Emission
= Total MSW Generated (kg/yr) x MSW Landfilled or
Dumped(%) x DOC in MSW (%) x Dissimilated DOC (80%) x
Concentration of CH4 in Landfill Gas (0.5 g/g) x Conversion
Factor (16 g CH4/12 g C) - Recovered CH4 (kg/yr)
where:
DOC
Dissimilated DOC =
Recovered CH4 =
degradable organic carbon;
portion of carbon in substrates that is converted to landfill gas
assumed to be 80% (Bingemer and Crutzen, 1987); and
amount of CH4 that is recovered through gas-recovery systems
and never emitted to the atmosphere.
The method is detailed in OECD (1991). It should be pointed out that two variables in
the equation (Total MSW Generated, MSW Landfilled, and the Fraction of MSW Landfilled or
Dumped) also are used in the AEERL method. The uncertainties of the IPCC/OECD
approach relative to the AEERL method are attributed to assumptions regarding the extent of
anaerobic decomposition (i.e., the fraction of dissimilated DOC). Many factors inhibit this
Page 8-12
-------
process, and this approach could overstate potential emissions. In addition, the methodology
as it is currently used by IPCC/OECD does not adjust for CH4 oxidation.
8.4 RESULTS
Table 8-3 gives country-specific waste-generated data, as well as CH4 emission
estimates using the IPCC/OECD and the AEERL methodologies. For reasons of comparison,
both methods use the same waste-generated data computed by AEERL. Therefore,
IPCC/OECD estimates may differ from official IPCC/OECD estimates. As shown in Table 8-3,
the global landfill CH4 emission estimate using the IPCC/OECD methodology is 57 Tg/yr.
Both estimates are based on 1990 data for the United States and some European countries.
Due to lack of reliable data, no attempt has been made to upgrade other country estimates to
1990 levels. Estimates have been adjusted for the amount of CH4 released from industrial
landfills and the amount of CH4 that is recovered for energy utilization or flared.
Based on U.S. EPA's regression model methodology, estimates of global landfill CH4
emissions range from 19 to 39 Tg/yr, with a mid-point of 27 Tg/yr. The estimate for the United
States (8-16 Tg/yr) has also been adjusted for the amount of CH4 that is recovered for energy
utilization and for the contribution from industrial landfills. AEERL estimates that 0.6 Tg/yr of
CH4 are emitted from industrial landfills (Doom et al., 1994). For other countries estimates of
CH4 emissions from industrial and hazardous waste landfills are not available. Emission
estimates for countries other than the United States have also been adjusted for the amount
of CH4 that may be recovered for energy utilization. Richards (1989) and Thorneloe (1992b)
estimate that worldwide there are 269 sites in 20 countries where landfill gas is recovered,
including 114 sites in the United States.
AEERL's estimate for U.S. landfill CH4 emissions differs from the estimate published in
EPA's report to Congress on domestic methane emissions (i.e., 8-12 Tg/yr) because of
differences in assumptions made about waste-in-place in U.S. landfills. There are many
factors that contribute to the uncertainty of these estimates. Future work to develop more
reliable information on waste quantities landfilled and waste composition will help to reduce
uncertainties. A midpoint estimate of 10 Tg/yr is used in this report for U.S. landfill methane
emissions.
Uncertainties in the AEERL estimate are expressed by giving lower- and upper-bound
values. Ranges have not been developed for the IPCC estimate. Due to the impossibility of
estimating errors associated with the assumptions made for most — if not all -- methods, a
mathematical approach in which individual errors are propagated is meaningless. Therefore,
the following method was adopted. The standard deviation in the emission factor, developed
from the landfill-gas-flow data, is 9% (Doom et al., 1993). Approximate 95% confidence
intervals are obtained by adding plus/minus two standard deviations to the estimate.
Consequently, it is assumed that errors in all other parameters amount to at least the same
error. Therefore, ranges are expressed by adding +36% to the emission estimate.
8.5 TRENDS
The potential control of CH4 emissions from municipal solid waste landfills has been
targeted by the United States and other countries as part of greenhouse gas reduction
programs designed to meet the goals of treaties signed at the United Nations Conference on
Page 8-13
-------
Environment and Development (UNCED) held in 1992. For example, the United States has
proposed regulations for MSW landfills that will result in a CH4 emission reduction of about 7
Tg/yr by the year 2000. Since such a reduction would be a significant step toward realizing of
the goals established at UNCED, and because this source is amenable to cost-effective
control measures, research designed to reduce the uncertainty associated with CH4 emission
estimates has a high priority (Thorneloe, 1993).
In general, the waste generated by developing countries has been projected to
increase over the next several decades; in contrast, the waste generated by developed
countries is expected to decline. This trend can be attributed to projections of higher
population increases in developing countries and not to increased per-capita waste
generation. Despite efforts toward source reduction and recycling, per-capita waste
generation is expected to increase in the United States (U.S. EPA, 1992a) and in other
industrialized countries. The much lower rates of population growth in industrialized countries
are expected to result in slower growth in MSW production, as compared to developing
countries (U.S. EPA, 1992b). However, this scenario will not be realized if per-capita income
decreases in developing countries. Recently, declining economic conditions have resulted in
reduced waste generation in Caracas, Venezuela; Mexico City, Mexico; and Buenos Aires,
Argentina (Bartone et al., 1991). As methods of MSW disposal change, there will be changes
in CH4 emissions. Trends in global waste management and their impacts on CH4 release are
discussed below.
8.5.1 Europe
In the future, some countries, including the Netherlands, Germany, and Denmark, plan
to increase the amount of MSW handled by recycling and incineration and to decrease the
amount placed in landfills. Landfills will be large, regional sites that also will be used to
dispose of incinerator ash. The decreases in the amount of MSW landfilled, coupled with
increases in landfill gas recovery and incineration, are anticipated to lead to reduced CH4
emissions and increased emissions of CO2 and other combustion gases.
Sanitary landfills or open dumps are used almost exclusively for MSW management in
Greece, Hungary, Portugal, Poland, Romania, Bulgaria, the former Yugoslavia, and the former
Soviet Union (Bartone and Haley, 1990; Curi, 1988; and Mnatsaknian, 1991). In the future,
Poland plans to prohibit open dumping in favor of sanitary landfilling. The former Soviet Union
hopes to establish an effective recycling program. No landfill-gas-recovery sites were
identified in these countries. A U.S. company (Natural Power) is negotiating a gas-recovery
contract with a landfill in Kiev in the Ukraine. This site has 9 million tonnes of waste. The
gas is to be used to generate electricity for the Ukraine capital (Thorneloe, 1992b). If this
project proves successful, additional projects may be initiated.
8.5.2 United States and Canada
Landfilling is the predominant MSW management method in both the United States
and Canada. Although there is a trend in both countries toward more recycling, more
incineration, and less landfilling, absolute MSW generation is projected to continue to grow
(U.S. EPA, 1992a; U.S. EPA, 1988; El Rayes and Edwards, 1991; and Alter, 1991). Until the
end of the century, it is not expected that the percent DOC in MSW will differ considerably
from current levels (U.S. EPA, 1993a). Based on this information and U.S. EPA (1992a) and
Willumsen (1990), it can be assumed that the current rate of landfill gas generation will
continue for several more decades.
Page 8-14
-------
The amounts of CH4 emitted to the atmosphere will decrease as more is controlled
through flaring or utilization (Thorneloe, 1992b; and Bonomo and Higginson, 1988). U.S.
landfills are currently recovering about 1.2 Tg of CH4 and are producing 344 megawatts of
power (Thorneloe, 1992a). The new rule for the control of nonmethane landfill gas, proposed
under the Clean Air Act Amendments, is expected to have a major impact on reducing landfill
CH4 emissions from both new and existing MSW landfills in the United States and may result
in an additional CH4 emission reduction ranging from 5 to 7 Tg (Federal Register, 1991; and
U.S. EPA, 1991).
8.5.3 Asia
Some Asian countries are upgrading their collection methods by introducing compactor
trucks and covered containers. These changes could serve to decrease the amount of
scavenging and increase the amount of MSW that is dumped. Economic constraints, in
tandem with a history of slow MSW management development, indicate that the use of
sanitary landfills will not increase markedly in the near future for most of Asia. However,
increases in population and, thus, total MSW will likely lead to increased CH4 emissions.
8.5.4 Latin America and the Caribbean Islands
In the future, Brazil hopes to build recycling and composting plants, and sanitary
landfills. Mexico also hopes to increase its number of sanitary landfills. Few landfill-gas-
recovery sites are currently operating in Brazil, Chile, and Mexico, although there seems to be
some interest in increasing the number of such sites in Brazil (Kessler, 1990; and Richards,
1988).
8.5.5 Africa and Middle East
*• In Africa's foreseeable future, the only MSW management methods reported to hold
promise for expanded and successful application are recycling, composting, or possibly biogas
recovery if markets and appropriate technologies can be developed to support these systems
(Cointreau, 1982; Betts, 1984; Oluwande, 1984; Vogler, 1984; and EI-Halwagi et al., 1988).
Much of the recyclable material in the MSW stream is currently being recovered, at least when
comparing the amount of material recycled in low-income countries with that of middle-income
and developed nations. However, because the recycling efficiency is low, recyclable materials
are still available in the waste streams, and because wages are low in the developing
countries, further recycling may be a viable waste management option (Wright et al., 1988).
8.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
8.6.1 Quantity and Composition of Waste- in-Place
Reliable waste-in-place estimates exist for only few countries. Where these data were
not available, waste-generation rates combined with waste management data were used.
Local waste-generation data are usually readily available; however, they typically apply to only
a certain city or region. Extrapolation of these data to national levels may be a source of
considerable uncertainty. Even if a reasonable estimate of rural waste generation can be
approximated, it is hard to assess how much of this waste will eventually be landfilled or
dumped. In certain rural areas of the world, it may well be that practically all rural MSW is
recycled. In the same fashion, waste composition data also usually apply to cities or regions
Page 8-15
-------
and have to be extrapolated. Fortunately, the countries that produce the most waste also
have the most extensive data on waste generation and composition. In fact, 57% of CH4 from
landfills is produced by six countries only (France, Germany, Italy, United Kingdom, Canada,
and the United States). Ongoing research by U.S. EPA will result in the refinement of these
estimates to account for waste composition in different geographical regions and the changing
waste composition in landfills occurring in countries adopting recycling programs (Barlaz,
1992; and Thorneloe, 1993).
8.6.2 CH4 Production of Waste in Open Dumps and Industrial Landfills
As discussed in section 8.2.3, assessment of the CH4 potential and generation from
landfilled waste is prone to considerable uncertainty. The uncertainty is enhanced for open
dumps. Two issues contribute to the difficulty of estimating CH4 production from open dumps:
both the physical characteristics of open dumps (size, configuration, temperature, moisture,
and compaction) and the quantity and composition of open-dumped waste are unknown.
Bhide et al. (1990) reported biogas recovery from two uncontrolled landfills in India. Each of
these sites was about 8 hectares in surface area and about 3-5 m deep. Neither site
contained any cover material, and the oldest of the two landfills accepted waste from 1971 to
1984, Most of the organic matter had decomposed by the time the tests were performed, but
biogas was obtained from a well 0.05 m in diameter at a rate of 0.240 cubic meter per hour
(rrvVhr) (CH4 content not presented). Waste had been deposited in the second site "only
recently," and the rate of biogas recovery was from 5 to 9 m3/hr. The CH4 content of the
biogas from the second site was 30-40%. The work of Bhide et al. (1990) suggests that
opens dumps are a source of CH4. Consequently, they have been included in the emission
estimates.
The CH4 production of other types of landfills, such as those containing industrial and
hazardous wastes, is not well understood. Definitions of "industrial waste" may vary. Some
authors include construction and demolition debris, thus reporting vast amounts of industrial
waste, with practically no DOC. Others distinguish between industrial and commercial waste
-- commercial waste being waste generated by offices and shops. This waste would typically
consist of paper and would, therefore, be a potential source of CH4. Industrial waste can also
consist of waste streams that will decompose under anaerobic conditions. Certain industrial
waste streams, such as that of the food industry, have a high organic content and are,
therefore, potentially significant sources of CH4. Landfills containing hazardous waste will
have a low CH4 production because of the low moisture content and the requirement that only
solid materials are accepted. In addition, chemicals in the waste stream may be toxic to the
microbes. The disposal of industrial and hazardous waste with MSW was common in the
United States through 1975 and is still common in most other countries. Waste streams in
developing countries are virtually uncontrolled, and mingling of MSW, industrial wastes, and
raw sewage in landfills is common (Cointreau, 1982).
s
8.6.3 Quantity of CH4 Emitted to the Atmosphere
As mentioned in section 8.2.3, not all CH4 produced in a landfill would be emitted to
the air. CH4 oxidation has been documented in landfill cover soil studied under laboratory
conditions (Whalen et al., 1990). However, there are no data on the quantitative significance
of CH4 oxidation for landfills. CH4 escaping through cracks in a landfill cover most likely will
not reside in the cover for a period sufficient to undergo significant oxidation. Assessment of
the amount of CH4 that may be oxidized on its way out of the landfill (i.e., 10%) is solely a
matter of expert judgment.
Page 8-16
-------
8.7 CONCLUSIONS
Global and country-specific estimates were developed for CH4 emissions from landfills
and open dumps using the proposed IPCC/OECD and the EPA/AEERL methodologies. The
estimates for global and U.S. CH4 emissions resulting from these two approaches are
presented in Table 8-4.
The United States is by far the largest contributor of global landfill CH4, accounting for
up to 50% of the emissions. The other geographic regions that are significant contributors are
Europe, accounting for about 20%, and Asia, with about 16%.
Because this source is amenable to cost-effective control through the use of the CH4, it
is under consideration by a number of countries for programs that would result in a reduction
in current emissions. The United States, for example, has proposed regulations under the
Clean Air Act for new and existing MSW landfills. These regulations are scheduled to be
promulgated in the spring of 1994. Approximately 7 Tg in the year 2000 are expected to be
reduced. If this CH4 is recovered for its energy potential, a further reduction of emissions
resulting from coal-fired power plants would also be achieved. The United States has also
adopted policies that will result in waste-minimization and recycling programs, which should, in
turn, reduce CH4 emissions.
TABLE 8-4
ESTIMATES OF METHANE EMISSIONS FROM LANDFILLS
FROM OECD/IPCC AND EPA/AEERL METHODOLOGIES
EPA/AEERL
IPCC/OECD
Range
Mid-Point
Global
United States
57
20
19-39
8-16
29
12
Other countries are considering programs either to regulate or to provide incentives
that encourage energy recovery. These countries include Canada, the United Kingdom,
Germany, the Netherlands, and Japan. These programs are expected to reduce global
emissions of CH4 from landfills. In addition, many industrialized countries are adopting
recycling and waste-minimization programs that decrease the amount of waste in landfills.
However, developing countries that are increasing their use of sanitary landfills are expected
to generate greater CH4 emissions in the future. It is important that larger sites are
encouraged towards use or control of the CH4 to continue a decrease in global CH4 emissions
from landfills and open dumps.
8.8 REFERENCES
Ahmed, M.F. 1986. Recycling of solid wastes in Dhaka. In Thome-Kozmiensky, K.J., ed.
Waste Management in Developing Countries, 1. EF-Verlag fur Energie und Umwelttechnik
GmbH, Berlin, Germany. 169-173.
Page 8-17
-------
Alter, H. 1991. The future course of solid waste management in the United States. Waste
Management & Research: v.9, 3-20.
Augenstein, D., and J. Pacey. 1990. Modeling Landfill CH4 Generation. International
Conference on Landfill Gas: Energy and Environment, 17 October 1990, Bournemouth,
United Kingdom.
Barlaz, M.A. 1988. Microbiological and Chemical Dynamics During Refuse Decomposition in
a Simulated Sanitary Landfill. Ph.D. Dissertation. Department of Civil and Environmental
Engineering, University of Wisconsin, Madison, Wisconsin.
Barlaz, M.A. 1991. Landfill Gas Research in the United States: Previous Research and
Future Directions. Proceedings of the Landfill Microbiology Research and Development
Workshop, November 20, 1991, United Kingdom Department of Energy, London, United
Kingdom.
Barlaz, M.A. 1992. Biodegredation of Individual Components of Municipal Solid Waste in
Landfills: Enhancement Opportunities and Research Needs. Proceedings of the Waste
Management Inc. Technology Conference, September 28-30, 1993, Oakbrook Illinois.
Barlaz, M.A, R.K. Ham, and D.M. Schaefer. 1990. CH4 Production from municipal refuse: A
review of enhancement techniques and microbial dynamics. CRC Critical Reviews in
Environmental Control. Vol. 19, Issue 6.
Barlaz, M.A., M.W. Milke, and R.K. Ham. 1987. Gas production parameters in sanitary
landfill simulators. Waste Management and Research 5:27.
Barlaz, M.A., D.M. Schaefer, and R.K. Ham. 1989. Mass balance analysis of decomposed
refuse in laboratory scale lysimeters. ASCE Journal of Environmental Engineering
115(6):1,088-1,102.
Barres, M., Y. Grenet, N. Millot, and A. Meisel. 1990. Sanitary landfilling in France. In Carra,
J.S., and R. Cossu, eds. International Perspectives on Municipal Solid Wastes and Sanitary
Landfilling. Academic Press, New York, New York. 1-14.
Bartone, C.R. 1990. Economic and policy issues in resource recovery from municipal solid
wastes. Resources Conservation Recycling 4:7-23.
Bartone, C.R., and C. Haley. 1990. The Bled symposium: Introduction. Resources
Conservation Recycling 4:1-6.
Bartone, C.R., L. Leite, T. Triche, and R. Schertenleib. 1991. Private sector participation in
municipal solid waste service: Experiences in Latin America. Waste Management and
Research 9:495-509.
Bateman, C.S. 1988. Landfill gas development in Australia. In Alston, Y.R., and G.E.
Richards, eds. Proceedings of the International Conference on Landfill Gas and Anaerobic
Digestion of Solid Waste. 4-7 October, Harwell Laboratory, Oxfordshire, United Kingdom.
156-161.
Page 8-18
-------
Beker, D. 1990. Sanitary landfilling in the Netherlands. In Carra, J.S., and R. Cossu, eds.
International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 139-155.
Betts, M.P. 1984. Trends in solid waste management in developing countries. In Holmes,
J.R., ed. Managing Solid Wastes in Developing Countries. John Wiley & Sons, Ltd.,
Chichester, United Kingdom. 291-302.
Bhide, A.D., S.A. Gaikwad, and B.Z. Alone. 1990. CH4 from land disposal sites in India. In
Proceedings of the International Workshop on CH4 Emissions from Natural Gas Systems, Coal
Mining and Waste Management Systems. Environment Agency of Japan, the U.S. Agency for
International Development, and the U.S. Environmental Protection Agency, Washington, D.C.,
9-13 April 1990.
Bhide, A.D., and B.B. Sundaresan. 1990. Solid Waste Management in Developing Countries.
14-21. Indian National Scientific Documentation Centre. New Delhi, India.
Bingemer, H.G., and P.J. Crutzen. 1987. The production of CH4 from solid wastes. Journal
of Geophysical Research 92(D2):2,181-2,187.
Bogner, J.E. 1990. Controlled study of landfill biodegradation rates using modified BMP
assays. Waste Management and Research 8:329-352.
Bonomo, L, and A.E. Higginson. 1988. International Overview on Solid Waste Management.
Harcourt Brace Jovanovich, New York, New York. 268.
Buivid, M.G. 1981. Fuel gas enhancement by controlled landfilling of municipal solid waste.
Resource Recovery and Conservation 6:3.
Carra, J.S., and R. Cossu, eds. 1990. International Perspectives on Municipal Solid Wastes
and Sanitary Landfilling. Academic Press, New York, New York. 1-14.
Christensen, T.H. 1990. Sanitary landfilling in Denmark. In Carra, J.S., and R. Cossu, eds.
International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 37-50.
Cointreau, SJ. 1982. Environmental Management of Urban Solid Wastes in Developing
Countries. A Project Guide, Urban Development Technical Paper Number 5. World Bank,
Washington, D.C. 1-17.
Cointreau, S.J. 1984. Solid Waste Collection Practice and Planning in Developing Countries.
John Wiley and Sons Ltd., Chichester, United Kingdom. 151-182.
Cossu, R. 1990a. Sanitary landfilling in Japan. In Carra, J.S., and R. Cossu, eds.
International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 110-138.
Cossu, R. 1990b. Sanitary landfilling in the United Kingdom. In Carra, J.S., and R. Cossu,
eds. International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 199-220.
Page 8-19
-------
Cossu, R., and G. Urbini. 1990. Sanitary landfilling in Italy. In Carra, J.S., and R. Cossu,
eds. International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 94-109.
Curi, K. 1988. Comparison of solid waste management in touristic areas of developed and
developing areas. In Proceedings of the 5th International Solid Waste Conference.
September 11-16, 1988, International Solid Waste and Public Cleansing Association,
Copenhagen, Denmark.
Diaz, L.F., and C.G. Golueke. 1987. Solid waste management in developing countries.
Biocycle 28(6):50-55.
Doom, M.R.J., LA. Stefanski, R.L. Peer, and D.L Epperson. 1994. Estimate of Methane
Emissions from U.S. Landfills. Prepared for U.S. EPA/AEERL, Research Triangle Park, North
Carolina. (Cleared peer review, to be published in 1994.)
EI-Halwagi, M.M., S.R. Tewfik, M.H. Sorour, and A.G. Abulnour. 1986. Municipal solid waste
management in Egypt: Practices and trends. In Thome-Kozmiensky, K.J., ed. Waste
Management in Developing Countries, 1. EF-Verlag fur Energie und Umwelttechnik GmbH,
Berlin, Germany. 283-288.
EI-Halwagi, M.M., S.R. Tewfik, M.H. Sorour, and A.G. Abulnour. 1988. Municipal Solid Waste
Management in Egypt: Proceedings of the 5th International Solid Waste Conference.
September 11-16, 1988, International Solid Waste and Public Cleansing Association,
Copenhagen, Denmark. 415-424.
El Rayes, H., and W.C. Edwards (B.H. Levelton & Associates, Ltd.). 1991. Inventory of CH4
Emissions from Landfills in Canada. Prepared for Environment Canada, Hull, Quebec. 25-69.
Emberton, J. R. 1986. The Biological and Chemical Characterization of Landfills.
Proceedings of Energy from Landfill Gas, 30-31 October 1986, Solihull, West Midlands, United
Kingdom.
EMCON Associates. 1982. CH4 Generation and Recovery from Landfills. Ann Arbor Science
Publishers, Inc., Ann Arbor, Michigan.
Ernst, A. 1990. A Review of solid waste management by composting in Europe. Resources
Conservation Recycling 4:135-149.
Ettala, M.O. 1990. Sanitary landfilling in Finland. In Carra, J.S., and R. Cossu, eds.
International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 67-77.
Federal Register. 1991 (May 30). 56(104):24,468-24,528.
Frantzis, I. 1988 (July). Recycling in Greece. BioCycle. 30-31.
Gadi, M.T. 1986. In Thome-Kozmiensky, K.J., ed. Waste Management in Developing
Countries, 1. EF-Verlag fur Energie und Umwelttechnik GmbH, Berlin, Germany. 188-194.
Page 8-20
-------
Gandolla, M. 1990. Sanitary landfilling in Switzerland. In Carra, J.S., and R. Cossu, eds.
International Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic
Press, New York, New York. 190-198.
Halvadakis, C.P. 1983. Landfill methanogenesis: Literature review and critique. Technical
Report No. 271. Department of Civil Engineering, Stanford University, Palo Alto, California.
Hayakawa, T. 1990. The status report on waste management in Japan - Special focus on
methane emission prevention. In Proceedings of the International Workshop on CH4
Emissions from Natural Gas Systems, Coal Mining and Waste Management Systems.
Environment Agency of Japan, the U.S. Agency for International Development, and the U.S.
Environmental Protection Agency, Washington, D.C., 9-13 April 1990. 509-523.
Holmes, J.R. 1984. Solid waste management decisions in developing countries. In
Managing Solid Wastes in Developing Countries. John Wiley and Sons Ltd., Chichester,
United Kingdom. 1-17.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
Jenkins, R.L., and J.A. Pettus. 1985. In Antonopoulos, A.A., ed. Biotechnological Advances
in Processing Municipal Wastes for Fuels and Chemicals. Argonne National Laboratory
Report ANL/CNSV - TM - 167. 419.
Kaltwasser, B.J. 1986. Solid waste management in medium-sized towns in the Sahel Area.
In Thome-Kozmiensky, K.J., ed. Waste Management in Developing Countries, 1. EF-Verlag
fur Energie und Umwelttechnik GmbH, Berlin, Germany. 299-307.
Kessler, T. 1990. Brazilian Trends in Landfill Gas Exploitation. STATEC Consultores s/c
Ltda, Sao Paulo, Brazil.
Kinman, R.N. et al. 1987. Waste Management and Research 5:13.
Lechner. 1990. Sanitary landfilling in Austria. In Carra, J.S., and R. Cossu, eds. International
Perspectives on Municipal Solid Wastes and Sanitary Landfilling. Academic Press, New York,
New York. 15-25.
Lohani, B.N., and N.C. Tanh. 1980 (May/June). Problems and practices of solid waste
management in Asia. J. Environ. Sci. 29-33.
Maniatis, K., and S. Vanhille. 1987. Solid waste management in Indonesia: Status and
potential. Resources Conservation Recycle 15(87):277-290.
Mei-Chan, L. 1986. Waste management in the Taiwan area. In Thome-Kozmiensky, K.J.,
ed. Waste Management in Developing Countries, 1. EF-Verlag fur Energie und Umwelttechnik
GmbH, Berlin, Germany. 299-307.
Mnatsaknian, R.A. 1991. Legislation and public control of waste sites in the U.S.S.R. In
Proceedings of the Third International Landfill Symposium, Sardinia, Italy. October 1991.
1,747-1,753.
Page 8-21
-------
Monney, J.G. 1986. Municipal solid waste management - Ghana's experience. In Thome-
Kozmiensky, K.J., ed. Waste Management in Developing Countries, 1. EF-Verlag fur Energie
und Umwelttechnik GmbH, Berlin, Germany. 321-327.
Mwiraria et al. 1991 (May 18). Municipal Solid Waste Management in Uganda and
Zimbabwe. Draft report for the United Nations Development Programme and the World Bank.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Oluwande, P.A. 1984. Assessment of solid waste management problems in China and
Africa. In Holmes, J.R., ed. Managing Solid Wastes in Developing Countries. John Wiley &
Sons, Ltd., Chichester, United Kingdom. 71-89.
Pacey, J. 1989. Enhancement of degradation: Large-scale experiments. In Christensen, T.,
R. Cossu, and R. Stegmann, eds. Sanitary Landfilling: Process Technology and
Environmental Impact. Academic Press, London, United Kingdom. 103-119.
Pairoj-Boriboon, S. 1986. State-of-the-art of waste management in Thailand. In Thome-
Kozmiensky, K.J., ed. Wa.ste Management in Developing Countries, 1. EF-Verlag fur Energie
und Umwelttechnik GmbH, Berlin, Germany. 208-219.
Parkin, G.F., and Owen, W.F. 1986. Journal of the Environmental Engineering Division,
>4SCE112(5):867.
Peer, R.L., S.A. Thorneloe, and D.L Epperson. 1993. A comparison of methods for
estimating global methane emissions from landfills. Chemosphere 26(1 -4):387-400.
Pohland, F.G. 1975. Sanitary Landfill Stabilization With Leachate Recycle and Residual
Treatment. EPA Grant No. R-801397 Georgia Institute of Technology, Atlanta, Georgia.
Pohland, F.G., and S.R. Harper. 1987. Critical Review and Summary of Leachate and Gas
Production from Landfills. EPA/600/2-86/073, PB86-240181. Hazardous Waste Engineering
Research Laboratory, Cincinnati, Ohio.
Population Reference Bureau, Inc. 1989. Demographic data and estimates for the countries
and regions of the world. World Population Data Sheet of the Population Reference Bureau,
Inc., Washington, D.C.
Rathje, W.L. 1991 (May). Once and future landfills. National Geographic. 117-134.
Rettenberger, G., and K. Weiner. 1986. Measures to improve the situation on the field of
sanitation and solid waste management in Juba, Sudan. In Thome-Kozmiensky, K.J., ed.
Waste Management in Developing Countries, 1. EF-Verlag fur Energie und Umwelttechnik
GmbH, Berlin, Germany. 289-298.
Richards, K.M. 1988. Landfill gas - A global review. In Houghton, D.R., R.N. Smith, and
H.O.W. Eggins, eds. Biodeterioration 7. Elsevier Applied Science, London, United Kingdom.
774-790.
Page 8-22
-------
Richards, K.M. 1989. Landfill gas: Working with Gaia. Biodeterioration Abstracts
3(4):317-331; and C.A.B. International 3(4). December 1989.
Scheepers, M.J.J. 1990. Landfill gas in the Dutch perspective. International Conference on
Landfill Gas: Energy and Environment '90, Session 3.2. October 1990. Sardinia, Italy
Shelton, D.R., and J.M. Tiedje. 1984. General method for determining anaerobic
biodegradation potential. Appl. Env. Microbiol. 47(4):850-857.
Stegmann, R. 1990. Sanitary landfilling in the Federal Republic of Germany. In Carra, J.S.,
and R. Cossu, eds. International Perspectives on Municipal Solid Wastes and Sanitary
Landfilling. Academic Press, New York, New York. 51-66.
Thorneloe, S. 1991. United States research on enhancing landfill gas production. Landfill
Gas Enhancement Test Cell Data Exchange. Final Report of the Landfill Gas Expert Working
Group, International Energy Agency. AEA-EE-0226. Harwell Laboratory, Oxfordshire, United
Kingdom.
Thorneloe, S. 1992a. Emissions and Mitigation at Landfills and other Waste Management
Facilities. Published for EPA Symposium on Greenhouse Gas Emissions and Mitigation
Research. August 18-20, 1992, Washington D.C. (Proceedings are in press).
Thorneloe, S. 1992b Landfill gas recovery/utilization - Options and economics. In
Proceedings of the 16th Annual Conference on Energy from Biomass and Wastes. March 5,
1992, Orlando, Florida. Institute of Gas Technology, Chicago, Illinois.
Thorneloe, S. 1993. Landfill gas and its influence on global climate change. Conference
Proceedings of the Fourth International Landfill Symposium, October 13, 1993, Cagliari, Italy.
UN (United Nations). 1989. City Profiles. Prepared by the United Nations Centre for
Regional Development and the Kitakyushu City Government. Kitakyushu, Japan.
UN (United Nations). 1990. 1988 Demographic Yearbook. Fortieth Issue. Department of
International Economic and Social Affairs, Statistical Office. New York, New York.
UNDP (United Nations Development Programme), The World Bank, and the Canadian
International Development Agency. 1987 (February). Master Plan for Resource Recovery
and Waste Disposal, City of Abidjan. Final Report. Prepared by Roche Ltd. Consulting
Group, Sainte-Foy, Quebec, Canada.
United Republic of Tanzania. 1989. Masterplan of Solid Waste Management for Dar es
Salaam. Volume II: Annexes. Ministry of Water, Department of Sewerage and Sanitation.
Prepared by HASKONING, Royal Dutch Consulting Engineers and Architects, Nijmegen,
Netherlands; and M-Konsult Ltd., Consulting Engineers, Dar es Salaam, Tanzania. 1-39,
47-49, 74-94, 109-124, and 142-145.
U.S. EPA (U.S. Environmental Protection Agency). 1988 (October). Report to Congress:
Solid Waste Disposal in the United States. EPA/530-SW-88-011.
Page 8-23
-------
U.S. EPA (U.S. Environmental Protection Agency). 1991 a. Air Emissions from Municipal
Solid Waste Landfills - Background Information for Proposed Standards and Guidelines. EPA-
450/3-90-011 a, U.S. EPA, Research Triangle Park, North Carolina.
U.S. EPA (U.S. Environmental Protection Agency). 1991b. Analysis of Factors Affecting
Methane Gas Recovery from Six Landfills. EPA-600/2-91-005, Air and Energy Engineering
Research Laboratory, Research Triangle Park, North Carolina.
U.S. EPA (U.S. Environmental Protection Agency). 1992a. Characterization of Municipal
Solid Waste in the United States: 1992 Update. EPA/530-R-92-019. Washington, DC
U.S. EPA (U.S. Environmental Protection Agency). 1992b. Development of an Empirical
Model of CH4 Emissions from Landfills. EPA-600/R-92-037, prepared for the U.S. EPA, Air
and Energy Engineering Research Laboratory, Research Triangle Park, North Carolina.
U.S. EPA (U.S. Environmental Protection Agency). 1993a. Effects of International Municipal
Solid Waste Management Trends on Global Greenhouse Gas Emissions. In press.
U.S. EPA (U.S. Environmental Protection Agency). 1993b. Report to Congress:
Anthropogenic Methane Emissions in the United States: Estimates for 1990. EPA-430-R-93-
003, Office of Air and Radiation, Washington, D.C.
Verrier, S.J. 1990. Urban waste generation, composition and disposal in South Africa. In
Carra, J.S.; and R. Cossu, eds. International Perspectives on Municipal Solid Wastes and
Sanitary Landfilling. Academic Press, Harcourt Brace Jovanovich, London, United Kingdom.
161-176.
Vogler, J.A. 1984. Waste recycling in developing countries: A review of the social,
technological, and market forces. In Holmes, J.R., ed. Managing Solid Wastes in Developing
Countries. John Wiley & Sons, Ltd., Chichester, United Kingdom. 241-266.
Whalen, S.C., W.S. Reeburgh, and K.A. Sandbeck. 1990. Rapid CH4 oxidation in a landfill
cover soil. Appl. Env. Microbiol. 56(11):3,405-3,411.
Willumsen, H. 1990. Landfill gas. Resources Conservation Recycling 4(90):121-133.
Wolfe, R.S. 1979. Methanogenesis. In Quayle, J.R., ed. Microbial Biochemistry, International
Review of Biochemistry. Vol. 21. University Park Press, Baltimore, Maryland.
Wolin, M.J., and T.L. Miller. 1985. In Dernsin, A.L., and N.A. Solomon, eds. Biology of
Industrial Microorganisms. Benjamin/Cumings Publ. Co. Inc., Menlo Park, California. 189-221.
World Bank. 1985. Metropolitan Area of Douala. Study of Waste Management and
Resource Recovery. Part A, Phase I. Prepared for the World Bank by Motor Columbus,
Consulting Engineers, Inc., CH-5401 Baden, Switzerland.
World Resources Institute. 1990. World Resources 1990-91. Oxford University Press, New
York, New York.
Page 8-24
-------
Wright, F., C. Bartone, and S. Arlosoroff. 1988. Integrated resource recovery: Optimizing
waste management. In Proceedings of the 5th International Solid Wastes Conference.
International Solid Wastes and Public Cleansing Association, Copenhagen, Denmark.
September 1988. 619-624.
Zehnder, A.J.B. 1982. Microbiology of CH4 bacteria. In Hughes, D.E., ed. Anaerobic
Digestion. Elsevier Biomedical Press B.V., Amsterdam, Netherlands. 45.
Page 8-25
-------
-------
CHAPTER 9
METHANE EMISSIONS FROM LIVESTOCK MANURE
9.1 SUMMARY
Methane is produced during the anaerobic decomposition of the organic material in
livestock manure. This report estimates that in 1990 global methane emissions from livestock
manure were about 14 teragrams (Tg)1 with a range of about 10-18 Tg. Three animal groups
account for about 80% of total emissions: swine, about 40%; nondairy cattle, about 20%; and
dairy cattle, about 20%. Liquid-based systems, such as lagoons and liquid/slurry storage,
account for about 60% of total emissions.
Three regions of the world contribute about 85% of total emissions: Europe (Eastern
and Western), about 40%; Asia and the Far East, about 30%; and North America, about 15%.
Methane emissions from livestock manure could increase significantly during the
coming decades. As the demand for livestock products increases, animal populations and
manure production will increase. Also, the use of liquid-based systems for managing livestock
manure (which promote methane production) may increase substantially because of growing
concerns over the direct land application of livestock manure and the increased adoption of
large, confined animal management practices.
This estimate of methane emissions from animal manure is lower than previous
estimates of 25 ± 5 Tg. Recent laboratory analyses indicate that the rate of methane
production is lower than previously estimated for several key manure management conditions,
including manure deposited on pastures and range. Additional laboratory and field
measurements are required to improve the basis for making these emission estimates.
9.2 BACKGROUND
Microbiological decomposition is a process in which microorganisms derive energy and
material for cellular growth by metabolizing organic material ~ for example, organic material in
livestock manure. When decomposition occurs in an oxygen-free environment (anaerobically),
methane (CH4) is produced. Because the quantity of livestock manure produced annually is
large, and because the manure is primarily composed of organic material, the potential for
CH4 emissions is great. However, only a portion of this emission potential is realized because
when the manure is kept in contact with oxygen (e.g., spread on fields) CH4 production is
minimal.
The principal determinants of CH4 production from livestock manure are the following:
Characteristics of the manure. Potential CH4 production is directly related to
the quantity of manure and the fraction of the manure available for
decomposition. These factors vary by animal species and their diet.
1 Teragram = 106 metric tonnes = 1012 grams.
Page 9-1
-------
Manure management system. The manure management system strongly
influences CH4 production from livestock manure. Manure management
systems that promote anaerobic (oxygen-free) decomposition will produce the
most CH4.
Climate. Temperature and rainfall affect both the rate and the total amount of
CH4 production in livestock manure. A warm and moist environment promotes
CH4 production.
9.2.1 Characteristics of the Manure
The composition of livestock manure, which is primarily a function of the animal
species and diet, determines its maximum CH4-producing capacity. For a given species, the
greater the energy content and digestibility of the diet, the greater the CH4-producing capacity
of the manure. For example, cattle fed a high-energy grain diet produce a highly
biodegradable manure. Cattle fed a roughage diet will produce a less biodegradable manure
containing more complex organics, such as cellulose, hemicellulose, and lignin. Under similar
conditions, the manure of cattle fed the high-energy, corn-based diet will produce about twice
as much CH4 per unit of volatile solids2 (VS) as the manure of the cattle fed a roughage diet.3
The CH4-producing capacity of livestock manure is generally expressed in terms of the
quantity of CH4 that can be produced per kilogram of volatile solids (VS) in the manure. This
quantity is commonly referred to as B0 with units of cubic meters of CH4 per kilogram VS (m3
CH4 / kg VS). B0 values for many common feeding conditions have been determined
experimentally.
9.2.2 Manure Management System
The characteristics of a manure management system determines how much of the
maximum CH4-producing capacity (B0) can be realized. These characteristics include:
Contact with oxygen. Under aerobic conditions where oxygen is in contact with
the manure, CH4 is not produced.
Moisture content. Liquid-based systems promote an oxygen-free environment
and anaerobic decomposition. In addition, water is required for bacterial cell
production and metabolism and acts as a buffer to stabilize pH. Moist
conditions lead to larger CH4 emissions than do dry conditions.
pH. Methane-producing bacteria are sensitive to changes in pH. The optimal
pH is near 7.0, but CH4 will be produced between 6.6 and 7.6. Deviation in pH
from 7.0 will decrease the rate of CH4 production.
a "Volatile solids" (VS) are defined as the organic fraction of the total solids (TS) in waste that will oxidize and
be driven off as gas at a temperature of 600°C. Total solids are defined as the material that remains after
evaporation of water at a temperature between 103° and 105°C.
3 The quantity of manure produced is also affected by the digestibility of the diet. At a given level of dry matter
intake, a higher-digestibility diet results in less manure excretion than a low-digestibility diet. However, dry-matter
Intake and digestibility are not independent, so that a simple inverse relationship between digestibility and manure
production is not generally observed under realistic production conditions.
Page 9-2
-------
f
Nutrients. Bacterial growth depends on the availability of nutrients, such as
nitrogen, phosphorus, and sulfur. Deficiency in one or more of these nutrients
will inhibit bacterial growth and CH4 formation. Animal diets typically contain
sufficient nutrients to sustain bacterial growth, and so nutrient availability is not
usually a limiting factor in CH4 production.
9.2.3 Climate
The climatic conditions in which manure decomposes have a strong influence on CH4
production. These conditions include:
Temperature. Methanogenesis in livestock manure has been observed
between 4° and 75°C. Temperature is one of the major factors affecting the
growth of the bacteria responsible for CH4 formation (Chawla, 1986). Methane
production generally increases with rising temperature within this range and
levels off between 60° and 75°C.
Moisture. For nonliquid-based manure systems, the moisture content of the
manure is determined by rainfall and humidity. Because the moisture content
of the manure will determine the rate of bacterial growth and manure
decomposition, moist climate conditions promote CH4 production.
9.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The approach described in OECD (1991) is recommended for estimating CH4
emissions from livestock manure. Emissions from this source depend on the quantity and
type of manure, the characteristics of the manure management system, and the climatic
conditions in which the manure decomposes. While limited data are available on which to
base emission estimates, a study recently prepared for the U.S. EPA (Safley et al., 1992) and
ongoing laboratory studies being conducted at Oregon State University (Hashimoto and Steed,
1993) provide an adequate basis for making initial estimates. Additional research is ongoing
to provide the necessary data for refining these estimates.
The steps used to estimate emissions are as follows:
Estimate the amount of volatile solids produced for each animal type (VS,),
using published statistics on animal populations, animal sizes, and manure
generation rates.
Estimate the maximum CH4-producing capacity for the manure from each
animal type (Boi), based on published literature values and animal diets.
Define the manure management systems in use, and for each system estimate
its CH4-producing potential (CH4 conversion factor, or MCFj).
For each animal type, estimate the fraction of total manure managed by each
manure management system (management system percentage, or
Page 9-3
-------
Estimate CH4 emissions for each animal type and manure management system
(TM,,) by multiplying the amount of volatile solids produced by the animal type
(VS,) by the CH4-producing capacity of the manure (Boi) by the CH4-producing
potential of the manure management system (MCFj) by the percentage of the
manure handled by the management system (MS%|,).4
Total CH4 emissions will be the sum over all animal types (/) and all manure
systems (/)•
Currently, Hashimoto and Steed (1993) are developing improved estimates of MCFs for key
manure management systems, including examining the variability of the MCF values with
temperature. The initial results of this investigation have been incorporated into this
assessment by assigning each country to one of three climate categories based on mean
annual temperature.5
Using this approach, total annual CH4 emissions (TM,) for animal type / in a particular
climate region is the sum of annual emissions over all applicable manure systems /:
= E MS/ • Bal
MS%
ij
(9.1)
where:
VS, =
total volatile solids produced annually (in kilograms) for animal type /;
maximum CH4-producing capacity per kilogram of VS for animal /;
CH4 conversion factor for manure system j in the climate region; and
percentage of the animal type /"s manure handled using manure
management system j.
The amount of volatile solids produced depends on the number of animals and their mass:
VS, =
TAM • VS
,
,
(9.2)
where:
A/, =
TAM =
number of animals;
typical animal mass in kilograms; and
average annual volatile solids production per unit of animal mass.
In many cases, either the typical animal mass (TAM) or volatile solids production per unit of
animal mass (vs,), or both, are not known. In these cases, total annual VS production per
head was estimated directly, so that the total annual VS production for the population of
animals can be estimated by:
* In Safley et al. (1992), MS% is referred to as WS% for waste system percentage.
5 The mean annual temperature is an imperfect predictor of the mean annual MCF. Based on Hashimoto and
Steed's preliminary analysis, the relationship between temperature and MCF is nonlinear and increasing in the
range between 0° and 30°C. This is particularly true for liquid-based systems. For liquid-based systems in many
temperate climates, using the mean annual temperature to predict the mean MCF will most likely understate the
true mean MCF. The preferred approach would be to use the mean monthly temperature to predict the mean
monthly MCF and to average these monthly MCFs for an annual figure. However, accurate monthly temperature
data for all the countries of the world were not readily available.
Page 9-4
-------
where:
= N,
vs.
(9.3)
A/, = number of animals; and
vs, = average annual volatile solids production per head.
Total annual CH4 emissions from all animals (TM) is estimated as the sum over all / animal
types: '
TM = £ TM,
(9.4)
The following sections describe the data collected by Safley et al. and Hashimoto and Steed
to implement this method.
9.3.1 Volatile Solids (VS) Production
Methane emissions from livestock manure are related to the amount of volatile solids
(VS) produced. The data required to estimate total VS production are shown in Equation 9.2:
the number of animals (A/,), their average size (TAMt), and their average VS production per
unit of animal size (vs;). Animal population data are available for all countries. For cattle and
buffalo, the animal size data used to estimate CH4 emissions from enteric fermentation
(Chapter 2) are used here also. For the other species in developed countries,6 data on animal
size and VS production per unit of animal size are generally available, and Equation 9.2 can
be implemented. For developing countries, however, animal size data are not generally
available, and total VS production must be estimated using Equation 9.3.
For each country, the animal populations were divided into the following 11 categories:
beef cattle,7 dairy cattle, swine, sheep, goats, chickens, ducks, turkeys, horses, donkeys, and
camels. Population statistics from FAO (1992) were used. For 13 developed countries, data
for estimating the typical animal mass (TAM) for each category were available from Meat and
Dairy Products (1988) and Taiganides and Stroshine (1971).8 For developed countries without
country-specific TAM data, the average of the values for the 13 countries was used.
Total VS production rates per unit of animal mass for the United States were used as
the basis for VS production for the developed countries. In some cases, small adjustments
were made to several of the rates to reflect differences in diets. Table 9-1 presents average
statistics for the TAM, manure production, and VS production estimates for the developed
countries.
6 For purposes of this analysis, developed countries include: Albania, Australia, Austria, Belgium, Bulgaria,
Canada, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Israel, Italy, Japan, the Netherlands, New
Caledonia, New Zealand, Norway, Poland, Portugal, South Africa, Sweden, Switzerland, United Kingdom, the
United States, the former Soviet Union, the former Yugoslavia, and the former Czechoslovakia.
7 The category "beef cattle" includes buffalo.
8 The 13 countries for which TAM data were obtained are: Australia, Austria, Belgium, Denmark, France, the
former West Germany, Ireland, Italy, the Netherlands, South Africa, the United Kingdom, the former Soviet Union,
and the former Yugoslavia. TAM data were not required for the United States because VS/head data were
available for animal types in the United States.
Page 9-5
-------
TABLE 9-1
LIVESTOCK MANURE PRODUCTION DATA FOR DEVELOPED COUNTRIES8
Animal Type
Beef Cattle"
Dairy Cattle
Swine
Sheep
Goats
Chickens
Ducks
Turkeys
Horses0
Donkeys
Camels
Typical Animal
Mass (TAM)
(kg)
375.0
550.0
59.0
67.0
64.0
1.1
1.4
6.8
450.0
300.0
450.0
Total
Kg/Day/
TAM
21.80
47.30
5.00
2.70
2.60
0.10
0.15
0.30
23.00
15.30
23.00
Manure Production
Kg/Day/
1000 Kg Mass
58
86
85
40
41
91
107
44
51
51
51
Volatile Solids
Percent of
Total Manure
Production
12.4
11.6
10.1
23.0
26.6
19.4
17.3
19.4
19.6
19.6
16.0
Production
Kg/Day/
1000 Kg Mass
7.2
10.0
8.6
9.3
10.8
17.6
18.5
8.6
10.0
10.0
8.2
a. Average values estimated for developed countries. Individual country data are used when
available.
b. Includes buffalo.
c. Includes mules.
Sources: Safley et al., 1992. TAM for cattle taken from Chapter 2 of this report.
With the exception of cattle and buffalo, estimates of the typical animal mass (TAM) for
each category were not made for the developing countries because of the lack of data.
Instead, manure and VS production per head were estimated directly using published
literature sources (Jain et al., 1981; Ramen et al., 1989; Singh et al., 1985; Gunaseelan,
1987; Lichtman, 1983; Gorkhali, 1984; and Chen et al., 1988). Table 9-2 summarizes the
manure production and VS production estimates by animal type for developing countries.
9.3.2 Maximum Methane-Producing Capacity (B0)
The maximum amount of CH4 that can be produced per kilogram of VS (B0) varies by
animal type and diet. Safley et al. found that little data are available describing the B0 values
for manures produced in countries other than the United States. Consequently, Safley et al.'s
B0 estimates for countries other than the United States were adopted here. In making their
estimates, Safley et al. considered the impact of the variation in livestock diets around the
world on the CH4-producing potential by estimating the energy content of the feed consumed
by the animals in different regions. Table 9-3 lists the B0 values adopted for developed and
developing countries.
Page 9-6
-------
TABLE 9-2
LIVESTOCK MANURE PRODUCTION DATA FOR DEVELOPING COUNTRIES
Animal Type
Cattle (nondairy)3
Dairy Cattle
Swine
Sheep
Goats
Chickens
Ducks
Turkeys
Horses and Mules
Donkeys
Camels
Total Manure
Production
(kg/head/day)
13.60
27.50
4.10
1.60
1.80
0.12
0.12
0.26
18.40
12.20
18.40
Volatile Solids Production
Percent of Total
Manure Production
12
12
10
23
27
19
17
19
20
20
16
Kg/Head/Day
1.70
3.20
0.41
0.37
0.49
0.02
0.02
0.05
3.70
2.40
2.90
a. Includes buffalo.
Sources: Safley et al., 1992. Estimates for cattle based on TAM data in
Chapter 2 of this report.
9.3.3 Descriptions of Manure Management Systems
Various manure management practices are in use throughout the world. In many parts
of the world, manure is spread on fields as a fertilizer. In other cases, manure is used as an
energy source. The following is a brief description of the major systems in use for managing
livestock manure.
Pasture/Range
Daily Spread
Solid Storage
Drylot
The manure from pasture- and range-grazing animals is allowed to lie
as is, and is not handled at all.
Manure is collected in solid form by some means, such as scraping.
The collected manure is applied to fields regularly (usually daily).
Manure is collected as in the daily spread system, but is stored in bulk
for a long period of time (months) before any disposal.
In dry climates animals may be kept on unpaved feedlots where the
manure is allowed to dry until it is periodically removed. Upon removal,
the manure may be spread on fields.
Page 9-7
-------
TABLE 9-3
Bn VALUES FOR DEVELOPED AND DEVELOPING COUNTRIES
Animal Type
Cattle (nondairy) and buffalo
Dairy Cattle
Swine
Sheep
Goats
Chickens
Ducks
Turkeys
Horses and Mules
Donkeys
Camels
United States
(m3 CH4 / kg VS)
0.17/0.333
0.24
0.36/0.47"
0.19/0.36°
0.17
0.30/0.34"
0.32
0.30
0.33
0.33
0.26
Developed Countries
(Non-U.S.)
(m3 CH4 / kg VS)
0.17
0.24
0.45
0.19
0.17
0.32
0.32
0.30
0.33
0.33
0.26
Developing
Countries
(m3 CH4 / kg VS)
0.10
0.13
0.29
0.13
0.13
0.24
0.24
0.24
0.26
0.26
0.21
a. The lower value is for cattle not in feedlots. The higher value is for feedlot cattle.
b. The lower value is for breeder hogs. The higher value is for market hogs.
c. The lower value is for sheep not in feedlots. The higher value is for feedlot sheep.
d. The lower value is for broilers. The higher value is for layers.
Source: Safley etal., 1992.
Deep-Pit Stacks
Utter
Paddock
Liquid/Slurry
With caged layers, the manure may be allowed to collect in solid form in
deep pits (several feet deep) below the cages. The manure in the pits
may only be removed once a year. This manure generally stays dry.
Broilers and young turkeys may be grown on beds of litter, and the
manure/litter pack is removed periodically between flocks. This manure
will not be as dry as with deep pits, but will still be in solid form.
Horses are frequently kept in paddocks where they are confined to a
limited area, but not entirely confined to their stalls. This system is
similar to pasture and drylot manure systems.
These systems are generally characterized by large concrete-lined tanks
built into the ground. Manure is stored in the tank for six or more
months until it can be applied to fields. To facilitate handling as a liquid,
water may be added to the manure.
Page 9-8
-------
Anaerobic Lagoon
Pit Storage
Anaerobic Digester
Burned for Fuel
Building Material
Anaerobic lagoon systems are characterized by flush systems that use
water to transport manure to lagoons. The manure resides in the
lagoon for periods from 30 days to over 200 days. The water from the
lagoon may be recycled as flush water or used to irrigate and fertilize
fields.
Liquid swine manure may be stored in a pit while awaiting final disposal.
The length of storage time varies, and for this analysis is divided into
two categories: less than one month or greater than one month.
The manure, in liquid or slurry form, is anaerobically digested to produce
CH4 gas for energy. The amount of CH4 produced depends on the
operating characteristics of the digester and the characteristics of the
manure. The digester effluent is often used as a fertilizer.
Manure is collected and dried in cakes and burned for heating or
cooking. This system is common in Asia and the Far East; in India it is
estimated that two-thirds of cattle manure is burned for fuel (NCAER,
1965).
Manure is used as a building material in South Asia. It is applied to
walls and floors to seal cracks and may have antiseptic properties.
9.3.4 Methane Conversion Factors (MCFs)
The extent to which the maximum CH4-producing capacity (B0) is realized for a given
livestock manure management system must be known to determine the amount of CH4 that is
emitted. This fraction is defined as the Methane Conversion Factor (MCF) for the manure
system. For example, a manure system that produces no CH4 emissions would have an MCF
of 0. A manure system that achieves the full potential CH4 emissions would have an MCF
of 1.
To assess the MCF values for the wide range of livestock manure management
systems defined in the previous section, two broad classifications of livestock manure-handling
systems can be defined, based on the total solids content of the manure:
Solid systems have a total solids content greater than about 20%.
Liquid/slurry systems have a total solids content less than 20%.
Liquid/slurry systems will typically cause the development of anaerobic conditions, which result
in CH4 production. The solid systems limit the development of anaerobic conditions and, thus,
limit the amount of CH4 that is produced from the manure.
Safley et al. reviewed the literature to investigate the appropriate range of MCF values
for manure management systems and found that data on MCF values were not available for
some systems. Consequently, they estimated values using the broad classifications of solid
versus liquid/slurry management. To improve the MCF estimates, the U.S. EPA is sponsoring
research to estimate better the MCFs for several key livestock manure systems (Hashimoto
and Steed, 1993). Preliminary findings from this research indicate that:
Page 9-9
-------
The estimated MCF value of dry in situ pasture, range, paddock, and solid
storage manure is 1-2%. The estimated MCF for drylot manure is 1-5%.
These values are lower than the 5-10% values used by Safley et al. However,
to date the new analysis has not considered the effect of moisture, for example,
from periodic rainfall or humidity. Consequently, these new values are most
representative for arid regions.
These estimates are consistent with the findings of Lodman et al. (1993) and
Williams (1993). For range-deposited manure, both studies report MCFs of less
than 1%; for feedlot-deposited manure, Lodman et al. (1993) estimate an
overall MCF of 5%. Lodman et al.'s results are based on U.S. beef cattle, and
Williams's results are based on Australian free-range dairy cows.
The MCF for liquid/slurry and pit storage varies between 10% at 10°C and 65%
at 30°C. Safley et. al. used a value of 20%.
The MCF for daily spread is less than 1 %, whereas the Safley et al. estimate
was 5%.
The MCF for anaerobic lagoons is about 90%. This estimate is based on
continuous CH4 measurements taken over a two and one-half year period at a
North Carolina dairy (Safley, 1991) and includes an assessment of surface
oxidation of the CH4. The MCF of 90% for anaerobic lagoons is the same
value as was used by Safley et al.
The MCFs for an individual country will depend on the temperature (see footnote 5). For
purposes of this analysis, three climate categories were defined, corresponding to the
temperature range used by Hashimoto and Steed (1993) to estimate MCFs:
Warm climates - having a mean annual temperature near or above 30°C;
Temperate climates - having a mean annual temperature near 20°C; and
Cool climates - having a mean annual temperature near or below 10°C.
Countries were assigned to each category based on Hammond Incorporated (1981). The
MCF values for each system and climate category are listed in Table 9-4.
9.3.5 Livestock Manure Management System Usage (MS%)
Safley et al.'s estimates of the use of the manure management systems was used
here. Safley et al. divided the world into eight primary regions, generally along the well-
recognized geographic and economic classifications used by the Food and Agriculture
Organization of the United Nations (FAO, 1992). Detailed information on the use of manure
management systems was collected for selected countries by contacting the Ministry of
Agriculture in each country. In addition, individual researchers in many countries provided
information. Countries for which manure system information was not obtained were assumed
to have manure system usage comparable to countries in their region with similar gross
national product (GNP) per capita values and similar climates. Table 9-5 lists the percentages
of manure managed by the major systems for each region of the world.
Page 9-10
-------
TABLE 9-4
METHANE CONVERSION FACTORS (MCFs) FOR LIVESTOCK MANURE SYSTEMS
MCFs Based on
Laboratory Measurement
Pasture, Range, Paddock3
Liquid/Slurry3
Pit Storage < 30 days3
Pit Storage > 30 days3
Drylot"
Solid Storage3
Daily Spread3
MCF at 30°C
2%
65%
33%
65%
5%
2%
1%
MCF at 20°C
1.5%
35.0%
18.0%
35.0%
1.5%
1.5%
0.5%
MCF at 10°C
1.0%
10.0%
5.0%
10.0%
1.0%
1.0%
0.1%
MCF Measured by
Long-Term Field Monitoring
Anaerobic Lagoon0
Average Annual MCF
90%
MCFs Estimated by Safley et al.
Litter"
Deep-Pit Stacking"
Anaerobic Digester (Chinese)8
Anaerobic Digester (Indian)6
Burned for Fuel'
Average Annual MCF
10%
5%
14%
5%
5-10%
a. Hashimoto and Steed, 1993.
b. Based on Hashimoto and Steed, 1993.
c. Safley et al., 1992, and Safley and Westerman, 1992.
d. Safley et al., 1992.
e. MCFs taken from Safley et al. (1992). Yancun et al. (1985) showed that typical
Chinese digesters leaked at least 14% of the CH4 produced. The typical floating-
cover digesters used in India do not appear to be as leak-prone as the Chinese
digesters (Stuckey, 1984; and Lichtman, 1983), so an MCF of 5% was adopted.
f. Methane emissions associated with combustion are not included in this estimate.
Page 9-11
-------
r
TABLE 9-5
LIVESTOCK MANURE SYSTEM USAGE
Animal
Type
Nondaiiy Cattle"
Dairy Cattle
Poultry"
Sheep
Swtne
Other Animals'
Anaerobic
Lagoons
0%
10%
5%
0%
25%
0%
Liquid
Systems3
1%
. 23%
4%
0%
50%
0%
Solid
Daily Storage
Spread &
North America
0%
37%
0%
0%
0%
0%
Drylots
14%
23%
0%
2%
18%
0%
Pastures,
Ranges, &
Paddocks
84%
0%
1%
88%
0%
92%
Used
for
Fuel"
0%
0%
0%
0%
0%
0%
Other
Systems0
1%
7%
90%
10%
6%
8%
Western Europe
Nondairy Cattle"
Dairy Cattle
Poultry"
Sheep
Swine
Other Animals'
Nondairy Cattle"
Dairy Cattle
Poultry0
Sheep
Swine
Other Animals'
Nondairy Cattle"
Dairy Cattle
Poultry"
Sheep
Swine
Other Animals'
Nondairy Cattle"
Dairy Cattle
Poultry"
Sheep
Swine
Other Animals'
0%
0%
0%
0%
0%
0%
8%
0%
0%
0%
0%
0%
0%
0%
0%
0%
55%
0%
0%
0%
0%
0%
0%
0%
55%
46%
13%
0%
77%
0%
39%
18%
28%
0%
29%
0%
0%
0%
0%
0%
0%
0%
0%
1%
9%
0%
8%
0%
0%
24%
0%
0%
0%
0%
Eastern Europe
0%
1%
0%
0%
0%
0%
Oceania
0%
0%
0%
0%
0%
0%
Latin America
0%
62%
0%
0%
2%
0%
2%
21%
1%
2%
23%
0%
52%
67%
0%
0%
0%
0%
0%
0%
0%
0%
17%
0%
0%
1%
0%
0%
51%
0%
33%
8%
2%
87%
0%
96%
0%
13%
1%
73%
27%
92%
100%
100%
3%
100%
0%
100%
99%
36%
42%
100%
0%
99%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
9%
1%
84%
11%
0%
4%
1%
0%
71%
27%
45%
8%
0%
0%t
98%
0%
28%
0%
1%
0%
49%
0%
40%
1%
(continued)
Page 9-12
-------
TABLE 9-5
LIVESTOCK MANURE SYSTEM USAGE (Continued)
Animal
Type
Nondairy Cattled
Dairy Cattle
Poultry6
Sheep
Swine
Other Animals'
Nondairy Cattle"
Dairy Cattle
Poultry6
Sheep
Swine
Other Animals'
Nondairy Cattle"
Dairy Cattle
Poultry6
Sheep
Swine
Other Animals'
Anaerobic
Lagoons
0%
0%
0%
0%
0%.
0%
0%
0%
0%
0%
0%
0%
0%
6%
1%
0%
1%
0%
Liquid
Systems3
0%
0%
0%
0%
7%
0%
Near East
0%
0%
1%
0%
32%
0%
Asia
0%
4%
2%
0%
38%
0%
Daily
Spread
Africa
1%
12%
0%
0%
0%
0%
Solid
Storage &
Drylots
3%
0%
0%
1%
93%
0%
Pastures,
Ranges, &
Paddocks
96%
83%
81%
99%
0%
99%
Used
for
Fuel"
0%
0%
0%
0%
0%
0%
Other
Systems0
0%
5%
19%
1%
0%
1%
and Mediterranean
2%
3%
0%
0%
0%
0%
and Far
16%
21%
0%
0%
1%
0%
0%
3%
0%
0%
68%
0%
East
14%
0%
0%
0%
53%
0%
77%
77%
71%
100%
0%
100%
29%
24%
44%
83%
0%
95%
18%
18%
0%
0%
0%
0%
40%
46%
1%
0%
7%
0%
2%
0%
28%
0%
0%
0%
0%
0%
52%
17%
0%
5%
a. Includes liquid/slurry storage and pit storage.
b. Includes anaerobic digesters and burned for fuel.
c. Includes deep-pit stacks, litter, and other systems.
d. Includes buffalo.
e. Includes chickens, turkeys, and ducks.
f. Includes goats, horses, mules, donkeys, and camels.
Source: Safley et al., 1992.
Page 9-13
-------
9.4 RESULTS
This report estimates that 1990 global CH4 emissions from livestock manure were
about 14 Tg. Table 9-6 summarizes the global distribution of emissions by region and animal
type. Table 9-7 presents the global distribution of emissions by region and livestock manure
management system. The major findings are that:
Of the 14 Tg, liquid livestock manure management systems (liquid/slurry
storage, pit storage, and anaerobic lagoons) account for about 8.6 Tg, or about
60% of total emissions from livestock manure. These systems are used at
confined, intensive livestock operations in developed countries and may provide
profitable opportunities to recover CH4 for use as a fuel. Currently, virtually no
CH4 is recovered from these systems for use as fuel.
Of the 14 Tg, three regions account for 85% of the total: Europe (Eastern and
Western), with 5.9 Tg (42%); Asia and the Far East, with 3.9 Tg (28%); and
North America, with 2.3 Tg (16%).
Of the 14 Tg, about 11.4 Tg are from three animal groups: swine, cattle (beef
and draft animals), and dairy cows.
Table 9-8 summarizes the emission contributions from the countries in each region with the
greatest emissions.
9.5 TRENDS
Future CH4 emissions from livestock manure will be driven by future levels of animal
production and manure management system usage. As discussed in Chapter 2 of this report,
increasing human population will lead to increasing production of milk, meat, hides, and other
livestock products. This trend is contingent on continued economic growth and is subject to
many factors. Parikh et al. (1988) project a 1.6% annual increase in milk and bovine/ovine
meat production through the year 2000. U.S. EPA (1989) projects a similar but slightly lower
growth rate. Growth rates are expected to be larger among developing countries because of
greater population growth and because of the high income elasticities of demand for livestock
products among the relatively poorer populations of most developing countries.
Some have questioned whether developing countries can achieve significant increases
in animal production because livestock feed availability may impose a constraint on production
increases. Parikh et al. (1988) address this issue and expect feed costs to increase as
supplies tighten in some regions. However, assuming that economic development continues,
the prices for livestock products are also expected to increase, substantially offsetting the
increased cost of feed.
Page 9-14
-------
TABLE 9-6
METHANE EMISSIONS BY ANIMAL TYPE AND REGION FOR 1990 (Tg CH4)
Animal
Type
Dairy Cattle
Nondairy
Cattle3
Swine
Poultry
Other Animals"
Total0
North
America
0.8
0.2
1.1
0.3
<0.1
2.3
Western
Europe
1.1
0.9
0.9
0.1
0.1
3.1
Eastern
Europe
0.5
0.9
0.9
0.2
0.3
2.8
Latin
Oceania America Africa
<0.1 0.1 <0.1
<0.1 0.3 0.1
0.1 0.2 <0.1
<0.1 0.2 <0.1
0.1 0.2 0.2
0.3 0.9 0.3
Near
East
&
Med.
0.1
0.1
<0.1
<0.1
0.1
0.3
Asia
&Far
East
0.4
0.8
2.0
0.4
0.3
3.9
Total0
2.9
3.2
5.3
1.3
1.2
13.9
a. Includes buffalo.
b. Includes sheep, goats, horses, mules, asses, and camels.
c. Totals may not add due to rounding.
TABLE 9-7
METHANE EMISSIONS BY REGION AND SYSTEM FOR 1990 (Tg CH4)
Manure
Management
System
Pasture/Range
Liquid/Slurry"
Solid Storage
Anaerobic
Lagoon
Drylot
Burned for Fuel
Daily Spread
Other Systems
Total3
North
America
0.1
0.5
0.1
1.4
0.0
0.0
0.0
0.1
2.3
Western
Europe
0.1
2.7
0.1
0.0
0.0
0.0
0.0
0.1
3.1
Eastern
Europe
0.1
1.0
0.3
0.6
0.0
0.0
0.0
0.7
2.8
Oceania
0.2
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.3
Latin
America
0.5
0.2
0.1
0.0
0.1
0.0
0.0
0.1
0.9
Africa
0.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
Near
East
&Med.
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.3
Asia
&Far
East
0.3
1.8
0.2
0.4
0.2
0.8
0.0
0.2
3.9
Total3
1.7
6.1
0.9
2.5
0.3
0.9
0.1
1.4
13.9
a. Totals may not add due to rounding. Values listed as 0.0 are less than 0.05.
b. Includes pit storage.
Page 9-15
-------
TABLE 9-8
METHANE EMISSIONS FOR MAJOR COUNTRIES FOR 1990 (Tg CH4)
Country
Canada
United States
Total
France
Germany (West)
Italy
Netherlands
Spain
United Kingdom
Other
Total
Germany (East)
Hungary
Poland
Former Soviet Union
Other
Total
Australia
New Zealand
Other
Total
Argentina
Brazil
Mexico
Other
Total
Ethiopia
Nigeria
Somalia
South Africa
Tanzania
Other
Total
Dairy
0.02
0.73
0.76
0.58
0.09
0.11
0.03
0.05
0.06
0.13
1.05
0.02
0.00
0.03
0.38
0.06
0.48
0.01
0.01
0.00
0.02
0.00
0.03
0.01
0.04
0.07
0.01
0.00
0.00
0.00
0.01
0.02
0.04
Sheep &
Cattle3 Swine Goats
0.03
0.17
0.20
North America
0.06
1.05
1.11
Western Europe
0.32 0.19
0.09
0.20
0.03
0.03
0.08
0.11
0.85
0.11
0.18
0.06
0.20
0.03
0.15
0.92
Eastern Europe
0.04 0.16
0.01 0.39
0.05 0.03
0.71
0.10
0.90
0.03
0.01
0.00
0.04
0.04
0.13
0.02
0.08
0.27
0.02
0.01
0.00
0.00
0.01
0.04
0.08
0.15
0.15
0.89
Oceania
0.12
0.02
0.00
0.14
Latin America
0.01
0.12
0.04
0.05
0.22
Africa
0.00
0.00
0.00
0.00
0.00
0.02
0.03
0.00
0.00
0.00
0.01 .
0.00
0.01
0.00
0.01
0.02
0.02
0.07
0.00
0.00
0.01
0.15
0.03
0.19
0.08
0.03
0.00
0.10
0.01
0.01
0.00
0.01
0.03
0.01
0.00
0.01
0.01
0.01
0.03
0.08
Poultry"
0.03
0.22
0.25
0.04
0.01
0.02
0.01
0.01
0.02
0.02
0.14
0.01
0.01
0.01
0.17
0.04
0.24
0.01
0.00
0.00
0.01
0.01
0.08
0.03
0.04
0.16
0.00
0.01
0.00
0.00
0.00
0.01
0.03
Other0
0.00
0.02
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.02
0.00
0.00
0.03
0.02
0.01
0.06
0.00
0.00
0.00
0.00
0.01
0.04
0.04
0.05
0.14
0.03
0.00
0.02
0.00
0.00
0.02
0.07
Total
0.15
2.19
2.34
1.15
0.30
0.51
0.14
0.31
0.21
0.44
3.05
0.22
0.41
0.16
1.58
0.39
2.75
0.24
0.07
0.01
0.31
0.08
0.41
0.14
0.27
0.89
0.07
0.03
0.03
0.03
0.03
0.14
0.33
(continued)
Page 9-16
-------
TABLE 9-8
METHANE EMISSIONS FROM MAJOR COUNTRIES FOR 1990 (Tg CH4)
(Continued)
Country
Dairy
Cattle3
Sheep &
Swine Goats
Poultry"
Other0
Total
Middle East & Mediterranean
Egypt
Sudan
Turkey
Other
Total
China
India
Indonesia
Japan
Viet Nam
Other
Total
0.01
0.01
0.02
0.02
0.06
0.05
0.18
0.01
0.13
0.00
0.04
0.4 1
0.01
0.02
0.01
0.01
0.06
Asia
0.08
0.50
0.02
0.05
0.01
0.09
0.77
0.00
0.00
0.00
0.00
0.00
& Far East
1.49
0.01
0.04
0.18
0.09
0.17
1.99
0.00
0.01
0.01
0.03
0.05
0.07
0.04
0.05
0.00
0.00
0.02
0.19
0.00
0.00
0.00
0.03
0.04
0.13
0.06
0.03
0.12
0.01
0.07
0.42
0.00
0.01
0.01
0.03
0.05
0.08
0.01
0.03
0.00
0.00
0.02
0.15
0.03
0.05
0.06
0.12
0.25
1.91
0.81
0.19
0.48
0.12
0.42
3.93
World Total
2.89
3.16
5.29
0.71
1.28
0.51
13.85
a. Nondairy cattle and buffalo.
b. Includes chickens, turkeys, and ducks.
c. Includes horses, mules, asses, and camels.
The level of production determines how much manure is produced; the systems used
to manage the manure determine how much CH4 is produced. Unlike livestock production,
forecasts of future livestock manure management practices are not available. However, if
current trends toward the increasing use of confined and intensive livestock production
systems continue, then there will likely be greater use of liquid-based manure-handling
systems, such as lagoons and liquid/slurry storage systems. Such systems are often
preferred for large-scale livestock production systems because they allow for the efficient
collection, storage, and, in some cases, treatment, of livestock manure. Because liquid
systems produce significantly more CH4 than solid systems, a shift toward liquid systems
would result in significantly higher emissions in the future. At this time, there are no plans for
collection of CH4 from liquid manure management systems on a large scale.
9.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
The estimates presented above should be regarded with caution. Some of the data on
which they are based are uncertain. At this time, insufficient information exists to provide a
statistical confidence limit for these emission estimates. The greatest uncertainty results from
the MCF assumptions for the manure management systems. While assumptions concerning
other factors are also uncertain, their contribution to the overall uncertainty is likely to be less.
Page 9-17
-------
To capture the uncertainty in these estimates, this chapter defines "high"- and "low"-case
emission estimates as follows:
High case. The MCFs for dry manure management systems are double that of
the base case. This is plausible, given that the dry-system measurements
performed by Hashimoto and Steed simulated arid conditions.
Low case. The MCFs for the liquid systems are half that of the base case.
This is plausible if liquid systems are much less anaerobic than indicated by
measurements in the United States.
Table 9-9 lists the MCF assumptions used to estimate the low and high cases.
TABLE 9-9
BASE-, HIGH-, AND LOW-CASE EMISSION ESTIMATE ASSUMPTIONS
Methane Conversion Factors (MCFs)
Relative to the Base Case
Management System
High Case
Low Case
Solid Systems
Pasture/Range
Drylot
Solid Storage
Liquid Systems
Liquid/Slurry Storage
Pit Storage, less than 1 month
Pit Storage, greater than 1 month
Anaerobic Lagoon
Double
Double
Double
Unchanged
Unchanged
Unchanged
Unchanged
Unchanged
Unchanged
Unchanged
Half
Half
Half
Half
As shown in Table 9-10, the range of emissions implied by these cases is about 10-18
Tg. The high-estimate assumptions have the greatest influence on the estimates for areas
with animals managed on dry systems, such as Latin America and Eastern Europe. The low-
case estimates have the largest impact on those areas that use liquid systems, including
North America and Western Europe.
The true uncertainty in the estimates may be larger because this range only reflects
the uncertainty in several key variables. Limitations of the analysis that must be considered
include:
Little information is available to assess the CH4 produced by pasture and range
manure. Because a large fraction of livestock manure is managed on pastures
and range, this creates uncertainty in the overall emission estimate.
The CH4-producing potential of liquid/slurry and pit storage manure systems
may be much greater than assumed in this report. Because of the widespread
use of these systems, total emissions may be underestimated.
Page 9-18
-------
TABLE 9-10
EMISSION ESTIMATE RANGES FOR 1990 (Tg CH4)
Region
Base Case High Case Low Case
North America
Western Europe
Eastern Europe
Oceania
Latin America
Africa
Near East & Mediterranean
Asia & Far East
2.3
3.1
2.8
0.3
0.9
0.3
0.3
3.9
2.8
3.4
3.9
0.5
1.6
0.6
0.4
4.9
1.4
1.7
2.0
0.2
0.8
0.3
0.3
2.8
Total
13.9
18.2
9.5
Limited data exist on the numbers and characteristics of livestock manure
systems in use in some parts of the world.
In some countries, limited information is available on livestock populations and
livestock characteristics, including manure production.
The U.S. EPA is sponsoring research to verify the MCFs for several key livestock
manure systems, including liquid/slurry storage, drylots, and pasture/range. In addition,
research is necessary to measure the CH4 capacity (B0) of livestock manure in developing
countries and to improve the characterization of livestock manure management systems
throughout the world.
9.7 CONCLUSIONS
The amount of CH4 emitted by animal manure depends on how the manure is
managed. Liquid manure management systems, including lagoons and liquid/slurry storage
pits, tend to promote CH4 formation and emission. Dry manure management systems tend to
have lower CH4 emissions. Because liquid management systems are used primarily in large-
scale confined livestock management facilities in developed countries, the developed countries
are the primary source of CH4 emissions from livestock manure.
The method used to estimate emissions of 10-18 Tg CH4 in this study is based on
previous methods discussed in OECD (1991) and Safley et al. (1992). This emission estimate
is lower than the previous estimates, however, because recent investigations indicate that
emission rates from dry systems are lower than estimated previously. Investigations are
continuing, however, and additional data are needed to reduce uncertainty in the estimates.
Page 9-19
-------
9.8 REFERENCES
Chawla, OP. 1986. Advances in Biogas Technology. Indian Council of Agricultural
Research, New Delhi, India.
Chen, T.H., D.L. Day, and M.P. Steinberg. 1988. Methane production from fresh versus dry
dairy manure. Biological Wastes 24:297-306.
FAO (Food and Agriculture Organization of the United Nations). 1992. FAO Yearbook:
Production 1991. vol. 44. FAO, Rome, Italy.
Gorkhali, H.G. 1984. Summary of the Nepal biogas program. In EI-Halwagi, M.M., ed.
B!ogas Technology, Transfer and Diffusion. Elsevier, New York, New York. 665-668.
Gunaseelan, V.N. 1987. Parthenium as an additive with cattle manure in biogas production.
Biological Wastes 21:195-202.
Hammond, Incorporated. 1981. Hammond World Atlas. Hammond, Incorporated, Maplewood,
New Jersey.
Hashimoto, A., and J. Steed. 1993. Methane emissions from typical U.S. livestock manure
management systems. Draft report prepared for ICF, Incorporated, under contract to the U.S.
Environmental Protection Agency, Washington, D.C.
Jain, M.K., R. Singh, and P. Tauro. 1981. Anaerobic digestion of cattle and sheep waste.
Agricultural Wastes 3:65-73.
Lichtman, R.J. 1983. Biogas Systems in India. Volunteers In Technical Assistance,
Arlington, Virginia.
Lodman, D.W., M.E. Branine, B.R. Carmean, P. Zimmerman, G.M. Ward, and D.E. Johnson.
1993. Estimates of methane emissions from manure of U.S. cattle. Chemosphere
26:189-199.
Meat and Dairy Products. 1988 (May). Commonwealth Secretariat Publications, Marlborough
House, London, United Kingdom.
NCAER (National Council of Applied Economic Research). 1965. Domestic Fuels in Rural
India. New Delhi, India. As cited in Odend'hal (1972).
Odend'hal, S. 1972. Energetics of Indian cattle and their environment. Human Ecology
1:3-22.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Parikh, K.S., G. Fischer, K. Frohberg, and O. Gulbrandsen. 1988. Towards Free Trade in
Agriculture. International Institute for Applied Systems Analysis. Martinus Nijhoff Publishers,
Boston, Massachusetts.
Page 9-20
-------
Ramen, P., V.V. Ranga Rao, and V.V.N. Kishore. 1989. A static scum-breaking net for
fixed-dome biogas plants. Biological Wastes 30:261-273.
Safley, LM. 1991 (January). Personal communication with Dr. Lawson Safley. Professor of
Biological and Agricultural Engineering. North Carolina State University. Raleigh, North
Carolina.
Safley, L.M., M.E. Casada, J.W. Woodbury, and K.F. Roos. 1992. Global Methane Emissions
from Livestock and Poultry Manure. EPA/400/1091/048, U.S. EPA, Washington, D.C.
Safley, L.M., Jr., and P.W. Westerman. 1992. Performance of a low temperature lagoon
digester. Bioresource Technology 41:167-175.
Singh, R., R.K. Malik, and P. Tauro. 1985. Anaerobic digestion of cattle waste at various
retention times: A pilot plant study. Agricultural Wastes 12:313-316.
Stuckey, D.C. 1984. Biogas: A global perspective. In EI-Halwagi, M.M., ed. Biogas
Technology, Transfer and Diffusion. Elsevier, New York, New York. 18-44.
Taiganides, E.P., and R.L. Stroshine. 1971. Impacts of farm animal production and
processing on the total environment. In Livestock Waste Management and Pollution
Abatement: The Proceedings of the International Symposium on Livestock Wastes, April
19-22, 1971, Columbus, Ohio. 95-98. American Society of Agricultural Engineers, St. Joseph,
Missouri.
U.S. EPA (U.S. Environmental Protection Agency). 1989. Policy Options for Stabilizing
Global Climate. Office of Policy, Planning, and Evaluation, Washington, D.C.
Williams, D.J. 1993. Methane emissions from manure of free-range dairy cows.
Chemosphere 26:179-187.
Yancun, C., H. Cong, and Liang Pusen, 1985. Development of a new energy village - Xinbu,
China. In El Mahgary, Y., and A.K. Biswas, eds. Integrated Rural Energy Planning.
Butterworths Publishing, Guildsford, United Kingdom. 99-108.
Page 9-21
-------
-------
CHAPTER 10
METHANE EMISSIONS FROM WASTEWATER
10.1 SUMMARY
Wastewater can be treated using aerobic and/or anaerobic technologies, or if
untreated, can degrade under either aerobic or anaerobic conditions. Methane is produced
when organic material in treated and untreated wastewater degrades anaerobically. Methane
emissions from these sources are estimated to range from 30 to 40 teragrams (Tg)1 per year.
This represents 8-11% of the total global anthropogenic methane emissions, estimated at 360
Tg/yr (IPCC, 1992). Methane from wastewater treatment is potentially amenable to cost-
effective control, and research by the Air & Energy Engineering Research Laboratory of the
U.S. Environmental Protection Agency (U.S. EPA/AEERL) is designed to reduce the
uncertainty associated with emission estimates for this source.
Global and country-specific estimates were developed for methane from domestic
wastewater. For industrial wastewater, only global estimates are presented, as adequate
country-specific data are not available. Industrial wastewater is estimated to be the major
contributor to methane from wastewater emissions, accounting for 26-40 Tg/yr. Domestic
wastewater is estimated to emit approximately 2 Tg/yr, with Asia accounting for 65%.
Substantial uncertainty in these estimates results from a lack of data characterizing
wastewater management practices, the quantities of wastewater that are subject to anaerobic
conditions, the extent to which methane is emitted, and flaring or utilization practices.
Revisions of these initial estimates are planned as field-testing efforts result in more reliable
emission factors and as better data become available on global wastewater management.
10.2 BACKGROUND
Highly organic waste streams, including domestic or commercial wastewater and
wastewater from such industries as food processing and pulp and paper plants, have a high
potential for methane (CH4) emissions. These waste streams quickly deplete available oxygen
as their organic matter decomposes. The concentration of organics in wastewater may be
expressed in terms of biochemical oxygen demand (BOD) in milligrams per liter (mg/l). A
standardized test of BOD is the BOD5, where the amount of oxygen consumed by the waste
under standard conditions is measured over a five-day period. The maximum, or ultimate,
BOD is denoted as BODy. Typically, the BOD5 would be 60-70% of the BODu- Untreated
domestic waste streams would have a BOD5 ranging from 110 to 400 mg/l. Food-processing
facilities, such as fruit-, sugar-, and meat-processing plants; creameries; and breweries can
produce untreated wastewater with a BOD as high as 10,000-100,000 mg/l. In the literature,
BOD may also be expressed as "BOD loading," in kg/capita/day, where the BOD was
multiplied by a wastewater production rate (liters per day per person).
Treatment of wastewater and its residual solids by-product (sludge) under anaerobic
conditions results in CH4 emissions (Figure 10-1). Treatment of wastewater in developed
countries typically occurs aerobically, using aerated impoundments. Digesters are often
1 Teragram = 106 metric tonnes = 1012 grams.
Page 10-1
-------
FIGURE 10-1
WASTEWATER TREATMENT SYSTEMS AND METHANE PRODUCTION
WASTEWATER IN
PRETREATMENT
screens
settling in storage lagoons
BIOLOGICAL TREATMENT
aerobic (shallow or aerated lagoons)
anaerobic (deeper, facultative lagoons)
ADVANCED TREATMENT
disinfection,
treatment of persistent pollutants
SLUDGE TREATMENT
dewatering,
secondary fermentation
SLUDGE OUT
Notes: Treated water can leave the process at various stages. Certain wastewater treatment systems might have
more or fewer stages.
Page 10-2
-------
used, and the gas is either flared or used. However, facultative (containing both anaerobic
and aerobic zones) and anaerobic lagoons may also be used for storage and treatment.
U.S. EPA estimated in 1987 that there are approximately 5,500 municipal waste
stabilization lagoons in the United States that treat 5.2 x 106 m3/day of wastewater (OMPC,
1987). The CH4 potential from these lagoons is not well understood, and little field data are
available. Industrial and commercial wastewater processes also use lagoons for treatment
and storage.
Methane production varies, depending upon temperature, retention time, BOD loading,
and lagoon maintenance. Facultative lagoons, the most common type, treat wastewater by
both anaerobic fermentation and aerobic processes. At the bottom of the lagoon, where an
anaerobic environment exists, organic matter is digested to CH4 and CO2. As these gases
bubble to the surface, much of the CO2 is adsorbed by algae and is used, along with nutrients
liberated during digestion, to produce algal biomass. Aerobic conditions, supported by algal
growth, are maintained near the surface. Between 20 and 30% of the BOD loading to a
facultative pond is anaerobically metabolized. As BOD loading increases and natural surface
aeration diminishes, facultative lagoons proceed to a more anaerobic state. This results in
higher CH4 production, providing that the temperature is higher than 15°C. Under these
conditions, a facultative lagoon may act more as an anaerobic pond, with possibly 95% of the
lagoon volume functioning anaerobically. Fermentation, and thus CH4 production, is negligible
at temperatures below about 15°C, at which point the lagoon serves principally as a
sedimentation tank (Gloyna, 1971).
The depth of the lagoon is also an important factor in CH4 production. Shallow
lagoons, one meter or less in depth, are not expected to produce large quantities of CH4
because the intake of oxygen from the surface, as well as the production of oxygen due to
photosynthesis, prevent the formation of a significant anaerobic zone. Facultative lagoons are
typically 1.2-2.5 meters deep; lagoons deeper than 2.5 meters are typically referred to as
anaerobic lagoons.
10.3 RECOMMENDED METHODOLOGY FOR ESTIMATING EMISSIONS
The methodology used to estimate CH4 emissions from wastewater is based on BOD
loading in the wastewater flow.2 The BOD of wastewater correlates with CH4 production
(Orlich, 1990; and Metcalf and Eddy, Inc., 1979). Wastewater with greater BOD
concentrations will yield more CH4 than wastewater with comparatively lower BOD
concentrations, given equivalent conditions. BOD is a commonly measured parameter, and
data are available on BOD loading rates for domestic wastewater and different kinds of
industrial wastewater. Consequently, data collection efforts focused on obtaining BOD
concentrations in wastewater. It is recognized that other parameters may be important, and
field work is under way by U.S. EPA/AEERL that will help determine what other variables
need to be considered in future refinements of these initial emission estimates. The field work
being conducted will include characterization of wastewater lagoons and septic sewage
systems (Thorneloe, 1992).
2 The draft methodology employed by the IPCC/OECD (IPCC, 1990) does not include CH4 emissions from
wastewater. The revised methodology, due to be released in late 1993, will include this emission source and will
recommend an inventory method that follows what is described here.
Page 10-3
-------
Applying BOD data required the definition of two additional factors. The first factor
expresses the CH4 yield per unit BOD loading and is estimated to be 0.22 kg CH4 per kg
BODS (Orlich, 1990). The second factor accounts for the fraction of the wastewater in which
anaerobic conditions will develop. In absence of country-specific data, values for this factor
are assumed. Many limitations were found in the available data, and research is underway to
reduce the uncertainties with the initial estimate for CH4 emissions from wastewater.
Two equations were developed to estimate wastewater CH4 emissions - one for
domestic wastewater and one for industrial wastewater.
Domestic wastewater:
(Country"
{Population,
kg BOD5l
capita day
365 days}
yr )
f 0.22 kg CH4°
kg BOD5
11 Fraction" j
\\AnaerobicaHy\ =
1 ( Digested )
a. Population Reference Bureau, Inc., 1991. (Used 1990 population.)
b. Gloyna, 1971; Mara, 1976; Mullick, 1987; and Viessman and Hammer, 1985.
c. Orlich, 1990; and Metcalf and Eddy, Inc., 1979.
d. Assignment of this fraction is solely a matter of professional judgment in the absence of country-
specific treatment data.
Data are available for BOD5 loadings in domestic wastewater for some countries
(Gloyna, 1971; Mara, 1976; Mullick, 1987; and Viessman and Hammer, 1985). These data
were used to develop estimates for other countries, and these values were adjusted for
population and BOD concentration.
Industrial wastewater:
World a Wastewater
Outflow m3
y
1,000 I'} ( Kg BOD/ } f0.22 kg CHj
I wastewater) I kg BOD5
| Fraction' |
\Anaerobically\ =
\ Digested )
yr
(10.2)
e. Carmichael and Strzepek, 1987.
f. Assumed 1 m3 = 1,000 kg = 1,000 I (conversion factor for volume).
g. From Table 10-1 of this Chapter.
h. Orlich, 1990; and Metcalf and Eddy, Inc., 1979.
i. Assignment of this fraction is solely a matter of professional judgment in the absence of country-
specific treatment data.
Since country-specific industrial wastewater volumes are not known at this time, the
estimated worldwide wastewater outflow for specific industries had to be used.
10.3.1 Assumptions and Data Used to Estimate Emissions
BOD Loading
The sources of information that were used to obtain BOD values for industrial
wastewater are summarized in Table 10-1. Many of the BOD values were reported as "BOD"
without clarifying whether these values were BOD5 or BODu. All unspecified BOD values were
assumed to be BOD5. This is considered a fairly safe assumption, since BOD measurements
are typically BODS.
Page 10-4
-------
TABLE 10-1
BIOCHEMICAL OXYGEN DEMAND (BOD) ESTIMATES
FOR VARIOUS INDUSTRIAL WASTEWATERS
Industry3
Iron and Steel
BOD
(kg/I)
0.001 b
References and Notes
No references for BOD were obtained. Used the value for
BOD
Nonferrous Metals 0.001b
Fertilizer 0.001b
Food & Beverages 0.035
Fruits/Vegetables 0.003
Cereals 0.001°
Meats 0.020°
Butter 0.003°
Cheese 0.003°
Cane Sugar 0.002"
Beet Sugar 0.01 Ob
Wine 0.135°
Beer 0.085°
Other Beverages 0.083°
Pulp and Paper 0.004b
Petroleum Refining 0.004°
(Petrochemical)
Textiles 0.001"
Rubber 0.001"
\
Miscellaneous" 0.002
in textile wastewaters, p. 67, Carmichael and Strzepek (1987),
since it was the lowest value obtained for industrial sources.
No references for BOD were obtained. Used the value for BOD
in textile wastewaters, p. 67, Carmichael and Strzepek (1987),
since it was the lowest value obtained for industrial sources.
No references for BOD were obtained. Used the value for BOD
in textile wastewaters, p. 67, Carmichael and Strzepek (1987),
since it was the lowest value obtained for industrial sources.
This value is an average of the following categories of the fruit
and beverage industry.
Barnes et al., (1984), p. 213.
U.S. EPA (1974a), pp. 39 & 40.
U.S. EPA (1975), pp. 58 & 60; U.S. EPA (1974c), pp. 39 & 41.
U.S. EPA (1974b), p. 59; and Barnes et al. (1984), p. 316.
U.S. EPA (1974b), p. 59; and Barnes et al. (1984), p. 316.
Barnes et al. (1984), p. 20.
Barnes et al. (1984), p. 12; and U.S. EPA (1974d).
Barnes et al. (1984), p. 73.
Barnes et al. (1984), p. 73.
Barnes et al. (1984), p. 73.
Carmichael and Strzepek (1987), p. 49; and Hall et al. (1988), as
cited in Torpy (1988), p. 20.
Average of values reported in Carmichael and Strzepek (1987),
pp. 33 & 36.
Carmichael and Strzepek (1987), p. 67.
No references for BOD were obtained. Used the value for BOD
in textile wastewaters, p. 67, Carmichael and Strzepek (1987),
since it was the lowest value obtained for industrial sources.
No BOD values were obtained. Used BOD reported for the
pharmaceutical industry in Carmichael and Strzepek (1987), p.
85.
a. Industries presented here are taken from Table 47, pp. 116 and 117 in Carmichael and Strzepek (1987).
b. Reported as BOD. This is assumed to be ultimate BOD.
c. Reported as BOD5.
d. Industries in this group were undefined.
Page 10-5
-------
For domestic wastewater, BOD5 loading can range from 0.023 to 0.091 kg/capita/day.
Domestic wastewater BOD5 loading has been reported from 0.023 to 0.045 kg/day in
developing countries and from 0.024 to 0.059 kg/day in developed countries (the lower value
was reported for rural France (Mara, 1976)). The BOD increases when substantial amounts of
kitchen wastes are discharged to sewers, for instance, as the result of using sink disposals.
Viessman and Hammer (1985) suggest a BOD5 loading for the United States of 0.045-
0.078 kg/capita/day, due to increases in BOD5 from additional loads of organic solids. In fact,
domestic wastewater BOD5 loading can be as high as 0.091 kg/capita/day according to Mullick
(1987, citing Gotass (1956) and Laak (1986)). BOD5 values of the organic load were adapted
from Mara (1976) for domestic wastewater for different geographic regions:
Africa
Asia, Middle East, Central & South America
N. America, Europe, Former USSR, Oceania
0.037 (kg/capita/day)
0.04 (kg/capita/day)
0.05 (kg/capita/day)
Country-Specific Fractions of Anaerobically Degrading Wastewater
Both aerobic and anaerobic methods are used to treat wastewater. Waste in untreated
wastewater may also degrade under either aerobic or anaerobic conditions. Therefore, only
the fraction of the wastewater that is prone to anaerobic conditions needs to be included in
the calculations. This fraction is assumed in the absence of country-specific data. The
following text describes wastewater treatment activities, or the absence thereof, based on the
available information in the literature. The geographic coverage is irregular, and much of the
information is anecdotal. Brief summaries are provided of country-specific data, along with
conclusions on how the data were used. More information is available for domestic
wastewater treatment than for industrial wastewater treatment. It is known that industrial
wastewater in developing countries is often discharged with domestic wastewater (UNEP,
1980). Therefore, for developing countries, any information available to assess the portion of
domestic wastewater that may undergo anaerobic digestion is also applied to industrial
wastewater.
Africa
Where wastewater is treated in Africa, it is most often directed to pit latrines and public
sewer systems. Latrines may be considered aerobic. In coastal cities, public sewer systems
often serve only to route sewage to a coastal outfall, thus preventing anaerobic conditions
from developing.
Ethiopia. In Addis Ababa, sewerage is provided by a combination of septic
tanks and cesspools.
Kenya. Coffee- and sugar-processing industries produce the bulk of the
industrial wastewater in Kenya (UNEP, 1980). Other small Kenyan industries
that are expanding include the dairy-, fruit-, and vegetable-processing and meat
industries. Wastewater stabilization ponds appear to be the preferred method
of treating industrial wastewater.
Tanzania. In 1973, Dodoma, Tanzania, had no sewerage systems, and
residents were using septic system, soak-away pits, and pit latrines. At that
time, the city planned to build waste-stabilization ponds (UNEP, 1980). Dar es
Salaam, Tanzania, has at least one series of waste-stabilization ponds.
Page 10-6
-------
Vacuum trucks are used to extract and haul sewage from central areas of the
city to the ponds. The city also has sewage outfall off the Indian Ocean. Pit
latrines, pour-flash latrines, and open defecation are common means of human
waste disposal in small villages and rural areas throughout the country (Huff,
1992).
Algeria. The World Bank (1985a) reported that in 1985 about 70% of the urban
population of Algeria had access to public sewerage, but only 25% of the rural
population had sewerage. Wastewater from the cities was discharged,
untreated, directly into the Mediterranean or into rivers.
Senegal. Public sewerage is provided only in some of the larger cities, but
merely serves a small portion of the population. The existing sewerage
networks do not function properly, as the lines are often blocked with debris
and solid waste. The remainder of the urban population uses pit latrines and
septic tanks. (World Bank, 1985b). The World Bank estimates that less than
20% of the rural Senegalese population has any provision for excreta disposal.
Where provision exists, it is usually by pit latrine.
Based on this sparse information, it is assumed that 10% of all domestic wastewater
and 10% of all industrial wastewater produced in Africa is treated or disposed of in such a way
that anaerobic conditions result.
Asia and Oceania
Based on the following information, it is not possible to accurately estimate the volume
of domestic and industrial wastewater subjected to anaerobic decomposition in Asia. The
disposal systems mentioned - e.g., pit latrines, vault systems, canals, and even rivers -- can
become anaerobic throughout portions or all of their respective volumes, given favorable
moisture, pH, and temperature. Further, it is clear that wastewater stabilization lagoons are in
use. Due to lack of any definitive data, it could be assumed that the fraction of Asia's
domestic and industrial wastewater subject to anaerobic decomposition may be comparable to
the African value of 10%. However, the literature does suggest that agriculture and food
industries, both of which are producing high BOD wastewater, are more prevalent in Asia than
in Africa. The literature also suggests that sanitation through pit latrines and vault systems --
systems that are potentially anaerobic -- is a more common practice in Asia than in Africa.
Therefore, the 10% value may be too low, so a fraction of 15% is adopted to represent the
portion of Asia's wastewater (domestic, as well as industrial) that decomposes anaerobically.
This percentage is also applied to Oceania, as information for that region was unavailable.
General observations have been made for various regions and economic cooperatives
within Asia. Schiller and Droste (1982) reported that the vacuum truck and vault systems
have been used in East Asia over the past century for municipal wastewater. Wastewater
with high BOD from certain agricultural industries is treated in lagoons (Jalal and Aziz, 1986).
The volume of wastewater was not reported.
Middle East. Mullick (1987) indicates that wastewater stabilization lagoons are
a favored treatment method in much of the Middle East. The AI-Hassa Oasis,
Saudi Arabia uses facultative stabilization ponds for its wastewater treatment
(Abderrahman and Shahalam, 1991). These lagoons have a potential to
develop anaerobic conditions.
Page 10-7
-------
India. Only 3 percent of India's total population is served by wastewater
treatment facilities (UNEP, 1980). Some areas of the country are better
equipped than others. In Kerala, a state along the southwest coast, septic
tanks serve 40% of the urban population, and latrines serve 52% (World Bank,
1985c). Only 37% of the rural population of Kerala has pit latrines, and the
remaining people have no sanitary facilities. Most of the wastewater discharge
in India results from domestic sewage even in the larger, more industrialized
cities. In Bombay, only 13% of the water pollution comes from industry; in
Calcutta, 11%; and in New Delhi, 10%. It is reported that the BOD loading in
the industrial effluent is very high (UNEP, 1980).
Pakistan. The UNEP (1980) reports that only 10% of Pakistan's population has
sewerage service. Generally the sewage is not treated and is discharged into
rivers or onto land for cultivation. A few existing treatment facilities do not work
to capacity because of a lack of skilled operators and inadequate maintenance.
Industrial wastewater is also discharged without treatment.
Thailand. Some of Thailand's wastewater is treated in anaerobic lagoons.
Orlich (1990) estimates that wastewater-stabilization ponds in Thailand emit 0.5
Tg CH4/yr.
Turkey. The UNEP (1980) reports that few towns in Turkey have completed
sewerage systems and, in most cases, sewage is discharged into private septic
tanks or directly into rivers and lakes without any treatment. Since 1978, new
industries have been required to treat their wastewater before discharge, but
industries in operation before that date still discharge their waste without
treatment. Izmir, Turkey, reported plans to develop a wastewater lagoon
system using a three-stage system of anaerobic ponds followed by facultative
and maturation lagoons. The lagoon system is expected to cover 2,000
hectares and will likely be the largest lagoon system in the world. It will have
the capacity to treat 1,520 million I/day when operating at full capacity. It is
estimated that this single lagoon may emit over 12,000 metric tonnes of CH4
per year (McNitt and Tekeli, 1988).
China. The total discharge of industrial wastewater and urban domestic sewage
in China is approximately 30 billion metric tonnes/yr, of which 80% is industrial
wastewater. Only 15% of the total wastewater flow receives treatment (GEMS,
1989). However, China has a history of using anaerobic digesters since the
1920s. These were originally developed to digest domestic wastes to provide
biogas for household energy. China's earliest industrial application of anaerobic
digestion may date to 1967, when it was applied to treat distillery waste.
China's application of anaerobic digestion to industrial wastewater is still mostly
pilot-scale, but successful treatment has occurred with such varied waste
streams as alcohol distillates, acetone and butanol dregs, slaughterhouse
wastes, paper pulp black liquid and furfuraldehyde, tannery waste, and assorted
others (Li, 1986).
Taiwan. Excreta from nearly half of the population of Taipei is collected by
carts and placed in vaults and then is sold as fertilizer, discharged into the
rivers, or dumped into local canals and drains. The remaining population and
Page 10-8
-------
industry use septic tanks or discharge directly into rivers (Schiller and Droste,
1982).
• '- «j - -T -V
North Viet Nam. The double-vault compost toilet has gained widespread
acceptance in north Viet Nam (Schiller and Droste, 1982).
Latin America
In Latin America the discharge of domestic sewage and industrial effluent is a major
source of water pollution. Between these two sources of contamination, domestic sewage is
usually the more severe source, particularly in large urban areas (UN, 1990). The entire region
generally lacks wastewater treatment plants, and virtually all municipal sewage and industrial
effluent is discharged into the nearest water course without treatment. Even in major cities,
the wastewater flows are only partly controlled (directed) through interceptor sewers and
planned outfalls (UN, 1990).
According to two reports, 41-66% of the total population in Latin America has access
to these sewer systems (Bartone, 1990; and PAHO, 1990). Bartone (1990) indicates that 90%
of the sewered wastewater is discharged directly into receiving waters without any kind of
treatment. Based on Bartone (1990) and assuming that 66% of the total population has sewer
service, then 59% of the sewered population's wastewater is discharged without treatment,
and only 7% is retained at treatment facilities, which most probably consist of lagoon systems
(Bartone, 1990; UNEP, 1980; and Yanez, 1980). The UN (1990) estimates that^less than 2%
of urban sewerage flows receives treatment. Also, the wastewater treatment plants in Latin
America often do not function properly or do not function at all (UN, 1990; arid Bartone, 1990).
Bartone states that a World Bank survey of 42 mechanical treatment plants showed that 33 of
them were out of service at the time of the survey.
Industrial wastewater effluent in the region is dominated by the pulp and paper,
chemicals, petrochemicals and petroleum refining, metal production (particularly iron and steel
production and nonferrous metal refining), food-processing (especially sugar and coffee),
fish-meal, electricity generation, and textiles industries (UN, 1990). The share of Latin
America's activity in these industries (1982) was: 17.7% of petroleum refining, 14.7% of
chemicals production, 11.4% of the total number of beverage industries, 8.7% of food
manufacturing, 7.1% of the iron and steel industries, 6.3% of nonferrous basic industries, and
5.4% of paper products industries (UN, 1990).
It was assumed that approximately 10% of all domestic wastewater in Latin America
degrades anaerobically. This is to account for emissions from wastewater treatment plants
and communities with sewer services that use septic systems, pit toilets, and cesspools. The
same percentage was adopted for industrial wastewater.
North America, Europe, and Australia
Wastewater in developed countries is not expected to be a major source of CH4.
Wastewater in developed countries is typically treated aerobically, using aerated
impoundments. However, anaerobic techniques have steadily been gaining favor in
wastewater treatment for a variety of industrial processes (Torpy, 1988). Very little is known
about the volume of domestic and industrial liquid waste that is anaerobically decomposed in
Europe and North America. The treatment methods expected to have the greatest potential
for CH4 production are anaerobic and facultative stabilization lagoons, septic tanks, cesspools,
Page 10-9
-------
and anaerobic sludge digesters. Stabilization lagoons have long been used for wastewater
from agricultural and food industries in the United States and Canada (Environment Canada,
1979).
The Organization of Economic Cooperation and Development (OECD) (1991) reports
that approximately 60% of the population of OECD countries and 73% of the North American
population Is served by wastewater treatment plants. UNEP (1980) indicates that
approximately 25 liters of digester gas per capita per day are produced in
mechanical-biological wastewater treatment plants in Western Europe. The fraction chosen to
represent the volume of anaerobically degraded wastewater in North America, Europe, and
Australia is 15%; this may overestimate the amount of wastewater anaerobically digested
within municipal lagoons and septic systems, but it may approach or underestimate the total
volume of anaerobically digested wastewater when consideration is given to industrial
wastewater treatment.
United States. In the United States, individual septic systems are the most
common treatment and disposal method for an estimated 25% of the
population not served by an established sewage authority (Hick, 1992).
Depending on temperature, retention time, and system configuration, these
septic systems could emit CH4to the atmosphere, either through cracks and
leaks in the tanks or through vent pipes in the systems, assuming that CH4 is
not completely oxidized in the surrounding soil.
An estimated 5,500 municipal waste stabilization lagoons in the United States
treat 5.2 x 106 m3/day of wastewater from 8% of the population served by
municipal wastewater treatment systems (OMPC, 1987). Mullick (1987) reports
that in Texas, alone, 470 wastewater treatment plants use one or more lagoons
to treat wastewater.
Anaerobic digestion is widely used in the United States in municipal wastewater
treatment plants to stabilize sludge. An estimated 65.5 x 10 m3/day of
wastewater flows through treatment plants using anaerobic digesters to treat
sludge (OMPC, 1992). Table 10-2 summarizes data for the United States.
Canada. Approximately 1,000 sewerage lagoons represent about half of the
existing wastewater treatment systems. Municipal wastewater lagoons are
quite prevalent in Canada, providing up to 50% of domestic wastewater
treatment (Townshend and Knoll, 1987). Anaerobic processes to treat industrial
wastewater are becoming increasingly popular.
Former West Germany. UNEP (1980) reports that West Germany uses more
than a thousand lagoons for wastewater treatment, and aerated lagoons are
becoming increasingly popular.
Australia. This country was among the first to advocate the advantages of
pond systems designed specifically to operate as anaerobic units. Municipal
and industrial wastewater have been treated in anaerobic lagoons in Australia
since the 1950s.
Page 10-10
-------
TABLE 10-2
SELECTED WASTEWATER TREATMENT METHODS IN THE UNITED STATES3
Wastewater Treatment Methods
Municipal Wastewater Lagoonsc
Septic Systems
Anaerobic Sludge Digestion
Processes
Flow
(1 x 10s m3/day)b
5.2
15.0
65.5d
Population Served
(%)
8
25
Unknown
a. See text for sources of information.
b. Estimated actual flow is given for septic systems and anaerobic sludge
digestion processes; estimated design flow is given for lagoons.
c. Includes lagoons receiving domestic and domestic/industrial
wastewater, with no pretreatment.
d. Actual flow represents the total flow associated with the treatment plant
at which the anaerobic digestion unit is located.
10.4 RESULTS
Approximately 10-15% of the world's domestic wastewater will undergo anaerobic
digestion, and BOD loading data for various geographical regions lie between 0.037 and 0.05
kg/capita/day. Applying these values, combined with country populations, to Equation 10.1
results in a CH4 emission estimate for global domestic wastewater of approximately 2 Tg/yr.
Country-specific CH4 emission estimates for domestic wastewater are given in Table 10-3.
Since country-specific industrial wastewater volumes are not known at this time, the
estimated worldwide volumes for specific industries were used. Information from Carmichael
and Strzepek (1987) was used to estimate the 1990 global industrial wastewater outflows for
specific industries. In their table, Carmichael and Strzepek presented a global estimate of
wastewater outflow in the year 2000. Table 10-4 presents the industrial wastewater outflows,
BODs, and CH4 emission estimates for both developing and developed countries. Based on
an assumed fraction of industrial wastewater that will undergo anaerobic digestion of 10% and
15% and the BOD loadings from Table 10-1, the resulting estimate of global CH4 emissions
from industrial wastewater ranges from 26 to 40 Tg/yr.
These numbers are comparable to Orlich's (1990) estimate. On the basis of the Thai
data, Orlich approximated that global CH4 emissions from wastewater lagoons would be
roughly 20-25 Tg/yr. Orlich's estimate was limited to lagoons, whereas the EPA/AEERL
estimates also include untreated wastewater.
10.5 TRENDS
In developing countries where sufficient land is available, lagoon systems using a
series of anaerobic, facultative, and aerobic treatment ponds will continue to be recommended
for wastewater treatment (Bartone, 1990; and UNEP, 1980). They are relatively inexpensive
Page 10-11
-------
TABLE 10-3
ESTIMATES OF GLOBAL AND COUNTRY-SPECIFIC METHANE
EMISSIONS FROM DOMESTIC WASTEWATER, 1990 (Tg/yr)
Country
Africa3
Egypt
Kenya
Morocco
Nigeria
South Africa
Sudan
Tanzania
Uganda
Other Africa
Total Africa
Asia"
China
North Korea
Viet Nam
Other Asia
Total Asia
South and Central America3
Argentina
Brazil
Colombia
Mexico
Venezuela
Other
Total South and Central America
North America
Canada
United States
Other North America
Total North America
Europe
France
Former German Democratic Republic
Italy
United Kingdom
Total Europe
Former Soviet Union
Oceania
Australia
Total Oceania
World Total
Emissions
0.02
0.01
0.01
0.04
0.01
0.01
0.01
0.01
0.09
0.21
0.54
0.01
0.03
0.92
1.50
0.01
0.05
0.05
0.03
0.01
0.02
0.13
0.02
0.15
0.04
0.21
0.01
0.01
0.01
0.01
0.04
0.18
0.01
0.07
2.30
a. 10% of BOD is assumed to be anaerobically degraded.
b. 15% of BODS is assumed to be anaerobically degraded.
Page 10-12
-------
TABLE 10-4
ESTIMATE OF GLOBAL METHANE EMISSIONS FROM INDUSTRIAL WASTEWATER
1990 (Tg/yr)
Developed Countries
Industry
Iron & Steela
Nonferrous Metals3
Fertilizer3
Food & Beverages
Pulp & Paper
Petroleum Refining
Textile
Rubber3
Miscellaneous
Total
Industrial
Wastewater"
Outflow
(mil m3/yr)
. 168,000
26,600
14,300
7,700
33,300
54,700
34,600
6,800
9,500
355,500
Percent of
BOD5
Anaerobically
Degraded
BOD5°
(kg/I)
0.001
0.001
0.001
0.035
Q.QQ4
0.004
0.001
0.001
0.002
10%
15%
Emissions
4
1
0
6
3
5
1
0
0
20
6
1
1
9
4
7
1
0
1
30
Developing Countries
Industrial
Wastewater
Outflow
(mil m3/yr)
56,000
8,850
4,500
2,600
11,100
18,200
11,450
2,300
3,200
118,200
Percent of
BOD5
Anaerobically
Degraded
BOD5°
(kg/I)
0.001
0.001
0.001
0.035
0.004
0.004
0.001
0.001
0.002
10%
15%
Emissions
1
0
0
2
1
2
0
0
0
6
2
0
0
3
2
2
1
0
0
10
Industrial
Wastewater
Outflow
(mil m3/yr)
224,300
35,000
19,000
10,300
44,400
73,200
46,100
9,100
12,700
474,100
Worldwide
Percent of
BOD5
Anaerobically
Degraded
BOD5C
(kg/I)
0.001
0.001
0.001
0.035
0.004
0.004
0.001
0.001
0.002
10%
15%
Emissions >
5
1
0
8
4
6
1
0
1
26
7
1
1
12
6
10
2
0
1
40
a. This group was identified in Carmichael and Strzepek (1987).
b. Carmichael and Strzepek (1987).
c. Sources of BOD values are given in Table 10-1.
-------
to build and operate and provide excellent effluent quality when properly operated. Should the
number of wastewater lagoons increase, the potential for future CH4 emissions from
wastewater treatment will also increase, assuming emissions are uncontrolled.
As mentioned before, state-of-the-art treatment of wastewater typically occurs
aerobically using aerated impoundments. Also, in other steps of the process, digester tanks
are often used, and the collected gas is either flared or used, therefore reducing CH4
emissions. Many "developed" countries practice limited wastewater treatment. Only in
Northern Europe and North America are wastewater disposal and treatment highly regulated
and, consequently, relatively advanced. As countries in other regions are required to develop
more stringent regulations and implement advanced wastewater treatment technologies (for
instance under European Community regulations), CH4 emissions per capita may decrease.
This reduction would not necessarily be nullified by economic growth or population increases.
With increasing population and industrial activity, water could become more scarce, thus
creating an economic incentive to reduce its usage. Also, improvements in production
techniques are nowadays aimed at waste minimization, which may lead to reduction of BOD
loadings in waste streams and, consequently, CH4 emissions.
10.6 MEASUREMENT, UNCERTAINTY, AND VERIFICATION ISSUES
The estimates developed using Equations 10.1 and 10.2 are considered initial
estimates to help identify the major sources of uncertainty and to target potentially cost-
effective control measures (Thorneloe, 1992). There are major uncertainties with these
estimates which are identified in this section. Research is underway to reduce these
uncertainties so that more reliable estimates can be made.
10.6.1 Activity Data: Quantities of Wastewater Flows
When using industrial wastewater data, researchers should give special care to the
definition of "wastewater," especially when data are pulled from different references. Sources
may or may not include cooling water and storm water when representing quantitative
wastewater data, whereas BOD or other qualitative data would typically not represent these
types of wastewater. For example, in 1980, about half of all wastewater generated by U.S.
manufacturing plants resulted from cooling operations (U.S. EPA, 1980). If the wastewater
flows from Table 10-4 include cooling water or other nonpolluted water, CH4 emission
estimates in this chapter will be overstated.
Specific volumes of domestic and industrial wastewater outflow for each country are
not available, as are the volumes that are biologically treated under anaerobic conditions.
Major data limitations exist for quantifying the fraction of wastewater subject to anaerobic
decomposition. Data are needed for specific volumes of wastewater undergoing anaerobic
decomposition. Data regarding the effect of recycling industrial wastewater outflow are also
limited.
A large volume of the world's wastewater is emitted to lakes, rivers, and oceans
without any kind of treatment. In many regions of the world, sewage and possibly also
industrial wastewater may be discharged into smaller lakes or into dry riverbeds (UNEP, 1980;
and WHO, 1987). No data are available regarding the extent to which anaerobic conditions
could develop under these circumstances (UNEP, 1980 and 1988).
Page 10-14
-------
Emissions from unsewered wastewater treated in septic systems, pit toilets, or
cesspools might be considerable. No data are available on CH4 emissions from these
systems. Depending on temperature, retention time, and system configuration, these septic
systems could emit CH4 to the atmosphere, either through vent pipes in the systems or
through cracks and leaks in the tanks, assuming that CH4 is not completely oxidized in
surrounding soil.
10.6.2 Methane Yield per Unit BOD Loading
The value of 0.22 kg CH4 per kg BOD5 used in section 10.3 is one of two estimates
that were found. Orlich (1990) used data from an inventory of lagoons in Thailand, which
presumably contained wastewater from agricultural industries (food and beverage industries).
Orlich's assumption is supported by an estimated yield of 0.25 kg CH4/kg BODy
(corresponding to 0.18 kg CH4/kg BOD5) for the anaerobic decomposition of glucose (Metcalf
and Eddy, Inc., 1979). As no other data exist, Orlich's estimate was adopted for all
wastewater. Differences in the physical and chemical conditions in lagoons have not been
accounted for. For example, this factor may be lower in moderate or cold climates. To
reduce the level of uncertainty in CH4 yields and other parameters, EPA/AEERL is conducting
a field testing program at lagoons, other wastewater treatment facilities, and septic sewerage
systems (Thorneloe, 1992). This will result in improved emission factors that will be used to
develop more reliable emission estimates.
10.6.3 Efficiency and Output of Wastewater Treatment Facilities
Wastewater that is accepted by treatment plants and supposedly treated aerobically
may still be subject to anaerobic conditions, due to poor functioning of the facilities.
Adjustment has not been made for (1) the amount of CH4 that is controlled through use or
flaring and (2) the amount of CH4 that is oxidized prior to atmospheric release. Both fractions
are potentially significant, particularly in such countries as the United States, where control is
required for reducing hydrogen sulfide (H2S) emissions. Data are being collected to account
for this in future estimates for this emission source.
10.6.4 BOD Values and Other Physical and Chemical Data
Data on physicochemical wastewater characteristics are limited, especially for country-
specific wastewater volumes. For industrial wastewater, BOD values reported for source
categories are averages of BOD values given for several process wastewater streams. For
iron & steel, fertilizer, and nonferrous metals, no BOD values were assumed because
published values were not available.
The estimate, presented in this chapter could be improved if data were obtained on the
chemical characteristics of process wastewater streams. For instance, a distinction can be
made between carbonaceous and nitrogenous oxygen demand. Wastewater with a high
nitrogenous oxygen demand or other noncarbon reductive ions is not a potential source of
CH4. The emission methodology does not account for such factors as temperature, pH, and
retention time, all of which influence the rate and extent of anaerobic decomposition and,
consequently, the CH4 emissions.
Page 10-15
-------
10.7 CONCLUSIONS
Global and country-specific estimates were developed for CH4 from domestic
wastewater. Domestic wastewater is estimated to emit approximately 2 Tg/yr, with Asia
accounting for 65%. For industrial wastewater only global estimates are presented, as
adequate country-specific data are not available. Industrial wastewater is estimated to be the
major contributor to CH4 emissions from wastewater, accounting for 26-40 Tg/yr. These
numbers are comparable to an estimate by Orlich (1990). On the basis of the Thai data,
Orlich reported that global CH4 emissions from wastewater lagoons would be roughly 20-25
Tg/yr. Orlich's estimate was limited to lagoons, whereas the EPA estimates attempt to include
untreated wastewater and septic sewage systems.
Substantial uncertainty in these estimates results from a lack of data characterizing
wastewater management practices, the quantities of wastewater that are subject to anaerobic
conditions, the extent to which CH4 is emitted, and flaring or utilization practices. EPA is
conducting a field-testing program to develop more reliable emission factors that will result in
reducing the uncertainty in these estimates. Country-specific activity data are also being
collected, along with information on current trends. This information will be used to revise
these initial estimates and to project future emissions.
It is expected that CH4 emissions from wastewater treatment in developed countries
may decrease because aerobic techniques, as well as CH4 utilization, are becoming
increasingly popular. Also, some countries require the control of H2S emissions, which results
in controlling CH4 emissions. The trend in developing countries is toward using facultative
lagoons for treatment. This may lead to an increase in CH4 emissions in developing
countries.
10.8 REFERENCES
Abderrahman, W.A., and A.B.M. Shahalam. 1991. Reuse of wastewater effluent for irrigation
in severely arid regions. Water Resources Development 7(4):235-246.
Barnes, D., C.F. Forster, and S.E. Hrudey, eds. 1984. Surveys in Industrial Wastewater
Treatment Vol. 10. Food and Allied Industries. Pitman Publishing, Inc., Marshfield,
Massachusetts.
Bartone, C.R. 1990. Water quality and urbanization in Latin America. Water International
15:3-14.
Carmichael, J.B., and K.M. Strzepek. 1987. Industrial Water Use and Treatment Practices.
Published for the United Nations Industrial Development Organization. Cassell Tycooly,
Philadelphia, Pennsylvania.
Environment Canada. 1979. Biological Treatment of Food Processing Wastewater Design
and Operation Manual. Economic and Technical Review. Report EPA, 3-WP-79-7. Water
Pollution Control Directorate.
Page 10-16
-------
GEMS (Global Environment Monitoring System). 1989. Global Freshwater Quality: A First
Assessment. Meybeck, M., D.V. Chapman, and R. Helmer, eds. Prepared for the World
Health Organization and the United Nations Environment Programme. Basil Blackwell, Inc.,
Cambridge, Massachusetts.
Gloyna, E.F. 1971. Waste Stabilization Ponds. World Health Organization, Geneva,
Switzerland.
Gotass, J.B. 1956. Compositing Sanitary Disposal and Reclamation of Organic Wastes.
World Health Organization (WHO) Monographs Series No. 31. WHO, Geneva, Switzerland.
Hick, T. 1992. Personal communication. National Small Flows Clearinghouse, West Virginia
University.
Huff, L.C. 1992. Personal observations. December 1991 - January 1992 while in Dar es
Salaam, Tanzania, and regions north.
IPCC (Intergovernmental Panel on Climate Change). 1992. Climate Change 1992: The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press,
Cambridge, United Kingdom.
Jalal, K.F., and M.A. Aziz. 1986. Pollution control from agro-based and small-scale industries
of the Economic and Social Commission for Asia and the Pacific (ESCAP) region. In Tay,
J.H., and S.L Ong, eds., Proceedings of the International Conference on Water and
Wastewater Management in Asia. February 1986. Singapore. 55-81.
Laak, R. 1986. Wastewater Engineering Design for Unsewered Areas. Ann Arbor Science
Publishing, Inc., Ann Arbor, Michigan.
Li, D.N. 1986. Status quo of anaerobic digestion on industrial wastewater in China. In Tay,
J.H., and S.L Ongi, eds., Proceedings of the International Conference on Water and
Wastewater Management in Asia. February 1986. Singapore. 65-81.
Mara, D. 1976. Sewage Treatment in Hot Climates. John Wiley & Sons, New York, New
York., 4-6, and 77.
McNitt, J.R., and A. Tekeli. 1988. Cleaning Up Izmir Bay. National Development Asia.
November/December 1988. 20-22.
Metcalf & Eddy, Inc. 1979. Wastewater Engineering: Treatment Disposal Reuse. 2nd
Edition. McGraw-Hill Book Company, New York, New York. 2, 3, 20, 64, and 621.
Mullick, M.A. 1987. Wastewater Treatment Processes in the Middle East. The Book Guild,
Sussex, United Kingdom.
OECD (Organization of Economic Cooperation and Development). 1991. OECD
Environmental Data Compendium 1991. Paris, France
OMPC (Office of Municipal Pollution Control). 1987. Report to Congress: Municipal
Wastewater Lagoon Study. PB88-154315. Washington, DC. 3-3,3-7,and 3-8.
Page 10-17
-------
OMPC (Office of Municipal Pollution Control). 1992. Conveyance, Treatment, and Control of
Municipal Wastewater, Combined Sewer Overflows, and Stormwater Runoff. Draft. U.S.
Environmental Protection Agency, Washington, D.C. 18 and 81.
Orlich, J. 1990. Methane emissions from landfill sites and wastewater lagoons. Presented at
the International Workshop on Methane Emissions from Natural Gas Systems, Coal Mining
and Waste Management Systems. Japan Environment Agency and the U.S. Environmental
Protection Agency. April 9-13, 1990, Washington, DC. 465-471.
PAHO (Pan American Health Organization). 1990. Health Conditions in the Americas.
Scientific Publication No. 524. Vol. I.
Population Reference Bureau, Inc. 1991. Demographic data and estimates for the countries
and regions of the world. World Population Data Sheet of the Population Reference Bureau,
Inc., Washington, D.C.
Schiller, E.J., and R.L. Droste. 1982. Water Supply and Sanitation in Developing Countries.
Ann Arbor Science Publishers, Ann Arbor, Michigan.
Thomeloe, S.A. 1992. Emissions and mitigation at landfills and other waste management
facilities. Published in Proceedings from EPA Symposium on Greenhouse Gas Emissions and
Mitigation Research. In press.
Torpy, M.F. 1988. Anaerobic Treatment of Industrial Wastewaters. Noyes Data Corporation,
Park Ridge, New Jersey.
Townshend, A.R., and H. Knoll. 1987. Cold Climate Sewage Lagoons. Report EPS 3/NR/1.
Proceedings of the 1985 Workshop, April 1987, Winnipeg, Manitoba, Canada.
UN (United Nations). 1990. The Water Resources of Latin America and the Caribbean -
Planning, Hazards and Pollution. United Nations Economic Commission for Latin America and
the Caribbean, Santiago, Chile.
UNEP (United Nations Environment Programme). 1980. Report of the International
Symposium on Wastewater Technology for Developing Countries. UNEP Report No. 4.
UNEP (United Nations Environment Programme). 1988. Environmental Guidelines for
Domestic Wastewater Management. UNEP Environmental Management Guidelines, No. 14.
U.S. EPA (Environmental Protection Agency). 1980. Research Summary -- Industrial
Wastewater. EPA-600/8-80-026. June 1980. Washington, D.C.
Viessman, Jr., and M.J. Hammer. 1985. Water Supply and Pollution Control. Harper & Row
Publishers, New York, New York. 612.
World Bank. 1985a. Staff appraisal report. Eleven Centers Water Supply Project. Senegal.
5 and 6. February 1, 1985.
World Bank. 1985b. Staff appraisal report. National Water Supply and Sewerage Project.
Algeria. 4 and 5. May 10, 1985..
Page 10-18
-------
World Bank, 1985c. Staff appraisal report. Kerala Water Supply and Sanitation Project.
India. 4 and 5. June 10, 1985.
WHO (World Health Organization). 1987. Wastewater Stabilization Ponds: Principle of
Planning & Practice. WHO EMRO Technical Publication No. 10, Alexandria, Egypt.
Yanez, F. 1980. Evaluation of the San Juan Stabilization Ponds Final Research Report of
the First Phase. Pan American Center for Sanitary Engineering and Environmental Sciences.
Page 10-19
-------
-------
CHAPTER 11
VERIFICATION OF METHANE EMISSIONS
11.1 INTRODUCTION
At the United Nations Conference on Environment and Development, held in Rio de
Janeiro in June 1992, 154 heads of state signed an international agreement to promote and
cooperate in activities that would mitigate future climate change and its potential impacts. The
ultimate objective of this agreement, The United Nations Framework Convention on Climate
Change (FCCC), is "stabilization of greenhouse gas concentrations in the atmosphere at a
level that would prevent dangerous anthropogenic interference with the climate system" (UN,
1992). The FCCC, however, does not define numerically the level at.which greenhouse gas
concentrations should be stabilized or the time period over which this stabilization should
occur. Except for a general goal of reducing emissions to 1990 levels by Annex I countries1,
no commitments for specific emission reductions by a certain date, either nationally or
globally, are contained in the FCCC.
Although specific emission reduction timetables are not contained within the FCCC, all
Parties to the Convention are required to develop and submit national inventories of annual
greenhouse gas emissions and sinks. These inventories are to be based on comparable
methodologies agreed upon by the Conference of the Parties (COP), and are to be updated
periodically. In fact, Parties to the Convention have begun to compile emission inventories in
order to comply with commitments and reporting requirements under Articles 4.1 (a) and
12.1 (a), and the process of developing an internationally agreed-upon emission inventory
methodology is underway through the Intergovernmental Panel on Climate Change (IPCC),
with technical assistance from the Organization for Economic Cooperation and Development
(IPCC, 1991 and 1993; OECD, 1991; and van Amstel, 1993).
The FCCC currently contains no measures for verification of national inventories
submitted to the COP, but an international system to review and verify national inventories of
methane (CH4) and other greenhouse gas emissions and sinks may prove useful for several
reasons. As discussed in previous chapters of this report, CH4 emission estimates for
individual sources are poorly constrained since they are based on flux measurements that are
either scarce or vary by up to several orders of magnitude. Moreover, CH4 emissions are
controlled by a wide variety of biogeochemical and socioeconomic factors that, in many cases,
are neither well understood nor well quantified. In addition, the activity data that are used to
extrapolate from individual flux measurements to national emission totals are often uncertain.
A verification system, through systematic exchange, review, and comparison of data, most
likely would improve scientific understanding of greenhouse gas sources and sinks. As a
result of the verification process, flux measurements, activity data, and relevant
biogeochemical and socioeconomic information, that perhaps had not been widely
disseminated previously, would become available to both the scientific and the policy
communities. Therefore, the verification process would promote greater sharing and
exchange of data, as well as increased dialogue among scientists and policymakers.
1 Annex I countries are the developed countries, including countries with economies in transition.
Page 11-1
-------
An international verification system also would contribute toward the ultimate objective
of the FCCC. By improving the quality of information about greenhouse gas sources and
sinks, a verification system would help policymakers set appropriate priorities for emission
reduction policies among different gases and among different countries. A verification system
would also be useful for monitoring progress toward the goal of reducing emissions to 1990
levels, at both the global and the national levels.
From a compliance standpoint, a verification system would be useful for three reasons
(e.g., Lewis, 1992): (1) to deter incorrect reporting, (2) to detect cases of incorrect reporting
and provide mechanisms for revising inventories, and (3) to increase confidence in the FCCC
and foster greater trust and collaboration among parties, thereby encouraging more nations to
sign and ratify the Convention. A system of verification would allow the Parties to address
potential problems promptly and in a coordinated and collaborative fashion. Also, such a
system would help determine how technical and financial assistance can be used to support
countries in improving their inventories.
An emission inventory verification system would be most effective if implemented
through a two-part structure (e.g., Fischer, 1992; and di Primio, 1992): (1) an internal,
national monitoring and assessment process and (2) an international verification process.
Through the national monitoring and assessment process, a country would compile, review,
and process information on greenhouse gas emissions and sinks with the ultimate purpose of
producing their implementation report -- i.e., their greenhouse gas inventory. The national
process would increase national capability for producing inventories, as well as help nations
determine what additional data should be acquired to improve national emission and sink
estimates.
The international component of a verification system would examine national reports
for reliability and completeness. Auxiliary data, not provided in the national implementation
reports, could also be employed by the international system in order to verify the reported
inventories. Substantial learning about greenhouse gas emissions and sinks would be
acquired through the review and exchange of national data and inventory estimation methods.
This process would most likely further the development of more accurate inventories as well
as improved inventory methods.
If the Parties to the Convention agree that an inventory verification system is useful,
they must decide what should be reviewed, how it will be verified, and the degree (or
intrusiveness) of review and verification required. These decisions will involve financial,
political, scientific, and technical considerations. For example, verification methods must be
robust in order to ensure confidence in the system, but costly data collection that does not
contribute significantly to the knowledge base must be avoided. The intrusiveness of the
verification system must not be so great as to be interpreted as encroachment on proprietary
political or commercial information and thereby create mistrust among the Parties to the
Convention. Although systematic review and monitoring of emission inventories are likely to
increase confidence in the FCCC, these activities may also undermine the process if Parties
to the Convention perceive them as too intrusive. Also, the ability to assess and verify
emissions from numerous sources is constrained by technological and scientific limitations.
For example, limited technological capabilities for measuring the area of savannas burned
annually preclude accurate estimations of trace gas emissions from biomass burning, while
limited scientific understanding of the factors that control CH4 production during anaerobic
decomposition precludes accurate assessments of CH4 emissions from waste management
practices.
Page 11-2
-------
Despite these difficulties, it is worthwhile at this point to discuss the scientific and
technological mechanisms available for verification of emission and sink estimates, to assist
nations in the interim as they develop their inventories as required by the FCCC. If countries
understand the methods for verifying their inventories, they will be better able to monitor and
verify their own calculations, as well as develop more accurate and complete inventories. And
if effective monitoring systems are in place nationally that include mechanisms for
independent verification, compliance with eventual emission reduction obligations is more
likely. Moreover, the establishment of accurate national verification systems will minimize the
need for an extensive and highly intrusive international monitoring and verification system.
The remainder of this chapter discusses scientific and technological mechanisms
available for verification of CH4 emission estimates. Since the many financial, political, and
institutional issues associated with emission verification could not be covered in this report,
the reader is referred to the small but growing body of literature on this subject for additional
information (e.g., Fischer et. al., 1990; Feldrnan, 1992; Ausubel and Victor, 1992; and U.S.
GAO, 1992aand 1992b).
11.2 VERIFICATION METHODOLOGIES
As documented in this report, CH4 is released as a by-product of a wide variety of
activities in the agricultural, energy, waste management, and industrial sectors. Because of
the great number and variety of emission sources, absolute verification of CH4 emission
estimates through comprehensive, direct measurement is essentially impossible. Some
sources are stationary point sources, and emissions from them can be measured fairly
accurately. Their emission rates, however, may vary significantly from region to region, so a
complete assessment would require an extremely detailed network of measurements (e.g.,
emissions associated with venting and leaking from oil and natural gas systems). Other
sources, either stationary or mobile, are too numerous and intermittent for continuous
measurement to be feasible (e.g., emissions from wood stoves, automobiles, and livestock).
Many CH4 sources, such as rice cultivation and savanna burning, are areal rather than point
sources, making comprehensive emission measurement physically and logistically impossible.
Mass-balance equations have been used successfully as a verification technique in
nuclear arms control agreements. This procedure balances an initial inventory against a final
inventory by accounting for all mass flows into and out of the system, and for all
transformations in the system within a given period. Emissions of greenhouse gases,
however, occur in an open system. Emissions from each source are released to the
atmosphere where they become mixed with those from other sources. It is not possible to
verify precisely the mass flows out of an individual greenhouse gas emission "system" after
the emissions have been released. As discussed in section 11.2.2, various atmospheric
measurements can be used to constrain estimates of emissions from various sources or
groups of sources (e.g., fossil sources), but at present, these measurements have limited
usefulness in determining either individual source strengths (e.g., CH4 from landfills) or total
emissions from small countries or regions. Because of these difficulties associated with
verification of emission inventories, due to both the varied nature and openness of emission
sources, greenhouse gas emission inventories have been described by Fischer et al. (1990)
as having "poor verification suitability."
Due to the technical, logistic, and financial difficulties of measuring CH4 emissions from
every source in a nation or region, national and regional emission estimates for each source
Page 11-3
-------
category are calculated by multiplying (1) a representative emission factor, which is derived
from theory, experiment, and/or expert judgment, by (2) an activity level, which is based on
national or regional statistics and expert judgment. The emission factors quantify the amount
of CH4 released per unit of activity — e.g., per animal, per tonne of municipal waste landfilled,
per kilogram of fuelwood burned. The activity levels are used to extrapolate from unit
emission factors to national or regional emission estimates for each source. For example, the
average annual CH4 emissions per head of a particular type of livestock (the "emission factor")
would be multiplied by the national population of that type of livestock (the "activity level") to
obtain the annual, national emissions of CH4 from that type of livestock.
Although it is impossible to verify a national emission inventory through direct
measurement or mass-balance calculations, an inventory can be verified through an
assessment of the plausibility, accuracy, and completeness of the calculations used to
estimate emissions. This direct, or "bottom-up," method of verification would include an
evaluation of the plausibility and completeness of the inventory results, the accuracy and
appropriateness of the methodologies used, the accuracy of the calculations, and the
accuracy and appropriateness of the numbers used in the calculations, especially the
emission factors and activity levels.
Additional information with which to verify national CH4 emission estimates can be
obtained through monitoring the atmospheric signatures of the emissions themselves. This
"top-down" monitoring approach works because the atmosphere does not mix trace gas
concentrations uniformly. Regions of strong emissions are clearly revealed by elevated
concentrations in the lower atmosphere. Also, the isotopic contents of atmospheric CH4 can
place bounds on individual source strengths, since isotopic contents vary among different
sources. Because winds spread the concentrations some distance from the source(s),
however, this approach is limited. National emission estimates for individual sources cannot
be verified accurately with this technique, especially if the country is small. This technique
can, however, provide cross-checks on total CH4 emissions from a broad region or a large
country located upwind from the measurement site. This is particularly true of regions with
relatively large CH4 emissions, such as Southeast Asia. Also, the use of isotope
measurements in this technique can provide constraints on estimates of national and regional
CH4 emissions from certain individual source categories (e.g., biomass burning, coal mining,
oil and gas systems). This technique can also be used to put bounds, albeit wide, on global
estimates of individual source strengths based on consistency with the atmospheric variations
(e.g., Fung et al., 1991). This top-down monitoring approach, therefore, is most useful as a
cross-checking device in verification of national CH4 emission inventories.
11.2.1 Direct Verification
"Bottom-up" verification of national inventories of CH4 emissions would entail a series
of steps through which the plausibility, accuracy, and completeness of the inventories would
be evaluated. These steps are as follows:
Evaluate the plausibility of the results.
Evaluate the inventory methods used.
Evaluate the accuracy of the calculations.
Evaluate the accuracy and appropriateness of the emission factors used.
Page 11-4
-------
Evaluate the accuracy and appropriateness of the activity levels used.
For this type of verification system to be successful, the documentation of the emission
estimates must be complete and transparent. Not only must emission estimates by source
category be provided, but also the methodologies used to estimate emissions must be clearly
defined. The data provided must be detailed enough so that the emission estimates can be
reconstructed. Scientific and technical references used to determine emission factors, as well
as data sources for activity data, must be cited and must be available to other scientists. If
those references are not published and internationally available, the original data that were
used in the calculations would need to be provided as well.
Evaluate the Plausibility of the Results
The plausibility, or reasonableness, of the emission inventory should be evaluated first.
All or some subset of the anthropogenic CH4 source categories will be included in the
inventory, depending upon the types of agricultural, energy, waste management, and industrial
activities that occur in the nation. Evaluation of whether all of the source categories that
should be included have been included will require some knowledge of national practices in
each of the relevant sectors. For example, a high emission estimate for domestic livestock
does not imply that the emission estimate for animal wastes necessarily should be high too.
Even if a nation has a large livestock population, if all of the animal wastes are managed
aerobically (e.g., spread on fields) then emissions from this source would be negligible.
Presumably, once methodologies are agreed upon internationally, there will be some minimum
cut-off for individual source estimates, so that if a country's emission estimate for a particular
source is less than, for example, 0.01 teragrams (Tg)2 CH4/yr, that country would not be
required to estimate emissions from that source.
Comparison of the emission inventory with those compiled by other groups provides an
independent, but somewhat risky, method of verification. Care must be taken to ensure that
the emission estimates for the same source and region are compared, or the comparison will
be meaningless. The year of inventory should also be tine same, but this variable is not as
important if the temporal variation in the activity level is not likely to be significant. Some
sources (e.g., biomass burning and rice cultivation) may, however, vary significantly from year
to year due to climatic, political, or social factors. Comparison of emission estimates from
different years for such sources could result in misleading results.
Significant differences between the subject inventory and other inventories for the
same source, region, and year could be the result of one or more of the following: (1) a
different methodology, (2) different emission factor(s), and/or (3) different activity level(s).
Determining which factor(s) is(are) responsible could indicate that more or less detailed
investigation is needed in the next steps in the verification process. For example, estimates of
CH4 emissions from cattle may differ between two countries in which livestock management
practices and production levels are similar because the population sizes (i.e., activity levels)
differ. Evaluation of this is simple. If both activity levels and emission factors vary, the
reasons for differing emission factors, which could be complex, will need to be reviewed and
evaluated as well.
: Teragram = 106 metric tonnes = 1012 grams.
Page 11-5
-------
If inventories for previous years are available, the inventory under examination can be
compared to the previous submissions. Any significant decline or increase in emissions
should be explained by the supporting information provided in the national report. Possible
reasons for a significant change in emissions include a change in the accepted emission
methodologies, new scientific research that results in altered emission factors, demographic or
economic changes that alter activity levels, and new or strengthened controls on emissions.
Last, national emission inventories for each individual source category should be
summed to obtain global emission estimates for each source. The global estimates should
not lie significantly outside the accepted range for each source.3 This could result, however,
from many countries submitting small, but consistent, over- or underestimates. Determining
which countries are responsible for the discrepancy would be difficult, unless the differences
are due to obvious errors or differences in assumptions. Also, the accepted ranges for most
Individual sources are quite broad, so this step will have limited usefulness. And if some
countries overestimate their emissions, and some underestimate their emissions, the
differences may cancel out. Atmospheric measurements of the concentration and isotopic
composition of CH4 provide information about the geographic distribution of emissions, and
may help resolve this issue (see section 11.2.2, entitled "Indirect Verification").
Evaluate the Inventory Methods Used
The methodology used should be compared with what has been agreed upon by the
COP. If it differs, an explanation of the reasons for the differences can justify a more
appropriate method and perhaps subsequently be used to improve the internationally
accepted methods. Substantial improvements to the methods should be expected as scientific
understanding of the biogeochemical and socioeconomic factors that control CH4 emissions
increases. Evaluation of inventory methods through a formal verification process would,
therefore, promote development of improved methods.
Although Parties to the Convention are required to use inventory methodologies that
are agreed upon by the COP, there may be more than one methodology to choose from for
an individual source category. For example, in chapter 6 of this report, a tiered approach to
estimating CH4 emissions from coal mining is recommended. A country would select a
particular tier, or methodology, depending upon the quantity and quality of available data with
which to implement the method.
Countries may also choose to revise agreed-upon methodologies, because of
particular socioeconomic or climatic conditions, or because new research results indicate that
such a change is necessary. For example, if an inventory methodology is based purely on
experiments in temperate countries, a tropical country may choose to alter the methodology
based on expert judgment.
For these reasons, the methodologies employed should be reviewed and evaluated for
their appropriateness, given the context of a particular country whose inventory is under
review.
3 The results of this step will not be meaningful unless all countries ratify the FCCC and submit an inventory, or if,
as part of the verification process, inventories are derived for those countries that do not submit them.
Page 11-6
-------
Evaluate the Accuracy of the Calculations
Evaluating the accuracy of the calculations is quite straightforward, assuming that the
implementation report is complete and sufficiently detailed. All the calculations must be
checked to ensure that the math is correct.
Evaluate the Accuracy and Appropriateness of the Emission Factors Used
Relatively few measurements have been made of CH4 emissions. Moreover, since
variables affecting emissions are numerous, varied, and sometimes poorly understood,
emission factors are highly uncertain. Development of emission factors is an ongoing
"science," and factors are revised and updated constantly in the scientific literature as new
information is published. As a first step, the emission factors used in the national inventories
should be evaluated against existing "default" factors in the inventory methods agreed upon
by the COP. Revisions to the internationally accepted methodologies, however, will lag
behind new scientific research results that indicate that factors should be revised. Therefore,
these "default" factors may, in some cases, be out of date - i.e., they may not be consistent
with current scientific understanding. Emission factors used in a national inventory may vary
significantly from the "default" factors because a country has developed new emission factors
based on scientific research and/or expert judgment, and has used them in the inventory
rather than the "default" factors. By essentially accelerating the dissemination of new
research results, the international review, evaluation, and sharing of emission factors in this
step in the verification process are likely to improve the accuracy of the factors used in the
methodology, as well as refine them to account for significant differences in technologies,
climate, management systems, and other variables.
Emission factors for a single activity vary among and within individual countries.
Therefore, choosing the correct emission factor for a specific activity requires an
understanding of not only the biogeochemical processes by which CH4 is produced and
emitted from that activity, but also the various management practices that affect emissions.
For example, production of CH4 in livestock varies not only among animal species, but also
among types within species - e.g., mature dairy cows, mature draft bullocks, and pre-weaned
calves. At the same time, CH4 production is affected by the physical and chemical
characteristics of the feed, the feeding level and schedule, the use of feed additives to
promote production efficiency, and the activity and health of the animal.
Evaluation of emission factors, therefore, should also include an evaluation of both the
biogeochemical and the socioeconomic or management variables used to derive the factors.
Information contained in scientific studies on the subject, as well as in other emission
inventories, should be used in this evaluation. Emission factors for certain sources are likely
to be similar to those in other countries with similar management practices, climatic conditions,
etc. For example, very few experimental measurements of CH4 flux from flooded rice fields
have been performed in Asia. Many Asian countries, therefore, will have to rely on estimates
derived in other, similar regions for this factor. Additional reports and data sources, such as
international and national agricultural production statistics, may be useful in comparing
variables. Outside sources of information, however, should be used with caution.
International data bases (e.g., the UN Food and Agriculture Organization (FAO) Agricultural
Production Statistics) sometimes do not agree with national or regional data bases. The
reasons for this are often not clear, since international data bases are usually based on data
provided by individual countries. National sources may be preferable, since they are usually
based on primary, rather than secondary, sources.
Page 11-7
-------
Evaluate the Accuracy and Appropriateness of the Activity Levels Used
Levels of activities that result in the emissions of CH4 are well quantified for some
sources because accurate national and international data bases exist, particularly in the
agriculture and energy sectors (e.g., livestock, rice cultivation, and fossil fuel production,
transport, and consumption). This is not true for all CH4 sources, in particular, biomass
burning, landfills, and liquid wastes. When specific data bases are not available, ancillary
data and expert judgment will have to be used, therefore, the process of compiling
inventories will indicate which data sets are inadequate for accurate assessments and which
are inconsistent. This process will point out what new data need to be collected within each
country, and may also instigate improved coordination of data collection and dissemination
among countries. This could be particularly useful for the collection and dissemination of
large, geographically broad, and expensive data sets, such as those compiled through remote
sensing.
Activity levels are based primarily on national compilations and statistics, which are
typically reported to a central repository. The reliability of the data and the plausibility of the
results should be evaluated. The data can be compared to international data bases, such as
the FAO Agricultural Production Statistics and the UN Energy Statistics, although as
discussed above, the results from such comparisons should only be used as general guides.
Nevertheless, interannual or regional variability in reported activity levels may indicate that
further data collection is recommended. Outside reports, such as fuelwood surveys by the
World Bank and nongovernmental organizations, national reports on rice cultivation or
livestock management, and regional studies of energy supply and demand, also can be useful
for cross-checks. Again, care must be taken to ensure that activity levels for the same activity
and year are being compared, since data bases are not always clearly defined. For example,
estimates of rice area harvested annually may or may not include upland rice areas. Upland
areas should not be counted, or should be assigned an emission factor of zero, when
estimating emissions because they are not flooded and, therefore, are not believed to produce
CH4.
11.2.2 Indirect Verification
The atmospheric distribution of CH4 reflects the geography of its sources and sinks.
Broad regions of elevated CH4 concentrations exist over strong source regions (e.g., Thorn et
al., 1993) because the distribution is not completely smoothed out by atmospheric circulation
(Fraser et al., 1986; and Steele et al., 1987). Mathematical models of tracer transport in the
atmosphere have been used to deduce objectively, from spatial and temporal distributions of
atmospheric concentrations, not only the total annual emissions and destruction of CH4, but
even rough seasonal and geographic distributions for these sources and sinks (Taylor et al.,
1991; Fung et al., 1991; Tie et al., 1991; Quay et al., 1991; and Brown, 1993). Thus, precise
measurements of atmospheric CH4 at a judiciously selected global network of sites can
provide valuable information for overall verification of total CH4 emissions from a particular
region.
While spatial variations of atmospheric CH4 are related to regional source strengths,
there is a limit to the information about the distribution of CH4 sources that can be extracted
from these variations. Atmospheric transport models, by themselves, yield CH4 source/sink
information that has been integrated over a few weeks and a few thousand kilometers.
Individual sources and sinks cannot be differentiated. For example, once CH4 emitted in
Europe from livestock, landfills, coal mining, and natural gas distribution systems mixes during
Page 11-8
-------
transport to a monitoring site in the North Atlantic Ocean, concentration measurements at the
monitoring site only will provide information about the sum of emissions from these sources.
Therefore, the attribution of an observed concentration profile to a particular distribution of
sources must make use of constraining data other than atmospheric CH4 concentrations (e.g.,
Fung etal., 1991).
One type of information that can be used to constrain direct estimates of emissions is
the isotopic content of CH4 in the atmosphere (Cicerone and Oremland, 1988; Tyler, 1986;
Wahlen et al., 1989; and Quay et al., 1991). Different CH4 sources have different isotopic
compositions, so the isotopic composition of atmospheric CH4 reflects the proportions of CH4
released by the different sources. Analysis of CH4's radiocarbon content (i.e., carbon-14
content, or 14C) provides a means of estimating fossil versus modern CH4 source strengths,
while its stable isotopic composition (i.e., the ratio of carbon-13 to carbon-12, or 13C/12C)
provides a means of distinguishing between biogenic and nonbiogenic sources.
Carbon-14 is a radioactive isotope, with a half-life of -5,700 years. As long as an
organism is alive, it exchanges carbon with its surroundings, and its 14C content equals that of
atmospheric .carbon dioxide. Once an organism dies, absorption of 14C ceases, and the
residual 14C continues to decay. The longer an organism has been dead, the lower its 14C
content. The formation of fossil fuels occurs on geologic time scales, so that CH4 associated
with these sources (oil, gas, and coal) contains zero 14C. The 14C content of atmospheric CH4
will thus reflect the fraction released from fossil carbon, and will decrease with increasing CH4
release from fossil fuels relative to release from nonfossil sources. Therefore, the 14C content
of atmospheric CH4 can place bounds on the emissions associated with the oil, natural gas,
and coal fuel cycles (Manning et al., 1990; Wahlen et al., 1989; and Quay et al., 1991).4
Similarly, measurement of the 13C/12C isotope ratio of atmospheric CH4 can place
bounds on biogenic and nonbiogenic source strengths (Cantrell et al., 1990; Stevens and
Engelkemeir, 1988; Tyler et al., 1988; and Quay et al., 1988). Sources of biogenic CH4 (i.e.,
CH4 produced by bacteria during anaerobic fermentation) include wetlands, rice cultivation,
livestock, solid and liquid wastes, and oceans and fresh waters. Nonbiogenic CH4 is derived
from thermal alteration of buried organic matter (i.e., fossil sources: oil, natural gas, and
coal)5 and from incomplete combustion during biomass burning. Methane from biogenic
sources has the lowest, or most negative, 13C/12C values, while CH4 from thermogenic sources
has mean values, and CH4 from biomass burning has the highest values (Quay et al., 1991).
Therefore, the 13C/12C ratio of atmospheric CH4 reflects the proportion of CH4 input from
biogenic versus nonbiogenic sources.6
Together, radiocarbon and stable isotope measurements of atmospheric CH4 can
narrow uncertainties about source strengths. For example, if a CH4 sample has a relatively
low 13C/12C ratio, but contains significant 14C, it is almost certainly of biogenic origin. Methane
from biomass burning would also have a significant 14C content, but would have a relatively
4 Methane from a few nonfossil sources (wetlands, bogs, rice fields, and tundra), is slightly depleted in 14C (e.g.,
Wahlen et al., 1989). Therefore, it contributes a small uncertainty to these calculations.
5 A small portion of natural gas deposits (-20%) contains biogenic CH4 (Rice and Claypool, 1981) - i.e., a small
portion is biogenic rather than nonbiogenic.
6 The 13C/12C isotope ratio of atmospheric CH4 is also a function of sink strengths because oxidation rates vary
between 13CH4 and 12CH4.
Page 11-9
-------
high 13C/12C ratio. If, however, a CH4 sample contains no 14C, it is of fossil origin. Data on
CH4 isotopes are thus a critical supplement to CH4 concentration measurements for identifying
and verifying the location and magnitudes of individual CH4 sources or groups of sources.
Available Measurements and Techniques
Measurements of atmospheric CH4 concentrations are obtained by analysis of air
samples collected at regular intervals from locations or sites established for long-term
monitoring, and occasionally during transects by aircraft or ships. Samples are usually
transported to a laboratory, where they are analyzed for CH4 content. Gas chromatography
with flame ionization detection is a widely accepted method of determining CH4 concentrations
in air samples, and yields a precision of 0.5% or better (Rasmussen and Khalil, 1981; and
Steele et al., 1987). Infrared absorption methods can also be used, although they are less
precise and more expensive than gas chromatography. The advantage of infrared absorption
methods is that high-frequency measurements can be made, which makes this technique
particularly useful for measuring CH4 concentrations from aircraft.
The existing network of long-term monitoring sites established by the Climate
Monitoring and Diagnostics Laboratory of the National Oceanic and Atmospheric
Administration (-30 sites) was designed to represent "clean" air (i.e., well-mixed, background
concentrations, rather than concentrations that are spatially or temporally distinct due to their
proximity to anthropogenic sources) (Steele et al., 1987). The strategy was formulated over
15 years ago (Komhyr et al., 1985). The purpose then was to capture the broad-scale
hemispheric and latitudinal gradients in atmospheric trace gas concentrations. The choice of
remote marine sites was dictated by sampling economy: spatial variability of trace gases near
urban or "polluted" sites is translated, via the winds, into temporal variability at a site. Recent
efforts in the last several years have focused on the addition of sites to narrow the estimates
of contributions from the various sources and of removals by sinks. For example, data from
around the Pacific rim, sampled by aircraft flown for dedicated research missions, or sampled
via collaborative programs with ships of opportunity, will not quantify individual sources, such
as CH4 from rice cultivation in China, but will put integrative constraints on the upwind source -
- i.e., constraints on the total regional strength of all sources from the Asian continent
(Dlugokencky et al., 1993).
Additional monitoring sites would improve understanding of regional source strengths.
Some strong regional sources, such as biomass burning in the Southern Hemisphere,
currently have no monitoring sites near enough to capture much of their seasonal signal.
However, as monitoring sites move closer to source regions, the CH4 signal becomes more
variable. Many years of data are required to obtain an average seasonal cycle, which can
then be used to constrain source strength estimates in atmospheric models of tracer transport.
Satellite capabilities for monitoring and mapping CH4 concentrations near the surface
do not exist at present. An instrument named MOPITT (Measurements of Pollution in the
Troposphere) (Drummond, 1992) proposed for the Earth Observing System (proposed launch
date of 1998) is purportedly able to measure CH4 concentrations near the ground. With a
target accuracy of 1 % for CH4 concentration, MOPITT will, at a minimum, provide repeated
global coverage, reveal CH4 "hot spots," and identify the scattered source regions. This
information can be used to test the plausibility of national emission inventories, as well as
cross-check tracer model inferences of regional sources/sinks.
Page 11-10
-------
Maintenance and Calibration of Standards
Current analytical techniques for measuring CH4 concentrations, such as gas
chromatography with flame ionization detection, require measurement of CH4 concentrations in
air samples relative to a "standard" of known CH4 content (Steele et al., 1987). Infrared
absorption methods, although absolute rather than relative, frequently use a reference gas to
improve the accuracy of the measurements. For CH4 measurements, working standards are
usually prepared in individual laboratories from natural air and are then intensively calibrated
against a primary set of standards that is maintained in the laboratory. The working standards
are used up over time, since they are destroyed in the measurement process. As a "batch" of
working standards is used up, a new batch is prepared and, again, intensively calibrated
against the laboratory's primary set of standards. Currently, laboratories around the world
either create their own primary standards or obtain them from one of several different sources.
There are no formal mechanisms for intercalibration of these standards among laboratories,
although some comparisons among different laboratories occur informally.
For ambient atmospheric CH4 measurements to be useful for verification of national
emission estimates, it is important that: (1) measurements made at different laboratories within
one country or in different countries must be capable of being compared quantitatively, and (2)
changes in the measurements made at different times at a single laboratory must be capable
of being interpreted as changes in sources or sinks. To achieve these goals, an international
standard scale -- i.e., a stringent set of "international primary" standards with known CH4
concentrations covering the range expected near ambient levels -- must be established and
maintained (WMO, 1981; and U.S. DOE, 1986).
These international primary standards would be propagated to "secondary" standards
that would then be used throughout the international measurement community. Individual
laboratories would use these secondary standards to calibrate their working standards. In this
way, CH4 measurements at different laboratories would all be made relative to working
standards derived from the same set of standards. Nonetheless, frequent intercalibrations
(i.e., about every two years) between laboratories would be necessary to ensure that
secondary standards are accurately propagated to working standards, especially since
laboratories have to switch from one "batch" of working standards to the next as they are used
up. The precision of the secondary standards should be ±0.5 ppbv. A protocol for
establishment of the primary standard and intercalibration of secondary standards (rather than
the current informal comparisons among different laboratories) needs to be investigated. Also,
the financial feasibility of propagating the international primary standards to secondary
standards would have to be studied in greater detail..
The internal consistency of the international standard scale would ensure, in part, the
quality of the measurements. Maintenance of the secondary standards at each laboratory
ensures that year-to-year changes in the CH4 concentration data record can be interpreted
correctly as changes in source (and/or sink) strengths and locations, rather than being
interpreted incorrectly due to the use of improperly calibrated standards. Moreover, the
establishment of an international CH4 standard scale for atmospheric measurements will allow
for quantitative comparison of measurements from one laboratory with those from other
laboratories around the world. Maintenance and calibration of standards, and intercomparison
of methods and results, are thus critical if useful atmospheric data for verification purposes are
to be obtained.
Page 11-11
-------
Standards should exist not only for CH4 concentration measurements, but also for the
determination of radiocarbon and stable isotopic contents. The radiocarbon content is given
by the decay activity of 14C, which is reported using the percent modern (pM) scale:
PM=(14Csamp,/4Cstandard)x100
The standard represents 95% of the 14C activity of an oxalic acid standard maintained by the
U.S. National Institute of Standards and Technology (NIST, formerly the U.S. National Bureau
of Standards). The NIST oxalic acid standard is used internationally as the standard for 14C
measurements.
The situation is more difficult for 13CH4. When determining isotopic contents of CH4 in
air samples, the CH4 in a sample is converted to CO2 and H2O, which are then measured for
isotopic content. Hence, among the standards that may be required for this determination are
a set of pure CO2 standards (gas or solid standard, such as barium carbonate) covering the
range of atmospheric 13CH4, and CH4 in whole air standards to compare sample preparation
methods. At present, there are no international isotopic standards for 13CH4.
Development of techniques for measuring CH4 concentrations that give better precision
than is currently available with gas chromatography with flame ionization detection (0.5% or
better) should be encouraged, particularly for application to propagating an international
standard scale to secondary standards. While methods for absolute calibration of CH4
standards are feasible (e.g., techniques used in isotope work to extract CH4 from whole air
samples (Lowe et al., 1991)), it is doubtful they can reach the precision of a relative method,
such as a gas chromatograph commonly used for atmospheric measurements. Nevertheless,
absolute calibration should be explored. If efforts to this end are successful, measurements
made with respect to a different scale could then be converted to the absolute scale.
Investigations of measurement techniques other than gas chromatography should be
encouraged to obtain higher precision and to propagate the international primary standard
scale to secondary standards.
Sampling and Reporting Strategy
There are several ways to obtain measurements of atmospheric CH4 and its isotopic
content for verification. Long-term monitoring sites provide information on seasonal, year-to-
year changes, and secular trends in the sources in a region. Regular or occasional transects
by aircraft or shipboard, or other intensive measurement campaigns, serve as spot checks
that can detect and quantify spatially mixed, in-situ emissions.
For the atmospheric data to be useful for verifying annual CH4 emissions from a
region, the measurements must be made with adequate temporal and spatial coverage. In
particular, measurement or sampling frequency must be great enough, and measurement
duration must be long enough, that the measurements define a full seasonal cycle. The
concentration data thus far have shown that the seasonal cycle is different in different places.
For example, at Cape Grim, Tasmania, the seasonal cycle consists of a winter maximum and
summer minimum. At Pt. Barrow, Alaska, however, there are double peaks in concentration,
one in February and the other in October. These seasonal cycles provide valuable
Information about strengths of some of the sources and sinks and are an important constraint
in atmospheric tracer models. A full year of atmospheric concentration data is necessary to
determine the annual mean at each station. The annual means at all the stations are
combined to define the latitudinal distribution of the atmospheric concentration, an important
Page 11-12
-------
constraint on source/sink locations in tracer models. For most locations, weekly samples are
sufficient to accomplish this. The samples should be collected from a well-defined wind
sector, such that samples are representative of well-mixed, large volumes of air, rather than a
local pollution source. At sites where the concentrations are highly variable, in-situ, high-
frequency measurements may be required. Since all sites have some interannual variability,
multiple years of measurements are required to derive accurate averages. Concurrent
measurements, such as meteorological data and other chemical species (such as CO2, CFCs,
aerosols) should be obtained to aid in the interpretation of the in-situ measurements.
Methane data can be collected and catalogued either by the World Meteorological
Organization's archive in Japan or by the Carbon Dioxide Information Analysis Center in Oak
Ridge, Tennessee. Ideally, the data should be reported based on the yet-to-be established
international CH4 standard, in one-year segments, and in a timely manner (i.e., within the next
calendar year). The data should also be scrutinized, and suspect data should be flagged
appropriately by the laboratory that made the measurements. For example, those samples
collected when the wind was out of sector or measured during a time of less than optimal
analytical performance should be indicated as such.
11.3 CONCLUSIONS
An international system to review and verify national inventories of CH4 emissions (and
emissions and sinks other greenhouse gases) should prove useful for several reasons. A
verification system, through the exchange, review, and comparison of data, would promote
dialogue and sharing of data among scientists and policymakers. This process is likely to
improve scientific understanding of the CH4 budget and refine national emission estimates,
thereby increasing confidence in the FCCC and the emission inventory process. Verification
of national emission inventories would also be useful in monitoring progress toward national
and global emission reduction goals.
Because of the great variety and number of CH4 emission sources, and because
emissions from different sources become mixed in the atmosphere, it is not possible to verify
national CH4 emission inventories precisely through direct measurement or mass balance
equations. An inventory can, however, be verified in terms of its plausibility, accuracy, and
completeness. This direct, or "bottom-up," method would include the following steps:
Evaluate the plausibility of the results.
Evaluate the inventory methods used.
Evaluate the accuracy of the calculations.
Evaluate the accuracy and appropriateness of the emission factors used.
Evaluate the accuracy and appropriateness of the activity levels used.
The results of this process would be used to improve national emission estimates, and
presumably would also be used to update and refine the internationally accepted inventory
methodologies.
Page 11-13
-------
Additional information with which to .verify national CH4 emission estimates can be
obtained from a "top-down" monitoring approach, in which the atmospheric signatures of the
emissions, in conjunction with a transport model, are used to test the plausibility of emission
estimates. Measurements of atmospheric CH4 concentrations and radioactive and stable
isotope contents can thus be used to provide cross-checks on total CH4 emissions from broad
regions or large countries and on national and regional CH4 emission estimates from certain
sources. This technique can also be used !to constrain estimates of global CH4 emissions
from individual sources or groups of sources. Atmospheric transport models that synthesize
the concentration and isotope data have proven useful in narrowing uncertainties about
individual CH4 source strengths.
i
The effectiveness and accuracy of this "top-down" monitoring technique in verification
of national, regional, and global emission estimates could be increased through a number of
research activities, including: (1) spatial expansion of the network of concentration measuring
sites; (2) greater temporal frequency of measurements at sites with seasonally varying source
strengths; (3) greater spatial coverage, and! accuracy, of isotope measurements; (4)
establishment of international standards, arid improved maintenance and calibration of
standards, both at individual laboratories arid internationally; and (5) standardized collection
and cataloguing of CH4 data at international centers.
11.4 REFERENCES
i
Verifik
Ausubel, J.H., and D.G Victor. 1992. Verification of international environmental agreements.
Annual Review of Energy and the Environment 17:1-43.
i
Brown, M. 1993. Deduction of emissions
-------
Feldman, D.L. 1992. Some lessons of the IAEA's nuclear non-proliferation regime for
confidence-building under a greenhouse gas convention. In di Primio, J.C., and G. Stein, eds.
A Regime to Control Greenhouse Gases, Proceedings of a Workshop, Bad Neuenahr, June
12-14, 1991. Forschungszentrum Julich GmbH, Julich, Germany. 79-90.
Fischer, W. 1992. Verification of international treaties for the protection of the environment -
Some empirical evidence. In di Primio, J.C., and G. Stein, eds. A Regime to Control
Greenhouse Gases, Proceedings of a Workshop; Bad Neuenahr, June 12-14, 1991.
Forschungszentrum Julich GmbH, Julich, Germany. 47-51.
Fischer, W., J.C. di Primio, and G. Stein. 1990. A Convention on Greenhouse Gases:
Towards the Design of a Verification System. Forschungszentrum Julich GmbH, Julich,
Germany. 43+ pp.
Fraser, P.J., P. Hyson, R.A. Rasmussen, A.J. Crawford, and M.A.K. Khalil. 1986. Methane,
carbon monoxide and methylchloroform in the Southern Hemisphere. Journal of Atmospheric
Chemistry 4:3-42.
Fung, I., J. John, J. Lerner, E. Matthews, M. Prather, L.P. Steele, and P.J. Fraser. 1991.
Three-dimensional model synthesis of the global methane cycle. Journal of Geophysical
Research 96:13,033-13,065.
IPCC (Intergovernmental Panel on Climate Change). 1991. Workshop on National
Inventories of Greenhouse Gas Emissions and Sinks. Proceedings of an International IPCC
Workshop, 5-6 December 1991, Geneva, Switzerland.
IPCC.
press.
1993. The Greenhouse Gas Inventory Reference Manual. Vol. 3, Review Draft. In
Komhyr, W.D., R.H. Gammon, T.B. Harris, L.S. Waterman, T.J. Conway, W.R. Taylor, and
K.W. Thoning. 1985. Global atmospheric CO2 distribution and variations from 1968-1982
NOAA/GMCC CO2 flask sample data. Journal of Geophysical Research 90:5,567-5,596.
Lewis, P.M. 1992. Experiences in verification - What can be learned for a greenhouse gas
convention? In di Primio, J.C., and G. Stein, eds. A Regime to Control Greenhouse Gases,
Proceedings of a Workshop, Bad Neuenahr, June 12-14, 1991. Forschungszentrum Julich
GmbH, Julich, Germany. 53-57.
Lowe, D.C., C.A.M. Brenninkmeijer, S.C. Tyler, and E.J. Dlugokencky. 1991. Determination
of the isotopic composition of atmospheric methane and its application in the Antarctic.
Journal of Geophysical Research 96:15,455-15,467.
Manning, M.R., D.C. Lowe, W.H. Melhuish, R.J. Sparks, G. Wallace, C.A.M. Brenninkmeijer,
and R.C. McGill. 1990. The use of radiocarbon measurements in atmospheric studies.
Radiocarbon 32:37-58.
OECD (Organization for Economic Cooperation and Development). 1991. Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from OECD Experts Meeting, 18-21
February 1991, Paris, France. Prepared for the Intergovernmental Panel on Climate Change.
OECD, Paris, France.
Page 11-15
-------
Quay, P.O., S.L. King, J.M. Landsown, and D.O. Wilbur. 1988. Isotopic composition of
methane released from wetlands: Implications for the increase in atmospheric methane.
Global Biogeochemical Cycles 2:385-397.
Quay, P.O., S.L. King, J. Stusman, D.O. Wilbur, L.P. Steele, I. Fung, R.H. Gammon, T.A.
Brown, G.W. Fan/veil, P.M. Grootes, and F.H; Schmidt. 1991. Carbon isotopic composition of
atmospheric methane: Fossil and biomass burning source strengths. Global Biogeochemical
Cycles 5:25-47. !
Rasmussen, R.A., and M.A.K. Khalil. 1981. Atmospheric methane (CH4): Trends and
seasonal cycles. Journal of Geophysical Research 86:9,826-9,832.
i
Rice, D.D., and G.E. Claypool. 1981. Generation, accumulation and resource potential of
biogenic gas. Bulletin of the American Association of Petroleum Geologists 65:5-25.
Steele, L.P., P.J. Fraser, R.A. Rasmussen, M.A.K. Khalil, T.J. Conway, A.J. Crawford, R.H.
Gammon, K.A. Maserie, and K.W. Thoning. 1987. The global distribution of methane in the
troposphere. Journal of Atmospheric Chemistry 5:125-171.
Stevens, C.M., and A. Engelkemeir. 1988. Stable carbon isotopic composition of methane
from some natural and anthropogenic sources. Journal of Geophysical Research 93:725-733.
Taylor, J.A., G. Brasseur, P. Zimmerman, and R. Cicerone. 1991. A study of the sources and
sinks of methane and methyl chloroform using a global three-dimensional Lagrangian
tropospheric tracer transport model. Journal of Geophysical Research 96:3,013-3,044.
Thorn, M., R. Bosinger, M. Schmidt, and I. Levin. 1993. The regional budget of atmospheric
methane of a highly populated area. Chemo$phere. In press.
Tie, X., D.N. Alyea, D.M. Cunnold, and C.Y.J. Kao. 1991. Atmospheric methane: A global
three-dimensional model study. Journal of Geophysical Research 96:17,339-17,328.
Tyler, S.C. 1986. Stable carbon isotope ratios in atmospheric methane and some of its
sources. Journal of Geophysical Research 91:13,233-13,238.
Tyler, S.C., P. Zimmermann, C. Cumberbatchm, J.P. Greenberg, C. Westberg, and J.P.E.C.
Darlington. 1988. Measurements and interpretation of 813C of methane from termites, rice
paddies and wetlands in Kenya. Global Biogeochemical Cycles 2:341-355.
UN (United Nations). 1992. United Nations framework Convention on Climate Change. May
9, 1992. U.N. Press, New York, New York.
U.S. DOE (U.S. Department of Energy). 1986. High Accuracy Standards, and Reference
Methodology for Carbon Dioxide in Air. DOE/PR-06010/31, U.S. DOE, Washington, D.C.
U.S. GAO (U.S. General Accounting Office). 1992a. International Agreements Are Not Well
Monitored. GAO/RCED-92-43, U.S. GAO, Washington, D.C.
U.S. GAO. 1992b. Strengthening the Implementation of Environmental Agreements.
GAO/RCED-92-188, U.S. GAO, Washington, D.C.
Page 11-16
-------
van Amstel, A.R., ed. 1993. Methane and Nitrous Oxide, Methods in National Emissions
Inventories and Options for Control. Proceedings of an International IPCC Workshop, 3-5
February 1993, Amersfoort, the Netherlands. RIVM, Bilthoven, Netherlands.
Wahlen, M., N. Tanaka, R. Henry, B. Deck, J. Zeglen, J.S. Vogel, J. Southon, A. Shemish, R.
Fairbanks, and W. Broecker. 1989. Carbon-14 in methane sources and in atmospheric
methane: The contribution from fossil carbon. Science 245:286-290.
WMO (World Meteorological Organization). 1981. WMO Environmental Pollution Monitoring
Programme, No. 5. Report of the WMO/UNEP/ICSU Meeting on Instruments, Standardization
and Measurement Techniques for Atmospheric CO2, 8-11 September 1981, WMO, Geneva,
Switzerland.
Page 11-17
-------
------- |