EPA/600/R 17/225 | August 2017
www.epa.gov/research
United States
Environmental Protection
Agency
Life Cycle Assessment of Cooking Fuel
Systems in India, China, Kenya, and Ghana
&ERA
Life Cycle Assessment of Cooking Fuel Systems In India, China,
Kenya, and Ghana - Data Analysis Tool
.ik
n
ggj
Q.
Compare Fuels
Customize Analyses
Explore Data
vm* vu Mini • •
1 1
Office of Research and Development
National Risk Management Research Laboratory
-------
VERG
Life Cycle Assessment of
Cookstoves and Fuels in
India, China, Kenya, and Ghana
Ben Morelli, Sarah Cashman, and Molly Rodgers
Eastern Research Group, Inc.
Lexington, MA 02421
Prepared for
Susan A. Thorneloe
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
Aucjus^^^
Contract No. EP-D-11-006
Work Assignment No. 5-10
-------
NOTICE
The U.S. Environmental Protection Agency through its Office of Research and Development
funded and managed the study described here under Contract EP-D-11-006 to Eastern Research
Group, Inc. This report has been subjected to the Agency's peer and administrative review and
has been approved for publication as an EPA document.
11
-------
ABSTRACT
Daily use of traditional cooking fuels and stoves in India, China, Kenya, and Ghana emits
harmful air pollutants that result in over a million premature deaths annually. Reducing pollution
from cookstoves is a key priority, as emissions from traditional cookstoves and open fires with
solid fuels are a major health concern and yield numerous environmental impacts. This project is
being conducted by the U.S. Environmental Protection Agency (US EPA) that aims to provide
life cycle assessment (LCA) data and tools that inform decisions regarding cookstove use and
fuel selection. This research expands the geographic scope of the Phase I study to include both
Kenya and Ghana. This work phase developed new stove use emission life cycle inventories
(LCIs) to conduct an uncertainty analysis for each fuel and stove type combination. The current
study also performs sensitivity analyses that test the effect of stove thermal efficiency, stove
technology use, electrical grid mix, forest renewability factor, and allocation approach on
environmental impacts of cookstove use. The study quantifies the effect that potential shifts in
the cooking fuel mix may have on the environmental impact of delivered cooking energy.
A normalized presentation of results is provided for each country, which helps to identify
the categories of environmental impact that are most strongly linked to the cooking sector. Study
results reinforce the findings of the Phase 1 study supporting the observation that the use of
traditional fuels and cookstoves contributes disproportionately to the environmental footprint of
cooking in developing nations. Normalized results further indicate that the cooking sector is a
dominant contributor for the countries of focus to national particulate matter formation potential
and black carbon (BC) and short-lived climate pollutant impacts, which are the two LCA
categories most strongly linked to human health impacts.
The study quantitatively demonstrates through the application of LCA that both cooking
fuel mix substitutions and stove technology upgrades provide promising avenues for reducing
particulate matter and BC emissions. India's results show that continued reliance on crop residue
and dung contributes disproportionately to particulate matter and BC environmental impacts. The
results suggest that a major environmental benefit in China could be realized by promoting
cooking fuel mix substitutions or stove technology improvements to replace the combustion of
coal powder in traditional stoves. Kenya and Ghana would benefit from adoption of improved
stove designs for both firewood and charcoal fuel. Use of improved charcoal kiln technology
also has the potential to significantly reduce the impact of charcoal use and production.
The study demonstrates the positive relative environmental results associated with
liquefied petroleum gas (LPG) and natural gas, which show a tendency to shift environmental
burdens away from indoor air pollutants and to other impact categories such as fossil fuel
depletion, fresh water eutrophication, and terrestrial acidification potential when substituted for
traditional fuels. Electric cookstoves demonstrate positive environmental performance in both
Kenya and Ghana, where they are characterized by relatively clean electrical grids. Without a
reduction in reliance on the use of coal as a source of electrical energy in India and China,
electric cookstoves are not able to match the environmental performance of other modern
cooking fuels. In addition to the selection of results presented in this report, dynamic results
workbooks are available to customize selection of national cooking fuel mix, stove technology
use, and sensitivity parameters to support the analysis of potential effects on environmental
impacts to support cookstove policy development in India, China, Kenya, and Ghana.
111
-------
FOREWORD
The United States Environmental Protection Agency (U.S. EPA) is charged by Congress
with protecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions leading to a
compatible balance between human activities and the ability of natural systems to support and
nurture life. To meet this mandate, EPA's research program is providing data and technical
support for solving environmental problems today and building a science knowledge base
necessary to manage our ecological resources wisely, understand how pollutants affect our
health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) within the Office of
Research and Development (ORD) is the Agency's center for investigation of technological and
management approaches for preventing and reducing risks from pollution that threaten human
health and the environment. The focus of the Laboratory's research program is on methods and
their cost-effectiveness for prevention and control of pollution to air, land, water, and subsurface
resources; protection of water quality in public water systems; remediation of contaminated sites,
sediments and ground water; prevention and control of indoor air pollution; and restoration of
ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's
research provides solutions to environmental problems by: developing and promoting
technologies that protect and improve the environment; advancing scientific and engineering
information to support regulatory and policy decisions; and providing the technical support and
information transfer to ensure implementation of environmental regulations and strategies at the
national, state, and community levels.
This publication was produced in support of ORD's Air, Climate, and Energy FY16-19
Strategic Research Action Plan. EPA, along with other federal partners, is working in
collaboration with the Global Alliance for Clean Cookstoves to conduct research and provide
tools to inform decisions about clean cookstoves and fuels in developing countries. EPA
previously completed a life cycle assessment (LCA) comparing the environmental footprint of
current and potential fuels and fuel mixes used for cooking within India and China (Cashman et
al. 2016). This study furthers the initial work by expanding the LCA methodology to include
new cooking mix and electrical grid scenarios, additional sensitivity analyses, uncertainty
analyses, and includes a normalized presentation of results. This phase of work also expands the
geographic scope of the study to include both Kenya and Ghana. Study results will allow
researchers and policy-makers to quantify sustainability-related metrics from a systems
perspective.
Cynthia Sonich-Mullin, Director
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
iv
-------
TABLE OF CONTENTS
Page
ES.l EXECUTIVE SUMMARY ES-1
ES.1.1 Introduction ES-1
ES.l.2 Phase II Project Approach ES-1
ES.1.3 Methodology ES-2
ES.1.4 Key Findings ES-6
ES. 1.4.1 Findings Common to all Countries Studied ES-10
ES.l.4.2 India ES-10
ES. 1.4.3 China ES-11
ES. 1.4.4 Kenya ES-11
ES.l.4.5 Ghana ES-12
ES.1.5 Report Organization ES-12
1. GOAL AND SCOPE DEFINITION 1-1
1.1 Introduction 1-1
1.2 Phase II Goal 1-1
1.3 Scope of the Study 1-2
1.3.1 Functional Unit 1-2
1.3.2 Geographic and Temporal Scope 1-2
1.3.3 T ransparency 1-3
1.3.4 Cooking Fuel Systems 1-4
1.3.5 System Boundary 1-7
1.3.6 Data Sources Summary 1-11
1.3.7 Life Cycle Impact Assessment Methodology and Impact
Categories 1-11
1.4 Quality Assurance 1-14
1.4.1 Data Quality Evaluation 1-15
1.4.2 Internal QA Review Procedures 1-17
2. COOKING FUEL AND STOVE DESCRIPTIONS AND METHODOLOGY 2-1
2.1 Fuel System Model Descriptions 2-1
2.1.1 Processed Fuel Heating Values 2-3
2.1.2 Electricity 2-5
2.1.3 Coal 2-7
2.1.4 Dung 2-7
2.1.5 Crop Residues 2-7
2.1.6 Firewood 2-8
2.1.7 Charcoal 2-9
2.1.8 Liquefied Petroleum Gas 2-9
2.1.9 Kerosene 2-10
2.1.10 Natural Gas 2-11
2.1.11 Coal Gas 2-11
2.1.12 Ethanol 2-12
2.1.13 Biogas 2-13
2.1.14 Biomass Pellets 2-13
v
-------
TABLE OF CONTENTS (Continued)
Page
2.2 Cookstove Descriptions 2-13
2.2.1 Stove Efficiency 2-15
2.2.2 Stove Technology Use by Country 2-16
2.2.3 Stove Emissions Data Sources and Methodology 2-19
3. METHODOLOGY FOR SCENARIO DEVELOPMENT AND
SENSITIVITY ANALYSES 3-1
3.1 Cooking Fuel Mix Scenario Development 3-1
3.1.1 India Cooking Fuel Mix Scenarios 3-4
3.1.2 China Cooking Fuel Mix Scenarios 3-7
3.1.3 Kenya Cooking Fuel Mix Scenarios 3-9
3.1.4 Ghana Cooking Fuel Mix Scenarios 3-12
3.2 Electrical Grid Scenario Development 3-14
3.2.1 India Electrical Grid Scenarios 3-16
3.2.2 China Electrical Grid Scenarios 3-19
3.2.3 Kenya Electrical Grid Scenarios 3-22
3.2.4 Ghana Electrical Grid Scenarios 3-23
3.3 Allocation Approach 3-25
3.3.1 Allocation to Crop Residues 3-27
3.3.2 Allocation to Biogas and Bioslurry 3-32
3.3.3 Biogas and Bioslurry LCI discussion 3-33
3.3.4 Electricity from Ethanol Production 3-35
4. METHODOLOGY FOR RESULTS COMPILATION 4-1
4.1 Biogenic Carbon Accounting 4-1
4.2 Non-Renewable Wood Fuel Calculations 4-1
4.3 Black Carbon and Short-Lived Climate Pollutants Calculations 4-2
4.4 LCA Model Framework 4-4
4.5 Monte Carlo Uncertainty Analysis 4-4
4.5.1 Uncertainty Modeling Documentation 4-5
4.6 Normalization 4-7
4.7 Results Presentation Format 4-8
5. UPDATED LCA RESULTS FOR INDIA 5-1
5.1 Cooking Fuel Mix Scenario Results - India 5-2
5.2 Baseline Normalized Results - India 5-7
5.3 Stove Efficiency Sensitivity - India 5-9
5.4 Electrical Grid Mix Sensitivity - India 5-13
5.5 Forest Renewability Factor Sensitivity - India 5-14
5.6 Allocation Approach Sensitivity - India 5-16
5.7 Stove Group Uncertainty Results - India 5-18
6. UPDATED LCA RESULTS FOR CHINA 6-1
6.1 Cooking Fuel Mix Scenario Results - China 6-2
vi
-------
TABLE OF CONTENTS (Continued)
Page
6.2 Baseline Normalized Results - China 6-5
6.3 Stove Efficiency Sensitivity - China 6-8
6.4 Electrical Grid Mix Results - China 6-10
6.5 Forest Renewability Factor Sensitivity - China 6-11
6.6 Allocation Approach Sensitivity - China 6-11
6.7 Stove Group Uncertainty Results - China 6-16
7. LCA RESULTS FOR KENYA 7-1
7.1 Cooking Fuel Mix Scenario Results - Kenya 7-2
7.2 Baseline Normalized Results - Kenya 7-6
7.3 Stove Efficiency Sensitivity - Kenya 7-9
7.4 Electricity Grid Mix Sensitivity - Kenya 7-11
7.5 Forest Renewability Factor Sensitivity - Kenya 7-12
7.6 Stove Group Uncertainty Results - Kenya 7-14
8. LCA RESULTS FOR GHANA 8-1
8.1 Cooking Fuel Mix Scenario Results - Ghana 8-2
8.2 Baseline Normalized Results - Ghana 8-5
8.3 Stove Efficiency Sensitivity - Ghana 8-8
8.4 Electricity Grid Mix Sensitivity - Ghana 8-10
8.5 Forestry Renewability Factor Sensitivity - Ghana 8-11
8.6 Stove Group Uncertainty Results - Ghana 8-13
9. KEY TAKEAWAYS BY COUNTRY AND STUDY CONCLUSIONS 9-1
9.1 Key Takeaways 9-1
9.1.1 Findings Common to all Countries 9-1
9.1.2 India 9-1
9.1.3 China 9-2
9.1.4 Kenya 9-2
9.1.5 Ghana 9-3
9.2 Conclusions 9-3
10. REFERENCES 10-1
APPENDIX A: BASELINE SINGLE COOKING FUEL RESULTS BY LIFE CYCLE
STAGE
APPENDIX B: COMPARISON OF RESULTS UPDATES BETWEEN PHASE I AND
PHASE II STUDY FOR INDIA AND CHINA
APPENDIX C: DATA QUALITY
vii
-------
LIST OF TABLES
Page
Table ES-1. LCIA Categories Considered in Phase II ES-5
Table 1-1. Abbreviated In-text References and Descriptions of Supplementary
Information 1-4
Table 1-2. Cooking Fuel List and Description 1-5
Table 1-3. Fuels Used for Cooking in Countries Studied 1-7
Table 1-4. Environmental Impact Category Descriptions and Units 1-13
Table 2-1. Emission Profile Sources for Each Stove Type by Study Country 2-2
Table 2-2. Heating Values of Cooking Fuels in India 2-3
Table 2-3. Heating Values of Cooking Fuels in China 2-4
Table 2-4. Heating Values of Cooking Fuels in Kenya 2-4
Table 2-5. Heating Values of Cooking Fuels in Ghana 2-5
Table 2-6. Current Electricity Grids in Covered Countries 2-6
Table 2-7. Stove Type and Efficiency by Nation 2-14
Table 2-8. Stove Technology Use and Aggregate Thermal Efficiency by Fuel 2-18
Table 2-9. Summary Table Showing Representative Stove Emissions 2-21
Table 3-1. Adoption of Improved Stove Technologies and Thermal Efficiency in India 3-2
Table 3-2. Adoption of Improved Stove Technologies and Thermal Efficiency in China 3-3
Table 3-3. Adoption of Improved Stove Technologies and Thermal Efficiency in Kenya 3-3
Table 3-4. Adoption of Improved Stove Technologies and Thermal Efficiency in Ghana 3-4
Table 3-5. Cooking Fuel Mix Scenario Names and Descriptions for India 3-4
Table 3-6. Cooking Fuel Mix Scenarios Evaluated for India 3-5
Table 3-7. Cooking Fuel Mix Scenario Names and Descriptions for China 3-7
Table 3-8. Cooking Fuel Mix Scenarios Evaluated for China 3-7
Table 3-9. Cooking Fuel Mix Scenario Names and Descriptions for Kenya 3-9
Table 3-10. Cooking Fuel Mix Scenarios Evaluated for Kenya 3-10
Table 3-11. Cooking Fuel Mix Scenario Names and Descriptions for Ghana 3-12
Table 3-12. Cooking Fuel Mix Scenarios Evaluated for Ghana 3-12
viii
-------
LIST OF TABLES (Continued)
Page
Table 3-13. Projected Electrical Grid Mix Contributions by Fuel for Ghana 2020 3-24
Table 3-14. Summary of Baseline and Sensitivity LCA Modeling and Allocation Options 3-26
Table 3-15. Crop-to-Residue Ratios, Production, and Fraction of Crop Residue Produced
- India and China 3-28
Table 3-16. Crop Residue Allocation Factors - India and China 3-30
Table 3-17. Allocation Percentages between Bioslurry and Biogas in India 3-33
Table 3-18. Allocation Factors for Biogas and Bioslurry in China 3-33
Table 4-1. Phase I and II Forest Renewability Factors 4-1
Table 4-2. Characterization Factors for BC eq 4-4
Table 4-3. Sources and Mechanisms of Uncertainty in Cookstove LCIs 4-5
Table 4-4. Sources and Mechanisms of Uncertainty in Crop Production 4-6
Table 4-5. Country Specific Normalization Factors (per person per year) 4-7
Table 4-6. Household Energy Use for Cooking per Year 4-8
Table 5-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - India 5-1
Table 5-2. Cooking Fuel Mix Scenario Technology Options (Figure Key) 5-2
Table 6-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - China 6-1
Table 6-2. Cooking Fuel Mix Scenario Technology Options (Figure Key) 6-2
Table 7-1. Summary Table of Single Cooking Fuel Results by Impact Category
(Impact/GJ Delivered Cooking Energy) - Kenya 7-1
Table 7-2. Cooking Fuel Mix Scenario Technology Options (Figure Key) 7-2
Table 8-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - Ghana 8-1
Table 8-2. Cooking Fuel Mix Scenario Technology Options (Figure Key) 8-2
IX
-------
LIST OF FIGURES
Page
Figure ES-1. Current cooking fuel mix in China, India, Kenya, and Ghana ES-3
Figure ES-2. Select fuel mix LCIA results for GCCP and PMFP under varying
technology assumptions in India. Scenario results shown relative to baseline
results ES-7
Figure ES-3. Select fuel mix LCIA results for GCCP and PMFP under varying
technology assumptions in China. Scenario results shown relative to baseline
results ES-8
Figure ES-4. Select fuel mix LCIA results for GCCP and PMFP under varying
technology assumptions in Kenya. Scenario results shown relative to baseline
results ES-9
Figure ES-5. Select fuel mix LCIA results for GCCP and PMFP under varying
technology assumptions in Ghana. Scenario results shown relative to baseline
results ES-9
Figure 1-1. Study system boundaries of the baseline scenario for covered countries 1-9
Figure 2-1. Range of reported stove thermal efficiencies by cooking fuel type 2-16
Figure 3-1. Potential future electrical grid mixes in India 3-18
Figure 3-2. Potential future electrical grid mixes in China 3-21
Figure 3-3. Potential future electrical grid mixes in Kenya 3-23
Figure 3-4. Potential future electrical grid mixes in Ghana 3-25
Figure 5-1. India GCCP cooking fuel mix scenario results 5-4
Figure 5-2. India PMFP cooking fuel mix scenario results 5-5
Figure 5-3. India CED cooking fuel mix scenario results 5-6
Figure 5-4. India normalized LCIA results 5-8
Figure 5-5. Effect of stove thermal efficiency improvement on PMFP traditional cooking
fuel results in India 5-10
Figure 5-6. Effect of stove thermal efficiency improvement on CED: traditional cooking
fuel results in India 5-12
Figure 5-7. Effect of electrical grid mix on GCCP impact of electric cookstoves in India 5-13
Figure 5-8. Effect of electrical grid mix on BC impact of electric cookstoves in India 5-14
Figure 5-9. Comparative effect of forest product renewability assumption on GCCP in
India 5-15
x
-------
LIST OF FIGURES (Continued)
Page
Figure 5-10. Effects of allocation methodology choice on sugarcane ethanol and biogas
GCCP impact in India 5-17
Figure 5-11. Effects of allocation methodology choice on sugarcane ethanol and biogas
CED impact in India 5-17
Figure 5-12. India GCCP uncertainty analysis results for traditional cooking fuels
modeled with various allocation approaches and stove technologies 5-20
Figure 5-13. India POFP uncertainty analysis results for traditional cooking fuels
modeled with various allocation approaches and stove technologies 5-21
Figure 5-14. India WDP uncertainty analysis results for select cooking fuels modeled
with various allocation approaches and stove technologies 5-22
Figure 6-1. China GCCP cooking fuel mix scenario results 6-3
Figure 6-2. China PMFP cooking fuel mix results 6-4
Figure 6-3. China CED cooking fuel mix results 6-5
Figure 6-4. China normalized LCA results 6-7
Figure 6-5. Cooking fuel form and stove thermal efficiency effects on GCCP of coal in
China 6-9
Figure 6-6. Effect of electrical grid mix on GCCP impact of electric cookstoves in China 6-10
Figure 6-7. Comparative effect of forest product renewability assumption on GCCP in
China 6-11
Figure 6-8. Effects of LCA allocation approach on crop residue GCCP impact in China 6-13
Figure 6-9. Effects of LCA allocation approach on crop residue PMFP impact in China 6-14
Figure 6-10. Effect of LCA allocation methodology on biogas and sugarcane ethanol
TAP impact 6-16
Figure 6-11. China GCCP uncertainty analysis results for improved stoves and modern
cooking fuels modeled with various allocation approaches 6-17
Figure 6-12. China PMFP uncertainty analysis results for select fuels modeled with
various allocation approaches and stove technologies 6-19
Figure 7-1. Kenya GCCP cooking fuel mix scenario results 7-4
Figure 7-2. Kenya CED cooking fuel mix scenario results 7-5
Figure 7-3. Kenya PMFP cooking fuel mix scenario results 7-6
Figure 7-4. Kenya normalized LCA results 7-8
XI
-------
LIST OF FIGURES (Continued)
Page
Figure 7-5. Kenya PMFP effect of stove thermal efficiency modeled with various stove
technologies 7-10
Figure 7-6. GCCP of electric cookstove with various electrical grid mix options in
Kenya 7-11
Figure 7-7. Comparative effect of forest product renewability on GCCP in Kenya
modeled for various stove technologies 7-13
Figure 7-8. Kenya GCCP uncertainty analysis results for wood-based cooking fuels
modeled with various allocation approaches and stove technologies 7-14
Figure 7-9. Kenya PMFP uncertainty analysis results for wood-based cooking fuels
modeled with various allocation approaches and stove technologies 7-15
Figure 7-10. Kenya TAP uncertainty analysis results for select cooking fuels modeled
with various allocation approaches and stove technologies 7-16
Figure 8-1. Ghana GCCP cooking fuel mix results 8-3
Figure 8-2. Ghana PMFP cooking fuel mix results 8-4
Figure 8-3. Ghana CED cooking fuel mix results 8-5
Figure 8-4. Ghana normalized LCA results 8-7
Figure 8-5. GCCP effects of stove thermal efficiency in Ghana for various kiln and stove
technologies 8-9
Figure 8-6. Ghana GCCP of electric cookstove use with various electrical grid mix
options 8-10
Figure 8-7. Comparative effect of forest product renewability assumption on GCCP in
Ghana 8-12
Figure 8-8. Ghana CED uncertainty analysis results for select cooking fuels modeled with
various stove technologies 8-14
Figure 8-9. Ghana POFP uncertainty analysis results for select cooking fuels modeled
with various stove technologies 8-15
Xll
-------
ACRONYMS AND ABBREVIATIONS
AD
Anaerobic Digester
AGB
Above Ground Biomass
AIS
Accelerated Improvement Scenario
BAU
Business As Usual
BC
Black Carbon
BCG
Boston Consulting Group
BGB
Below Ground Biomass
BrC
Brown Carbon
CCS
Carbon Capture and Storage
CED
Cumulative Energy Demand
CFC
Chlorofluorocarbons
ch4
Methane
CIS
Continued Improvement Scenario
CN
China
CO
Carbon Monoxide
co2
Carbon Dioxide
co3
Carbonate
DME
Dimethyl Ether
DS
Domestic Supply
EC
Elemental Carbon
eq
Equivalent emissions
EIA
Energy Information Administration
EOL
End of Life
ERG
Eastern Research Group
FAO
Food and Agriculture Organization
FC
Final Consumption
FEP
Freshwater Eutrophication Potential
FDP
Fossil Fuel Depletion
GACC
Global Alliance for Clean Cookstoves
GCCP
Global Climate Change Potential
GDP
Gross Domestic Product
GH
Ghana
GHG
Greenhouse Gas
GJ
Gigajoule
GSF
Gold Standard Foundation
HAPs
Hazardous Air Pollutants
HHV
Higher Heating Value
IEA
International Energy Agency
IGCC
Integrated Gasification Combined Cyc
Xlll
-------
ACRONYMS AND ABBREVIATIONS
IN
India
IOCL
Indian Oil Corporation Limited
IPCC
Inter-Governmental Panel on Climate Change
IRENA
International Renewable Energy Agency
ISO
International Organization for Standardization
kg
Kilogram(s)
LBNL
Lawrence Berkeley National Laboratory
LCA
Life Cycle Assessment
LCI
Life Cycle Inventory
LCIA
Life Cycle Inventory Assessment
LHV
Lower Heating Value
ISO
International Standards Organization
K
Potassium
KE
Kenya
LPG
Liquefied Petroleum Gas
N
Nitrogen
NGCC
Natural Gas Combined Cycle
nh3
Ammonia
NMVOC
Non-Methane Volatile Organic Compound
NOx
Nitrogen Oxides
N03
Nitrate
N20
Nitrous Oxide
NPK
Nitrogen, Phosphorus and Potassium
NREL
National Renewable Energy Laboratory
NRMRL
National Risk Management Research Laboratory
OC
Organic Carbon
ODP
Ozone Depletion Potential
ONGC
Oil and Natural Gas Corporation
ORD
Office of Research and Development
P
Phosphorus
POFP
Photochemical Oxidant Formation Potential
PM2.5
Particulate Matter, <2.5 micrometers
PMFP
Particulate Matter Formation Potential
PM10
Particulate Matter, <10 micrometers
QA
Quality Assurance
QAPP
Quality Assurance Project Plan
SI
Supplementary Information
SLCPs
Short-Lived Climate Pollutants
S02
Sulfur Dioxide
xiv
-------
ACRONYMS AND ABBREVIATIONS
SOx
Sulfur Oxides
TAP
Terrestrial Acidification Potential
TERI
The Energy and Resources Institute
U.S. EPA
United States Environmental Protection Agency
USD A
United States Department of Agriculture
US LCI
United States LHCFCife Cycle Inventory
VOC
Volatile Organic Carbon
WDP
Water Depletion Potential
WISDOM
Woodfuel Integrated Supply/Demand Overview Mapping
xv
-------
Executive Summary
ES.l EXECUTIVE SUMMARY
ES.1.1 Introduction
Cookstove use in developing countries affects millions of lives daily with far-reaching
consequences. Reducing pollution from cookstoves is a key priority as emissions from
cookstoves and open fires with solid fuels are a major health concern and contribute to numerous
environmental impacts. The U.S. Environmental Protection Agency (U.S. EPA) is conducting
research to provide data and tools that inform decisions regarding clean cookstoves and fuels for
developing countries. Toward this end, the U.S. EPA previously completed a life cycle
assessment (LCA) comparing the environmental footprint of current and potential fuels and
cooking fuel mixes used for cooking within India and China (Cashman et al. 2016). This report
builds on the original India and China cookstove LCA (referred to throughout this report as the
"Phase I study"), by expanding the LCA methodology to include new cooking fuel mix and
electrical grid scenarios, additional sensitivity analysis, uncertainty analysis, and a normalized
presentation of results. This phase of work also expands the geographic scope of the study to
include both Kenya and Ghana.
The scope expansion of the Phase II work was conducted to develop new data on stove
combustion emissions and present uncertainty results so that policies aimed at reduction of
environmental impacts in the cooking sector can be pursued with greater confidence.
Implementation of sensitivity analyses regarding stove thermal efficiency, cooking fuel mix,
electrical grid mix, forest renewability, and the choice of LCA allocation approach further
bolster the ability of the research to provide robust guidance tools to stakeholders considering
changes to the current cooking fuel mix and stove technology for the countries studied.
To support decision-making the study focuses on delivering information to stakeholders
such as insights into the potential variation in emissions associated with a given stove type (e.g.,
firewood burned in a traditional stove), the potential benefits of adopting improved stove designs
or implementing specific cooking fuel mix substitutions (e.g., charcoal replaces crop residue),
and the influence that renewability of forestry practice has on global climate change potential
(GCCP) of wood-based cooking fuels.
ES.l.2 Phase II Project Approach
The Phase II report is intended to provide an understanding of the types of data and other
information resulting from the use of LCA to evaluate fuels and stoves for India, China, Kenya,
and Ghana. Plans are to develop a LCA calculator that provides access to all of the data and
information available through this research. This will allow for a site-specific analysis to
evaluate LCA environmental tradeoffs by stakeholders interested in furthering emission
reductions for this source category (i.e., emissions from the wide array of stoves and fuels).
The number of countries and the breadth of included cooking fuels, stove types, fuel
mixes, and sensitivity analysis prohibits the possibility of including or discussing all generated
results within the report itself. Rather, the Phase II report is intended to document pertinent
methods regarding study assumptions and present select results that provide the most
comprehensive and engaging perspective on the findings of the second phase of work. A series
of four dynamic results workbooks is included along with detailed supplementary information
(SI) files that comprehensively document the full range of research, tabular results, and figures
ES-1
-------
Executive Summary
that were generated for Phase II. Table 1-1 in the main report introduces the associated
supplementary files, while Section 4.7 presents an introduction to results as presented both in this
report and in the associated results workbooks for each country.
Detailed discussion and figure presentation in the main report focuses on the following
impact categories:
• GCCP
• Cumulative Energy Demand (CED)
• Particulate Matter Formation Potential (PMFP)
• Black Carbon (BC) and Short-Lived Climate Pollutant Potential.
Results for other impact categories are presented when the findings are of particular
interest or the results do not follow the same trends as seen for the four primary impact
categories.
ES.1.3 Methodology
This LCA investigates both current fuels and those with market potential for cookstoves
in India, China, Kenya, and Ghana. The current national cooking fuel mix for each country,
including potential fuels considered but not currently utilized in measurable quantities, is
illustrated in Figure ES-1. The Phase II report focuses on the environmental impact of cooking
fuels included in both current and projected future cooking fuel mixes along with select results
from the sensitivity analysis that highlight key national trends. The following life cycle stages
are analyzed for each fuel system:
• Production of the cookstove fuel feedstock, including all stages from extraction or
acquisition of the fuel feedstock from nature through production into a form ready for
processing into cooking fuel (e.g., cultivation and harvesting of sugarcane, extracting
crude oil from wells).
• Processing of the fuel into a form ready to be used in a cookstove.
• Distribution of fuels from the production site to the processing location and on to a
retail location or directly to the consumer. Distribution also includes bottling for fuels
stored in cylinders.
• Use of the fuel via combustion of the fuel or use of electricity in a cookstove,
including disposal of any combustion wastes or residues (e.g., ash).
Cookstove production and distribution, human energy expended during collection of
fuels, and the production, preparation, consumption, and disposal of food and food wastes are
outside the boundaries of this project. A previous LCA examining production of fuel-efficient
cookstoves found that the use phase significantly dominates life cycle greenhouse gas (GHG)
emissions regardless of the combusted cooking fuel type utilized (Wilson 2016); therefore, it is
reasonable to exclude processes associated with stove production and distribution from the study
scope.
ES-2
-------
Executive Summary
China12
Biomass Pellets. Sugarcane Ethanol,
0% _ 0%
Electricity, 11%
Natural
Gas, 2%
Biogas,
0%
\_ Crop
Residue,
12%
Kerosene,
0.3%
Kenya6
Electricity.
1%
LPG, 5%
Kerosene.
12%
Biogas,
1%
Firewood,
15%
Biomass
Pellets,
0%
Sugarcane
Ethanol, 0%
Charcoal,
17%
Firewood.
65%
Natural Gas, 0%
Biomass
Pellets, 0%
LPG. 25%
Electricity
0°/c
Coal. 1%
Duns. 11%
Biosas. 0%
Crop
Residue.
9%
ugarcane
Ethanol.
Charcoal,
1%
Sugarcane
Ethanol,
0%
Electricity, 0.3%
Biogas,
no/„
Firewood,
49%
Ghana7
Biomass
Pellets,
0%
Crop
Residue,
0.4%
Charcoal.
31%
Firewood,
46%
¦ Coal
¦ Dung
¦ Crop Residue
Firewood
¦ Charcoal
• Kerosene
¦ LPG
Natural Gas
¦ Electridtv
¦ Sugarcane Ethanol
Biogas
¦ Biomass Pellets
Sources: 'Dalberg 2014,2NBSC 2008,3Dalberg 2013b, 4Gov. of India 2014, 5Venkataraman et al. 2010, «KNBS 2012,
7GLSS6 2014
Figure ES-1. Current cooking fuel mix in China, India, Kenya, and Ghana.
Results of the LCA are expressed in terms of a common functional unit. As this analysis
is a comparison of different fuels used to provide cooking energy, an energy functional unit is a
proper basis of comparison. Therefore, the LCA results are based on useful energy delivered for
cooking: 1 gigqjoule (GJ) of useful energy delivered to the pot for cooking.
ES-3
-------
Executive Summary
This study investigates bio-based and fossil-based cooking fuels, as well as electricity (a
mix of fuel types) currently used at a measurable level of capacity in India, China, Kenya, and
Ghana, as depicted in Figure ES-1. Cooking fuels not currently used or used in only small
quantities but with future market potential in these four countries are also assessed. Four future
cooking fuel mix scenarios were developed for each nation studied (displayed in Table 3-6,
Table 3-8, Table 3-10, and Table 3-12 of Section 3.1 of the main report) through review of
public sources discussing possible changes in the fuels used within these countries, as well as
through an analysis of past trends observed within the four study countries. The scenarios focus
on a continuation of current trends, feasible substitution of modern and improved fuels, and a
Diverse Modern Fuels scenario for each nation that explores a more dramatic departure from the
current cooking fuel mix.
National estimates of stove technology use were developed for each country and are
applied to both current and future cooking fuel mixes. A future improved stove technology mix
that assumes full adoption of improved stove designs is included in the analysis and serves to
highlight the relative potential environmental benefit of stove technology upgrades both in
combination with and as opposed to a strategy that focuses on cooking fuel mix substitutions.
Adoption refers to the future use of improved stove technologies or fuel forms in place of current
alternatives. Each stove group, which is defined as a unique combination of fuel type, stove type
(traditional, improved, modern) and country of use, has an associated current and future
improved stove thermal efficiency value based on records of stove performance drawn from the
literature. Results are run for both estimates of stove thermal efficiency to estimate the current
potential variability that exists within each stove group. Impact results for electric cookstoves are
generated for both current and potential future electrical grid mix scenarios.
Two forms of methodology-related sensitivity analysis are also included within the study.
The first of these involves the choice of LCA modeling conventions that are used to allocate
environmental impact among multi-output processes. The effect of allocation approach selection
on environmental impact of cooking with crop residue, biogas, and sugarcane ethanol are
explored in the sensitivity analysis. Table 3-14 lists both the baseline and sensitivity allocation
approaches considered in this study, and Section 3.3 describes each method in detail. The second
methodology-related sensitivity involves the national forest renewability factor that is used to
estimate GHG emissions for each country. Table 4-1 lists both Phase I and Phase II forest
renewability factors considered in the sensitivity analysis. As is described in Section 4.1, for
biomass, only carbon dioxide (CO2) emissions from non-renewable wood products are
considered to contribute to GCCP.
Uncertainty results were calculated using baseline assumptions for each stove grouping.
Uncertainty ranges help to quantify variability in the environmental performance of a given stove
grouping and to demonstrate the potential overlap in environmental performance that exists
between stove groupings. The nature of this study, which looks at national average
environmental emissions of numerous cookstove options, necessarily encompasses many sources
of uncertainty that affect result calculations at various levels of implementation, which range
from consideration of a single life cycle stage for a given fuel all the way up to aggregate
national cooking sector results. Section 4.5 introduces Monte Carlo uncertainty analysis and the
study parameters that contribute to uncertainty in baseline results.
ES-4
-------
Executive Summary
The Phase II study also includes a normalized presentation of life cycle impact
assessment (LCIA) results for each country. Normalization compares cooking sector impacts to
national estimates of characterized environmental impact in each country. Normalized results
provide an indication of the relative contribution that the cooking sector makes to environmental
impact across the different categories considered in this study. Section 4.6 describes
normalization and documents the country-specific statistics and normalization factors that were
used in the analysis.
The environmental analysis was conducted in accordance with the following voluntary
international standards for LCAs:
• International Standards Organization (ISO) 14040: 2006, Environmental management
- Life cycle assessment - Principles and framework (ISO 2010a); and
• ISO 14044: 2006, Environmental management - Life cycle assessment -
Requirements and guidelines (ISO 2010b).
The majority of life cycle inventory (LCI) data were extracted from existing studies in
publicly available academic literature. An LCI is an accounting of the material, energy, and
water inputs and the product, waste, emission, and water outputs for a product or process
(Baumann and Tillman 2004). Detailed unit process LCI data were entered into openLCA
software (GreenDelta 2016) to calculate the LCIA results. LCIA is the process of translating
emissions data contained in an LCI into environmental loads, which help users to interpret
cumulative environmental impacts of the studied system (Baumann and Tillman 2004). Table
ES-1 lists ten impact assessment indicators included in this study and the units in which they are
reported.
Table ES-1. LCIA Categories Considered in Phase II
Impact/Inventory Category
Abbreviation
Unit
Global Climate Change Potential
GCCP
kg C02 eq
Cumulative Energy Demand
CED
MJ
Water Depletion Potential
WDP
m3
Black Carbon and Short-Lived Climate Pollutants
BC
kg BC eq
Particulate Matter Formation Potential
PMFP
kg PM10 eq
Terrestrial Acidification Potential
TAP
kg SO2 eq
Freshwater Eutrophication Potential
FEP
kg P eq
Photochemical Oxidant Formation Potential
POFP
kg NMVOC eq
Ozone Depletion Potential
ODP
kg CFC-11 eq
Fossil Depletion Potential
FDP
kg oil eq
This suite of indicators addresses global, regional, and local impact categories of
relevance to the cookstove sector such as energy demand driving depletion of bio-based and
fossil-fuel resources, and GHG and BC emissions causing both long-term and short-term climate
effects. Of particular concern, are those impact categories that directly impact human health.
ES-5
-------
Executive Summary
These categories include emissions resulting in BC, particulate matter formation, and
photochemical oxidant formation, all of which can lead to eye irritation, respiratory disease,
increased risk of infection, and cancer (Goedkoop et al. 2008). Table 1-4 in Section 1.3.8
provides a description of each impact category. Results for most impact categories are calculated
using the ReCiPe impact assessment methodology (Goedkoop et al. 2008). For the category of
GCCP, contributing elementary flows are characterized using factors reported by the
Intergovernmental Panel on Climate Change (IPCC) in 2013 with a 100-year time horizon (IPCC
2013). BC and co-emitted species are characterized to BC-equivalents (eq) based on a novel
method released by the Gold Standard Foundation (GSF) (GSF 2015). CED and WDP are also
included as inventory indicators. CED and WDP inventory indicators are not associated directly
with environmental impacts but rather provide a cumulative count of energy and water resources
used within the study system. CED includes both renewable and non-renewable energy sources.
ES.1.4 Key Findings
Select LCA results pertaining to cooking fuel mix scenarios for each country, which
include the potential effect of improved stove technology adoption, are presented in Figure ES-2,
Figure ES-3, Figure ES-4, and Figure ES-5 for India, China, Kenya, and Ghana, respectively.
Potential future cooking fuel mix results in each figure are presented relative to current fuel mix
impacts for both GCCP and PMFP. Section 3.1 introduces the developed projections for future
cooking fuel mix scenarios.
One of the more positive findings of the Phase II work involves the discovered sensitivity
of BC and PMFP impact to projected changes in both stove technologies use and cooking fuel
mix substitutions. This finding is clearly demonstrated in Figure ES-2 through Figure ES-5 for
PMFP by the dramatic drop in bar height associated with all potential future cooking fuel mix
scenarios, relative to the current cooking fuel mix scenario. This finding indicates that multiple
policy approaches can be expected to produce desirable results concerning the two impact
categories that are most strongly affected by activity in the cooking sector. All four nations still
rely not only on traditional cooking fuel sources but also on traditional cookstoves with low
associated thermal efficiencies. Uncertainty results, presented as part of the sensitivity analysis,
support findings from the baseline results asserting the significance of emission reductions
achievable through adoption of improved stove designs and cooking fuel forms. The term fuel
form recognizes that energy feedstocks are not limited to use in just a single configuration. As an
example, coal powder can be compressed into honeycomb briquettes or transformed into coal
gas, leading to variable environmental performance from a single energy source.
Cooking fuel mixes also demonstrate that in many cases the magnitude of impact for a
given country and indicator tends to be driven predominantly by one or two traditional fuels and
stove technologies. For example, PMFP emissions in India are disproportionately associated with
dung and crop residue combustion, and the use of these two fuels for cooking is exacerbated by
reliance on traditional stoves. In China, reliance on various forms of coal to provide
approximately one-third of cooking energy produces a disproportionate share of impact in
several impact categories. Traditional cooking fuel sources are associated with a wide range of
stove types and even fuel forms, and the results shown present an opportunity to reduce
environmental burdens without the necessity of abandoning a traditional fuel source. Substituting
honeycomb coal briquettes for coal powder in China, for example, presents an opportunity to
ES-6
-------
Executive Summary
reduce GCCP and PMFP impacts of coal-derived cooking energy by approximately 70 and 97
percent, respectively.
CD
53
120%
100%
3
!_
-------
Executive Summary
100% —
90% —
80% —
70% ¦ ¦ ¦
|||l||i -
111 ¦ 11 h 11
Current
Adv Diverse
Increased Adv
Current
Adv Diverse
Adv Diverse
Biomass & Mcxlern
Electricity Bioinass &
Biomass & Modern
Biomass & Modem
Electricity Fuels
Electricity
Electricity Fuels
Electricity Fuels
Baseline
Current Tech and Eff
Imp Tech and Eff
Baseline
Current Tech and Eff
Imp Tech and Eff
Global Climate Change Potential
Particulate Matter Formation Potential
¦ Coal Powder
¦ Coal Briquettes
¦ Coal Honeycomb
¦ Crop Residue
Firewood
¦Kerosene
¦ Biomass Pellets
¦ Electricity
¦ LPG
Natural Gas
Coal Gas
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
Figure ES-3. Select fuel mix LCIA results for GCCP and PMFP under varying technology
assumptions in China. Scenario results shown relative to baseline results.
(Axis abbreviations: Adv=Advanced, Tech=stove technology, Eff=stove efficiency,
Imp=improved)
ES-8
-------
Executive Summary
120%
= 100%
80%
o
¦- 60%
40%
t 20%
0%
V
Current
Baseline
Ghana Diverse
Transition Modern
Fuels
Current Tech and Eff
Ghana Diverse
Transition Modern
Fuels
Imp Tech and Eff
Global Climate Chanse Potential
Current
Baseline
Ghana Diverse
Transition Modem
Fuels
Current Tech and Eff
Ghana Diverse
Transition Modem
Fuels
Imp Tech and Eff
Particulate Matter Formation Potential
Firewood
¦ Charcoal from Wood
¦Biomass Pellets I Kerosene
¦ LPG
¦ Electricity
¦ Biogas from Cattle Dung ¦ Sugarcane Ethanol
Figure ES-4. Select fuel mix LCI A results for GCCP and PMFP under varying technology
assumptions in Kenya. Scenario results shown relative to baseline results.
(Axis abbreviations: Tech=stove technology, Eff=stove efficiency, Imp=improved)
gS 120%
110<"/- ¦ ¦ ¦
1 80% —¦ 1 H ¦
60% -M 1 1 1
lllll 1|,
8 0%
£ Current Moderated Diverse Moderated Diverse Current Moderated Diverse Moderated Diverse
Current
Moderated Diverse
Moderated Diverse
Current
Moderated Diverse
Moderated Diverse
Growth Modem
Growth Modem
Growth Modem
Growth Modem
Fuels
Fuels
Fuels
Fuels
Baseline
Current Tech and Eff
Imp Tech and Eff
Baseline
Current Tech and Eff
Imp Tech and Eff
Global Climate Change Potential
Particulate Matter Formation Potential
Firewood
¦ Crop Residue
¦ Charcoal from Wood
¦ Biomass Pellets
¦ Kerosene
¦ LPG
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Duns
Figure ES-5. Select fuel mix LCIA results for GCCP and PMFP under varying technology
assumptions in Ghana. Scenario results shown relative to baseline results.
(Axis abbreviations: Tech=stove technology, Eff=stove efficiency, lmp=improved)
ES-9
-------
Executive Summary
ES. 1.4.1 Findings Common to all Countries Studied
While the LCA results' values differ for all country and fuel combinations, certain trends
were found to be common across all four countries studied:
• Normalized results for all countries show that BC and PMFP impact categories are
strongly linked to the cooking sector. Results also show that impacts in these two
categories are sensitive to the projected future fuel mix and stove technology shifts
considered in this study, indicating multiple pathways by which to reduce impacts
attributable to the cooking sector.
• Utilization of modern cooking fuels such as liquefied petroleum gas (LPG), natural
gas, biomass pellets, and ethanol resulted in significant reductions in PMFP and BC,
categories strongly linked to the cooking sector.
• Normalized results confirm that traditional fuels pose a significant risk to human
health (e.g., due to PMFP). The possibility of using renewably sourced wood fuel in
combination with the adoption of improved or pelletized stoves could significantly
reduce hazardous emissions while still allowing the use of traditional biomass
resources.
• The sensitivity analysis shows that a significant range in potential environmental
impact exists between the worst and best performing cookstoves within a given stove
type (e.g. firewood traditional).
• Biogas and biomass pellets hold significant potential to reduce household air
emissions attributable to the cooking sector.
• Updated LCI information for the agricultural production of sugarcane indicates
significant upstream environmental impacts associated with ethanol production.
Findings unique to each country studied are highlighted in the subsequent sections.
ES. 1.4.2 India
• Normalized BC impacts in India are high relative to other nations and are
disproportionately influenced by the use of dung and crop residues in the current
cooking fuel mix.
• The current, coal heavy electricity mix in India and high electrical grid losses
contribute to the poor performance of electric cookstoves relative to other modern
fuel options.
• Realizing further GCCP impact reductions will be a challenge for India as the country
moves to adopt modern fossil-based cooking fuels. GCCP of the current Indian
cooking fuel and stove technology mix is at minimum 35 percent lower than that
realized by the other nations studied due to a continued reliance on biomass fuels,
relatively high baseline forest renewability, and an absence of significant
ES-10
-------
Executive Summary
contributions from stoves that exhibit particularly poor performance such as
traditional coal powder and charcoal cookstoves, which drive up GCCP impact in the
other countries.
ES. 1.4.3 China
• The one-third of cooking fuel energy that is produced from coal disproportionately
contributes to normalized PMFP and BC impacts in China. Potential reductions in
environmental impact realized by switching from coal powder to advanced forms of
coal consumption such as honeycomb briquettes or coal gas provide a robust option
for consistently improving performance of the cooking sector across all impact
categories.
• The current coal-heavy electricity mix in China results in poor performance of
electric cookstoves for most impact categories assessed relative to other modern fuel
options. Upgrades to China's electricity sector will be required for electric cookstoves
to achieve environmental impact scores in line with, or better than, other modern
fuels.
ES. 1.4.4 Kenya
• Scenario results show that reductions in Kenyan cooking sector emissions, compared
to other countries studied, are more sensitive to adoption of improved stove
technologies and thermal efficiencies, when holding cooking fuel mix constant. This
is because Kenya currently relies heavily on three-stone fires and traditional wood
stoves, which are associated with low thermal efficiencies and notable air emissions
during cookstove use.
• Forest renewability is important in determining if the best performing wood-based
options, biomass pellets and improved firewood stoves, can compete with the GCCP
of modern liquid and gas fuel options. This is especially true for Kenya, which has
the lowest forest renewability among the four study nations.
• Low availability of renewable wood resources in Kenya indicates that following
Ghana's lead in pursuing increased charcoal use as a means of improving urban air
quality could lead to significant pressure on other environmental impact categories
and forest resources. While charcoal may serve to reduce emissions in the household,
it does not reduce cumulative emissions across the supply-chain and serves as an
inefficient use of forest resources.
• The current electricity grid in Kenya has the lowest GCCP of all nations studied due
to the prevalence of hydropower and geothermal energy in their electrical grid mix.
The electricity grid sensitivity results show that all the future electrical grid mixes
yield further reductions in GCCP. However, Kenya currently has the lowest national
electrification rate, 23 percent (World Bank 2012). of any of the four countries
studied, which poses a challenge for use of electric cookstoves.
ES-11
-------
Executive Summary
ES. 1.4.5 Ghana
• Ghana demonstrates the second highest sensitivity to improvements in stove
technology and efficiency, following Kenya, indicating the potential to reduce
cooking sector emissions even in the absence of fuel mix shifts.
• Of the four study nations, Ghana is most heavily reliant on charcoal energy as a
source of cooking fuel (GLSS6 2014). Significant improvements in environmental
performance are possible through improved charcoal stove and kiln technology
adoption. However, even assuming the most optimistic adoption of charcoal
technology, this fuel demonstrates consistently poor environmental performance
relative to other cooking options and places a heavy burden on forest resources.
• Normalized CED of Ghana's cooking sector is significantly higher than that realized
for other nations, which is due largely to inefficient energy conversion in charcoal
kilns and the LPG refining process, as well as lower overall national per capita energy
consumption for all sectors compared to national per capita energy use in the other
study countries.
Results presented in this report are only a subset of the full results available in the
supporting files and were selected to highlight key trends while serving as a guide for
interpreting the accompanying result workbooks.
ES.1.5 Report Organization
The remainder of this report is organized as follows:
• Section 1: Goal and Scope Definition - Discusses the overall study goal, scope, and
boundaries, and describes the LCIA categories addressed in the study.
• Section 2: Cooking Fuel and Stove Descriptions and Methodology - Describes
details of the cooking fuel LCI models and documents cookstove efficiency, national
stove technology use, and sources of emissions information.
• Section 3: Methodology for Scenario Development and Sensitivity Analysis -
Describes cooking fuel and electrical grid mix scenario development and allocation
approaches used within the sensitivity analysis.
• Section 4: Methodology for Results Compilation -Provides documentation of LCA
methodology-related modeling choices, sensitivity analysis, and results presentation.
• Section 5: Updated LCA Results for India - Presents selected LCA results and
discussion for India.
• Section 6: Updated LCA Results for China - Presents select LCA results and
discussion for China.
• Section 7: LCA Results for Kenya - Presents select LCA results and discussion for
Kenya.
ES-12
-------
Executive Summary
• Section 8: LCA Results for Ghana - Presents select LCA results and discussion for
Ghana.
• Section 9: Key Takeaways by Country and Study Conclusions - Presents a brief
summary of key findings specific to each country and highlights trends observed both
across and between nations.
• Section 10: References - Lists references used in this LCA.
• Appendix A: Baseline Single Cooking Fuel Results by Life Cycle Stage - Presents
LCIA results by fuel type and life cycle stage for all countries.
• Appendix B: Comparison of Results Updates between Phase I and Phase II
Study for India and China - Presents a comparison of findings between the Phase I
and Phase II study.
• Appendix C: Data Quality - Reports on the quality of data utilized in the Phase II
study.
ES-13
-------
Section 1—Goal and Scope Definition
1. GOAL AND SCOPE DEFINITION
1.1 Introduction
Cookstove use in developing countries affects millions of lives on a daily basis with far-
reaching consequences. Reducing pollution from cookstoves is a key priority as emissions from
cookstoves and open fires with solid fuels are a major health concern and contribute to numerous
environmental impacts. The United States Environmental Protection Agency (U.S. EPA) is
conducting research to provide data and tools that inform decisions regarding clean cookstoves
and fuels for developing countries. Toward this end, EPA previously completed a life cycle
assessment (LCA) comparing the environmental footprint of current and potential fuels and fuel
mixes used for cooking within India and China (Cashman et al. 2016). This report builds on the
original India and China cookstove LCA (referred to throughout this report as the "Phase I
study"), by expanding the LCA methodology to include new cooking mix and electrical grid
scenarios, additional sensitivity analyses, uncertainty analyses, and a normalized presentation of
results. This phase of work also expands the geographic scope of the study to include both Kenya
and Ghana.
1.2 Phase II Goal
The goals of Phase II are to expand the geographic scope of the Phase I study, to facilitate
the comparison of the current fuel mix with potential future changes to the fuel mix, and to
examine the potential environmental effects of stove technology and efficiency upgrades from a
life cycle perspective. The study is conducted in accordance with International Standards
Organization (ISO 14040 and 14044, the international standards for conducting LCA studies
(ISO 2010a, 2010b). Additional goals of the Phase II study are to:
1. Determine the life cycle environmental burdens associated with commonly used
cooking fuels in Kenya and Ghana (GACC 2017a);
2. Expand the analysis of cooking fuel impacts in India and China to include updates
based on the stove technology mix, building on our earlier study;
3. Perform sensitivity analyses on allocation methodology, calculation of
greenhouse gas (GHG) impacts from non-renewable forestry practices, and stove
efficiency;
4. Assess new cooking fuel and electrical grid mix scenarios for consideration of
future shifts; and
5. Perform uncertainty analyses and calculate normalized results.
This study provides comparative data to inform policy decisions based on an LCA of
changes in cooking fuels and stoves on the local and global scale. Environmental issues
surrounding cooking fuels are identified, along with opportunities to address these issues based
on the choice of cooking fuel and stove technology. By providing results according to life cycle
stage, the study gives more specific insight regarding interventions capable of reducing energy
use, water consumption, or impacts associated with environmental emissions (e.g., emissions
released to air, water, and land). While the study does assess a wide array of environmental
indicators, it is not intended to be referenced in isolation, and readers will benefit from
1-1
-------
Section 1—Goal and Scope Definition
considering this study in concert with other research, especially regarding social and economic
impacts and potential strategies for implementing sustainable cookstove projects.
Audiences that benefit from information developed through this research include local
and national governments in China, India, Kenya, and Ghana, donors and investors (e.g.,
strategic planners), and researchers (e.g., sustainability scientists).
1.3 Scope of the Study
This section discusses the scope of the study required to accomplish the goals presented
above. The LCA components covered include the functional unit, fuel systems, study boundaries,
scenario development, impact assessment methods and data quality requirements.
1.3.1 Functional Unit
Results are expressed in terms of a common functional unit, which is defined in relation
to the shared functionality of the products under study. This common unit allows fair
comparisons to be made between the studied options. As this analysis is a comparison of
different fuels used to provide energy for cooking, a functional unit of 1 gigajoule (GJ) of useful
energy delivered to the pot for cooking is used as the basis of comparison. Useful energy refers
to energy that goes into work and is not lost (e.g., through transmission, distribution, or heat
losses at the cookstove).
1.3.2 Geographic and Temporal Scope
Current and projected future cooking fuel and stove technology use within India, China,
Kenya, and Ghana constitutes the geographic and temporal scope of this analysis. India and
China, covered in the Phase I study, were initially selected because they are both Global Alliance
for Clean Cookstoves (referred to throughout this report as "the Alliance") focus countries.
Kenya and Ghana were added in this phase of work to expand the study scope to include African
countries. Kenya and Ghana are also Alliance focus countries. Focus countries are those for
which the Alliance mobilized resources to grow the global market for clean cookstoves between
2012 and 2014. The Alliance selected focus countries as top priorities for clean cookstoves based
on the size of the impacted population, the maturity of the market in each country, the magnitude
of need, and the strength of the partner.
While the selection of the geographic scope for this project aligns with regions of focus
established by the Alliance, this project does not explicitly define or categorize stoves as clean
cookstoves, and instead concentrates specifically on providing quantitative, comparative LCA
results that can be used by interested stakeholders to achieve diverse aims in the cooking sector.
In both India and China, approximately half of each country's population currently uses
traditional cooking fuels, and over a million annual premature deaths are attributed to hazardous
air pollutants (HAPs) released from combustion of these fuels. Consumption of traditional
cooking fuels, in combination with rapid rates of urbanization and industrialization, has
contributed to the countries' resource depletion, deforestation, desertification, and biodiversity
loss. According to the United Nations Convention to Combat Desertification, nearly 40 percent
1-2
-------
Section 1—Goal and Scope Definition
of the Asian continent is arid, semi-arid, and dry sub-humid land, with 27 percent of China's
land being desertified. Deserts are expanding in both China and India (UNCCD 2015).
Kenya, located in eastern Africa, is the seventh largest country by population in Africa
(World Bank 2014b). Over 80 percent of the population in Kenya rely on some form of solid
biomass as their cooking fuel. Firewood use is particularly dominant in rural and peri-urban
areas and among those with low incomes (GVEP International 2012a, SID 2015), while charcoal
and kerosene are more commonly used in Kenyan urban areas. Ghana is Africa's 14th most
populous country, with the population evenly divided between urban and rural areas (World
Bank 2014b, ADP 2012). Over 40 percent of Ghana's population relies on unprocessed
firewood, with over 30 percent using charcoal (GLSS6 2014). In Ghana, there are over 21
million people affected by HAP emissions from cookstoves, with over 13 thousand deaths
attributed to these emissions per year (GACC 2017b). Similarly, in Kenya, there are over 36
million people affected by HAP emissions from cookstoves, with over 15 thousand deaths
attributed to these emissions per year (GACC 2017c). Overuse of wood-based fuels in both
Ghana and Kenya have also accelerated deforestation in these countries (Energy Commission
2010, Dalberg 2013a).
1.3.3 Transparency
The methods, standards, tools, and data used in this study are clearly documented in the
report body, appendices, and supplementary information (SI). Detailed supplementary files,
documenting the development of custom life cycle inventory (LCI) information for Phase II, are
provided as an accompaniment to the main report. Table 1-1 lists the associated SI files,
abbreviated in-text references, and a short description of the information contained in each.
Accompanying results files document the LCA model output by impact category for each
study scenario. The dynamic results templates allow stakeholders to explore the full breadth of
results that were calculated for Phase II, including customizing input parameters to assess
impacts on results for the following:
• Cooking fuel mixes,
• Stove technology use,
• Stove thermal efficiency,
• Electricity grid, and
• Forest renewability factor.
Using the supplementary files in combination with this study report will allow interested
parties to explore and recreate the LCA results.
1-3
-------
Section 1—Goal and Scope Definition
Table 1-1. Abbreviated In-text References and Descriptions of Supplementary
Information
File Name
In Text
Reference
File Description
EPA Phase II Cookstove LCA Results
-IN
Result Files
One result file is available for each study
country (IN = India, CN = China, KE = Kenya,
GH = Ghana). The fdes are dynamic and allow
the user to explore various results presentations,
allowing specification of custom fuel mixes and
parameters considered within the sensitivity
analysis. Raw LCA results as exported from
openLCA are included in these files.
EPA Phase II Cookstove LCA Results
-CN
EPA Phase II Cookstove LCA Results
-KE
EPA Phase II Cookstove LCA Results
-GH
SI1. Stove Use and Emissions
Supplementary Information
SI1
Documents stove emissions information, stove
thermal efficiency values, and estimates of stove
technology use for each country studied.
SI2. Stove LCI Supplementary
Information
SI2
Documents specific stove records and
associated emissions values for each stove
grouping, as described in Section 2.2.
Documentation of standard deviation used in the
Monte Carlo analysis for each pollutant is also
available in the file.
SI3. Fuel Mix Scenario Supplementary
Information
SI3
Documents primary literature sources and
calculations used to estimate current and
potential future fuel mix scenarios for each
study country.
SI4. Cookstove Electricity Scenario
Supplementary Information
SI4
Documents primary literature sources and
calculations used to specify current and
potential future electricity grid mix scenarios for
each country studied.
SI5. Crop Residue Supplementary
Information
SI5
Documents primary literature sources and
calculations used to develop crop production
LCI information and allocation factors used in
the sensitivity analysis.
SI6. Charcoal Kiln Supplementary
Information
SI6
Documents primary literature sources and
calculations used to develop the average and
improved performance charcoal kiln unit
processes for Kenya and Ghana.
SI7. Biogas Modeling Supplementary
Information
SI7
Documents primary literature sources and
calculations used to develop the bioslurry land
application LCI and biogas allocation factors.
1.3.4 Cooking Fuel Systems
This LC A considers the main cooking fuels currently used in India, China, Kenya, and
Ghana, as well as several emerging cooking fuel options such as sugarcane ethanol, biomass
1-4
-------
Section 1—Goal and Scope Definition
pellets, and electricity. Table 1-2 provides a list of the cooking fuels considered in this study,
accompanied by a brief description of each fuel. This table also lists the countries for which each
fuel is considered.
Table 1-3 lists the current cooking fuel mix in each of the four countries. This study also
considers possible cooking fuel mix shifts for the countries covered, as discussed in Section 3.
Detailed profiles of individual fuels, fuel heating values, and stove thermal efficiencies by
cooking fuel type are presented in Section 2.1 and Section 2.2.1, while the fuel mix scenarios are
described in Section 3.1.
Throughout this report, and the literature in general, the terms traditional, improved, and
modern are used as categorical descriptions for both cookstoves and cooking fuels. The
International Energy Agency (IEA) uses the terms traditional, intermediate, and modern to
describe fuel groupings. IEA notes that the use of these terms is not meant to imply a ranking
and refers instead to how well established a fuel is within a given nation (IEA 2006). This report
does not utilize the term intermediate and instead refers to all fuels as either traditional or
modern, with traditional fuels having a longer history of use. In this study, coal, charcoal,
firewood, crop residues, dung, and kerosene are all considered to be traditional fuels. Liquefied
petroleum gas (LPG), natural gas, coal gas, sugarcane ethanol, biogas, and electricity are referred
to as modern fuels.
The terms traditional and modern, as well as the additional term improved, can also refer
to stove technologies. This usage follows directly from the use of these terms in reference to the
fuels themselves. As noted, traditional fuels have a longer history of use, and over time the
original cookstove technologies used to combust these fuels have seen enhancements to increase
thermal efficiency, cooking quality, and to decrease a user's exposure to irritating or harmful
emissions. Stoves burning traditional fuels that incorporate features designed to accomplish these
goals are referred to as improved. Examples of such features include insulated combustion
chambers, chimney flues, and pot skirts. All stoves that burn modern fuels are considered to be
modern.
While the terminology used in this study is generally associated with improved
performance along the progression from traditional to improved fuels and cookstoves, this
terminology does not imply a strict quantitative improvement in any single metric of stove
performance. Detailed results of this and other studies should be consulted prior to making
statements or assumptions regarding the relative environmental performance of individual fuels
and cookstoves.
Table 1-2. Cooking Fuel List and Description
Fuel Type
Fuel
Description
Countries
A solid fossil fuel used widely for heating and
India
Coal
cooking, especially in China (GACC 2015).
Coal powder is the most popular form of coal
China
Traditional
used today.
Dung
Dried animal waste, usually from cows, is used
as an inexpensive fuel in rural areas.
India
India
1-5
-------
Section 1—Goal and Scope Definition
Table 1-2. Cooking Fuel List and Description
Fuel Type
Fuel
Description
Countries
Crop
Residue
Unprocessed biomass harvested as a by-product
of food production in agricultural regions. Crop
residues can include straws, stems, stalks,
leaves, husks, shells, peels, etc.
China
Ghana
Firewood
An unprocessed, solid wood fuel that is one of
the largest energy sources in all four study
nations. Much of the firewood used is manually
gathered from local forests.
India
China
Kenya
Ghana
Charcoal
A product made of carbonized firewood.
Carbonization is the process of burning
firewood in a low oxygen environment such as
a traditional earthen mound kiln to increase the
fuel energy density and decrease transport
weight.
India
Kenya
Ghana
Kerosene
Also referred to as paraffin, a liquid fossil fuel
product often derived from crude oil and used
for heating, lighting, and cooking.
India
China
Kenya
Ghana
Modern
LPG
A gas, which is a co-product of the production
of natural gas and/or crude oil (GACC 2015).
LPG is most widely used by urban residents and
is experiencing expanded use in all four
countries studied.
India
China
Kenya
Ghana
Natural
Gas
A fossil fuel-derived gas-that is piped to
customers via a centralized distribution
pipeline. Natural gas use is limited to urban
areas and is not yet a prevalent cooking fuel in
the countries studied.
China
India
Coal Gas
A gaseous fuel that is a product of the coal
gasification process.
China
Electricity
Electrical energy in each country is assumed to
be produced via the centralized electrical grid
according to the national average fuel mix for
the year 2013.
India
China
Kenya
Ghana
Ethanol
A liquid fuel is produced through the distillation
of various agricultural products or wood.
Sugarcane is the considered feedstock in this
study due to its prevalent production in India,
Kenya, and China.
India
China
Kenya
Ghana
Biogas
A methane-rich gas produced through the
anaerobic digestion of organic wastes. Biogas
can be generated from animal, human-and
kitchen wastes, as well as some crop residues.
India
China
Kenya
Ghana
Biomass
Pellets
Highly densified biomass material. Assumed to
be derived from wood in this study.
India
China
Kenya
Ghana
1-6
-------
Section 1—Goal and Scope Definition
Table 1-3. Fuels Used for Cooking in Countries Studied
Fuels Used for Cooking
India1'2'3
China4'5
Kenya6
Ghana7
Fuel5
Fuel Type
(%)
(%)
(%)
(%)
Coal
Traditional
1.15
28.9
-
-
Dung
10.6
-
-
-
Crop Residue
8.90
12.0
-
0.412
Firewood
49.0
14.7
64.9
45.9
Charcoal
1.15
-
17.0
31.5
Kerosene
3.20
0.300
11.6
0.157
LPG
Modern
25.2
31.1
5.02
21.7
Natural Gas
-
2.40
-
-
Electricity
0.400
10.6
0.803
0.315
Biogas
0.400
-
0.703
-
| Total
100
100
100
100
Sources and Notes: 1 Dalberg 2013b,2 Gov. of India 2014,3 Venkataraman et al. 2010, 4NBSC 2008,5
Dalberg 2014,6KNBS 2012,7 GLSS6 2014
5 Coal gas, biomass pellets and ethanol are not present in the current cooking fuel mix of any country.
1.3.5 System Boundary
The following life cycle stages are included for each cooking fuel system:
• Production of the cooking fuel feedstock, including all stages from extraction or
acquisition of the feedstock material from nature through production into a form
ready for processing into cooking fuel (e.g., cultivation and harvesting of sugarcane,
extracting crude oil from wells).
• Processing of the fuel into a form ready to be used in a cookstove.
• Distribution of fuels from the processing location to a retail location or directly to
the consumer. Distribution also includes bottling for fuels stored in cylinders (e.g.,
LPG).
• Use of the fuel via combustion of the fuel or use of electricity in a cookstove,
including disposal of any combustion wastes or residues (e.g., ash).
Figure 1-1 provides the general study system boundaries for all countries covered. Fuel
production and processing consists of all necessary steps, beginning at resource extraction, which
are required to make the fuel ready for use in a cookstove. For ethanol, this includes impacts for
growth and harvesting of the sugarcane. For crop residues in the baseline results, burdens begin
at collection of the biomass from the field, with all cultivation burdens assigned to the primary
food crop. The effects of allocating a share of crop production impacts to crop residue are
examined as part of the sensitivity analysis. Impacts of firewood harvesting are not included in
the study, however carbon dioxide (CO2) emissions do contribute to global climate change
potential (GCCP) for the fraction of forest products produced using non-renewable practices. In
1-7
-------
Section 1—Goal and Scope Definition
the case of electricity, power generation, as well as transmission and distribution losses, are
included within the system boundaries. Additionally, this study accounts for transportation
requirements between all life cycle stages within the boundaries of this study. Specific
processing steps included in the analysis are described in greater detail for individual fuels in
Section 2.1.
Cookstove production and distribution, human energy expended during collection of
fuels, and the production, preparation, consumption, and disposal of food and food wastes are
outside the boundaries of this project. The rationale for excluding these stages is discussed in the
next section (Section 1.3.5.1).
The use phase includes the combustion of the cooking fuels and associated stove
emissions. The types and quantities of air emissions associated with fuel use depend on the fuel's
elemental composition (e.g., average fixed carbon, ash content, and volatile matter) and the
cookstove technology or technology mix (e.g., thermal efficiency) for each country, which
affects the quantity of fuel that must be consumed to deliver 1 GJ of cooking energy. At fuel
end-of-life (EOL), solid residues from the combustion of cookstove fuels (bottom ash and carbon
char) are disposed. The major components of these wastes are determined by the type of fuel
combusted. For example, biomass fuel combustion typically results in ash containing silica,
alumina, calcium oxides, sodium, magnesium, and potassium. These wastes are assumed to be
disposed of by land application, whereby the wastes in question are spread out over a landscape,
often as an agricultural amendment, to be assimilated by the environment.
1-8
-------
Section 1—Goal and Scope Definition
( Food Production,
| Preparation, and
I Consumption I
Elementary Inputs from Nature
•Water
•Raw Materials
Intermediate Inputs
•Treated Water
•Energy
•Economic Goods
| Cookstove
, Production
Electrical Grid
Fuels**
Coal
Oil
Hydroelectric
Natural Gas
Nuclear
Biomass
Wind & Solar
Waste
Geothermal
Cookstove Fuels**
Unprocessed Solid Biomass Mix
1. Firewood
2. Crop Residues
3. Dung
Processed Solid Biomass
1. Biomass Pellets
2. Charcoal
rye
luid and Gas
1.
LPG
2.
Kerosene
3.
Natural Gas
4.
Coal Gas
5.
Biogas
^6.
Ethanol
1.
Coal Powder
2.
Coal Briquettes
3.
Honeycomb Coal
Briquettes
L
A
I
Distribution
Distribution
Fuel Production
Transport*
| Transport*
_
1 y
\
Generation of
Electricity
Use of
Electricity/Fuel
in Cookstoves
« ; Within Study Boundary
— — Outside Study Boundary
Elementary Outputs to Nature
•Water
•Airborne Emissions
•Waterborne Emissions
Intermediate Outputs
•Waste water to be Treated
•Economic Goods
•Solid Waste to be Managed
1 Food Wastes ¦
*Human energy expenditures are not included.
** Not all fuels are utilized in every study nation, please refer to Sections 1.3.4 for current fuel and electricity use and Sections 3.1 and 3.2 for potential future
cooking fuel and electricity grid fuel use in each country.
Figure 1-1. Study system boundaries of the baseline scenario for covered countries.
1-9
-------
Section 1—Goal and Scope Definition
1.3.5.1 System Components Excluded
The following components of each system are not included in this study.
Cookstove Production and Distribution. All burdens associated with production and
distribution of the cookstoves themselves are excluded from the analysis, as the focus of the
study is production and use of cooking fuels. A previous LCA examining production of fuel-
efficient cookstoves found that the use phase significantly dominates life cycle GHG emissions
regardless of cooking fuel type (Wilson 2016). Life cycle impacts of the stove relative to fuel
production and use are assumed to be negligible.
Human Energy Expended During the Collection or Use of Fuels. This analysis does
not include human biological energy or emissions. Shifts in the mix of fuels may decrease the
overall human energy and emissions expended during the distribution phase in some cases (e.g.,
shifting to fuels with higher energy density that are easier to transport or that do not require
consumer transport such as electricity). While outside the scope of this study, there are important
benefits to reducing the time and effort associated with the collection of solid biofuels,
particularly as they relate to women's health and safety. The benefits and burdens of such
changes would be better captured by qualitative or analytical methods apart from LCA.
Food and Food Wastes. The focus of this study is the provision of cooking energy and is
intended to be independent of the food itself. All burdens associated with production,
preparation, storage, consumption, and disposal of the food being prepared are excluded from the
analysis.
Capital Equipment and Infrastructure. Energy and wastes associated with the
manufacture of capital equipment and infrastructure are excluded from this analysis, including
equipment to manufacture buildings, motor vehicles, and industrial machinery, as well as roads
and electricity distribution infrastructure used to distribute fuels throughout the supply chain and
to end users. In general, these types of capital equipment and infrastructure are used to produce
and deliver large quantities of product output over a useful life of many years. Thus, energy and
emissions associated with the production of these facilities and equipment generally become
negligible when allocated over the total amount of output or service over their useful lives
(Berglund 2006).
Stove Stacking. The transition from one cooking system to another does not always
occur instantaneously. In communities that are undergoing transitions to a new cooking fuel
type, field observations indicate that very often individual homes will initially use a mixture of
new and traditional cooking systems. This phenomenon, known as 'stove-stacking,' allows
households to take advantage of the differences that exist between the stove and cooking fuel
combinations that they employ. While this would ultimately affect the pace of change and the
attendant shift in environmental impacts, it represents a dynamic force operating at a household
level (Hiemstra-van der Horst and Hovorka 2008) that lies outside the study scope. This study
focuses on scenarios encompassing the national cooking fuel mix, which could include
households using a mixture of fuels, although this was not explicitly considered when developing
the cooking fuel scenarios.
1-10
-------
Section 1—Goal and Scope Definition
Effect of Stove Operational Practice. Practical and environmental performance of a
cookstove is determined not only by the selection of fuel and stove technology, but also based on
stove operational conditions. The result of many cookstove emissions tests are associated with
operation of a cooking stove under ideal or laboratory conditions. A small number of studies
look at field conditions. The availability of field study results is not extensive enough to provide
a comprehensive source of data if considered on its own. Given this, the results of both field and
laboratory studies have been combined when compiling LCI information for each stove
grouping, and thereby contribute to the variation represented in the uncertainty results. As such,
it beyond the scope of this project to quantify the effect of variation in operational practice on
environmental performance, which should be considered when interpreting results.
Human Exposure to Emissions. This study does not include the detailed analysis of
human exposure to cookstove emissions that would be required to accurately estimate the human
health impacts of household cooking. For example, if cookstoves burning solid fuels are more
commonly used outdoors than liquid/gas alternatives, then there may not be a one-to-one
relationship between cookstove emissions and human exposure between these alternatives.
Similarly, upstream emissions at the point of manufacture can be expected to have a lower
exposure factor than emissions from the cookstove themselves, thereby affecting the potential for
human health impact. Such differences that affect human exposure to cooking emissions are
outside of the scope of this study, and must be considered separately.
1.3.6 Data Sources Summary
The majority of stove emission data was extracted from existing studies in publicly
available academic literature. Much of this research has been supported by US EPA Office of
Research and Development's small-scale combustion evaluation program in collaboration with
the Global Alliance, the World Bank, and other research efforts to support reducing health and
environmental pollution associated with cookstove use. The SI contains detailed LCI data for
the life cycle stages modeled for each cooking fuel system. Baseline LCI data for China and
India from the Phase I study can be found in Appendix A of that document (Cashman et al.
2016). Data were assembled according to the procedures established in the project Quality
Assurance Project Plan (QAPP) "Quality Assurance Project Plan for Comparative Life Cycle
Assessment of Cooking Fuel Options in China andlndia", approved August 25, 2014. Data
quality and data requirements are covered in more detail in Section 1.4.
1.3.7 Life Cycle Lmpact Assessment Methodology and Lmpact Categories
Life Cycle Impact Assessment (LCIA) helps with interpretation by consolidating a
lengthy LCI into a smaller number of relevant indicators. LCIA is defined in ISO 14044 Section
3.4 as the "phase of life cycle assessment aimed at understanding and evaluating the magnitude
and significance of the potential environmental impacts for a product system throughout the life
cycle of the product (ISO 2010b)." In the LCIA phase, the inventory of emissions is first
classified into categories in which the emissions may contribute to impacts on human health or
the environment. Within each impact category, the emissions are then normalized to a common
reporting basis, using characterization factors that express the impact of each substance relative
to a reference substance.
1-11
-------
Section 1—Goal and Scope Definition
Characterization factors are defined to quantify the impact potential of LCI results. There
are two main methods to develop LCIA characterization factors. The 'midpoint' method links
LCI results to categories of commonly defined environmental concerns like eutrophication
potential and global climate change potential. The 'endpoint' method further models the
causality chain of environmental stressors to link LCI results to environmental damages (e.g.,
final impacts to human and ecosystem health). ISO standards allow the use of either method in
the LCIA characterization step. Overall, indicators closer to the inventory result (midpoint
indicators) have a higher level of scientific consensus, as less of the environmental mechanism is
modeled. Conversely, endpoint and damage-oriented characterization models inevitably include
more aggregation, or more assumptions (e.g., about fate and transport, exposures/ingestion, etc.).
To reduce uncertainty in communication of results, this LCA focuses on indicators at the
midpoint level.
1.3.7.1 Scope of Impact Assessment
This study addresses global, regional, and local impact categories of relevance to the
cookstove sector such as air emissions leading to human health issues, energy demand driving
depletion of bio-based and fossil-fuel resources, and GHG and black carbon (BC) and short-lived
climate pollutant emissions causing both short-term and long-term climate effects. For most of
the impact categories examined, the ReCiPe impact assessment method is utilized to represent
global conditions (Goedkoop et al. 2008). Characterization factors, which are developed based
on established impact pathways, form the basis of impact assessment methods such as ReCiPe.
An impact pathway is a series of quantifiable relationships that can be used to link LCI emissions
to units of environmental impact (e.g. kg C02-eq for GCCP). Characterization factors in ReCiPe
were originally developed for global or European conditions and are not specific to any of the
study countries. The characterization factors used in this study are associated with ReCiPe's
hierarchist cultural perspective, which makes characterization assumptions based on what
ReCiPe's authors consider to be standard policy perspectives and time horizons (i.e., consensus
model) for the included impact categories (Goedkoop et al. 2008). Currently, no established
LCIA method exists specifically for the scope of India, China, Kenya, or Ghana.
For the category of GCCP, a global impact, contributing elementary flows are
characterized using factors reported by the Intergovernmental Panel on Climate Change (IPCC)
in 2013 with a 100-year time horizon (IPCC 2013). Considerations for biogenic carbon
accounting are covered in Section 4.1 and Section 4.2. BC and co-emitted species are
characterized to BC - equivalents (eq) based on a novel method released by the Gold Standard
Foundation (GSF) (GSF 2015). A detailed discussion of the BC methodology is presented in
Section 4.3. Cumulative energy demand (CED) and water depletion are also included as
inventory indicators. Energy and water inventory indicators are not associated directly with
environmental impacts but rather provide a cumulative count of energy and water resources used
within the study system. CED includes both renewable and non-renewable energy sources.
A summary of the LCI and LCIA categories and methods used in this study is presented
in Table 1-4. While this study focuses on environmental impacts and does not include impact
categories that focus exclusively on human health, several included impact categories are closely
associated with both environmental and human health impacts. These emission types include
emissions leading to BC, particulate matter formation potential (PMFP), and photochemical
1-12
-------
Section 1—Goal and Scope Definition
oxidant formation potential (POFP), all of which can lead to eye irritation, respiratory disease,
increased risk of infection, and cancer. Linking these emissions definitively to human health
impacts would introduce a higher level of uncertainty to the study results. Human health impacts
are dependent not only on emission quantities but also on the fate and transport of the emitted
substances and the concentrations and pathways by which organisms are exposed to these
substances. These detailed types of exposure information are not tracked in an LCI, requiring
additional assumptions about the environmental mechanism to be made by the developer of the
LCIA methodology. While human health impacts are not explicitly estimated by this study,
pertinent impact categories related to known human health impacts of cookstove use are included
in the analysis.
Table 1-4. Environmental Impact Category Descriptions and Units
Impact/Inventory
Category
Description
Unit
Global Climate
Change Potential
The GCCP impact category represents the heat trapping capacity of
GHGs over a 100-year time horizon. All GHGs are characterized as
kg CO2 equivalents according to the IPCC 2013 5th Assessment
Report global warming potentials.
kg CO2
eq
Cumulative
Energy Demand
The CED indicator accounts for the total usage of non-renewable
fuels (natural gas, petroleum, coal, and nuclear) and renewable fuels
(such as biomass and hydro). Energy is tracked based on the heating
value of the fuel utilized from point of extraction, with all energy
values summed together and reported on a megaioule (MJ) basis.
MI
Water Depletion
Potential (WDP)
WDP results, in alignment with the ReCiPe impact assessment
method, are based on the volume of fresh water inputs to the life
cycle of the assessed fuels. Water may be used in the product,
evaporated or returned to the same or different water body or to land.
If the water is returned to the same water body, it is assumed that the
water is returned at a degraded quality, which constitutes a
consumptive use. Water consumption includes evaporative losses
from establishment of hydroelectric dams.
m3
Black Carbon and
Short-Lived
Climate Pollutants
BC, formed by incomplete combustion of fossil and bio-based fuels,
is the carbon component of particulate matter (PM2.5) that most
strongly absorbs light and thus has potential short-term (e.g., 20-year)
radiative forcing effects (e.g., potential to contribute to climate
warming). Organic carbon (OC) is also a carbon component of PM
and possesses light-scattering properties typically resulting in climate
cooling effects. PM from the cookstove sector is typically released
with criteria pollutants such as carbon monoxide (CO), nitrogen
oxides (NOx), and sulfur oxides (SOx), which may result in additional
warming impacts or exert a cooling effect on climate. This indicator
characterizes all PM and co-emitted pollutants to BC equivalents
depending on the relative magnitude of short-term warming or
cooling impacts. The BC method is based on the novel GSF method
(GSF 2015).
kg BC eq
Particulate Matter
Formation
Potential
PMFP results in health impacts such as effects on breathing and
respiratory systems, damage to lung tissue, cancer, and premature
death. Primary pollutants (including PM2.5) and secondary pollutants
kg PM10
eq
1-13
-------
Section 1—Goal and Scope Definition
Table 1-4. Environmental Impact Category Descriptions and Units
Impact/Inventory
Category
Description
Unit
(e.g., SOx and NOx) leading to PM formation are characterized here
as kg PMio eq based on the ReCiPe impact assessment method.
Terrestrial
Acidification
Potential (TAP)
TAP quantifies the acidifying effect of substances on their
environment. Important emissions leading to terrestrial acidification
include sulfur dioxide (SO2), NOx, and ammonia (NH3). Results are
characterized as kg SO2 eq according to the ReCiPe impact
assessment method.
kg SO2
eq
Freshwater
Eutrophication
Potential (FEP)
Freshwater eutrophication assesses the potential impacts from
excessive loading of macro-nutrients to the environment and eventual
deposition in freshwater. Pollutants covered in this category are all
phosphorus (P)-based (e.g., phosphate, phosphoric acid, elemental P),
with results characterized as kg P eq based on the ReCiPe impact
assessment method.
kg P eq
Photochemical
Oxidant
Formation
Potential
The POFP (e.g., smog formation) results determine the formation of
reactive substances that cause harm to human health and vegetation.
Results are characterized here to kg of non-methane volatile organic
compounds (NMVOCs) eq according to the ReCiPe impact
assessment method. Some key emissions leading to POFP include
CO, methane (CH4), NOx, NMVOCs, and SOx.
kg
NMVOC
eq
Ozone Depletion
Potential (ODP)
Measures stratospheric ozone depletion. Important contributing
emissions include chlorofluorocarbon (CFC) compounds and halons.
It is likely that ozone depletion is of lower importance for cookstoves
fuels compared to other impact categories. There will be differences
between stove options as fossil fuels generate ozone depleting
emissions within their supply chain that are absent in the biomass
options. However, the ODP category has become less critical
following the regulation of the worst offending ozone-depleting
chemicals.
kg CFC-
11 eq
Fossil Depletion
Potential (FDP)
Fossil fuel depletion captures the consumption of fossil fuels,
primarily coal, natural gas, and crude oil. All fuels are normalized to
kg oil eq based on the heating value of the fossil fuel and according to
the ReCiPe impact assessment method.
kg oil eq
1.4 Quality Assurance
In accordance with the project's Quality Assurance Project Plan (QAPP) entitled Quality
Assurance Project Plan for Comparative Life Cycle Assessment of Cooking Fuel Options in
China and India approved by EPA on August 25, 2014, ERG collected or adapted existing data
to develop: (1) cooking fuel production LCIs, (2) cookstove use LCIs, (3) national cooking fuel
mix scenarios, (4) stove technology use estimates, (5) electrical grid mix scenarios, and (6) forest
renewability factors that encompass the main data requirements for this study (ERG 2014). The
collected data sources include peer-reviewed literature, government and NGO reports, and
national survey information. ERG evaluated the collected information for completeness,
accuracy, and reasonableness. In addition, ERG considered publication date and
accuracy/reliability when reviewing data quality. Finally, ERG performed conceptual,
1-14
-------
Section 1—Goal and Scope Definition
developmental, and final product internal technical reviews of the openLCA model and
supplementary files used to document the compilation and development of information listed
above. The remainder of this section first outlines the study data quality evaluation, followed by
a discussion of quality assurance procedures implemented.
1.4.1 Data Quality Evaluation
ISO standards 14040 and 14044 detail various aspects of data quality and data quality
analysis. These ISO Standards state: "descriptions of data quality are important to understand the
reliability of the study results and properly interpret the outcome of the study (ISO 2010a,
2010b)." These ISO Standards list three critical data quality criteria: time-related coverage,
geographical coverage, and technology coverage. The following subsections discuss these three
critical data quality criteria and the typical specifications associated with high quality data.
Additional data quality criteria evaluated include data source reliability and completeness.
The geographic scope of the study encompasses cooking fuel use in India, China, Kenya,
and Ghana. However, some cooking fuels or upstream inputs to fuel production/processing are
imported from other regions of the world. High quality data and information for geography-
dependent processes (e.g., energy production) were obtained from country-specific articles and
databases. Data for technology-based processes are based on the most recent average country-
specific technology mix (e.g., the current production methods China employs for mining and
processing coal). It is more difficult to evaluate data quality for future technologies not yet in use
or that currently have a small market share. When more specific information was not available,
data quality for future technological processes was based on current technological processes used
in the same country. For example, for a scenario with increased use of natural gas to produce
electricity in China, the future natural gas production is modeled assuming China will produce
natural gas in the future using the same methods it currently employs.
'High quality temporal data' typically refers to data that are less than six years from the
reference period. The wide scope of this project and the nature of national data for these four
countries has necessitated the establishment of a reference period, rather than a single reference
year. The reference period for country-specific information such as cooking fuel mix, stove
technology use, and electrical grid mix is believed to be most representative of the period
between 2008-2013. The use of data representative of a date prior to the reference period has
been required where more recent information is unavailable. Projected scenarios such as the
future electrical and cooking fuel mix scenarios, are generally associated with a 20- to 30-year
time horizon. This period should be used only for general guidance in interpretation. In cases
where the supporting literature provides a specific time period estimate, that estimate has been
provided. Standardization of temporal scope has been sought to ensure that differences in study
results are focused on material and process differences for the fuels (and associated stove
efficiencies) rather than as a result of disparities in data quality available between nations.
Table 1-5 presents the data quality criteria ERG used when evaluating the data collected.
Not all data quality criteria are applicable for every data source referenced in this project. Actual
data quality scores are documented in Appendix C. Data fields for which a data quality criterion
is not relevant are clearly noted, and the specific SI file used to document the data requirement is
listed to facilitate review of the underlying references and calculations. ERG documented
1-15
-------
Section 1—Goal and Scope Definition
qualitative descriptions of source reliability, completeness, temporal correlation, geographical
correlation, and technological correlation. Additional notes on how data quality criterion were
applied to the datatypes utilized in this project are included in Appendix C-l.
Table 1-5. Data Quality Rubric
Quality
Metric
Data Quality Criteria
Quality Estimate
Data verified based on measurements.
High
Data Source
Reliability
Data verified based on some assumptions and/or standard science and
engineering calculations.
Medium High
Data verified with many assumptions, or non-verified but from quality
source.
Medium
Qualified estimate.
Medium Low
Non-qualified estimate.
Low
Representative data from a sufficient sample of sites over an adequate
period, with records for all necessary inputs/outputs.
High
Data
Completeness
Smaller number of sites, but an adequate period.
Medium High
Sufficient number of sites, but a less adequate period.
Medium
Smaller number of sites and shorter periods or incomplete data from
an adequate number of sites or periods.
Medium Low
Representativeness unknown or incomplete data sets.
Low
Less than 3 years of difference to year of study/current year.
High
Temporal
Data Quality
Less than 6 years of difference.
Medium High
Less than 10 years of difference.
Medium
Less than 15 years of difference.
Medium Low
Age of data unknown or more than 15 years of difference.
Low
Data from area under study.
High
Average data from larger area or specific data from a close area.
Medium High
Geographical
Data from area with similar production conditions.
Medium
Data Quality
Data from area with slightly similar production conditions.
Medium Low
Data from unknown area or area with very different production
conditions.
Low
Data from technology, process, or materials being studied.
High
Data from a different technology using the same process and/or
materials.
Medium High
Technological
Data Quality
Data on related process or material using the same technology.
Medium
Data or related process or material using a different technology.
Medium Low
Data or poorly related process or material using a different
technology.
Low
1-16
-------
Section 1—Goal and Scope Definition
1.4.2 Internal QA Review Procedures
ERG developed SI files containing all the necessary information and data required to
execute the compilation of LCI information and the establishment of sensitivity scenarios. All SI
files were reviewed by a team member knowledgeable of the project, but who did not develop
the input files. The reviewer ensured the accuracy of the data transcribed into the input files, the
technical soundness of methods and approaches used and the accuracy of the calculations.
ERG input all LCI data developed into the openLCA software (GreenDelta 2016). A
team member knowledgeable of the project, but who did not develop the model, reviewed the
openLCA model to ensure the accuracy of the data transcribed into the software. The openLCA
model was also reviewed to ensure that each elementary flow (e.g., environmental emissions,
consumption of natural resources, and energy demand) was characterized under each impact
category for which a characterization factor was available. The draft final fuel system models
were reviewed prior to calculating results to make certain all connections to upstream processes
and weight factors were valid. LCIA results were then calculated by generating a contribution
analysis for the selected cookstove product system based on the defined functional unit of 1 GJ
of heat delivered for cooking. Similarly, after the LCIA results were generated and exported
from openLCA to results spreadsheets, the generated spreadsheet results were reviewed by a
team member who did not calculate the results. The ERG reviewer compared the spreadsheet
LCIA results against generated results in the final openLCA model. All Phase II results were also
compared to similar results from Phase I (if applicable) to ensure any changes were reasonable
and accounted for. Differences between Phase 1 and Phase II LCIA results are documented in
Appendix B of this report.
1-17
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
2. COOKING FUEL AND STOVE DESCRIPTIONS AND METHODOLOGY
This section describes the method for constructing the relevant cooking fuel LCIs and
identifies stove technologies used in the LCA model. The term LCI refers to the inventory of
relevant energy and material inputs and environmental releases associated with a cooking fuel
across all life cycle stages assessed.
2.1 Fuel System Model Descriptions
For India, cookstove fuel system modeling assumptions are based largely on work
conducted by Singh and colleagues (2014a/b) for all cookstove fuels except sugarcane ethanol
and biomass pellets. Sugarcane ethanol production in India is derived from a study by
Tsiropoulos and colleagues (2014), with fuel combustion emission values coming from
laboratory tests carried out at the Aprovecho Research Center (Berick 2006, MacCarty 2009).
For the Chinese cooking fuels, fuel modeling data are primarily from work by Zhang and
colleagues (2000). For both China and India, biomass pellet production is from work by
Jungbluth and colleagues (2007), while combustion of the pellets is modeled based on emission
and stove efficiency profiles from Jetter et al. (2012). Cooking fuel data for Kenya and Ghana
are largely drawn from the work of Afrane and Ntiamoah (2011).
Stove emission profiles drawn from the literature, which are used to develop use phase
LCI data for each stove-fuel combination, are summarized in Table 2-1. In addition to listing the
sources used, this table specifies the number of emission records that were aggregated to derive a
specific stove LCI unit process. Not all emission records provide a quantity for each pollutant of
interest in this study, please see SI2 for a more detailed presentation of the sources utilized for
each stove group and the pollutants covered by each. Each of these unit processes is defined by
the fuel type, stove type, and country or region. When possible, emission profiles that were
specific to a study country were used. For cases in which an emission profile was not available
for a specific cooking fuel and country combination, all stove emission profiles available were
aggregated into LCI unit processes to represent a global average. This global average was then
used as a proxy dataset for the country and fuel combinations lacking specific regional emission
profiles. These global average emissions profiles were, however, linked to upstream fuel
extraction, production and processing LCI specific to the country of interest.
Documentation of the processed cookstove fuel heating values is provided in the next
section, followed by a discussion on the supply chain for each fuel. Upstream processes such as
transport and ancillary material inputs are modeled using information from the National
Renewable Energy Laboratory's (NREL's) US Life Cycle Inventory (US LCI) Database and
Ecoinvent v2.2. The US LCI is a publicly available LCI source specific to US conditions (NREL
2012) and Ecoinvent v2.2 is a private Swiss LCI database with data for many global unit
processes (Ecoinvent Centre 2010). Where possible, these upstream databases are adapted to the
geographic scope of interest, i.e., by linking process electricity requirements to the country-
specific electricity grid mix.
2-1
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-1. Emission Profile Sources for Each Stove Type by Study Country
Fuel
Stove Type
Country
Stove
Emission
Records
Sources
Coal
Traditional
India
1
1
Coal, Powder
Traditional
China
6
3,12
Improved
China
3
3,12
Coal, Briquette
Improved
China
4
3,12
Coal,
Traditional
China
8
3,12,15,16
Honeycomb
Improved
China
13
3,12,17
Dung
Traditional
India
11
1,2,3,6
Improved
India
4
3,6
Crop Residue
Traditional
India
14
1,2,3,6
China
12
3,11,15
Global
26
1,2,3,6,11,15
Improved
India
8
3,6
China
26
3,12,13,14,18,19
Firewood
Three-stone
India
4
3,6
Global
4
7,8
Traditional
India
14
1,2,3,6
China
21
4,3,10,11
Global
16
4,5
Improved
India
16
3,4,5,6
China
35
3,12,13,14
Global
28
4,7
Charcoal
Traditional
Global
5
4,8,9
Improved
India
2
1,3
Improved
Kenya
4
7
Improved
Ghana
2
7
Kerosene
Improved,
Pressure
India
2
3,6
Improved, Wick
India
2
3,6
Improved
China
4
3,12
Global
8
3,6,12
LPG
Modern
India
2
3,6
China
4
3,12
2-2
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-1. Emission Profile Sources for Each Stove Type by Study Country
Fuel
Stove Type
Country
Stove
Fmission
Records
Sources
Global
6
1,3,6
Natural Gas
Modern
India, China
3
3,12
Coal Gas
Modern
China
2
3,12
Electric
Modern
All Countries
-
-
Sugarcane
Ethanol
Modern
All Countries
4
21
Biogas
Modern
India, China
3
1,6,3
Kenya,
Ghana
1
20
Biomass Pellets
Modern
All Countries
4
7, 22
Sources: 1 Singh et al. 2014a/b,2 Saud et al. 2012,3 Zhang et al. 1999, 4 Bhattacharya et al. 2002b,5 Bhattacharya
et al. 2002a,0 Smith et al. 2000,7 .letter et al. 2012,8 Sweeney 2015,9 Booker 2012,10 Shen et al. 2013, 11 Shen et
al. 2012b,12 Zhang et al. 2000, 13 Shen et al. 2012a,14 Wang et al. 2009,15 Shen et al. 2010b,10 Shen et al. 2010a,
17Zhi et al. 2008, 18 Cao et al. 2008, 19 Wei et al. 2014,20 Afrane and Ntiamoah 2011,21 MacCarty 2009,22 Carter
et al. 2014
2.1.1 Processed Fuel Heating Values
Table 2-2, Table 2-3, Table 2-4, and Table 2-5 list the lower and higher heating values
(LHV, HHV) for cooking fuels in India, China, Kenya, and Ghana, respectively. Lower heating
values in combination with stove thermal efficiency values are used to calculate the quantity of
fuel required per GJ of delivered cooking energy. HHVs are used to calculate CED results within
the fuel system unit processes. Associated cookstove thermal efficiencies for each country and
fuel combination are provided in Table 2-7.
Table 2-2. Heating Values of Cooking Fuels in India
Fuel Type
Lower Heating
Value (LHV)
MJ/kg
Higher Heating Value
(HHV)
MJ/kg
Source1
Firewood
14.0
15.8
Singh et al. 2014a
Crop Residue
12.8
14.6
Singh et al. 2014a
Dung Cake
11.9
13.3
Singh et al. 2014a
Charcoal Briquettes
from Wood
27.4
27.9
Singh et al. 2014a
Biomass Pellets
16.5
17.8
letter et al. 2012
Ethanol from
27.0
29.7
GREET 2008,
Sugarcane
MacCarty 2009
Biogas from Dung
18.2
19.9
Singh et al. 2014a,
calculated2
2-3
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-2. Heating Values of Cooking Fuels in India
Fuel Type
Lower Heating
Value (LHV)
MJ/kg
Higher Heating Value
(HHV)
MJ/kg
Source1
LPG
45.2
48.9
Singh et al. 2014a,
calculated
Natural Gas
51.3
56.8
Zhang et al. 2000,
calculated2
Kerosene
42.9
49.0
Singh et al. 2014a
Hard Coal
11.8
12.3
Singh et al. 2014a,
calculated2
1 Where two sources are listed, first refers to LHV. Second refers to HHV.
2 HHV is calculated based on ratio of HHV/LHV for similar fuel as documented in SI1. HHVx = LHVx*(HHVy/LHVy)
Table 2-3. Heating Values of Cooking Fuels in China
Lower Heating Value
Higher Heating
Source1
Fuel Type
(LHV)
Value (HHV)
MJ/kg
MJ/kg
Firewood
15.3
17.3
Zhang et al. 2000, calculated2
Crop residue
16.1
18.3
Zhang et al. 2000, calculated2
Biomass Pellets
16.5
17.8
Jetter et al. 2012
LPG
49.0
53.0
Zhang et al. 2000, calculated2
Kerosene
43.3
49.5
Zhang et al. 2000, calculated2
Natural Gas
51.3
56.8
Zhang et al. 2000, calculated2
Coal Gas
43.8
48.0
Zhang et al. 2000, calculated2
Honeycomb Coal
19.2
20.3
Zhang et al. 2000, calculated2
Coal, Powder
27.3
28.8
Zhang et al. 2000, calculated2
Coal, Briquette
13.9
14.6
Zhang et al. 2000, calculated2
Biogas from Dung
18.2
19.9
Singh et al. 2014a, calculated2
1 Where two sources are listed, first refers to LHV. Second refers to HHV.
2 HHV is calculated based on ratio of HHV/LHV for similar fuel as documented in SI1. HHVx = LHVx*(HHVy/LHVy)
Table 2-4. Heating Values of Cooking Fuels in Kenya
Cooking Fuel Type
Lower Heating
Value (LHV)
MJ/kg
Higher Heating
Value (HHV)
MJ/kg
Source1
Firewood
14.0
15.8
Singh et al. 2014a
Charcoal, Average Kiln
29.6
30.4
calculated2, Pennise et al. 2001
Charcoal, High-Performing
Kiln
30.2
31.0
calculated2, Pennise et al. 2001
Kerosene
42.9
49.0
Singh et al. 2014a
LPG
45.8
49.6
Afirane and Ntiamoah 2011,
calculated2
Ethanol
27.0
29.7
GREET 2008, MacCarty 2009
2-4
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-4. Heating Values of Cooking Fuels in Kenya
Cooking Fuel Type
Lower Heating
Value (LHV)
MJ/kg
Higher Heating
Value (HHV)
MJ/kg
Source1
Biogas from Dung
17.7
19.4
Afrane and Ntiamoah 2011,
calculated2
Wood Pellets
16.5
17.8
Jetter et al. 2012
1 Where two sources are listed, first refers to LHV. Second refers to HHV.
2 LHV is calculated based on ratio of LHV/HHV for similar fuel as documented in SI1. LHVx = HHVx*(LHVy/HHVy)
Table 2-5. Heating Values of Cooking Fuels in Ghana
Cooking Fuel Type
Lower Heating
Value (LHV)
MJ/kg
Higher Heating
Value (HHV)
MJ/kg
Source1
Crop Residue
12.8
14.6
Singh et al. 2014a
Firewood
14.0
15.8
Singh et al. 2014a
Charcoal, Average Kiln
25.7
26.2
Afrane and Ntiamoah 2011,
calculated2
Charcoal, High-Performing
Kiln
30.2
31.0
Calculated3, Pennise et al. 2001
Kerosene
42.9
49.0
Singh et al. 2014a
LPG from Crude Oil
45.8
49.6
Afrane and Ntiamoah 2011,
calculated2
Ethanol
27.0
29.7
GREET 2008, MacCarty 2009
Biogas from Dung
17.7
19.4
Afrane and Ntiamoah 2011,
calculated2
Wood Pellets
16.5
17.8
Jetter et al. 2012
1 Where two sources are listed, first refers to LHV. Second refers to HHV.
2 HHV is calculated based on ratio of HHV/LHV for similar fuel as documented in SI1. HHVx = LHVx*(HHVy/LHVy)
3 LHV is calculated based on ratio of LHV/HHV for similar fuel as documented in SI1. LHVx = HHVx*(LHVy/HHVy)
2.1.2 Electricity
The electricity mix for each country is based on the average 2013 electricity mix as
reported by the IEA(2013a-d). The electricity modules include estimates of generation,
transmission, and distribution losses, which are substantial, and amount to 26, 17, 18, and 26
percent, respectively, for India, China, Kenya, and Ghana. The mix of fuels in the electrical grid
is summarized for all four nations in Table 2-6. Potential future changes in the electrical fuel mix
are presented in Section 3.2.
2-5
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-6. Current Electricity Grids in Covered Countries
Fuels:
2013 India
Electrical
Grid
(%)us
2013
China
Electrical
Grid
(%)2
2013 Kenya
Electrical
Grid
(%)3
2013 Ghana
Electrical
Grid
(%)4
Coal
72.8
75.5
-
-
Oil
1.94
0.119
30.7
25.6
Natural Gas
5.46
1.66
-
10.4
Biofuels
1.83
0.703
2.02
-
Nuclear
2.87
2.05
-
-
Hydro
11.9
16.9
44.4
64.0
Solar
0.288
0.284
0.011
0.023
Wind
2.81
2.59
0.203
-
Geothermal
-
2.00E-3
22.6
-
Waste
0.112
0.226
-
-
Total Production
100
100
100
100
Distribution Losses6
26.0%
17.2%
18.4%
26.7%
Sources and Notes: 1IEA 2013a,2IEA 2013b, 3IEA 2013c, 4IEA 2013d
5 Percentages based on total Gigawatt hours electricity produced from each fuel.
0 Calculation: (DS-FC)/DS x 100, where DS = domestic supply and FC = final consumption
India Electricity Grid: As of 2013, coal-fired electricity generation constituted the
majority of India's electrical grid, providing over 70 percent of all electricity (Table 2-6).
Hydropower and gas comprise twelve and five percent of the grid mix, respectively. Indian
distribution losses consume approximately 26 percent of generated electricity.
China Electricity Grid: The electrical fuel mix in China is comprised of just over 75
percent coal with hydroelectric providing a further 17 percent. The remaining five percent of
China's electricity grid is generated from a mix of natural gas, nuclear, oil, biomass, and
renewables. Electricity losses in the Chinese system amount to 17 percent of generated
electricity.
Kenya Electricity Grid: Three fuels supply the majority of Kenyan electricity.
Hydroelectric is the largest source of power providing nearly 45 percent of electrical energy. Oil
and geothermal provide approximately 31 and 23 percent of electricity, respectively. Kenya is
one of a few East African nations that possess significant geothermal resource potential (IRENA
2013). Electricity distribution losses in Kenya total just over 18 percent.
Ghana Electricity Grid: Hydroelectric provides approximately 64 percent of Ghana's
electricity. Oil provides a further 26 percent of electricity with the final ten percent being
generated by natural gas. Ghana's electrical losses are high, with nearly 27 percent of electricity
being lost before reaching the final consumer.
2-6
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
2.1.3 Coal
Coal is a widely used cooking fuel in China with nearly 30 percent of cooking energy
being provided by coal powder, coal briquettes, and honeycomb briquettes. Coal sees only
limited use in India and is not used in Kenya or Ghana as a cooking fuel at this time.
In India, coal for cookstove use is modeled as produced in an open cast surface mine.
Surface mines account for over 80 percent of total coal production in India, and almost 100
percent of the coal grades used for cooking. The consumption of coal for cooking is primarily in
areas near coal mines, with an average transport distance of 100 km (rail). Coal is combusted in a
metal stove. The coal ash remaining after combustion, as well as the mining overburden, is
assumed to be disposed in landfills.
In China, coal is used in a variety of forms, including unprocessed, washed and dried,
powdered, formed into briquettes, or formed into honeycomb briquettes. Coal is combusted in
metal and brick stoves (both traditional and improved) which have efficiencies assumed to range
from 14 percent - 37 percent depending on the fuel/stove technology combination (Zhang et al.
2000). Coal transportation is adapted from the incoming transport within the "hard coal mix at
regional storage" unit process from Ecoinvent 2.2. Coal is transported approximately 30 km by
barge, 51 km by train, and 100 km by light duty diesel vehicle from the distributor to retail. The
coal ash remaining after combustion, as well as the mining overburden, is assumed to be
disposed in landfills. The process also includes estimated emissions due to leaching from coal
heaps into groundwater at storage sites.
2.1.4 Dung
Dung is a low cost traditional source of cooking fuel in India where it provides over ten
percent of cooking energy. Dung is not widely used in the other nations studied. In the LCA
model of this study, the dung of stall fed cattle and buffaloes is converted into dung cake
primarily by women who mix the manually collected dung with residual feed (e.g., straw, wood
chips) (Singh et al. 2014a). Dung cake is combusted in a traditional mud stove with a low
thermal efficiency. The remaining ash after combustion is modeled as land-applied. Dung cake is
a significant fuel source for cooking only in India. All CO2 emissions associated with dung cake
combustion are assumed to be associated with biogenic carbon and therefore not to contribute to
GCCP, as described in Section 4.1.
2.1.5 Crop Residues
Residues from crops such as rice, wheat, cotton, maize, millet, sugarcane, jute, rapeseed,
mustard, and groundnut are burned by households in India and China. Crop residues are not used
as a cooking fuel in Kenya and contribute less than one percent of cooking energy in Ghana.
Country-specific estimates of crop production practices were not developed for the African
nations. In India and China, crop residues are modeled as manually collected and air dried but
not further processed prior to combustion in cookstoves. In India, 95 percent of crop residues are
assumed to be combusted in traditional mud stoves (Smith et al. 2000). In China, 75 percent of
crop residues are combusted in improved stoves (IARC 2010), while the remaining crop residues
are burned using traditional stove designs. In both countries, the ash remaining after stove use is
2-7
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
assumed to be land-applied. The three major biomass crops for India and China are used to
estimate the environmental impacts associated with crop production. Rice and wheat are
modeled for both countries with the addition of sugarcane and maize for India and China,
respectively.
On average in India and China, ten and 29 percent of crop residues are returned to the
soil, respectively (IARI 2012, Wang et al. 2013). Additionally, 19 and eight percent of crop
residues in India and China are burned in the open on agricultural fields (IARI 2012, Food and
Agriculture Organization (FAO)STAT 2016). Emissions associated with field burning of
biomass are included in the analysis.
Fertilizer and water use input values are included for each crop. National average
estimates of energy and agricultural chemical use are included (FAOSTAT 2016). Emissions of
nitrous oxide (N2O), nitrate, ammonia, and phosphorus are calculated based on fertilizer
application rates (Wang et al. 2014, Xia and Yan 2011), with additional values drawn from the
literature where available. Methane emissions associated with rice production are estimated
using the IPCC method (2006). All CO2 emissions associated with crop residue combustion are
assumed to be associated with biogenic carbon and therefore not to contribute to GCCP, as
described in Section 4.1. Detailed documentation of crop production LCI development are
available in SI5.
2.1.6 Firewood
Firewood is the predominant cooking fuel in India, Kenya and Ghana where it provides
49, 65, and 46 percent, respectively, of cooking energy. Firewood is also a common cooking fuel
in China where it provides 15 percent of total cooking energy demand. Typical tree species used
for firewood in India are acacia, eucalyptus, sheesham and mango. In the baseline model, 24
percent of firewood cooking fuel in India is estimated to be non-renewable, based on trends in
forest land area, renewable biomass generation on forest land, and demand for cooking firewood
as discussed in Section 4.2 (Drigo 2014). In China, cooking firewood is harvested from mature
trees or large branches (e.g., eucalyptus, acacia, oak, pine, poplar, and willows), obtained
manually from local forest and sun-dried. All carbon in firewood is assumed to come from
biogenic sources; however, due to the prevalence of non-renewable forestry practices in the four
study nations, not all CO2 emissions from firewood combustion are considered carbon neutral.
The percentage of CO2 emissions that count towards GCCP is determined by the country specific
forest renewability factor, which represents the ability of forests to re-sequester the combusted
carbon based on national forest regrowth, as presented in Table 4-1.
Firewood is assumed to be collected manually and combusted using the stove
technologies listed in Table 2-8. The remaining ash is assumed to be land applied. Ash
production for firewood in Kenya and Ghana is based on an average ash content of 3.3 percent
(Afrane and Ntiamoah 2011).
2-8
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
2.1.7 Charcoal
Charcoal in lump form is a widely used cooking fuel in Kenya and Ghana where it
provides 17 and 32 percent, respectively, of the current national cooking energy. Charcoal
provides approximately one percent of cooking energy in India and is not used in China.
In India, charcoal is produced from wood in a traditional earth mound kiln. The charcoal
yield from the kiln is modeled as 30 percent, and kiln combustion residuals are land-applied. The
firewood is assumed to be collected and brought to the charcoal kiln manually. Charcoal is
modeled as combusted in a metal Angethi stove. Charcoal is an informal manufacturing sector in
India, and it is assumed that charcoal is used for cooking only by those living near charcoal kilns
(Singh et al. 2014a).
Charcoal in Ghana in the baseline model is produced in an earth mound kiln, with 4.9 kg
wood required per kilogram (kg) charcoal output (Afrane and Ntiamoah 2011). Charcoal is
assumed to be transported 483 km by single unit truck based on the average distance between
forested areas and large urban population centers in Ghana. Charcoal in Kenya is also produced
in an earth mound kiln. The dry wood yield of Kenyan kilns is higher than the dry wood yield for
Ghana, with 3.2 kg of wood being required per kg charcoal produced (Pennise et al. 2001).
Charcoal is transported from the kiln to end users 323 km via single unit truck based on the
average distance between forested areas and main population centers in Kenya.
Similar to firewood, all carbon in charcoal is assumed to come from biogenic sources;
however, due to the prevalence of non-renewable forestry practices in the four study nations, not
all CO2 emissions from the kiln process and charcoal combustion are considered carbon neutral.
The percentage of CO2 emissions that count towards GCCP is determined by the country specific
forest renewability factor, which represents the ability of forests to re-sequester the combusted
carbon based on national forest regrowth, as presented in Table 4-1. For all countries with
charcoal use, ash remaining after combustion is land applied to nearby agricultural fields.
2.1.8 Liquefied Petroleum Gas
LPG is a common cooking fuel in urban markets in all four study countries. This fuel
comprises between 21 and 31 percent of the current cooking fuel mix in India, China, and
Ghana. Use of LPG in Kenya is still limited with five percent of cooking fuel energy being
provided by this source.
In India, 21 percent of LPG is assumed to be produced from natural gas and 79 percent
from crude oil (MPNG 2014). For Indian LPG from natural gas, natural gas extraction is based
on drilling, metering, testing and servicing of oil wells and production data of the Oil and Natural
Gas Corporation (ONGC), the largest oil company in India. Eighty-four percent of natural gas in
India comes from offshore sources and 16 percent is from onshore sources. LPG production is
based on the scenario of an LPG production line of the ONGC Uran Gas fractionating plant
located near Mumbai, India. Natural gas is transported to the gas fractionating plant by pipeline
(500 km from onshore, 250 km from offshore). Processing requirements are allocated to the
outputs from LPG production on a direct mass basis. The bottling stage is modeled based on the
per-day production scenario of Indian Oil Corporation Limited (IOCL) Barkhola bottling plant
2-9
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
located in Assam, India. This plant is one of the recent state-of-the art bottling plants
commissioned by IOCL and is considered representative of bottling plants in India. LPG is
bottled in steel cylinders (Singh et al. 2014a). Incoming transport of natural gas to the bottling
plant is 60 percent by rail (1000 km) and 40 percent by heavy duty vehicle (500 km). The bottled
LPG is then transported 750 km by heavy duty diesel vehicle to the distributor and 100 km by
light duty diesel vehicle from the distributor to retail.
For the 79 percent of LPG produced from crude oil, the India model considers only the
domestic production of refined petroleum fuels. The exclusion of overseas crude oil is not
expected to impact findings significantly because only the extraction stage is impacted (not the
refining stage), and Indian companies engage in extraction of crude oil following globally
accepted practices and operational standards -equivalent to overseas oil companies (Singh et al.
2014a). Onshore crude oil is 30 percent of refinery inputs and is transported 1000 km by rail to
the refinery; offshore crude oil makes up 70 percent of the inputs and is first transported 500 km
to the port, then 60 percent is transported 1000 km by rail to refineries and 40 percent is
transported 500 km to refineries by heavy duty diesel vehicle (Singh et al. 2014a). Mass
allocation is used to partition petroleum refining burdens to different refinery products. Once the
LPG reaches the bottling plant, the supply chain is equivalent to that modeled for the natural gas
LPG supply chain.
LPG production for China is based on two Swiss refineries for the year 2000. Electricity
grid mix and rail transport are adapted to the China geographic scope. The bottling stage is
simulated based on the model created for India.
LPG in Kenya is modeled as 100 percent derived from crude oil. The crude oil is
assumed to be produced in Algeria and transported to Kenya by ship (8,445 km). LPG is bottled
in steel cylinders and transported 750 km by truck and 100 km by van within Kenya.
LPG in Ghana is modeled as produced 100 percent from crude oil. The crude oil is
produced in Nigeria. LPG is either refined in Nigeria and imported to Ghana, or crude oil is
imported to Ghana and the LPG is refined at Ghana's only refinery (Tema Oil Refinery) (Afrane
and Ntiamoah 2011). The transport from Nigeria to Ghana is modeled as 433 km by ship. LPG is
bottled in steel cylinders and transported 750 km by truck and 100 km by van within Ghana.
2.1.9 Kerosene
Kerosene is used widely in India and Kenya, where it constitutes approximately three and
12 percent of the current cooking fuel mix in each country, respectively. Use of kerosene
contributes less than 0.3 percent of cooking energy in both China and Ghana.
For the India kerosene model, only domestic production of petroleum refining products is
considered. The exclusion of overseas crude oil is not expected to impact findings significantly
because only the extraction stage is affected (not the refining stage), and Indian companies
engage in extraction of crude oil following globally accepted practices and operational standards
equivalent to overseas oil companies. Onshore crude oil (30 percent of refinery inputs) is
transported 1000 km by rail to the refinery; offshore crude oil (70 percent of the inputs) is first
transported 500 km to the port, then 60 percent is transported 1000 km by rail to refineries and
2-10
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
40 percent is transported 500 km to refineries by heavy duty diesel vehicle. Mass allocation is
used to partition petroleum refining burdens to different refinery products. Thirty percent of
kerosene is assumed to be transported 1000 km by rail, while the remaining 70 percent travels
the same distance by way of heavy duty diesel vehicle. All kerosene is transported in a light duty
diesel vehicle 100 km from the distributor to retail. Similar to LPG, the bottling stage is
simulated based on the per-day production scenario of the IOCL Barkhola bottling plant located
in Assam, India. Kerosene is bottled in steel cylinders (Singh et al. 2014a).
For China, production of petroleum products is adapted to the China geographic scope
using a refinery dataset in Ecoinvent (Ecoinvent Centre 2010). The data set includes all flows of
materials and energy for throughput of one kilogram of crude oil in the refinery. The multi-
output process 'crude oil, in refinery' delivers the co-products gasoline, bitumen, diesel, light fuel
oil, heavy fuel oil, kerosene, naphtha, propane/butane, refinery gas, secondary sulfur, and
electricity. The impacts of processing are allocated to the different products on a mass basis.
Electricity grid mix and rail transport are adapted to the China geographic scope. The bottling
stage is simulated based on the per-day production scenario of the IOCL Barkhola bottling plant
located in Assam, India. Kerosene is bottled in steel cylinders. Incoming transport to the bottling
plant is 60 percent rail (1000 km) and 40 percent heavy duty vehicle (500 km). All bottled
kerosene is modeled as being transported 750 km by heavy duty diesel vehicle to the distributor
where it travels a further 100 km by light duty diesel vehicle from the distributor to retail.
Kerosene in Kenya and Ghana is modeled as 100 percent derived from crude oil
(Ecoinvent Centre 2010). In Ghana, the crude oil is modeled as produced in Nigeria and shipped
to Ghana, where it is refined to kerosene at Ghana's only refinery Tema Oil Refinery (Afrane
and Ntiamoah 2011). The crude oil transport from Nigeria to Ghana is modeled as 433 km by
ship. For Kenya, the crude oil for kerosene is assumed to be produced in Algeria and transported
to Kenya by ship 8,445 km (Ecoinvent Centre 2010). The Ghana kerosene refining process is
adapted for Kenya conditions in this study (Afrane and Ntiamoah 2011). For both Kenya and
Ghana, the kerosene supply-chain and bottling models are equivalent to those applied for LPG in
those countries.
2.1.10 Natural Gas
Natural gas does not currently provide a significant amount of cooking energy in any of
the study countries but has the potential to see increased use in the future. Natural gas extraction
is based on Russian production data and long-distance pipeline transport of natural gas to China.
Energy requirements for operation of the gas pipeline network are adapted from an Italian
company data set in Ecoinvent for delivery of natural gas to consumers via pipelines (Ecoinvent
Centre 2010). The total leakage rate, modeled as 1.4 percent for long-distance pipeline transport,
is based on European data (Ecoinvent Centre 2010). The electricity grid mix and rail transport
are adapted for the Chinese and Indian geographic scope.
2.1.11 Coal Gas
Goal gas is produced through coal gasification and delivered directly to consumers via a
pipeline network (Zhang et al. 2000). The process technology used in this model, coal gas
produced from coke oven gas, is adapted from Ecoinvent for the Chinese geographic scope
2-11
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
(Dones et al. 2007). Coal gas is transport from the plant to rural consumers via a long-distance
pipeline network. Coal gas is assumed to be burned in a traditional gas range. Use of coal gas as
a cooking fuel in India, Kenya and Ghana is considered unlikely due to the current low
prevalence of coal as a heat source for cooking in these countries.
2.1.12 Ethanol
Ethanol is not yet a widely used source of cooking energy in any of the four countries
studied. Ethanol is being proposed as a potential clean and efficient source of cooking energy
that utilizes agricultural and food waste products widely available in these nations.
Ethanol production and processing in India is modeled based on the data provided by
Tsiropoulos and colleagues (2014). In India, sugarcane cultivation practices are almost
exclusively manual, with the exception of plowing, which is modeled as partially mechanized in
some states. Pre- and post-harvest burning of straw is not practiced in most of India. Sugarcane is
transported 12 km by truck to the sugarcane mill. The output products of the conventional sugar
mill are sugar, molasses, and electricity from surplus bagasse. Conventional mills represent 75
percent of the sugar production in India. Bagasse provides all necessary energy requirements at
the mill as well as surplus electricity, which is considered a useful co-product to replace grid
electricity in India. Sugarcane ethanol is then produced from the molasses. This study considers a
weighted average of ethanol distilleries as standalone distilleries and as adjacent to sugar
refineries. Molasses is transported on average 75 km to the ethanol plant. Sugarcane ethanol
production energy is also provided by bagasse. The model is based on a hydrous ethanol yield
(for 95 percent ethanol by volume) of 84.7 liters/tonne of cane and an ethanol density of 0.789
kg/L. All ethanol is assumed to be transported 750 km by heavy duty vehicle to the distributor
and 100 km by light duty vehicle from the distributor to retail. Sugarcane ethanol combustion
emissions are based on laboratory testing rather than field results (e.g., actual measurements
from cookstoves in use within India). Sugarcane ethanol production in China is based on the
Indian unit process described above, modified for the Chinese electrical grid mix. Ethanol in
Kenya is assumed to be produced in India and transported to Kenya (see India model
assumptions). Transport to Kenya is modeled as 4,409 km (by ship) based on the distances
between major ports in the two countries.
For Ghana, sugarcane ethanol is assumed to be produced in Brazil, the largest global
producer of ethanol from sugarcane. Sugarcane production is modeled as 80 percent manual and
20 percent mechanical harvest (Macedo et al. 2008). Ethanol is produced directly from the cane
(i.e., cannot be converted first to molasses). Ethanol is produced via a fermentation route using
energy from the bagasse. Electricity is co-produced with ethanol, but no credit for exported
electricity is applied in the model. Ethanol is transported by ship from Brazil to Ghana (5,177
km).
Sugarcane ethanol combustion emissions are based on laboratory testing. The sugarcane
ethanol combustion emission profiles are the same for all countries evaluated (Berick 2006,
MacCarty 2009).
2-12
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
2.1.13 Biogas
Biogas sees only modest current use in India and Kenya where it provides less than one
percent of cooking energy. This study considers a two-cubic meter household type fixed dome
anaerobic digester (AD) operating in continuous feeding mode for 350 days/year and ten years of
operational life (UN 2007). The AD is loaded with 19.3 kg/day of fresh dung mixed with small
quantities of water to produce 1.31 rnVday of biogas (Singh et al. 2014a). Leakage is the source
of fuel production emissions. Approximately one percent of biogas (CH4) generated is assumed
to leak from the system (Afrane and Ntiamoah 2011, Borjesson 2006). Digested slurry is a useful
co-product and is stored for application in land farming. The AD is located at the home where
the fuel is used (distributed through piping running from the digester to the home). Feedstock
amounts and biogas yields at the household level were available specifically for the Ghana scope
based on questionnaires and field measurements within the country (Afrane and Ntiamoah,
2011). Biogas production in China is based on the Indian unit process as described above.
2.1.14 Biomass Pellets
Biomass pellets are not yet widely used as a cooking fuel within any of the countries
studied but are proposed as a cleaner and more efficient use of biomass resources (Cashman et
al. 2016). Biomass pellets are based on wood feedstock, and forest renewability factors specific
to each nation are assumed (Table 4-1). Manual collection and small-scale mechanized
pelletization is modeled for all nations using the appropriate national electricity grid (IEA 2013a-
d). Pelletization processing energy and distribution transport are adapted from Austria and
central Europe (Jungbluth et al. 2007). Incoming transport to pelletization (rail and truck) is
included. The inventory for emissions from biomass pellet combustion is based on laboratory
testing results. Similar to firewood and charcoal, all carbon in biomass pellets from wood is
assumed to come from biogenic sources; however, due to the prevalence of non-renewable
forestry practices in the four study nations, not all CO2 emissions from the pellet combustion are
considered carbon neutral. The percentage of CO2 emissions that count towards GCCP is
determined by the country specific forest renewability factor, which represents the ability of
forests to re-sequester the combusted carbon based on national forest regrowth, as presented in
Table 4-1. Ash remaining following combustion of biomass pellets is disposed of via land
application.
2.2 Cookstove Descriptions
The choice of which stove technology to use or to promote, in addition to the selection of
the fuel itself, is a critical determinant of the life cycle environmental impacts of an integrated
cooking system as was demonstrated in Phase I of this study. This phase incorporates a
significant amount of new emissions data for many stove groups, as compared to Phase I, and
examines the effect of increased adoption of improved stove technologies alongside shifts in the
cooking fuel mix as part of the scenario analysis. Adoption refers to the future use of improved
stove technologies or fuel forms in place of current alternatives.
Stove efficiency and emissions data as reported in the literature are always associated
with cookstove use in a specific context. That context includes the specific stove model used for
the study as well as the fuel type. For each fuel type, there are several parameters that contribute
2-13
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
to variation in operational performance. An example of these parameters for firewood would be
the heating value associated with a specific tree species or moisture content of the feedstock that
is used to perform the study. The laboratory or field testing setup is also a part of the study
context and contributes additional uncertainty to the results when they are considered to be
representative of national average cookstove use. For this study, individual stove efficiency and
emissions results from the literature are grouped together into stove groupings that correspond to
the level of detail available on stove use within the nations of study.
Table 2-7 presents stove groupings along with a record of the information that was
available for each. A stove group, in this study, is defined by a unique combination of stove type,
fuel type, country, and current average efficiency. In the case where a specific country is listed,
the average of all available thermal efficiency values was taken as the estimated thermal
efficiency that defines that grouping. Thermal efficiency values for the global region are
calculated in excel as the 20th percentile value of the sampled thermal efficiencies. While the
average was considered as an option for the global region, it was found that this yielded thermal
efficiency estimates appreciably higher than those for India and China. This appears to be
attributable to a tendency in the literature to focus stove emission testing efforts on more
advanced versions of improved cookstoves than on models that are expected to be widely
employed in practice. In other words, the availability of global stove emission test results
provides more information on emissions from potentially deployable technologies than it does on
those in current use. Use of the 20th percentile value is found to produce results that are more in
line with those observed for India and China and is believed to be more representative of the
stoves used in both Kenya and Ghana.
Table 2-7. Stove Type and Efficiency by Nation
Stove
Current
Efficiency
Sample
Fuel
Type
Country
Efficiency
Range
Size (n=)
Source(s)
Hard Coal
Angethi
India
16%
16%
1
1
Coal, Powder
Traditional
China
10%
7-14%
3
8
Improved
China
17%
17-18%
3
8
Coal, Briquette
Improved
China
32%
27-37%
2
8
Coal,
Traditional
China
20%
16-23%
2
8
Honeycomb
Improved
China
45%
44-47%
2
8
Dung
Traditional
India
9%
8-9%
3
4
Improved
India
11%
10-13%
2
4
Crop Residue
Traditional
India
11%
10-12%
3
1,4
China
11%
11%
1
8
Global25
11%
4-18%
6
1,4,8,18
Improved
India
16%
11-22%
4
4
China
17%
15-19%
2
9
Firewood
Three-
India
18%
17-18%
2
4
stone
Global
13%
12-15%
4
5,6
Traditional
India
17%
13-23%
4
1,4,17
China
12%
12%
1
2
Global
11%
9-18%
23
2,3,9,15,1
6
Improved
India
24%
20-29%
13
2,3,4,17
2-14
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-7. Stove Type and Efficiency by Nation
Fuel
Stove
Type
Country
Current
Efficiency
Efficiency
Range
Sample
Size (n=)
Source(s)
China
16%
13-24%
4
8
Global
19%
11-50%
69
2,5,10,13,
14,15,18,
19,20
Charcoal
Traditional
Global
14%
12-22%
4
2,6,7
Angethi
India
18%
18%
1
1
Improved
Kenya
25%
23-27%
4
5
Ghana
23%
23%
2
5
Kerosene
Improved,
Pressure
India
47%
47%
2
1,4
Improved,
India
50%
50%
1
4
Wick
Improved
China
45%
42-49%
2
8
Global25
46%
37-52%
8
1,4,8,10,2
0,21
LPG
Modern
India
55%
54-57%
2
1,4
China
47%
42-54%
3
8
Global24
49%
42-75%
11
10,20,21
Natural Gas
Modern
China
57%
54-61%
2
8
Coal Gas
Modern
China
46%
46%
1
8
Electricity
Modern
Global
59%
57-80%
4
11,12
Ethanol
Modern
India
53%
53%
1
21
Kenya
46%
40-52%
2
10
Global
49%
43-66%
4
10
Biogas
Modern
India
56%
55-57%
2
1,4
Modern
China
56%
55-57%
2
1,4
Modern
Global
55%
32-57%
5
1,21
Pellet, Wood
Modern
Global
35%
35-53%
6
5, 23
Sources and Notes: 1 Singh et al. 2014a/b,2 Bhattacharya et al. 2002b,3 Bhattacharya et al. 2002a, 4 Smith et al. 2000,5 .Tetter et
al. 2012,6 Sweeney 2015,7 Booker 2012,8 Zhang et al.'2000,9 Afrane andNtiamoah 2012,10 GACC 2016,11 Schaetzke 1995,
12 EC 2011 13 .Tetter and Kariher 2009,14 Winrock 2009, 15 AED 2008,16 AED 2007, 17 Bailis et al. 2007,18 Collivignarelli et al.
2010, 19 Robinson 2013, 20 MacCarty et al. 2010,21 CES 2001, 22 Berick 2006,23 Carter et al. 2014
24 Current average thermal efficiency set as the average of India/China (IN/CN) due to the wide range of reported values, which
skew towards high thermal efficiency.
25 Average of reported thermal efficiencies used to derive current thermal efficiency, as opposed to 20th percentile, due to the
presence of low values and a better match with the IN/CN average.
The following sections outline stove characteristics considered for each country and the
methodology and data sources used to compile emissions data and assemble cookstove LCIs.
2.2.1 Stove Efficiency
As is indicated in the preceding section, the records of stove emissions and performance
included in this study exhibit a range of thermal efficiency values for each fuel. The Phase ILCA
for China and India indicated that stove thermal efficiency is a driving parameter for the overall
2-15
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
life cycle environmental impacts of cookstove fuels (Cashman et al. 2016). Stove efficiency not
only affects the emissions at point of use but also the overall fuel quantity required to produce
the functional unit. Changes in the fuel quantity required to deliver 1 GJ of energy to the pot
impact all upstream life cycle stages (e.g., higher stove thermal efficiencies result in less fuel
being extracted, processed, and transported). Figure 2-1 presents a box plot depicting the range
of thermal efficiency values for each fuel type as compiled from the literature. Figure 2-1 also
displays the average current thermal efficiency value for fuels by country used in the baseline
model when applicable. These presented ranges incorporate both improved and traditional
stoves. Stove technology use by country is covered in subsequent sections. Documentation of
values and references used to create this figure are available in Table 2-7 and SI1.
90%
80%
O
J 70%
o
§ 60%
| 50%
£ 40%
> 30%
55 20%
10%
0%
T
T
-0-
A
T
~
V
- w
&
IX
cf
y
<$>
India Average
China Average
> Ghana Average
~ Kenya Average
v x Ghana Average
Figure 2-1. Range of reported stove thermal
2.2.2 Stove Technology Use by Country
The rate of adoption of improved cookstoves and the specific model of cookstove used
varies widely between nations. This section presents the current use of traditional, improved, and
modern cookstove technologies to provide cooking energy by fuel type in India, China, Kenya,
and Ghana. Table 2-8 presents this information along with the national average thermal
efficiency value for each fuel, which is a weighted average of stove thermal efficiency associated
2-16
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
with underlying stove technology use. Stove technology use for each country was adapted from
the references listed in Table 2-8. Assumptions based on noted references used to develop the
documented stove technology mixes are documented in SI1. Information pertaining to current
national stove technology use was found to be limited, see Appendix C for more detail on related
stove technology use data quality considerations.
2-17
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-8. Stove Technology Use and Aggregate Thermal Efficiency by Fuel
(Stove Technology Use is Presented as a Fraction of National Cooking Energy)
Stove Use Technology Use
Aggregate Efficiency
i
Fuel
Stove Type
India1
China2 4
Kenya5 9
Ghana1012
India
China
Kenya
Ghana
Coal
Traditional
1.2%
17.9%
_
_
15.5%
21.8%
Improved
_
11.0%
_
_
Dung
Traditional
10.0%
_
_
_
8.7%
Improved
0.6%
_
_
_
Crop Residue
Traditional
8.4%
2.8%
_
0.4%
11.4%
15.8%
Improved
0.5%
9.2%
_
_
Firewood
Three-stone
4.6%
_
51.9%
36.7%
Traditional
41.5%
3.4%
11.5%
9.2%
17.5%
17.5%
13.5%
13.5%
Improved
2.9%
11.3%
1.5%
_
Charcoal
Traditional
1.2%
_
7.6%
20.3%
17.5%
20.0%
16.9%
Improved
_
_
9.3%
11.2%
Kerosene
Pressure
1.5%
_
_
_
Wick
1.7%
_
_
_
48.6%
45.3%
45.9%
45.9%
Improved
_
0.3%
11.6%
0.2%
LPG
Modern
25.2%
33.5%
5.0%
21.7%
55.3%
47.0%
49.2%
49.2%
Electricity
Modern
0.4%
10.6%
0.8%
0.3%
58.5%
58.5%
58.5%
58.5%
Biogas
Modern
0.4%
_
0.7%
_
56.2%
47.6%
47.6%
47.6%
Total
100%
100%
100%
100%
Sources and Notes: 1 Smith et al. 2000,2IARC 2010,3 Dalberg 2014, 4 NBSC 2008,5 Githiomi et al. 2012,6 SEI2016,7 Dalberg 2012,8 Clough 2012,9 Dalberg
2013a,10 Energica 2009,11 GLSS6 2014,12 ADP 2012
13 For efficiency of a particular stove-fuel combination see Table 2-7
2-18
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
India: Nearly 50 percent of cooking energy in India is derived from firewood, and over
85 percent of this cooking is done using traditional mud stoves. Improved woodstoves have not
yet been widely adopted, representing less than three percent of total cookstove use. Traditional
mud stoves are also the main combustion technology used for cooking with both dung cake and
crop residues. Dung burned in traditional stoves exhibits the lowest thermal efficiency of all
stove-fuel combinations at just nine percent. Coal and charcoal are assumed to be burned in the
traditional Angethi stove, which has a reported thermal efficiency of 15 and 18 percent for the
two fuels, respectively. Among kerosene users in India there is a relatively even split between
the use of wick and pressure stoves. LPG, electricity, and biogas are all burned in modern stoves.
China: The International Agency for Research on Cancer (IARC 2010) conducted a
survey in rural China finding that 77 percent of biomass stoves are classified as improved. The
remaining 23 percent of biomass is burned in traditional stoves with an average thermal
efficiency of only twelve percent. Thirty-eight percent of coal is burned in improved cookstoves
with the remaining 62 percent of this feedstock being burned in traditional cookstoves. These
percentages are applied to all forms of coal-based fuel including powder, briquette, and
honeycomb briquette. Kerosene, LPG, electricity, and biogas are all burned in modern stoves.
Kenya: Approximately 52 percent of all Kenyan households cook their meals with wood
fuel over an open three-stone fire, which constitutes 80 percent of firewood users. A further 18
percent of firewood is consumed in traditional stoves, signaling very limited adoption of
improved wood stoves in Kenya. Improved cookstoves occupy a larger proportion of the
charcoal market where they are used preferentially in urban areas, providing 55 percent of
cooking energy from this fuel source. Kerosene, LPG, electricity, and biogas are all burned in
modern stoves with thermal efficiencies above 45 percent.
Ghana: Over 65 percent of charcoal cooking is done in a traditional stove known as a
coal pot. The coal pot is simple in design consisting of an open vessel on a base into which the
charcoal is placed. The vessel has a slotted bottom for air flow and the cooking pot is placed
directly on the charcoal for cooking. Firewood is also used heavily in Ghana with nearly all users
burning firewood over a traditional three-stone fire or mud stove. Crop residue is used only
marginally and is assumed to be consumed using traditional stoves. Kerosene, LPG, and
electricity are all burned using modern stoves with thermal efficiency values between 46 and 59
percent.
2.2.3 Stove Emissions Data Sources and Methodology
Stove emissions data from the literature were compiled into LCI unit processes according
to the groupings established in Table 2-1. Sufficient information was found to create country-
specific stove emissions and efficiency data for both India and China. Information specific to
Kenya and Ghana was limited and for most stove groupings, it was necessary to use the average
of the remaining data discovered for developing countries as a proxy. In a few cases, this
averaging was also necessary for India and China. The averaging was done in the interest of
including the full breadth of possible emissions and efficiency data. The stove emission profiles
for Kenya and Ghana are not specific to those nations; however, this generalized stove emission
information was incorporated into upstream fuel production, fuel mix, and stove technology use
scenarios that are specific to both countries. A sample of selected representative stove emissions
2-19
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
is provided in Table 2-9. The references cited in the table are sources of information used for the
full stove LCI, and not all references contain a record for each individual pollutant species.
An extensive database of all stove emission information was compiled and reviewed
according to the project QAPP. This database was filtered according to the criteria that define the
established groups: (1) country, (2) fuel type, and (3) stove type. All data fields for which an
emission value is present were extracted from the database and used as the basis of LCI
emissions, which are documented in SI2. Stove thermal efficiency and fuel heat content are used
to transform emission values that are reported on the basis of fuel consumed (e.g., g/kg) and not
on the basis of delivered heat. Where possible, these values were drawn from the original study
itself and are reported in the database. In cases where the original study does not report either
stove thermal efficiency or fuel heat content, the average of all reported values that correspond to
a specific stove group was used in the calculation.
To ensure a fair comparison of environmental impacts among different stoves, fuels, or
countries, it is important that the same scope of inputs and emissions be considered across
options. This type of comparison requires a line to be walked between inclusion of the most
detailed available information, which may be available only for certain stoves, fuels, or
countries, and a desire to establish fair comparisons between stove groups and countries. With
this interest in mind, the authors identified a list of key pollutants that are necessary to ensure a
complete inventory for each stove grouping. These pollutants include: CO2, CH4, N2O, CO,
NOx, SO2, PM, NMVOCs, and ash. BC emissions are estimated on the basis of PM emissions as
described in Section 4.3. Other emissions are included in the LCI for each stove, when available,
but these emissions are not necessarily available for all stove groupings. To ensure that the full
inventory of priority pollutants is present for each stove grouping, it has been necessary in some
instances to use records of pollutant emissions from other countries or the closest available stove
grouping to fill holes in some of the stove emission inventories. A record of precisely what
emissions are included or excluded for each stove group and the source of proxy emission values
is available in Appendix C and SI2. A discussion of the approach to estimating stove emission
uncertainty is included in Section 4.5.1.
2-20
-------
Section 2—Cooking Fuel and Stove Descriptions and Methodology
Table 2-9. Summary Table Showing Representative Stove Emissions
Stove Grouping
Country
CO2
CO
PM (>2.5<10)
CH4
NOx
SO2
Source(s)
Coal Powder, Traditional
China
979
38
2.5
3.9
0.93
4.90
3,10
Coal Powder, Improved
China
642
32
0.8
0.7
0.3
0.1
3,10
Coal Briquette, Improved
China
364
4.5
0.03
0.00
0.1
0.3
3,10
Honeycomb Coal, Traditional
China
326
17
no estimate
no estimate
no estimate
no estimate
3,10,13,14
Honeycomb Coal, Improved
China
284
7.3
0.5
0.4
0.1
0.1
3,10,15
Dung Cake, Traditional
India
991
51
23
9.7
0.8
0.3
1,2,3,6
Dung Cake, Improved
India
800
23
no estimate
2.6
0.2
no estimate
3,6
Crop Residue, Traditional
India
806
39
11
4.2
0.8
0.2
1,2,3,6
Crop Residue, Traditional
China
892
73
4.9
no estimate
no estimate
no estimate
3,9,13
Crop Residue, Improved
India
480
37
no estimate
4.2
0.1
no estimate
3,6
Crop Residue, Improved
China
539
47
4.9
2.0
0.6
0.04
3,10,11,12,16,17
Firewood, three-Stone
India
534
22
no estimate
2.2
0.05
no estimate
3,6
Firewood, Traditional
India
614
32
4.6
1.8
0.4
0.2
1,2,3,6
Firewood, Traditional
China
660
31
1.2
2.3
0
no estimate
4,3,8,9
Firewood, Improved
India
358
19
no estimate
2.1
0
no estimate
3,4,5,6
Firewood, Improved
China
606
34
1.5
1.4
0.5
0.01
3,10,11,12
Charcoal, Improved
Kenya
337
28
0.5
2.0
no estimate
no estimate
7
Charcoal, Improved
Ghana
346
36
0.9
1.7
no estimate
no estimate
7
Kerosene, Modern, Pressure
India
145
2.9
no estimate
0.1
0.01
no estimate
3,6
LPG, Modern
India
126
0.6
no estimate
0.00
0.01
no estimate
3,6
LPG, Modern
China
146
0.3
no estimate
0.02
0.0
0.00
3,10
Biogas, Modern
India
144
0.2
0.2
0.1
0.01
0.1
1,3,6
Sources: 1 Singh et al. 2014a,b,2 Saud et al. 2012,3 Zhang et al. 1999, 4 Bhattacharya et al. 2002b,5Bhattacharya et al. 2002a,0 Smith et al. 2000,7 .Tetter et al. 2012,8 Shen et
al. 2013,9 Shen et al. 2012b,10 Zhang et al. 2000,11 Shen et al. 2012a,12 Wang et al. 2009,13 Shen et al. 2010b, 14 Shen et al. 2010a,15 Zhi et al. 2008,16 Cao et al. 2008,
17 Wei et al. 2014,18 Sweeney 2015
2-21
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3. METHODOLOGY FOR SCENARIO DEVELOPMENT AND SENSITIVITY
ANALYSES
3.1 Cooking Fuel Mix Scenario Development
As a result of Phase I, it was determined that differences in environmental impact that
existed between individual cooking fuels were greater than the differences in environmental
impact realized between the fuel mix scenarios analyzed, indicating that more dramatic
departures from the current fuel mix than were explored in Phase I may be required to realize
appreciable environmental gains. Given this, a Diverse Modern Fuel scenario has been
developed for each country, which represents a relatively dramatic departure from the status quo.
The remaining three fuel mix projections are based both on literature sources and analysis of past
trends in fuel mix development as recorded in SI3. Notable sources used to construct the future
cooking fuel mix scenarios include government surveys, IEA World Energy Outlook reports, and
other peer-reviewed studies. Result files for each country allow users to specify a custom fuel
and stove technology mix based on their own understanding of potential developments in the
cooking sector for each nation.
Underlying the potentially dramatic shifts in cooking fuel mix over the next 20 to 30
years are a series of major socioeconomic changes. Both India and China are experiencing a
period of rapid economic expansion, with rates of gross domestic product (GDP) growth
exceeding seven percent per annum. GDP growth rates in Kenya and Ghana are appreciably
lower at approximately five and four percent, respectively (World Bank 2014a). The population
of all four nations is expected to expand over this period, with India projected to become the
most populous country in the world sometime between 2030 and 2040. Currently, the population
of India is 1.25 billion and is expected to grow to 1.6 billion people by 2040. The Chinese
population is now over 1.35 billion and will top 1.5 billion by the year 2030. While Kenya and
Ghana's total populations are significantly lower at approximately 44 and 26 million,
respectively, they each exhibit growth rates of greater than two percent (World Bank 2014b,c).
Access to electricity within the four countries also varies dramatically. China reports that
over 99 percent of the population currently has access to electricity (IEA 2007). However, only
79 percent of the Indian population and 64 percent of Ghana's population have access to this
basic service. Access within Kenya is particularly low with an electrification rate of only 23
percent (World Bank 2012). By 2030, 96 percent of India is projected to have access to
electricity; reliability should increase dramatically, and losses in the electricity grid are expected
to fall (IEA 2007). The governments of both Kenya and Ghana also have plans to dramatically
increase generation capacity.
Within the context of economic and population growth and development lies a dynamic
landscape of fuel resources and relative fuel costs. It is well understood that as incomes rise, the
reliance on traditional, solid biomass fuels begins to fall in favor of more convenient liquid and
gas options (Malla and Timilsina 2014). Advanced biomass may also provide an attractive
alternative if adequate supply chains can be established that provide for much of the convenience
of other advanced fuel options while still retaining a traditional character and flavor.
3-1
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Assumptions regarding the increased uptake of improved stove types and advancements
in thermal efficiency are applied to future cooking fuel scenarios as a form of sensitivity
analysis. This aspect of the present study affords a better understanding of the potential gains to
be made by fuel mix substitutions versus adoption of more advanced stove technologies. Values
describing current and future cookstove technology use scenarios are presented in Table 3-1
through Table 3-4 for India, China, Kenya, and Ghana, respectively.
References for current stove use and current thermal efficiency values can be found in
Table 2-7 and Table 2-8. Future stove use scenarios assume that 100 percent of cooking fuels are
burned in either improved or modern cookstove designs. This assumption serves as an upper
bound regarding the possible adoption of improved technologies. The proposed future thermal
efficiency is assumed equal to the maximum reported stove thermal efficiency value for each
stove group.
Table 3-1. Adoption of Improved Stove Technologies and Thermal Efficiency in India
Current
Future
Current
Future
Fuel
Stove Type
Stove
Stove
Thermal
Thermal
Use1
Use
Efficiency2
Efficiency2
Coal
Traditional
100%
100%
16%
16%
Dung
Traditional
95%
0%
9%
9%
Improved
5%
100%
11%
13%
Crop Residue
Traditional
95%
0%
11%
12%
Improved
5%
100%
16%
22%
Firewood
Three-stone
9%
0%
18%
18%
Traditional
85%
0%
17%
23%
Improved
6%
100%
24%
29%
Charcoal from
Wood
Traditional
100%
0%
18%
18%
Kerosene
Improved, Wick
54%
54%
50%
50%
Improved,
Pressure
46%
46%
47%
47%
LPG
Modern
100%
100%
55%
57%
Natural Gas
Modern
-
100%
-
61%
Electricity
Modern
100%
100%
59%
80%
Sugarcane Ethanol
Modern
100%
100%
-
53%
Biogas
Modern
-
100%
-
56%
Biomass Pellets
Modern
-
100%
-
53%
1 Estimates of current stove technology use adapted from Smith et al. 2000
2 Supporting documentation and calculations available in SI1.
3-2
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-2. Adoption of Improved Stove Technologies and Thermal Efficiency in China
Current
Future
Current
Future
Current Stove
Use Reference2
Fuel
Stove Type
Stove
Stove
Thermal
Thermal
Use
Use
Efficiency1
Efficiency1
Coal, Powder
Traditional
62%
0%
10%
14%
IARC 2010
Improved
38%
100%
17%
17%
IARC 2010
Coal, Briquette
Improved
100%
100%
32%
37%
-
Coal,
Traditional
62%
0%
20%
23%
IARC 2010
Honeycomb
Improved
38%
100%
45%
47%
IARC 2010
Crop residue
Traditional
23%
0%
11%
11%
IARC 2010
Improved
77%
100%
17%
19%
IARC 2010
Firewood
Traditional
23%
0%
12%
12%
IARC 2010
Improved
77%
100%
16%
24%
IARC 2010
Kerosene
Improved
100%
100%
45%
49%
-
LPG
Modern
100%
100%
47%
54%
-
Natural Gas
Modern
100%
100%
57%
59%
-
Coal Gas
Modern
-
100%
-
46%
-
Electricity
Modern
100%
100%
59%
80%
-
Sugarcane
Ethanol
Modern
-
100%
-
53%
-
Biogas
Modern
-
100%
-
56%
-
Biomass Pellets
Modern
-
100%
-
53%
-
1 Supporting documentation and calculations available in SI1.
2 Estimates of current stove technology use adapted from the listed reference(s)
Table 3-3. Adoption of Improved Stove Technologies and Thermal Efficiency in Kenya
Current
Future
Stove Use
Current
Future
Current Stove
Use Reference2
Fuel
Stove Type
Stove
Use
Thermal
Efficiency1
Thermal
Efficiency1
Firewood
Three-stone
80.0%
0%
13%
15%
Githiomi et al.
2012, SEI2016
Traditional
17.7%
0%
11%
18%
Dalberg 2012
Improved
2.3%
100%
19%
27%
Clough 2012
Charcoal
Traditional
45%
0%
14%
18%
Clough 2012
Improved
55%
100%
25%
26%
Clough 2012
Kerosene
Improved
100%
100%
52%
52%
-
LPG
Modern
100%
100%
49%
53%
-
Electricity
Modern
100%
100%
59%
69%
-
Sugarcane Ethanol
Modern
-
100%
-
53%
-
Biogas
Modern
100%
100%
48%
52%
-
Biomass Pellets
Modern
-
100%
-
53%
-
1 Supporting documentation and calculations available in SI1.
2 Estimates of current stove technology use adapted from the listed reference(s)
3-3
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-4. Adoption of Improved Stove Technologies and Thermal Efficiency in Ghana
Fuel
Stove Type
Current
Stove
Use
Future
Stove
Use
Current
Thermal
Efficiency1
Future Thermal
Efficiency1
Current Stove
Use Reference2
Crop
Residue
Traditional
100%
0%
11%
18%
Firewood
Three-stone
80%
0%
13%
15%
Energica 2009
Traditional
20%
0%
11%
18%
Energica 2009
Improved
0
100%
-
27%
Energica 2009
Charcoal
Traditional
64%
0%
14%
22%
Energica 2009
Improved
36%
100%
23%
23%
Energica 2009
Kerosene
Improved
100%
100%
46%
52%
-
LPG
Modern
100%
100%
49%
55%
-
Electricity
Modern
100%
100%
59%
80%
-
Sugarcane
Ethanol
Modern
-
100%
-
53%
-
Biogas
Modern
100%
100%
48%
52%
-
Biomass
Pellets
Modern
-
100%
-
53%
-
1 Supporting documentation and calculations available in SI1.
2 Estimates of current stove technology use adapted from the listed reference(s)
3.1.1 India Cooking Fuel Mix Scenarios
Table 3-5 provides a name and basic description for each of the five cooking fuel mix
scenarios developed for India. Table 3-6 introduces the cooking fuels that comprise each
scenario and compares those values to the baseline (current) cooking fuel mix. A full description
of each scenario is provided in the subsections that follow. Details regarding fuel mix
development and documentation are available in SI3.
Table 3-5. Cooking Fuel Mix Scenario Names and Descriptions for India
Scenario
Scenario Name
Scenario Description
(1)
Current
Current fuel mix, recent year
(2)
BAU 2040
Projected 2040 fuel mix adapted from IEA 2015
(3)
Improved Biomass
Assumes increased use of improved biomass options such as
biogas, biomass pellets, and ethanol
(4)
Increased Electricity
Electricity use displaces the use of LPG, kerosene, and the
traditional biomass fuels
(5)
Diverse Modern Fuels
Promotes a balanced use of modern fuels and improved
stove technologies
3-4
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-6. Cooking Fuel Mix Scenarios Evaluated for India
BAU
Improved
Increased
Diverse
Fuel Type
Current
2040
Biomass
Electricity
Modern Fuels
d)1'2
(2)3
(3)4
(4)4
(5)
Hard Coal
1.2%
0.80%
0.90%
0.90%
-
Dung Cake
11%
4.5%
2.8%
2.2%
1.0%
Crop Residue
8.9%
3.8%
2.0%
1.6%
1.0%
Firewood
49%
21%
13%
17%
5.0%
Charcoal from Wood
1.2%
3.3%
6.6%
3.3%
6.6%
Kerosene
3.2%
-
2.0%
6.0%
-
LPG
25.2%
52%
48%
38%
38%
Natural Gas
-
6.8%
-
-
3.0%
Electricity
0.40%
3.2%
8.4%
25%
20%
Sugarcane Ethanol
-
-
6.0%
-
6.0%
Biogas from Cattle
Dung
0.40%
2.2%
4.4%
2.2%
4.4%
Biomass Pellets
-
3.3%
6.6%
3.3%
15%
TOTAL4
100%
100%
100%
100%
100%
Sources: 1 Dalberg 2013b,2 Venkatarman et al. 2010,3IEA 2015, 4IEA 2007
4 Columns may not total 100 due to rounding, unrounded numbers available in SI
3.1.1.1 Current Fuel Mix (India Cooking Fuel Mix Scenario 1)
The current cooking fuel mix estimate for India is based on 2011 census data collected by
the Government of India. Nearly 70 percent of India's population, mostly in rural areas, still rely
on dung, crop residues, and firewood to provide their cooking energy. Firewood contributes just
over 49 percent of the total cooking fuel mix, while dung and crop residues contribute eight and
nine percent, respectively. Coal and charcoal together make up only 2.4 percent of the cooking
fuel mix. LPG is used extensively at the national level, providing cooking energy for
approximately 25 percent of households. Kerosene is used in much more limited quantities (three
percent of the fuel mix). Unlike China, electricity is used only sparsely, providing 0.4 percent of
cooking energy.
3.1.1.2 Potential Future Scenarios
Business-as-Usual (BAU) 2040 (India Cooking Fuel Mix Scenario 2)
The IEA projected cooking fuel mix for 2040 sees a 45 percent decrease in reliance on
biomass fuels as compared to the current baseline scenario. Traditional fuels are expected to
comprise under 30 percent of the overall fuel mix. The IEA reference does not specify the type
of traditional fuel, so the original ratios of coal, firewood, crop residue, and dung use have been
maintained and applied to the lower percentage of cooking energy provided by traditional fuel
sources. Reliance on direct combustion of fossil fuels nearly doubles, in this scenario, increasing
to comprise 59 percent of the fuel mix. LPG provides the majority of fossil-based heat with a
seven percent contribution from natural gas, a fuel that was not considered in the Phase I study.
Electricity use increases by a factor of eight but still provides only three percent of the fuel mix
(IEA 2015).
3-5
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Improved Biomass (India Cooking Fuel Mix Scenario 3)
The Improved Biomass scenario is adapted from the IEA's cooking energy projections
for the year 2030 (IEA 2007). The original IEA 2030 fuel mix is similar to the fuel mix projected
by the IEA for the year 2040 (IEA 2015), with a higher expectation for the increased use of
electricity. In the IEA 2030 scenario, reliance on electricity as a cooking fuel rises to just over
eight percent. The original values projected by IEA have been adjusted to provide greater
differentiation with the 2040 scenario, exploring the effect of a more aggressive transition from
traditional to improved biomass sources. Expectations for LPG use have been maintained as
provided in the original scenario, with 47 percent of the fuel mix being provided by LPG. The
original IEA scenario values suggested that kerosene would rise to constitute eight percent of the
cooking fuel mix. This scenario assumes that the use of sugarcane ethanol will instead increase
to six percent of the cooking fuel mix, thereby leaving kerosene to contribute near its current
level of use. Assumed contributions from biogas, charcoal, and biomass pellets have doubled in
comparison to the 2040 scenario with fuel mix contributions of 4.4, 6.6, and 6.6 percent,
respectively. These increases are offset be a decreased reliance on firewood, crop residue, and
dung cake. Together, these three fuels constitute approximately 18 percent of the fuel mix.
Increased Electricity (India Cooking Fuel Mix Scenario 4)
The Increased Electricity scenario is based on a combination of the IEA 2030 and 2040
projections. However, with electricity access expanding to 96 percent of the population by 2030,
this scenario was developed to explore the effect of greater adoption of electric stove technology
than the IEA is projecting. The likelihood of this switch depends on the relative cost of
electricity versus other advanced fuel options, particularly LPG. In this scenario, it is assumed
that the projected increase in LPG and Kerosene use is reduced by 20 percent with the difference
being made up by the adoption of electric stoves. Reliance on the traditional biomass fuels is also
decreased by nearly ten percent in favor of electricity and kerosene use. Kerosene use is set at six
percent of the cooking mix in this scenario. Levels of advanced biomass use associated with the
IEA 2040 scenario are maintained here, with low reliance on crop residue and dung cake from
the 2030 scenario. Overall, electricity makes up just over 25 percent of the cooking fuel mix in
this scenario.
Diverse Modern Fuels (India Cooking Fuel Mix Scenario 5)
The Diverse Modern Fuels scenario presents a dramatic departure from the current
cooking energy mix, which takes cues from the LCA results that came out of the Phase I study.
The LCA fuel results showed that both biogas and biomass pellets were the best performing
cooking fuels in most impact categories. LPG also performed relatively well, especially in the
PM and BC impact categories, which are so crucial to human health. LPG is also an incredibly
convenient fuel with an attractive package of incentives being offered by the Government of
India. LPG is expected to comprise a large portion of any future cooking fuel mix. Together,
LPG and natural gas contribute 41 percent of the cooking fuel mix in the Diverse Modern Fuels
scenario. It is assumed that 20 percent of the cooking fuel mix is provided by electricity. Biogas
and charcoal are adopted at the same rate as specified in the Improved Biomass scenario. The
main difference from the other scenarios is the dramatic adoption of pelletized biomass fuel,
which increases to provide 15 percent of the cooking fuel mix. The potential economic savings
of a switch from LPG to biomass pellets has been demonstrated in some contexts (Thurber et al.
3-6
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
2014). It is assumed that a minimum amount of firewood, crop residue, and dung use will
continue past 2030. The minimal amount of coal use that previously existed is eliminated
entirely.
3.1.2 China Cooking Fuel Mix Scenarios
Table 3-7 provides a name and basic description for each of the five cooking fuel mix
scenarios for China. Table 3-8 introduces the cooking fuels that comprise each scenario and
compares those values to the current cooking fuel mix. No more recent values for the cooking
fuel mix were able to be found since the release of the Phase I study, so the Phase I and Phase II
current fuel mix estimates are identical. A full description of each scenario is provided in the
subsections that follow. Details regarding fuel mix development and documentation are available
in SI3.
Table 3-7. Cooking Fuel Mix Scenario Names and Descriptions for China
Scenario
Scenario Name
Scenario Description
(1)
Current
Current fuel mix, recent year
(2)
BAU 2030
2030 BAU cooking fuel projections
(3)
Increased Electricity
Electricity use displaces the use of LPG and Coal
(4)
Advanced Biomass and
Electricity
Coal use offset by adoption of electricity and advanced
biomass technology
(5)
Diverse Modern Fuels
Promotes a balanced use of modern fuels and improved
stove technologies
Table 3-8. Cooking Fuel Mix Scenarios Evaluated for China
Current
d)1'2
BAU
Increased
Advanced
Biomass &
Electricity
(4)3
Diverse Modern
Fuel Type:
2030
(2)3
Electricity
(3)3
Fuels
(5)3
Coal
29%
24%
12%
15%
5.9%
Coal Powder
14%
12%
5.9%
7.3%
3.0%
Coal Briquettes
7.2%
5.9%
3.0%
3.6%
1.5%
Honeycomb Briquettes
7.2%
5.9%
3.0%
3.6%
1.5%
Crop Residue
12%
5.2%
5.2%
2.0%
-
Firewood
15%
6.4%
6.4%
2.4%
-
Kerosene
0.30%
-
-
-
-
LPG
31%
45%
38%
33%
46%
Natural Gas
2.4%
-
-
-
15%
Coal Gas
-
12%
12%
9.3%
2.9%
Electricity
11%
8%
27%
19%
15%
Biogas
-
0.62%
0.60%
6.0%
6.0%
Biomass Pellets
-
-
-
13%
9.0%
TOTAL3
100%
100%
100%
100%
100%
Sources: 1 Dalberg 2014,2 NBSC 2008,3 adapted from Mainali et al. 2012
3 Columns may not total 100 due to rounding, unrounded numbers available in SI
3-7
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3.1.2.1 Current Baseline Scenario (China Cooking Fuel Mix Scenario 1)
The current fuel mix scenario for China is the same as the scenario used in the Phase I
report. More than half of China's population, mostly in rural areas, still rely on traditional solid
fuel feedstock for their cooking needs. The current fuel mix in China is dominated by the use of
three fuels: LPG, coal, and biomass. Each of these fuels comprises slightly less than one third of
the total fuel use. Nearly 11 percent of the population uses electricity as a cooking fuel. Only a
small percentage of the population uses kerosene or natural gas.
3.1.2.2 Potential Future Scenarios
BAU 2030 (China Cooking Fuel Mix Scenario 2)
The 2030 projections are a BAU scenario that projects no major policy changes out to the
year 2030. Despite this conservative approach, the BAU 2030 cooking fuel mix is dramatically
different from the cooking fuel mix that existed in 2005, the beginning of the author's study
period (Mainali et al. 2012). By 2030, biomass (firewood and crop residues) are expected to
contribute only a combined nine percent of the cooking fuel mix. Reductions in coal use are far
less substantial. Still nearly a quarter of cooking energy is provided by some form of coal. Coal
gas is expected to provide 12 percent of cooking energy, particularly in urban areas where it is
distributed via pipeline. LPG use increases to provide 45 percent of cooking energy, while
electricity use holds flat at approximately ten percent. Biogas is expected to contribute a small
portion, 0.6 percent, of cooking energy in rural areas.
Increased Electricity (China Cooking Fuel Mix Scenario 3)
The Increased Electricity scenario examines the effect of more widespread adoption of
electricity for household cooking. There are two major factors providing a rationale for this
scenario. The first is the existing presence of nearly universal access to electricity resources
throughout urban and rural China (World Bank 2012). The second factor is the observation of the
2030 BAU scenario that shows a continued, significant reliance on coal energy for cooking in
both urban and rural households. There is little indication in the literature that China has plans of
dramatically scaling back coal production in the foreseeable future. In fact, the majority of grid
projections in China for the period 2030 to 2050 rely on coal for between 47 and 73 percent of
electrical energy. There is, however, a possibility that China will pursue an aggressive upgrade
of their coal electricity generating technology, with the possible inclusion of carbon capture and
sequestration (IEA 2010, Zhou et al. 2011). Advanced coal burning technologies such as
supercritical generators and Integrated Gasification Combined Cycle (IGCC) coal plants have the
potential to reduce coal use while simultaneously cutting harmful air emissions. This cooking
fuel scenario assumes that 27 percent of cooking energy is supplied by electricity. This shift
allows a 50 percent reduction in the direct combustion of coal in households, combined with a 16
percent decrease in the use of LPG. The remainder of cooking energy demand is consistent with
the values projected by Mainali et al. (2012).
Improved Biomass and Electricity (China Cooking Fuel Mix Scenario 4)
The baseline 2030 BAU scenario values are adjusted in this scenario to explore the effect
of policy support for advanced biomass stove use, combined with a 25 percent reduction in LPG
3-8
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
use in favor of electricity. The 2030 BAU scenario shows a two-thirds reduction in the use of
fuelwood and crop residue by the year 2030. This scenario supposes that a combination of
factors works to hold the contribution of firewood and crop residues within the cooking fuel mix
to a constant level. A recent survey conducted by The World Bank (2013) indicates that the
production of advanced biomass stoves has increased rapidly since 2005. The increased thermal
efficiency of these stoves allows the same delivery of heating energy, while decreasing the
required demand for biomass. Biogas use is scaled up to utilize two-thirds of the national biogas
potential, which is estimated to be approximately nine percent of cooking energy (World Bank
2013). The scenario also assumes that one quarter of the increase in modern fuel use, as it is
modeled in the 2030 BAU scenario, accrues to electricity instead of LPG. Like the Increased
Electricity scenario above, this shift would allow China to continue leveraging their significant
domestic coal resources while reducing in-household exposure to HAPs. Combined, these shifts
facilitate nearly a two-thirds reduction in reliance on solid coal combustion in the household.
Diverse Modern Fuels (China Cooking Fuel Mix Scenario 5)
The Diverse Modern Fuels scenario assumes that a balanced mix of modern fuels and
advanced biomass options are adopted instead of coal, dung, and firewood combustion. The
scenario assumes a 75 percent reduction in coal use, which leaves six percent of households still
reliant on this fuel source. A five percent increase in the use of electricity is assumed over the
projected ten percent contribution from the BAU 2030 scenario. It is assumed that 100 percent of
biomass fuel consumption projected by the BAU 2030 scenario is consumed in advanced pellet
stoves. Four percent of households cook with biogas, and the remainder of cooking energy is
provided by modern liquid and gas options that tend to be favored as household incomes increase
(Malla and Timilsina 2014). Reliance on coal gas is assumed to be limited to just three percent of
cooking energy. Unlike the BAU projections, this scenario assumes a large increase in natural
gas use, which offsets a portion of LPG production. This switch is supported by IEA projections
that show both domestic production and imports of natural gas increasing significantly between
now and 2030 (IEA 2007). The ratio of LPG to natural gas adoption is assumed to be three-to-
one. Together, these two fuels account for 61 percent of cooking energy.
3.1.3 Kenya Cooking Fuel Mix Scenarios
Table 3-9 provides a name and basic description for each of the five fuel mix scenarios
for Kenya. Table 3-10 introduces the cooking fuels that comprise each scenario and compares
those values to the current cooking fuel mix. A full description of each scenario is provided in
the subsections that follow. Details regarding fuel mix development and documentation are
available in SI3.
Table 3-9. Cooking Fuel Mix Scenario Names and Descriptions for Kenya
Scenario
Scenario Name
Scenario Description
(1)
Current
Current fuel mix, recent year
(2)
BAU 2030
Applies current trends to the 2030 urban/rural population
(3)
Ghana Transition (for
Models future cooking fuel mix shifts in Kenya based on
Ghana's fuel mix development since the mid-1990s when
Kenya)
biomass and LPG use rates were like those found in Kenya
today
3-9
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-9. Cooking Fuel Mix Scenario Names and Descriptions for Kenya
Scenario
Scenario Name
Scenario Description
(4)
Slow Transition
Based on a slower transition to modern fuels and improved
cookstoves than is in indicated by the Ghana Transition (for
Kenya) scenario
(5)
Diverse Modern Fuels
Promotes a balanced use of modern fuels and improved stove
technologies
Table 3-10. Cooking Fuel Mix Scenarios Evaluated for Kenya
Fuel Type
Current
(D1
BAU 2030
(2 )2
Ghana
Transition
(for Kenya)
(3)2
Slow
Transition
(4 )2
Diverse Modern
Fuels
(5)2
Firewood
65%
68%
46%
56%
11%
Charcoal
17%
16%
27%
21%
17%
Kerosene
12%
10%
1.3%
6.2%
1.3%
LPG
5.0%
4.4%
24%
14%
36%
Electricity
0.80%
0.70%
1.0%
1.0%
13%
Biogas
0.70%
0.70%
1.0%
1.0%
3.0%
Biomass Pellets
-
-
-
-
19%
TOTAL3
100%
100%
100%
100%
100%
Sources: 1 KNBS 2012,2 GVEP 2012a, CBS 2002
3 Columns may not total 100 due to rounding, unrounded numbers available in SI
3.1.3.1 Current Baseline Scenario {Kenya Cooking Fuel Mix Scenario 1)
Firewood is the predominant cooking fuel used in Kenya today, providing nearly 65
percent of national cooking energy. Charcoal, another wood-based fuel, provides a further 17
percent of cooking energy. The two main references for the current cooking fuel mix disagree
regarding current reliance on kerosene use with estimates of both five and 12 percent (KNBS
2012, Dalberg 2013a). Both fuel mixes refer to the data year 2009, with the national statistics
being the preferred source for this phase of work. Use of the 12 percent kerosene estimate
establishes a conservative baseline, and the discrepancy signals that Kenya may be in the process
of accelerating adoption of modern fuels such as LPG. Electricity and biogas provide 0.8 and 0.7
percent of cooking energy, primarily in urban areas.
3.1.3.2 Potential Future Scenarios
BAU 2030 (Kenya Cooking Fuel Mix Scenario 2)
The BAU 2030 scenario is included to provide a conservative estimate of future cooking
energy needs. The scenario applies current urban and rural cooking fuel use patterns to the
projected urban and rural populations in the year 2030 in the absence of other pressures on the
cooking fuel mix, as documented in SI3. The shifts in fuel mix do not exceed a few percentage
points for any given fuel.
3-10
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Ghana Transition (Kenya Cooking Fuel Mix Scenario 3)
The cooking fuel mix for this scenario is created using the recent experiences of Ghana in
combination with Kenya's own data characterizing the early stages of a shift towards LPG and
charcoal use. Charcoal use has been slowly rising in Kenya since the late 1980s when this fuel
provided approximately seven percent of cooking energy (CBS 2002). Projecting forward along
the same linear trend line, this scenario estimates that charcoal use could contribute 27 percent of
cooking energy by the year 2030. This is the same level of market penetration that Ghana
reached in the late 1980s and 1990s (GLSS2 2008, GLSS3 1995). LPG use in Kenya has risen
rapidly from very low usage to comprise approximately five percent of the Kenyan cooking fuel
mix (2010). A similar transition was observed in Ghana between the late 1980s and the turn of
the 21st century. Modeling Kenya based on the experience of Ghana following that period, the
use of LPG can be expected to rise to nearly 25 percent of the overall cooking fuel mix. The use
of kerosene is assumed to decline to five percent of the national cooking fuel mix. The remainder
of the previously mentioned increases in fuel consumption are offset by decreased reliance on
firewood, whose use is projected to fall to 46 percent of the cooking fuel mix by the year 2030.
This trend is consistent with Ghana's experience, where gains in LPG and charcoal contributed
to a decrease in demand for firewood.
Slow Transition (Kenya Cooking Fuel Mix Scenario 4)
The stability of Kenya's cooking energy mix and their slower pace of urbanization over
the past three decades is an indicator that Kenya may move more slowly away from firewood
than Ghana's experience suggests. The Slow Transition scenario uses the same approach as that
developed in the previously described Ghana Transition (for Kenya) scenario, but the rate of
conversion to charcoal and LPG use is cut in half. In this scenario, kerosene use declines more
slowly to provide six percent of the cooking fuel mix, consistent with the slow decline in
kerosene use that Kenya has experienced thus far. Still, LPG use triples its contribution to the
cooking fuel mix, rising to provide 14 percent of total cooking energy. Charcoal use increases by
25 percent to provide 21 percent of cooking energy. It is assumed that the marginally used
alternative fuels, electricity and biogas, see an increase in use but remain as minor contributors to
the total fuel mix. To compensate for increased use of charcoal and modern fuels, the use of
firewood decreases by 14 percent to contribute approximately 56 percent of the national cooking
fuel.
Diverse Modern Fuels (Kenya Cooking Fuel Mix Scenario 5)
The Diverse Modern Fuels scenario is designed to explore the potential benefits and
burdens of a more rapid shift to a diverse portfolio of modern cooking fuels and improved stove
designs. Use of LPG, electricity, and biomass pellets are all assumed to rise significantly in this
scenario. African nations such as South Africa and Zimbabwe, which, respectively, generate 85
and 73 percent of urban cooking energy from electricity, indicate that widespread adoption of
electricity as a cooking energy is possible. However, it would take time to reach this level of
market penetration, so this scenario assumes that 30 percent of urban cooking energy is provided
by electricity. Electricity use is assumed to expand much more modestly in rural areas,
eventually comprising five percent of the rural cooking fuel mix. LPG use rises to provide
approximately 36 percent of national cooking energy, while reliance on kerosene falls to just 1.3
percent. Charcoal use remains nearly constant with its observed level of use in the current
3-11
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
cooking fuel mix. Charcoal use is assumed to continue expanding in rural areas to satisfy 22
percent of total rural cooking energy, while its use falls in urban areas in favor of other modern
fuels. All firewood is assumed to be eliminated from use in urban areas with a fraction of that
demand being replaced by the use of biomass pellets. The use of biomass-based fuels decreases
only modestly in rural areas, but the adoption of wood pellet stoves increases dramatically to
provide 22 percent of rural cooking energy. This shift increases the efficiency of biomass
utilization while simultaneously providing a rural employment opportunity for those workers
who used to satisfy the urban charcoal market.
3.1.4 Ghana Cooking Fuel Mix Scenarios
Table 3-11 provides a name and basic description for each of the four fuel mix scenarios
for Ghana. Table 3-12 introduces the cooking fuels that contribute to each scenario and compares
those values to the current cooking fuel mix. A full description of each scenario is provided in
the subsections that follow. Details regarding fuel mix development and documentation are
available in SI3.
Table 3-11. Cooking Fuel Mix Scenario Names and Descriptions for Ghana
Scenario
Scenario Name
Scenario Description
(1)
Current
Current fuel mix, recent year
(2)
BAU 2030
Applies current trends to the 2030 urban/rural population
(3)
Moderated Growth
Reflects a slowdown in the current growth of LPG
(4)
Fast Growth
Based on a continued rapid growth in LPG use
(5)
Diverse Modern Fuels
Promotes a balanced use of modern fuels and improved stove
technologies
Table 3-12. Cooking Fuel Mix Scenarios Evaluated for Ghana
Moderated
Diverse Modern
Current
BAU 2030
Growth
Fast Growth
Fuels
(1)
(2)
(3)
(4)
(5)
Biomass
46%
39%
26%
14%
16%
Firewood
46%
38%
26%
14%
16%
Crop Residue
0.41%
0.30%
0.37%
0.26%
-
Charcoal
32%
35%
32%
23%
20%
Kerosene
0.2%
-
-
-
-
LPG
22%
26%
40%
61%
30%
Electricity
0.32%
-
1.9%
1.9%
21%
Biomass Pellets
-
-
-
-
13%
Total1
100%
100%
100%
100%
Sources: All scenarios based on GLSS1 through GLSS6, GSS 2012, and Dalberg 2013a
1 Columns may not total 100 due to rounding, unrounded numbers available in SI
3.1.4.1 Current Baseline Scenario (Ghana Cooking Fuel Mix Scenario 1)
The current cooking fuel mix in Ghana relies on unprocessed firewood for 43 percent of
cooking energy, charcoal for 33 percent, and LPG for 23 percent. LPG is primarily used in urban
3-12
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
areas (GLSS6 2014). Electricity, kerosene, and crop residues contribute the remaining one
percent of cooking energy.
3.1.4.2 Potential Future Scenarios
BAU 2030 (Ghana Cooking Fuel Mix Scenario 2)
The BAU 2030 scenario is included to provide a conservative estimate of future cooking
energy needs. The scenario applies current urban and rural cooking fuel use patterns to projected
urban and rural populations in the year 2030 in the absence of other pressures on the cooking
fuel mix, as documented in SI3. Reliance on traditional biomass falls to provide approximately
39 percent of cooking energy, while charcoal use increases to 35 percent. LPG use also increases
a few percentage points to provide nearly 26 percent of cooking energy.
Moderated Growth (Ghana Cooking Fuel Mix Scenario 2)
The Moderated Growth scenario assumes that the use of firewood continues to fall at a
rate just below that observed over the period from 1980 to the present. This scenario predicts that
the contribution of firewood to the cooking energy mix drops to just over 25 percent by the year
2030. In this scenario, charcoal use is predicted to continue its pattern of flat growth over the
same period. LPG use is expected to continue to rise as it has in the recent past, but the pace of
growth slows down. In the Moderated Growth scenario LPG is assumed to provide 40 percent of
cooking energy. There are several reasons supporting the possibility of such a scenario. The
literature recognizes a complex set of factors, which contribute to a household's selection of
cooking fuel (Malla and Timilsina 2014). The realities of cost, taste preference, and fuel
availability are but a few of the many factors that could challenge the current rapid growth in
LPG fuel use. In particular, Ghana has a long history of cooking over firewood and charcoal,
both of which lend a desirable flavor to many traditional dishes. As the experience of other
countries has shown, the complete elimination of traditional fuels can be a long process. On top
of this, it is important to consider the stated goals of Ghana's government and other
organizations to enhance access to improved cookstoves and efficient kiln technology (GEC
2006).
Fast Growth (Ghana Cooking Fuel Mix Scenario 3)
Since 1987, the share of LPG in the national fuel mix has risen from 0.8 to 22 percent.
Charcoal use has increased at a more moderate pace rising from 26 to 31 percent by the year
2000 and has remained relatively flat since that time. All the while, the use of unprocessed
firewood has continued its steady decline from 70 to 40 percent of the national cooking fuel mix.
If these trends were to continue unabated, the use of LPG could provide over 60 percent of
cooking energy by the year 2030. For this to occur, the decrease in reliance on solid wood fuel
would have to quicken slightly. In this scenario, it is assumed that the use of wood in urban areas
drops to zero from its current level of approximately 14 percent. Rural wood use would need to
drop far more dramatically, from 75 to 14 percent. This scenario also assumes that charcoal use
in urban areas is reduced in favor of LPG, while it holds roughly constant in rural regions. There
is evidence that charcoal is being displaced by LPG in urban areas. Between the 5th and 6th
Ghana Living Standards Surveys, reliance on charcoal in urban areas dropped by nearly 20
percent with LPG absorbing much of that energy demand. Both Ghana's government and
3-13
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
international organizations are supporting the shift to modern fuels, and intervention from these
actors is likely to be required if such a scenario is to be realized.
Diverse Modern Fuels (Ghana Cooking Fuel Mix Scenario 4)
The Diverse Modern Fuels scenario is designed to explore the potential benefits and
burdens of a more rapid shift to a diverse portfolio of modern cooking fuels and improved stove
designs. LPG use is projected to grow to comprise 30 percent of the cooking fuel mix, which
represents a moderated rate of LPG growth in favor of other modern fuels. Reliance on wood
resources falls to below 30 percent of cooking fuel energy, and 45 percent is assumed to be
consumed in improved biomass pellet stoves. Charcoal use is assumed to drop, providing
approximately 20 percent of cooking energy in the year 2030. Use of electricity as a cooking fuel
rises dramatically to comprise 20 percent of the cooking fuel mix, as it offsets charcoal and LPG
use.
3.2 Electrical Grid Scenario Development
The fuel mix that underlies the electricity grid is a key factor in the environmental impact
of electric powered cookstoves and upstream manufacturing associated with cooking fuels that
require industrial processing. As a sensitivity analysis within this study, a range of projected grid
mixes for India, China, Kenya, and Ghana have been included, following a review of the
available literature. Important factors that influence the adoption of specific fuels include
population and economic growth, changes in the relative cost between fuels and generation
technologies, and national and international government policies concerning environmental
management and the trade of goods. Scenarios included in the sensitivity analysis range from
those based on a moderate BAU perspective to scenarios that embrace climate change mitigation
and dramatically pursue electricity production based upon renewable fuels.
It is not just electricity fuel mix that is expected to change in the coming decades.
Dramatic shifts in generation technology promise to wring more kilowatt hours out of each unit
of fuel burned. In particular, the pursuit of advanced coal burning technologies such as
supercritical, ultra-supercritical, and IGCC generators could have a dramatic effect on emissions
even if the proportion of the grid fueled by coal remains high. The possibility of carbon capture
and storage (CCS) also provides an attractive option for countries that have ample coal resources
and well established production chains for these commodities. The possibility of adopting
advanced generation technologies is also considered in these scenarios.
A standard generating technology has been assumed for each fuel that is consistent with
those used in the first phase of this study. For example, the vast majority of coal-based power
plants in India and China today rely on subcritical generation technology. The following list
describes adaptations made to base coal and natural gas generation technologies to reflect
advancements expected to figure prominently into the electricity scenarios over the next ten to 30
years. Documentation of emission adjustments incorporated in the LCI for each electricity
generation unit process are included in SI4.
3-14
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Natural Gas - Efficient
A number of the scenarios generated by The Energy and Resources Institute (TERI) in
India predict the uptake of advanced gas combustion technologies. Natural gas combined cycle
(NGCC) is one example of such advancements. As foreseen by TERI (2006), the efficient natural
gas unit process is modeled as having a 39.4 percent generator efficiency as compared to current
natural gas generator efficiency of 34.5 percent.
Natural Gas + CCS
Natural Gas with CCS is modeled as being based on generation technology with a
thermal efficiency of 39.4 percent. Due to the energy penalty of CCS, the effective thermal
efficiency is reduced to 33.5 percent., which constitutes an approximate 15 percent increase in
fuel demand per delivered kWh. The CCS system is assumed to capture 90 percent of CO2
emissions, while NOx emissions increase by a factor of 1.15. Total life cycle carbon emissions
are reduced by 79 percent per unit of delivered energy (Odeh and Cockerill 2008).
Coal Supercritical
Supercritical coal power plant technology has an associated thermal efficiency of 39.6
percent. This increase in efficiency drives a 10.6 percent reduction in life cycle CO2 emissions
relative to subcritical generation. Increased combustion efficiency also decreases emission of
NOx, SOx, and PM by 84, 88, and 88 percent, respectively, relative to subcritical reactors (Odeh
and Cockerill 2008).
Coal Ultra-Supercritical
Ultra-supercritical generation is modeled as having a thermal efficiency of 43 percent,
which yields a 13 percent reduction in life cycle CO2 emissions. Due to a lack of data, the
emission reductions for NOx, SOx, and PM are modeled as being the same as those associated
with the supercritical reactor.
Coal IGCC
Power plants using IGCC technology use a combination of gas and steam turbines to
achieve higher electrical efficiency per unit of fuel. This increased efficiency leads to a reduction
in GHG emissions as a result of burning less fuel. The thermal efficiency of an IGCC reactor is
assumed to be 37.2 percent. The combustion process is also more efficient and reductions of 96,
96, and 99 percent are achieved, as compared to conventional technology, for SOx, NOx, and
PM, respectively (Odeh and Cockerill 2008, Beer 2005).
Coal + CCS
CCS technology is assumed to be paired with supercritical generating facilities. The
addition of CCS facilities and additional emission control features, necessary to control SOx for
the benefit of efficient CCS, yields an effective reduction in thermal efficiency. Therefore, more
fuel must be burned per kWh of electricity produced, but emissions per unit fuel combustion are
dramatically reduced CCS facilities are assumed to be able to sequester 90 percent of
combustion-related CO2 emissions. SOx and PM emissions are reduced to just 0.12 percent and
3-15
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
six percent of those associated with subcritical coal generation technology. NOx emissions are
also reduced compared to conventional technology without CCS, but they increase slightly
relative to the supercritical reactor without CCS.
The following subsections describe the projected electrical grid mixes for each nation
studied that are considered in this report.
3.2.1 India Electrical Grid Scenarios
Figure 3-1 shows nine potential future Indian electrical grid mixes and compares them to
the most recent IEA estimate of the Indian electrical grid mix for 2013 (IEA 2013a). Projections
for the years 2021-2050 have been made by TERI, the U.S. Energy Information Administration
(EIA), the IEA, and researchers at Imperial College in London. Scenarios for the year 2021 and
more conservative fuel mix shifts are grouped to the left of the figure with those for later years
and those encompassing more dramatic changes to the structure of the underlying fuel mix being
grouped to the right. Generation technologies considered in this study are also depicted by
changing the pattern of the bar while keeping the color constant, which allows readers to see
both shifts in the fuel mix and the generation technology used.
Seventy-three percent of the current Indian electrical grid is fueled by the burning of coal
in subcritical generators. Hydropower and natural gas provide a further 12 and five percent of
electricity, respectively. Oil/diesel, nuclear and renewables each provide between two and three
percent of electricity (IEA 2013a).
TERI is an Indian research group that has produced five projections for the electricity
fuel mix between 2021 and 2031. The TERI scenarios are based on generating capacity, so load
factors taken from the IEA for India were used to estimate electricity production from each
source. All TERI scenarios foresee continued reliance on coal-based electricity over the next 15
years. Their 2021 BAU scenario projects that the coal share of the fuel mix drops to just over 50
percent, with the difference being made up by hydro, natural gas, and nuclear. In this scenario,
hydropower provides 22 percent of all electricity, natural gas 16 percent, and reliance on nuclear
doubles to provide nearly seven percent of electricity. Limited adoption of supercritical generator
technology is anticipated. If India continues along this track, the share of coal in the electricity
fuel mix is again expected to rise to almost 70 percent by the year 2031 to satisfy a nearly
twofold increase in national energy demand. Use of renewables such as wind and biomass are
expected to drop from their current level of three percent to provide less than 0.5 percent of
electricity by 2031. Reliance on hydropower, natural gas, and nuclear all rise slightly in the 2031
BAU scenario (TERI 2006).
TERI's hybrid and efficiency scenarios for the year 2021 project a more rapid adoption of
advanced coal generation technology, with their models showing a preference for IGCC
generators. The use of nuclear increases to provide nearly 15 percent of electricity in the hybrid
scenario, while hydropower supplies approximately 25 percent of electricity in both scenarios.
The efficiency scenario projects lower adoption of nuclear and continues to rely on coal for 49
percent of electricity. Over two-thirds of coal electricity is generated by the more efficient IGCC
power plants. The use of natural gas also increases in both the hybrid and efficient scenarios to
comprise between 15 and 16 percent of the fuel mix. The 2031 TERI Efficiency scenario also
3-16
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
projects that reliance on coal will pick up between the years 2021 and 2031 to meet a rapidly
increasing demand for electricity. While coal is projected to provide just under 65 percent of
electricity in 2031, over 60 percent of that electricity comes from more efficient IGCC power
plants. Reliance on hydropower, natural gas, nuclear, and renewables shows a relative decrease
as compared to the efficient 2021 scenario (TERI2006).
The EIA 2030 scenario is similar in its predicted changes to what is observed in the TERI
BAU scenario for 2021. Reliance on subcritical coal power plants contracts to provide 58 percent
of electricity in 2030. Nuclear and natural gas use expand to provide over 11 and eight percent of
electricity, respectively. Reliance on hydropower and alternative renewables does not change
significantly from the hydropower and alternative renewables currently in use. The IEA 2050
scenario is similarly conservative with the use of coal expected to persist at a level near 70
percent. Hydropower is displaced partially by natural gas in the IEA 2050 scenario (IEA 2010).
Both the Low Carbon and Blue Map 2050 scenarios represent more radical departures
from the current state of electrical generation in India. Both predict a widespread embrace of
wind and solar technology. The Low Carbon scenario indicates that fully 42 percent of India's
electricity could be provided by solar energy in the year 2050 (Gambhir et al. 2012). The use of
coal disappears completely in this scenario while use of nuclear, natural gas, and wind energy all
rise. The IEA Blue Map scenario projects that 18 percent of electricity will still be generated
using coal, but the majority of this 18 percent is subject to CCS. The use of nuclear rises to
provide over 26 percent of electricity. Natural gas use rises to provide 16 percent of electricity
with the adoption of more efficient turbines for two-thirds of this generating capacity (IEA
2010). Hydropower still provides approximately ten percent of electricity in both grids.
3-17
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
2013 Grid TERI BAU - TERI Hybrid - TERI TERIBAU- EIA-2030 TERI IEA-2050 Low Carbon - IEA Blue Map
(IEA) 2021 2021 Efficiency- 2031 Efficiency- 2050 - 2050
2021 2031
¦ Nuclear
¦ Oil/Diesel
¦ Coal, Subcritical
^ Coal, CCS
\ Coal, Supercritical
Coal, IGCC
¦ Gas
% Gas, CCS
: Gas, Efficient
¦ Hydro
Biomass
Geothermal
Wind
Solar
Figure 3-1. Potential future electrical grid mixes in India.
3-18
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3.2.2 China Electrical Grid Scenarios
The future of the Chinese electricity sector is of great interest to the international
community and has been reported extensively. Ten potential future grid mixes were found in the
literature as reported by the IEA, the U.S. EIA, Lawrence Berkeley National Laboratory
(LBNL), and the Boston Consulting Group (BCG). Figure 3-2 depicts these potential future
Chinese grid mixes and compares them to the most recent IEA estimate of the Chinese electrical
grid mix for 2013 (IEA 2013b). Scenarios for the year 2030 and more conservative fuel mix
shifts are grouped to the left of the figure, while those for the year 2050 with more dramatic
changes to the structure of the underlying fuel mix grouped to the right. Generation technologies
considered in this study are also depicted by changing the pattern of the bar while keeping the
color constant, which allows us to see both shifts in the fuel mix and the generation technology
used.
Currently, over 75 percent of Chinese electricity is produced using coal energy.
Hydropower is the next largest contributor providing almost 17 percent of electrical energy.
Renewables, natural gas, and nuclear each provide between 1.5 and three percent to round out
the rest of the grid.
The baseline and slow-shift BCG scenarios, the EIA 2030 and the IEA 2050 scenarios are
conservative in the sense that they do not predict dramatic departures from the current structure
of the Chinese grid. This is particularly true of the IEA 2050 scenario, given the longer
timeframe available to affect a shift. Use of natural gas, nuclear, and renewables all expand in
these four scenarios, but reliance on subcritical coal generation remains high, between 60 and 70
percent. Reliance on hydropower contracts slightly in these four scenarios, dropping from 15
percent in the current scenario to between six and 12 percent.
The LBNL Continued Improvement Scenario (CIS) 2030 scenario is interesting in that,
while the underlying fuel mix remains similar to the current scenario in China, we see a near
complete departure from subcritical generation technology towards supercritical and ultra-
supercritical generators. Combined use of renewables such as solar, wind, and biomass triples,
while the use of nuclear reactors expands six-fold to provide 13 percent of the electricity
demand. The LBNL Accelerated Improvement Scenario (AIS) anticipates a similar shift in
generation technology but combines that shift with a more rapid replacement of coal-fired
generation with alternative options. Under the AIS scenario renewables, natural gas,
hydropower, and nuclear expand to provide 12, 4, 17, and 19 percent of electricity, respectively
(Zhou et al. 2011).
The BCG Clean 2030 scenario anticipates that subcritical coal technology still provides
over 50 percent of China's electricity. Use of renewables expands to nearly 11 percent with over
three quarters of that electricity produced by wind. In this scenario, use of nuclear and natural
gas rises 13 and 15 percent, respectively. Reliance on hydropower falls by nearly half (Michael
et al. 2013).
The LBNL CIS 2050 scenario anticipates that 47 percent of electricity is generated by
coal with the major portion of that electricity generation occurring in ultra-supercritical
generators. Use of nuclear is projected to rise to 24 percent. Reliance on natural gas remains
3-19
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
relatively flat over time providing just under two percent of China's electricity. Reliance on
hydropower falls only slightly to between 11 and 12 percent. Other renewables, particularly
wind, constitute 14 percent of the fuel mix.
Both the LBNL AIS and IEA Blue Map scenarios are dramatic in their departure from the
status quo, although they follow two distinct pathways. The LBNL AIS scenario predicts the
adoption of nuclear technology throughout China, with over 50 percent of all electricity being
produced from this fuel in the year 2050. Reliance on coal is projected to drop to less than ten
percent of the total fuel mix. Renewables expand to provide 20 percent of all electricity, while
reliance on hydropower remains roughly constant with 16 percent of electricity being provided
by this source in the year 2050. The IEA Blue Map scenario also sees the use of nuclear power
increase, but more modestly, to the point that nuclear power provides 25 percent of electricity in
the year 2050. Use of coal drops to 15 percent and the major portion of that coal-generated
electricity is paired with CCS technology. Use of alternative renewables such as wind and solar
is projected to rise to provide 22 percent of all electricity. In this scenario, the use of natural gas
expands significantly to provide 23 percent of electrical energy (IEA 2010).
3-20
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
2013 Grid BCG Slow EIA - 2030 LBNL CIS - BCG Base - LBNL AIS - BCG Clean - IEA Baseline LBNL CIS - LBNL AIS - IEA Blue
(IEA) Shift-2030 2030 2030 2030 2030 -2050 2050 2050 Map - 2050
¦ Nuclear 11 Oil ¦ Coal, subcritical c Coal, supercritical ® Coal, ultra-supercritical = Coal, CCS ¦ Gas Gas, CCS ¦ Hydro Biomass Wind Solar
Figure 3-2. Potential future electrical grid mixes in China.
3-21
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3.2.3 Kenya Electrical Grid Scenarios
Currently, 23 percent of Kenyan households are connected to the electrical grid, while
less than one percent of households utilize electricity as their primary cooking fuel (World Bank
2012). Electricity access is concentrated in urban areas. There is a massive potential for
increased electricity demand, and with so much new generating capacity being built, the
possibility of shifting towards a dramatically different fuel mix is quite high. In Kenya, total
installed capacity is expected to increase from 1.3 GW in 2011 to between 17 and 30 GW in the
year 2031, a potential 23-fold increase (ROK 2011). Figure 3-3 depicts five potential future
Kenyan grid mixes and compares them to the most recent IEA estimate of the Kenyan electrical
grid mix for 2013 (IEA 2013c). Scenarios for the year 2030 are grouped to the left of the figure,
while those for the year 2040 are grouped to the right.
The current Kenyan grid is fueled by hydropower, fuel oil, and geothermal energy, with
each, respectively, providing 44, 31, and 23 percent of electricity. Biofuels provide the remaining
two percent.
The three 2031 scenarios were developed by the Kenyan government using a least cost
approach that adds generating capacity as is required to minimize the long run marginal cost of
electricity. The Kenyan government's model selects the least cost option from among a list of
project plans subject to a number of constraints such as the maximum number of a given plant
type that could feasibly be built annually. The base case (Least Cost - 2031) respects their full
list of constraints and is required to supply 18.9 additional GW of generating capacity by the
year 2031. The results of this model run indicate that reliance on geothermal energy increases by
nearly a third, the use of oil declines by nearly two-thirds, and the contribution from
hydroelectricity drops over tenfold. Wind, coal, natural gas, and nuclear technology increase
from current marginal levels to satisfy the remaining demand. Nuclear capacity increases the
most dramatically to supply 30 percent of electrical energy in the year 2031 (ROK 2011).
Republic of Kenya's (ROK's) low and high demand scenarios were developed using the
same least cost approach and vary only in the amount of generating capacity that must be
provided. The low demand scenario realizes slightly lower deployment of wind, hydropower,
and coal technology than does the base case, while the use of nuclear and geothermal satisfy a
larger portion of electricity demand. The high demand scenario relies heavily on nuclear
technology, which provides 49 percent of electrical energy in the year 2030. The contribution of
wind and geothermal energy are lower relative to the other scenarios developed by the Kenyan
government with these options providing eight and 22 percent of electricity, respectively. All
three scenarios are dramatically different from the 2013 grid mix with significant decreases in
reliance on oil and hydropower.
The McKinsey & Company 2040 scenario decides to avoid nuclear as a potential source
of electrical energy, based on their judgment that this option is not feasible in Sub-Saharan
Africa due to high up-front costs, potential community opposition, and lack of local trained
professionals in that sector (Castellano et al. 2015). The Kenyan government's least cost model
excluded solar as an option. By 2040, the McKinsey model foresees oil fueling only four percent
of Kenya's electricity production, down from 25 percent today. They too project a rapid decrease
in reliance on hydropower, but the drop is less dramatic, with 30 percent of electricity still being
3-22
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
provided by this source in the year 2040. Reliance on geothermal also falls, to provide only nine
percent of electrical energy. McKinsey expects the costs of solar to be much more competitive
by the year 2040, which facilitates its rising contribution to the grid mix. Use of natural gas rises
dramatically from zero today to provide almost half of Kenya's electricity in 2040 (Castellano et
al. 2015).
100%
3
•c
>
£
o
O
c
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
ranges have been used to create three scenarios in addition to a fourth low carbon-renewable
scenario as shown in Figure 3-4.
Table 3-13. Projected Electrical Grid Mix Contributions by Fuel for Ghana 2020
Fuel Source
Grid Mix Contribution (%)
Hydropower
39-49
Thermal
41-51
Nuclear
3-8
Renewables
5-11
Source: GEC 2006
The current electrical grid mix is dominated by hydropower, which provides 64 percent
of all electricity. Oil-powered generation supplies a further 26 percent of electrical energy, with
the remaining ten percent being provided by natural gas.
Ghana's government projects that reliance on hydropower will drop from its current high
level to provide between 39 and 49 percent of electricity in the year 2020. Thermal capacity will
expand with a preference for natural gas and coal power plants as an option to replace Ghana's
current reliance on fuel oil/diesel. Nuclear energy is projected to emerge in Ghana and to provide
between three and eight percent of electricity. Other renewables are expected to expand to
provide between five and 11 percent of electricity. These projections were initially made in the
year 2006, and based on the 2013 grid, it appears that this transition is happening more slowly
than expected, which likely pushes back the expected transition dates from 2020 to 2030 and
beyond.
The thermal scenario represents the maximum projected reliance on thermal generating
sources such as natural gas and coal and assumes that two-thirds of thermal power is fueled by
natural gas. This scenario also assumes adoption of renewables on the lower end of the projected
range, seven percent. No nuclear reactors are assumed to be built in this scenario.
The nuclear scenario assumes an eight percent reliance on nuclear energy for electricity
production. All other fuel sources are increased proportionally to their minimum projected value
in the grid mix. Natural gas is supposed to be the only source of fossil fuel-based thermal power
and it supplies 44 percent of all electricity. Hydropower provides 42 percent of electricity while
non-hydro renewables supply the remaining five percent of electricity demand.
The renewable scenario assumes that 49 percent of electricity is produced by
hydropower, which is the highest level foreseen by Ghana's energy commission (GEC 2006).
The lowest projected level of reliance on thermal generation is projected for this scenario, with
11 percent of electricity coming from non-hydro renewables.
As was done for Kenya, a Low Carbon scenario was included to represent a more
dramatic departure from the current electricity grid mix in line with the IEA Blue Map scenarios
for India and China. In this scenario, 26 percent of electricity is provided by non-hydro
renewables. The international renewable energy agency (IRENA 2013) reports that Ghana has
significant renewable energy resources, which are more than capable of satisfying Ghana's
3-24
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
demand. Given the extended time horizon for this scenario, electricity demand is assumed to
increase significantly. The share of hydropower drops to provide 35 percent of electricity under
this scenario as the best dam locations become developed favoring other fuel options. Nuclear is
assumed to provide 15 percent of Ghana's electricity, a rate that is twice the maximum
contribution projected by the government for the year 2020. The remaining 15 percent of
electricity demand is satisfied by natural gas. Half of gas production capacity is assumed to rely
on efficient power plant technology.
100%
90%
80%
g 70%
M 60%
o 50%
U
g 40%
p
tx 30%
20%
10%
0%
¦ Coal ¦ Oil ¦ Gas = Gas, efficient ¦ Nuclear ¦ Hydro se Solar PV I Wind
Figure 3-4. Potential future electrical grid mixes in Ghana.
3.3 Allocation Approach
Several cooking fuels examined such as crop residues, ethanol, and biogas are produced
by multi-output processes. Allocation is required for partitioning burdens among the various co-
products. Table 3-14 lists the baseline allocation approach and the allocation and modeling
conventions employed in the sensitivity analysis.
2013 Grid Ghana EC Ghana EC Ghana EC Low Carbon -
Thermal - 2020 Renewable- Nuclear-2020 2030+
2020
3-25
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-14. Summary of Baseline and Sensitivity LCA Modeling and Allocation Options
Baseline
Sensitivity Allocation
Fuel Type
Allocation Approach
Approach
Crop Residue
Cut-off
Physical Allocation
Economic Allocation
System Expansion
Sugarcane Ethanol
Physical Allocation
Economic Allocation
System Expansion
Biogas
Cut-off
Economic Allocation
System Expansion
ISO 14044 suggests that allocation be avoided either by using system expansion or by
breaking up manufacturing into multiple unit processes. This scenario is not always possible,
making allocation necessary. No single allocation approach is suitable for every scenario. The
method used for handling product allocation varies from one system to another, but the choice of
allocation is not arbitrary. ISO 14044, Section 4.3.4.2 states that "the inventory is based on
material balances between input and output. Allocation procedures should therefore approximate
as much as possible such fundamental input/output relationships and characteristics (ISO
2010b)." Under Phase I of this study, the baseline method used for modeling multi-output
product processes with one primary product and one or more unavoidable co-products was the
"cut-off approach. Using this approach, all burdens are assigned to the primary product
(Baumann and Tillman 2004).
Physical allocation is generally recommended within LCA studies due to its ease and
reproducibility. However, physical relationships do not always lead to a fair allocation of
environmental burdens between products and co-products. The allocation of impacts between
food crops and co-produced crop residues is a classic example where physical allocation does not
lead to allocation fractions that reflect the underlying drivers of environmental damage. The
residue portion of the plant in many cases has mass equal or greater than the food crop itself.
Economic allocation is a third means of distributing environmental burden between
products and co-products. This method assumes that economic demand is the driving factor
behind supply-chain activities and their attendant environmental impacts. The relative economic
value between products and co-products can be used to allocate environmental impact. This
scheme is complicated by crop residues, for example, often considered to be free by farmers,
which supports the previous use of the cut-off method. However, considering crop residues 'free'
neglects their value as cooking fuels or the value of alternative uses such as for animal feed or
soil fertilization and conditioning. Use of economic allocation requires the determination of a
price for all allocated products and co-products, a process that can be difficult for goods such as
crop residues and gathered firewood that often have no widely available market price. Market
price may also fluctuate widely, while the physical properties of the biomass remain unchanged.
A number of valuation strategies can be employed to set a price in the absence of market pricing.
3-26
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
• Substitute Value: Substitute use valuation assumes that the value of the good in
question is equal to the value of an equivalent good for which a market price is
available.
• Alternative Use Value: Alternative use valuation of a product assumes that the value
of a product in one application is equivalent to a known value for that product in
another application.
• Labor Value: Labor use valuation assumes that the value of a given non-market
good is equal to the labor cost necessary to obtain that good.
In addition to both physical and economic allocation, and as recommended by ISO
14044, this study employs system expansion in the sensitivity analyses to avoid the allocation of
environmental burdens between products and co-products. By expanding the system boundaries,
both the environmental impacts of co-product production and the impacts and avoided impacts of
co-product use are attributed to the main product. The net effect of this choice can be either
positive or negative, with net positive effects leading to the attribution of environmental credits.
3.3.1 Allocation to Crop Residues
Both Phase I and II of this project employ the cut-off method as the baseline LCA
modeling convention for crop residue production. The cut-off method considers crop residues as
a necessary by-product of food crop production and thereby attributes to them none of the
environmental impacts of agricultural processes.
The use of physical and economic allocation as well as system expansion is included in
Phase II to explore both the magnitude of the potential contribution of agriculture to cooking
impacts and alternate approaches to allocating environmental burdens between food crops and
crop residues. The use of crop residues as a cooking fuel is not common in Kenya and Ghana,
and this allocation sensitivity was not carried out for the African countries.
While a case can be made for allocating environmental burdens solely to foodstuffs, there
is a rationale for considering crop residues and the goods and services that they produce to be
valuable co-products. The sheer magnitude of these biological resources is difficult to ignore. In
India and China combined, upwards of 1.5 billion metric tons of crop residue are estimated to be
produced per annum (see Table 3-15). For centuries, this material has served as fodder for
animals and as a valuable soil amendment. Likewise, we understand from our initial study both
the importance and impact of these materials as cooking and home heating fuels. Increasingly,
there are competing interests for these materials as bio-based feedstocks for paper, textiles, and
chemicals, all of which serve as valid reasons to allocate a portion of the environmental footprint
of agriculture for crop residues such as straw and corn stover.
3-27
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-15. Crop-to-Residue Ratios, Production, and Fraction of Crop Residue Produced -
India and China
India
China
Fraction of
Fraction of
Crop-to-
2012
total Crop
2012
total Crop
Residue
Production
Residue
Production
Residue
Crop
Ratio1
(10,000 tons)2
(%)5
(10,000 tons)2
(%)5
Wheat
1.00
9,490
15%
12,100
13%
Rice3
1.21
15,800
30%
20,400
25%
Corn
2.00
2,230
5%
20,600
43%
Sugarcane4
1.50
36,100
17%
12,300
4%
Beans
1.00
3,140
7%
1,300
2%
Tubers
3.00
5,130
8%
3,290
3%
Cotton
2.00
582
3%
684
2%
Oil Crops
0.240
4,020
13%
3,440
7%
Totals
-
76,500
-
74,100
-
Notes and Sources: 1 FAO 2012,2 Zhenhong 2001,3 includes both straw and husk, 4 sugarcane tops are the available residue,
5 column value may not add to totals due to rounding.
System Expansion
The system expansion approach considered in the sensitivity analysis is based on the
assumption that a hierarchy of uses exists for crop residues that include use as animal fodder, soil
conditioner, cooking fuel, and field burning. Field burning is assumed to be the least preferable
use of crop residues, and occurs only in the absence of demand for higher uses. The burning of
crop residue as a cooking fuel is assumed to replace field burning, which would be the
alternative use of those crop residues if they were not utilized for cooking. In India and China,
respectively, approximately 19 and eight percent of crop residues are being disposed of by way
of burning on the field (FAOSTAT 2016). Field burning is considered a form of waste disposal.
Because crop residues are not being fully utilized for beneficial purposes, any use of crop
residues for cooking fuel can be assumed to avoid the necessity of crop burning, and can be
credited with the environmental benefit of this avoided action (Weidema 2000).
Physical Allocation
Physical allocation is not generally used to allocate between food crops and their co-
products, as physical relationships are generally not assumed to drive agricultural inputs, farming
practices, and their attendant environmental benefits and impacts. Physical allocation is included
in this project as part of the sensitivity analysis to develop the fullest possible understanding of
the potential distribution of impacts between food crops and crop residues. That is, the physical
allocation approach is likely to provide the upper bound of impact results for crop residues. The
full range of reported crop-to-residue ratios are used in these calculations. The national average
portion of residues re-incorporated into the soil or burned on the field is subtracted from crop
residue production and is allocated no impacts. This choice is made to reflect the national
average context within each nation studied.
3-28
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Economic Allocation
All economic allocations are performed on values transformed to 2014 U.S. dollars. This
study assumes that firewood is the fuel that crop residues would substitute for. Values associated
with the firewood substitution are based on the market price of firewood in India. The cost of
firewood is first converted to dollars/MJ-delivered and then this value is applied to crop residues,
thereby accounting for the difference in energy content and stove combustion efficiency that
exists between these feedstocks. The labor value is based on estimated levels of effort necessary
to collect the substitute fuel, firewood. Differences in stove thermal efficiency and heat content
of the two fuels are again corrected. For India, the value of labor is based on the current national
minimum wage in India of 160 Rupees per day, which is approximately 0.32 dollars per hour
assuming an eight-hour workday (Jadhav 2015). The alternative use allocation is based on the
substitute value of crops for use as fertilizer. The value of various fertilizers in India has been
determined on the basis of U.S. dollars per kg of nitrogen, phosphorus, and potassium (NPK).
All nutrients are assumed to contribute equally to the value of a fertilizer. The equivalent nutrient
content of each crop residue type is calculated, and the value per kg of NPK calculated for
various fertilizers is applied to the residue to determine a range of estimated values realized by
the farmer in reduced fertilization costs. Low and high estimates of crop residue value calculated
using the above methods are used in combination with high and low estimates of crop residue
production per kilogram of crop production, respectively, to capture the full potential range of
allocation fractions. The average of all economic allocation values is considered in the sensitivity
analysis. This same method of economic allocation was applied for the China scope.
Table 3-16 shows the allocation percentages determined for major biomass crops in India
and China. The underlying data and calculations associated with both physical and economic
allocation are documented in SI5.
3-29
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Table 3-16. Crop Residue Allocation Factors - India and China
India
China
India
China
India
China
Allocation
Wheat
Wheat
Rice
Rice
Sugarcane
Maize
Approach1
Wheat
Straw
Wheat
Straw
Rice
Straw
Rice
Straw
Sugarcane
Tops
Maize
Stover
Physical, residue
low
59%
41%
61%
39%
69%
31%
72%
28%
93%
7%
49%
51%
Physical, residue
average
47%
53%
50%
50%
53%
47%
56%
44%
88%
12%
44%
56%
Physical, residue
high
33%
67%
41%
59%
45%
55%
47%
53%
83%
17%
Economic, coal
substitution, residue
low, low value
96%
4%
95%
5%
98%
2%
97%
3%
97%
3%
88%
12%
Economic, coal
substitution, residue
high, high value
73%
27%
87%
13%
87%
13%
90%
10%
92%
8%
84%
16%
Economic, labor
value, residue low.
low value
99%
1%
97%
3%
99%
1%
94%
6%
92%
8%
89%
11%
Economic, labor
value, residue high.
high value
90%
10%
92%
8%
95%
5%
98%
2%
99%
1%
93%
7%
Economic,
alternative use.
residue low, low
value
99%
1%
98%
2%
100%
0%
99%
1%
100%
0%
96%
4%
Economic,
alternative use.
residue high, high
value
93%
7%
95%
5%
95%
5%
94%
6%
98%
2%
94%
6%
Economic, average
92%
8%
94%
6%
96%
4%
95%
5%
96%
4%
91%
9%
1 Allocation approach labels list the valuation method used, the estimate of crop residue production used (low/high), and where applicable whether the low, high or average
valuation estimate is used to calculate the associated allocation factor.
3-30
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3.3.1.1 Crop Residue LCI Modeling
The LCI of each crop aims to capture the national average agricultural production
practices within each country. It is the environmental effect of this LCI that is being allocated
between crops and crop residues in the sensitivity analysis. An uncertainty range has been
developed for each input and output with the crop LCIs, which is meant to capture the breadth of
climatic and cultural practices affecting these values within each study country.
Yield per hectare is perhaps the dominant determinant of emissions per kg of crop output,
which is the basis of this LCI. This determinant is true both for a specific crop and also between
crop types. The large sugarcane biomass yields per hectare and the comparatively low emissions
per kg of product are the most striking example of this phenomenon.
As described briefly in Section 2.1.5, estimates of nitrogen (N), phosphorus (P), and
potassium (K) fertilization specific to each crop type were drawn from the literature (Wang et al.
2014, Xia and Yan 2011). Water use estimates are also specific to each crop type, and are based
on blue water consumption, which includes both ground and surface water. The use of soil water
is not considered in this study as agricultural systems tend to yield greater flows to blue water
systems than the natural ecosystems that they replace (Huang et al. 2013). As a conservative
estimate, all irrigation water is considered to be consumed in this study.
Nitrogen losses to air and water are particularly dependent upon management and
environmental factors. Nitrogen leaching rates are a function of fertilizer application rates,
methods of soil incorporation, soil type and quality, precipitation rate, and temperature (Gao et
al. 2016). Several methods have been used to estimate nitrogen runoff based on applied fertilizer.
The simplest methods assume that a fraction of the applied nitrogen makes it into waterways. It
is suggested that dryland crops in China lose approximately four percent of applied nitrogen to
leaching (Hu et al. 2011). The work of Wang et al. (2014) develops a regression equation for
nitrate, N2O, and ammonia losses due to fertilization. This equation, originally developed for use
with maize in China, is used to calculate nitrogen emissions for all dryland crops in both India
and China, using crop- and country-specific fertilization rates. A similar approach is used to
calculate N2O and ammonia emissions from rice production (Xia and Yan 2011). Information on
the above calculation procedures is included in SI5.
No feasible method for calculating phosphorus runoff was discovered on the basis of
applied fertilizer. Therefore, estimated values found in the literature specific to each crop are
incorporated into the LCI. Low, average, and high estimates of yield are used to calculate a range
of potential phosphorus runoff values. Phosphorus adsorbs to soil particles much more strongly
than does nitrogen, and it can build and persist in the soil over many years. Although some
fraction of adsorbed phosphorus is expected to be lost through windborne erosion, an estimate of
this value is not included in the LCI.
Methane production is a serious concern during the production of rice using flooded
fields. This study uses the IPCC (2006) method to estimate a range of potential CH4 emissions
for rice production. A base CH4 emission factor (kg per day) is adjusted using scaling factors
associated with length of growing season, water regime, and the incorporation of crop residues.
Detailed calculation of CH4 emissions during rice production is provided in SI5.
3-31
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
3.3.2 Allocation to Biogas and Bioslurry
Bioslurry1 refers to the residual solids that remain following feedstock degradation to
biogas in an anaerobic digester (AD). In a cooking system based on biogas, animal manure and
other organic wastes are diverted from alternative pathways into the digester. These alternative
pathways could include direct incorporation of residues or manures into the soil, composting, or
use of residues as fuel or fodder. This project focuses on the use of cattle dung as an AD
feedstock.
An LCI of bioslurry was created to facilitate the allocation of environmental impacts
from ADs between both biogas and bioslurry. The cut-off method, which is used as the baseline
modeling convention in both Phase I and II, assumes that 100 percent of the impacts associated
with biogas production are allocated to the cooking fuel. Given the potential alternative uses of
bioslurry listed above, there is a strong case to be made for allocating a portion of AD
operational impacts to the bioslurry, as well as to the biogas. This study employs economic
allocation and system expansion as part of a sensitivity analysis to quantify the effect of
allocation on the environmental impacts of biogas production. Physical allocation is not
considered, as it does not provide an accurate representation of the comparative value of the
products.
3.3.2.1 System Expansion
Biogas is the main product, with bioslurry and its land application considered as a
valuable co-product. Both the impacts of bioslurry land application and avoided fertilizer
production are considered within the biogas unit process. Determination of the avoided products
is addressed for N, P, and K content individually with urea, single-superphosphate, and
potassium chloride being used as avoided products for each nutrient, respectively. By including a
separate avoided product for each nutrient individually, it is possible to exactly match the
nutrient content contained in the bioslurry with that present in the avoided products. Several
studies have shown that the fertilizer value of nutrients derived from bioslurry are comparable to
those supplied by commercial fertilizers (Nkoa 2014, Mikled et al. 2002).
3.3.2.2 Economic Allocation
No market price for either biogas or bioslurry is available. In the absence of a market
value for biogas, a substitute value was calculated based on the price per MJ delivered for LPG.
Low, medium, and high substitute value estimates for bioslurry are calculated based on a range
of fertilizer values and bioslurry NPK content. The average nutrient content of bioslurry was
used to perform the allocation. If packaged for sale, the nutrient content of bioslurry would read
7:4.4:4.9, corresponding to N, P2O5, and K2O content as a percentage of dry weight, respectively.
Assumed fertilizer and LPG prices are presented in SI7.
1 List of additional terms for bioslurry: biogas digestate, fermenter slurry, biogas slurry, and digested slurry.
3-32
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
Calculated Allocation Factors
Calculated allocation factors for India and China are presented in Table 3-17 and Table
3-18. Details regarding allocation factor calculation are included in SI7.
Table 3-17. Allocation Percentages between Bioslurry and Biogas in India
Allocation Approach
Biogas
Bioslurry
Cut-off Method
100%
0%
Economic Allocation, slurry low
88%
12%
Economic Allocation, slurry high
20%
80%
Economic Allocation, average
55%
45%
Table 3-18. Allocation Factors for Biogas and Bioslurry in China
Allocation Approach
Biogas
Bioslurry
Cut-off Method
100%
0%
Economic Allocation, slurry low
85%
15%
Economic Allocation, slurry high
16%
84%
Economic Allocation, average
51%
49%
3.3.3 Biogas and Bioslurry LCI discussion
This section describes the bioslurry LCI data that are allocated as part of the sensitivity
analysis. An uncertainty range associated with each LCI entry is employed to capture the
variation that is inherent in both biogas and bioslurry production and use.
Bioslurry is a good source of NPK. Additionally, it provides some micronutrients
necessary for plant production such as zinc and manganese. In addition to the positive
components of bioslurry, there are concerns about the heavy metal content of some manures and
the bioslurry that is produced from them. Estimates of nutrient, micronutrient, and the heavy
metal content of bioslurry that would be applied to agricultural fields are included in the LCI
During AD operation, between 20 and 30 percent of the organic matter is converted into
biogas. This change in mass is the direct consequence of biogas production, and it is reported
that between 0.19 and 0.67 cubic meters of biogas are produced per kg of dry matter exiting the
AD (Kalia and Singh 1998, Singh et al. 2014a, Adelekan 2014, Plochl and Heirmann 2006,
Poeschl et al. 2012).
Bioslurry can also vary widely in its nutrient content. For example, values found in the
literature suggest that the nitrogen content of digested slurry can vary between 0.05 and 1.8
percent of bioslurry mass on a wet basis (Gurung 1997 and Kumar et al. 2015, respectively).
These values depend upon variables such as feed quality, cattle health, and dung water content to
name but a few (Nkoa 2014). Controlling for variation in water content is particularly important.
3-33
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
All LCI values have been standardized to the dry matter content of the bioslurry when it exits the
AD. In some cases, the dry matter content of a sample is not reported, and when this is true, the
average dry matter content of the bioslurry profiles compiled has been applied so that the
information can be included in the LCI.
There are also many elements of AD operation and bioslurry management that ultimately
affect nutrient availability to agricultural crops. After being expelled from the animal, nitrogen
starts being lost through ammonia volatilization. Time to collection, weather, and storage
practices can greatly affect these losses (Nkoa 2014). Little nitrogen is generally assumed to be
lost during the actual process of anaerobic digestion. However, one source reported a potential
loss between three and ten percent of total nitrogen content (Gurung 1997). Further losses to
volatilization can come during subsequent storage of the digested slurry or during and
immediately following field application. The amount of time elapsed, application method, extent
of soil incorporation, temperature, and precipitation all have a significant effect on loss rates
following digestion. The literature suggests that ammonia losses of 20 to 35 percent are possible
during field application alone (Makadi et al. 2012).
These losses, apart from the minor losses in the digester itself, are also applicable to other
forms of organic and inorganic fertilizer that may be used as an alternative to bioslurry.
Quantitative evaluation of the average relative losses between various fertilizers and application
methods is beyond the scope of this analysis. However, we discuss the ways in which the AD
affects both rates of volatilization and plant utilization efficiency.
Contradictory results are present in the literature regarding the relative potential nutrient
losses of digested and undigested slurry. Most references reviewed indicate that the digestion
process increases the potential for ammonia volatilization due to its increased share of the
nitrogen fraction and the increase in pH associated with digestion2 (Nkoa 2014). Others suggest
that undigested slurry3 tends to lose more nitrogen to volatilization during field application while
digested slurry loses more during storage (Smith et al. 2007). The authors hypothesize that the
decreased solids content of digested slurry allows quicker infiltration, thereby reducing
volatilization after field application. On average, the digestion process increases the ammonium
(NH4+) content of bioslurry, in relation to the fresh manure that was used as a feedstock, by 25
percent (Arthurson 2009). Ammonium is one of two plant-available forms of nitrogen, the other
being nitrate (NO3).
Many authors have used the increased share of total nitrogen attributable to ammonium to
suggest that rates of nutrient utilization are higher for bioslurry as compared to rates in un-
digested manure. The work of Smith et al. (2007) suggests that while a higher rate of nutrient
utilization for bioslurry may be true in the short-term, in the long term the relative rate of nutrient
utilization evens out between digested and undigested manure because, over time, nitrogen in un-
digested manure is mineralized and becomes available to plants. A good number of other studies
indicate that the fertilizer value of bioslurry is greater than (Somasundaram et al. 2007, Mikled et
al. 2002, Ahmad and Jabeen 2009) or equal to (Haraldsen et al. 2011, Nkoa 2014) that of a
comparable quantity of manure or mineral fertilizer. In this study, we consider that the
2 Not all references support this supposed increase in pH (e.g., Smith et al. 2007)
3 Manure mixed with water.
3-34
-------
Section 3—Methodology for Scenario Development and Sensitivity Analyses
fertilization value of digested manure, undigested manure, and commercial fertilizers is
equivalent per unit of applied nutrient. The organic matter content of both digested and
undigested manure is expected to improve soil tilth and moisture retention when compared to
mineral fertilizer applications. However, this benefit is not quantified.
The research of Smith et al. (2007) also indicates that there is no discernible effect of
slurry digestion on annual emission of N2O, a potent greenhouse gas. N2O emissions are
considered to be equivalent between digested and undigested slurry per kg of applied nitrogen.
Other authors have shown that the use of digested slurry leads to reduced N2O emissions per unit
of applied nitrogen (Nkoa 2014). The range of this reduction was between 17 and 71 percent
(Borjesson and Berglund 2006). The absolute level of these reductions is highly dependent upon
soil type, application method, and local weather, as is the magnitude of N2O emissions generally.
The work of Koster et al. (2015) indicates that lower N2O emissions are due to the lower amount
of labile carbon that is available for denitrification in digested cattle waste4.
Finally, this study considers the differential effect of digestion on the potential for
nutrient runoff from agricultural fields, and again the results are mixed. The review by Nkoa
(2014) suggests that given the state of current research, we can expect similar nitrogen runoff
emissions at a given application rate regardless of fertilizer type. Nitrate nitrogen is the
predominant species contributing to nitrogen runoff. Ammonium, on the other hand, is a minor
contributor, which indicates a low potential for short-term increase in nitrogen runoff attributable
to the ammonium increase during digestion. However, this ammonium can oxidize to nitrate over
time.
In general, phosphorus runoff is determined by soil type, application rate, and weather
conditions following application (Radcliffe et al. 2015). No references specific to bioslurry field
application have been found to indicate that there is an expected influence on phosphorus
leaching to surface and groundwater beyond what is typical for other phosphorus additions.
3.3.4 Electricity from Ethanol Production
Electricity is often a co-benefit of ethanol production. The Phase I study did not include a
credit for grid electricity displaced by electricity co-produced with ethanol. Inclusion of this
credit, by application of the system expansion modeling approach, could decrease the overall
environmental impacts for ethanol. Bagasse at the mill provides excess energy that can be
exported as electricity. A sensitivity analysis covers incorporation of the electricity credit of the
electrical grid mix for the relevant country. Ethanol is assumed to be produced from molasses.
Refined sugar is also an output of molasses production. In all cases, the allocation between
molasses and sugar is conducted on a mass basis.
4 Soil carbon is necessary for cell growth during denitrification.
3-35
-------
Section 4—Methodology for Results Compilation
4. METHODOLOGY FOR RESULTS COMPILATION
This section discusses key methodological considerations for transforming the LCI data
compiled into the environmental impact results presented in this report.
4.1 Biogenic Carbon Accounting
In biomass fuel systems, CO2 is removed from the atmosphere and incorporated into the
plant material that is harvested from the forest or field. This (biogenic) carbon is stored in the
material throughout the life of the product until that fuel is combusted or degrades, at which
point the carbon is released back into the environment. Combustion and degradation releases are
predominantly in the form of CO2 and CH4. This study, in alignment with the IPCC
methodology, assumes a zero net impact for biogenic carbon that is removed from the
atmosphere in the form of CO2 and later returned to the atmosphere (e.g., as CO2 emissions from
the combustion of biomass cookstove fuels). That is, if the carbon removed from the atmosphere
is returned to the atmosphere in the same form, the net impact GCCP is zero. Impacts associated
with the emission of biogenic carbon in the form of CH4 are included since CH4 was not
removed from the atmosphere and its GCCP is 28 times that of CO2 when applying the IPCC
2013 100a LCIA method. The one exception is the CO2 emissions from non-renewable wood
fuel associated with deforestation in the four countries assessed and, therefore, long-term
reduction of global CO2 sinks. The method used to calculate the non-renewable portion of wood
for cooking fuel is described in the next section.
4.2 Non-Renewable Wood Fuel Calculations
In the GHG analysis, the carbon dioxide emissions for the portion of biomass fuel from
unsustainable wood supplies are considered non-renewable and are therefore incorporated into
the overall GCCP results. This phase of work uses the methodology described by Bailis et al.
(2015) to calculate forest renewability factors. Using the Yale Woodfuel Integrated
Supply/Demand Overview Mapping (WISDOM) database (Drigo 2014), the Bailis method
developed a spatially explicit assessment of woodfuel supply and demand based on the relative
woody biomass supply and regrowth compared to demand. The Bailis study found its results for
non-renewable forestry were lower than those previously published in the literature. Phase I of
the cookstoves research relied on the renewable wood calculation outlined by Singh and
colleagues (2014a), which is described in subsequent paragraphs. Table 4-1 lists the Phase I and
Phase II baseline forest renewability factors. Because the methods for determining such
renewability factors are novel and uncertain, a sensitivity analysis is included in this study to
assess the effect of methodology assumption on overall results.
Table 4-1. Phase I and II Forest Renewability Factors
Country
Phase II Forest Renewability
Factor
(% renewable)
Phase I Forest Renewability
Factor
(% renewable)
India
76.3
59.2
China
77.8
57.5
Kenya
36.1
0
Ghana
70.6
0
4-1
-------
Section 4—Methodology for Results Compilation
Using the Singh method, the biomass stock in m3 for each country (from FAO 2010
Table 10) was multiplied by the regional factor for tonnes of above-ground biomass (AGB) per
m3 (from FAO 2010 Table 2.18) to calculate the tonnes of AGB. The amount of below-ground
biomass (BGB) was calculated by multiplying the tonnes of AGB by the regional factor for
BGB/AGB (from FAO 2010 Table 2.18). The amount of dead wood was then calculated using
the regional factor for dead-to-live biomass ratio (from FAO 2010 Table 2.18) applied to the
total AGB and BGB. Next, the average annual increase or decrease in forest land for each
country was calculated based on the carbon stocks in living forest biomass reported for each
country in 2000 and 2010 (from FAO 2010 Table 11). The annual firewood supply potential for
each country was then calculated as the total weight of AGB and dead wood multiplied by
country-specific factors for the percent accessibility to forests (from the Yale WISDOM
Database (Drigo 2014)) and the country-specific average annual change in forest land.
The annual demand for firewood cooking fuel (tonnes) for each country was calculated
based on the country-specific cooking energy demand per household multiplied by the number of
households using wood for cooking fuel, divided by the cooking energy per kg of firewood
(calculated as the lower heating value of firewood multiplied by stove efficiency). For India,
11.0 MJ of cooking energy are consumed per household per day (Habib et al. 2004), with 105
million rural households and 16 million urban households using wood for cooking fuel (Singh et
al. 2014a). In China, 13.6 MJ of cooking energy are consumed per household per day (Zhou et
al. 2007), with over 131 million rural households and over nine million urban households using
wood for cooking according to World Bank statistics. In Kenya, 12.5 MJ of cooking energy are
consumed per household per day (IEA 2014, GVEP International 2012a), with over five million
households using wood for cooking (GVEP International 2012b). In Ghana, 13.6 MJ of cooking
energy are consumed per household per day (IEA 2014, with almost three million households
using wood for cooking (GVEP International 2012c). Finally, the renewable percentage of
cooking firewood was calculated as the annual firewood supply potential divided by the total
annual demand for cooking firewood. The percentage of annual firewood demand that cannot be
met by the annual firewood supply potential was considered non-renewable.
4.3 Black Carbon and Short-Lived Climate Pollutants Calculations
This section summarizes key physical parameters considered in the approach to include
the differences in potential amounts of BC, OC, and other co-emitted species produced from use
of the investigated cookstove/fuel technologies. BC and co-emitted species are formed by
combustion of fossil and bio-based fuels (e.g., diesel, coal, crop residues).
Per the Gold Standard Framework method (GSF 2015), fuel production, transport, and
consumption life cycle phases are included in the inventory and impact assessment. An inventory
of BC and OC is based on the quantity of PM (less than or equal to 2.5 microns of aerodynamic
diameter-PM2.5) released for each inventory step in the cookstove fuel/technology life cycle. In
many cases, LCI data sources do not specify the type of PM emissions (e.g., outputs are reported
as 'particulate matter' or 'particulate matter, unspecified'). For upstream process inventories
where PM emission speciation is not provided, no BC and/or OC emission factors are applied.
However, co-emitted species emission factors for these processes are included. In the foreground
cookstove fuel combustion, BC and OC emission factors based on quantity of PM released (e.g.,
4-2
-------
Section 4—Methodology for Results Compilation
per fraction reported as PM2.5) are applied. Where no size distinctions between PM emissions
have been made in LCI data sources, all PM emissions from fuel combustion are assumed to be
of the fine particle variety (e.g., of less than or equal to 2.5 microns in size).5
Carbon in PM2.5 emissions takes the following forms: 1) organic carbon; 2) elemental
carbon (EC), which usually includes soot; and 3) carbonate ion (CO3"2). Methods that measure
light absorption in PM2.5 assume that the light absorbing component is BC and partitioning of EC
and OC is somewhat arbitrary. Though some components of OC may be light-absorbing (e.g.,
brown carbon or BrC), most researchers presume that OC possesses light-scattering properties
(e.g., producing climate cooling effects). Because there is high uncertainty and lack of consensus
on the ratio of the BrC class of OC particles for each fraction of OC, analyzing impacts of BrC in
OC is excluded in this analysis and instead, focus is placed on the EC or soot portion and the OC
portions of the PM2.5 emissions. In other words, BC emissions may be estimated by assuming
that only the EC portion of the PM2.5 emissions contributes to BC release and subsequent
positive radiative forcing, while OC emissions are assumed to contribute to negative radiative
forcing. This approach requires estimating the PM2.5 emission amount and source-specific EC-to-
PM2.5 and then the BC-to-OC ratio for each of the fuel/stove technologies being investigated in
the study.
Potential climate forcing impacts resulting from BC/OC and co-emitted species include
direct, albedo, and other indirect effects. Overall, most estimates indicate BC yielding a net
warming effect on climate, but co-emitted species can have some offsetting effects, as discussed
below. Species co-emitted with BC/OC such as CO, NMVOCs, NOx, and SO2 are precursors to
the formation of sulfate and/or organic aerosols in the atmosphere. These aerosols affect
reflectivity and other cloud properties and have a cooling affect.
BC and other short-lived climate pollutants (SLCPs) such as the aforementioned co-
emitted species are distinguished from other climate-forcing emissions (e.g., GHGs) in that their
atmospheric lifetime is not as long-lived, so potential impacts are estimated on a shorter time-
scale and can be very geographic and seasonally dependent (unlike long-lived, well-mixed
GHGs). However, short-lived forcing effects of BC are substantial compared to effects of long-
lived GHGs from the same sources, even when the forcing is integrated over 100 years. The
GCCP of BC and co-emitted species included in this approach are calculated using GCCP 20-
year BC eq. factors from IPCC 2013 as summarized in Table 4-2.
5 Per (2008) "Secondary PM and combustion soot tend to be fine particles (PM2 5), whereas fugitive dust is mostly
coarse particles".
4-3
-------
Section 4—Methodology for Results Compilation
Table 4-2. Characterization Factors for BC eq
Included in GSF 2015
GCCP (20)
per IPCC 2013
BC eq
BC
2421
1
Warming Effects
NOx
16.7
0.00690
CO
5.9
0.002
NMVOC
14
0.006
Cooling Effects
OC
-244
-0.1
S04 (-2)
-141
-0.058
Source: GSF 2015.
4.4 LCA Model Framework
All LCI unit processes developed for this work (summarized in SI1-SI7) are intended for
publication in the US Federal LCA Digital Commons Life Cycle Inventory Unit Process
Templates (in Microsoft Excel® format) (United States Department of Agriculture (USD A) and
U.S. EPA 2015). To build the life cycle model, the unit processes were entered into the open-
source openLCA software (Version 1.5.0, GreenDelta 2016. Quality assurance (QA) reviews
were completed for the openLCA model to ensure that all inputs and outputs, quantities, units,
and metadata were correctly entered. Associated metadata for each unit process are recorded in
the openLCA unit processes.
Once all necessary data were imported into the openLCA software and reviewed, system
models were created for each fuel and country combination. The models were QA-reviewed to
ensure that each elementary flow (e.g., environmental emissions, consumption of natural
resources, and energy demand) was characterized under each impact category for which a
characterization factor was available. The draft final system models were also QA-reviewed
prior to calculating results to make certain all connections to upstream processes and weight
factors were valid. LCIA results were then calculated by generating a contribution analysis for
the selected fuel product system based on the defined functional unit of 1 GJ of delivered heat
for cooking.
4.5 Monte Carlo Uncertainty Analysis
An important issue to consider when using LCI study results is the reliability of the data.
In a complex study with literally thousands of numeric entries, the accuracy of the data and how
it affects conclusions is truly a complex subject, and one that does not lend itself to standard
error analysis techniques. Techniques such as Monte Carlo analysis can be used to study
uncertainty, but a lack of uncertainty data or probability distributions for key parameters, which
are often only available as single point estimates, continues to pose a challenge.
Monte Carlo analysis is a statistical procedure used to simulate the potential range of
results in each impact category based on underlying uncertainty distributions attached to
individual flows of input materials, energy, and emissions to nature. Five thousand simulations
were conducted for each stove grouping based on the distributions associated with each flow
value. The important concept that is highlighted by the uncertainty analysis is that for any study
4-4
-------
Section 4—Methodology for Results Compilation
dealing with a general functional unit such as the national average GJ of delivered heat from a
given stove grouping, there is not one number that accurately quantifies environmental impact.
Instead, multiple parameters can be varied at once to estimate the potential range in
environmental impact scores. Stove emission and crop production uncertainty information
incorporated in the Monte Carlo analysis are described below. Ranges on other parameters (e.g.,
emissions from the charcoal kiln and from bioslurry land application) are applied to the
uncertainty analysis as well.
4.5.1 Uncertainty Modeling Documentation
Stove Uncertainty
Stove use phase emission estimates are a compilation of emission testing results from the
literature as reported in Table 2-1. Emission values from these studies are classified into stove
groupings defined by fuel type, stove type, and country. Stove use phase emissions are modeled
using a lognormal distribution. For emissions that have six or greater recorded emission
estimates for a given stove grouping, the geometric standard deviation of the emission values is
used in the Monte Carlo analysis. For each country and pollutant combination, a proxy standard
deviation is calculated based on the stove grouping with the greatest recorded number of
emission estimates. This value is used for stove groupings that have less than six recorded
emission estimates for a given pollutant in combination with the geometric average of the
available emission values. The range of recorded stove thermal efficiencies is used in
combination with the fuel LHV to calculate a triangular distribution for stove fuel, ash
production, transport, and other associated inputs and outputs. A geometric standard deviation of
1.05 is used for embodied energy flows that contribute to CED. Table 4-3 lists factors expected
to contribute to uncertainty in the cookstove emissions data.
Table 4-3. Sources and Mechanisms of Uncertainty in Cookstove LCIs
Category
Source
Mechanism
Fuel Characteristics
Heat Content
Specific fuels used in the emissions studies
used to compile stove emission LCIs vary over
a given range.
Moisture Content
Moisture content is variable for a given fuel
type and affects thermal and combustion
efficiency.
Stove Characteristics
Thermal Efficiency
Varies over a given range within the assigned
stove groupings.
External Factors
Operator Practice
Fuel placement, ventilation control, cooking
pot, and cooking practices affect thermal
transfer and combustion efficiency.
Climate
Humidity, wind, and air temperature affect
combustion and thermal transfer of heat.
Combustion
Emissions
Testing uncertainty. Should be a dependent
factor.
4-5
-------
Section 4—Methodology for Results Compilation
Crop Residue Production
Description of national average crop production practice and environmental impact is a
particular challenge. Table 4-4 lists parameters that contribute to uncertainty in estimates of
national average crop production. Multiple estimates were sought for each LCI flow. All input
and output flows were recalculated so that they are reported per kg of crop produced. The
average of these values is taken as the average flow value used in the baseline results. To cover
the widest possible range of uncertainty, a national estimate of low and high yield for each crop
is used to recalculate the LCI flow per kg of crop production. The lowest and highest values for
each LCI flow are taken as the lower and higher ends of a triangular distribution, while the
average value is taken as the peak, or most likely, flow. This approach assumes that even the
lowest fertilization rate can correspond to the highest yield, or that the highest fertilization rate
can correspond to the lowest yield. This assumption is justifiable in light of the independent
nature of many of the factors that affect crop yields and thereby impacts per kilogram of crop
production. Any single agricultural practice, no matter how ideal, cannot guarantee a successful
crop. Appropriate rates of fertilization and pest management can be undone by an early frost or
lack of rain. While this assumption holds in any given year, it is expected that over the long run
the lowest and highest values cannot persist and that impacts tend towards the average.
Table 4-4. Sources and Mechanisms of Uncertainty in Crop Production
Category
Source
Mechanism
Location Dependent
Climate
Temperature, day length, precipitation patterns,
and the frequency of extreme weather events all
have a direct effect on crop yields and emissions to
air that is independent of agricultural inputs
applied.
Soil Type
Varies widely within a country or region and has a
direct effect on crop yields and emissions to air
and water that result from fertilizer application.
Topography
Affects erosion and runoff rates which has an
indirect effect on emissions to land and water.
Management
Farm Size
The range of farm sizes within a given nation has a
direct effect on the level of mechanization, soil
management practices, and a range of other factors
that affect the quantity of agricultural inputs,
emissions, and yields.
Crop Variety
Selection
Selection of varieties within a given crop type
(e.g., maize, rice) affect the necessary rates of
irrigation and fertilization at a given location. For
rice, variety selection is of particular importance to
CH4 emissions where the time to maturity and
inundation requirements have a direct relationship
to CH4 production.
Farming Practice
A wide range of production methods exist at the
national level. Heavy tilling, no-till, organic
production, and pesticide application are a few of
the production practices that can affect LCI input
and output values.
4-6
-------
Section 4—Methodology for Results Compilation
4.6 Normalization
Normalization is a process of standardizing impact scores in all categories so that the
relative contribution of impact scores associated with the functional unit can be judged relative to
total national or global emissions contributing to impacts in a given category. Table 4-5 lists
normalization factors for each country and impact category. Normalization allows us to better
assess the significance of impact categories by comparing to benchmarks at the national level.
Table 4-5. Country Specific Normalization Factors (per person per year)
Impact
. India
China
Kenya
Ghana
category
unit
Factor
Source
Factor
Source
Factor
Source
Factor
Source
Global Climate
Change
kg C02 eq
6,890
1
6,890
1
6,890
1
6,890
1
Energy Demand
MJ
28,300
2
96,000
2
19,600
7
14,800
9
Fossil
Depletion
kg oil eq
1,290
1
1,290
1
1,290
1
1,290
1
Water
Depletion
m3 eq
631
3
408
3
52.2
3
52.2
3
Particulate
Matter
Formation
kg PM2.5 eq
14.1
1
14.1
1
14.1
1
14.1
1
Photochemical
Oxidant
Formation
kg NMVOC
eq
56.7
1
56.7
1
56.7
1
56.7
1
Freshwater
Eutrophication
kg N eq
0.29
1
0.29
1
0.29
1
0.29
1
Terrestrial
Acidification
kg SO2 eq
38.2
1
38.2
1
38.2
1
38.2
1
Ozone
Depletion
kg CFC-11
eq
0.04
1
0.04
1
0.04
1
0.04
1
Black Carbon
and Short-Lived
Climate
Pollutants
kg BC eq
0.92
4,5
1.44
6
1.36
8
3.33
8
Sources: 1 Goedkoop et al. 2008,2 adapted from Enerdata 2016,3 adapted from FAO 2016, 4 adapted from Sloss 2012,5 adapted
from Paliwal et al. 2016,0 adapted from Wang et al. 2012,7 adapted from IEA 2016,8 adapted from U.S. EPA 2012,9 adapted
form GEC 2015
Normalized results are calculated by multiplying environmental impact per GJ of cooking
energy by national cooking energy expenditures (Table 4-6) and dividing by the appropriate
normalization factor from Table 4-5. This calculation produces results in units of person
equivalent emissions for all impact categories. Normalized results scaled by country-specific
cooking energy use provide key context for the relative changes in impacts from shifting fuel
choice under specific future cooking fuel mix scenarios.
4-7
-------
Section 4—Methodology for Results Compilation
Table 4-6. Household Energy Use for Cooking per Year
Country
GJ/Household/year
Sources
India
4.02
Habib et al. 2004 per
Venkataraman et al. 2010
China
4.95
Zhou et al. 20071
Kenya
4.56
IEA2014, GVEP 2012a
Ghana
4.96
IEA2014
Country
# Households
Sources
India
267,006,110
calculated
China
437,741,935
ASTAE 2013
Kenya
8,870,000
calculated
Ghana
6,475,000
calculated
Country
Household Size (ppl.)
Source
India
4.91
Dalberg 2013b
China
3.1
TekCarta 2015
Kenya
5
GVEP International 2012a
Ghana
4
ADP 2012
Notes: Ppl = number of people; values may be converted from original source to be
shown on an annual basis.
1 Includes energy for both cooking and water heating from Table 7 of Zhou et al. 2007.
Values combined for urban and rural population based on population statistics in
WorldBank2013.
4.7 Results Presentation Format
Results presented in Sections 5 through 8 of this report do not include all possible results
tables and figures for each impact category. Rather, the report includes a more focused analysis
on new information associated with Phase II. The full breadth of information generated from this
study is available in the appendices and SI 1-7 (see SI file descriptions in Section 1.3.3),
including country-specific Excel® workbooks that contain all of the results and figures available.
Each workbook presents results for both individual fuels and current and potential fuel mixes.
Within the workbook, custom fuel mix, stove technology mix, stove efficiency, electricity grid
mix, LC A modeling approach, and forestry renewability fraction can be customized by the user
to compare the associated changes in environmental impacts. The charts that are included in each
results workbook are described below:
• Single Fuel Results by Impact Category - Bar charts present impact scores broken
down by life cycle stage for each cooking fuel type. Results for each cooking fuel
type are aggregated according to stove technology mix, stove efficiency, electricity
grid mix, and forest renewability fraction, which can be customized by the user.
• Single Fuel Results as a Percent of Maximum Impact6 - Bar charts show the
impact of each fuel type relative to the fuel with the greatest impact in each category.
Results are aggregated according to the life cycle stage in which they occur.
6 Results are dependent on the underlying stove technology mix, stove efficiency, electricity grid mix, LCA
modeling approach to allocation, and forest renewability fraction.
4-8
-------
Section 4—Methodology for Results Compilation
• Relative Fuel Mix Scenario Results6 - Aggregate impact scores for both current and
potential fuel mixes are presented relative to the cooking fuel mix with the greatest
impact score in each category. Results are depicted according to cooking fuel type.
• Fuel Mix Scenario Results6 - Bar charts compare fuel mix impact scores for both
current and potential fuel mixes, aggregated according to fuel type.
• Normalized Results6 - Bar charts show person equivalent emissions for each impact
category presented according to fuel type.
• Stove Uncertainty Analysis Results - Impact scores for each stove group are
presented according to impact category with error bars showing the estimated
uncertainty range for each stove.
• Stove Efficiency Sensitivity - A bar chart shows comparative results for both current
and future improved stove thermal efficiency assumptions for each stove group.
• Electricity Grid Sensitivity - Comparative results for the electric cooking stove are
presented according to the underlying electricity grid mix. Users are able to toggle
between current and future improved stove thermal efficiency and impact category
assessed.
• Cooking Scenario Sensitivity - Bar charts show the comparative effect on aggregate
fuel mix impact score from the range of available stove technologies, stove thermal
efficiencies, and electrical grid mix selections. Results are aggregated according to
fuel type.
• Forest Renewability Fraction Sensitivity - A bar chart shows the comparative
effect of the assumed forest renewability fraction on impact scores for stove groups
where fuel is derived from forest products.
• LCA Modeling Sensitivity - A bar chart shows the comparative effect of LCA
modeling choices for selected cooking fuels. Results are presented separately for each
impact category.
LCIA results for ten impact categories were calculated for this study. Summary
discussion and figures for each country include results for all impact categories. Detailed
discussion and figures focus on the following four impact categories:
• Global Climate Change Potential
• Cumulative Energy Demand
• Particulate Matter Formation Potential
• Black Carbon and Short-Lived Climate Pollutant Potential.
The above impact categories were selected due either to general interest or their strong
connection with the cooking sector as revealed in the normalized results analysis. This selection
is not meant to imply that other impact categories are of less importance. Select results for
4-9
-------
Section 4—Methodology for Results Compilation
impact categories, beyond the four impact categories listed above, are included within the
sensitivity and uncertainty analysis results sections to highlight trends of interest and to
demonstrate the full breadth of results available in the SI. The presentation of results in Sections
5 through 8 provides a foundation on which interpretation of results for other impact categories,
reported in the results workbooks, is possible. All readers are encouraged to explore the full
range of results for all impact categories.
Baseline LCA results highlight the general trends in environmental impact associated
with different cooking fuel mix interventions. The uncertainty and sensitivity analyses provide a
measure of how robust these trends are and the actual gains that could be made by adopting
various strategies designed to reduce emissions and increase efficiency in the generation and use
of household cooking energy.
While the study has justified its specification of the fuel mix scenarios included, these
cooking fuel mix scenarios are not intended to strictly define perception of what is either
possible or most likely. All results should be viewed as a starting point for understanding current
environmental impacts in relation to future possibilities, with an eye towards what is technically
possible, and a focus on identifying key levers that are available to achieve improvement in the
environmental and human health outcomes of the cooking sector.
4-10
-------
Section 5—Updated LCA Results for India
5. UPDATED LCA RESULTS FOR INDIA
Table 5-1 presents summarized India LCA results for all cooking fuel types and impact
categories on the basis of 1 GJ of cooking energy delivered. The results are representative of
baseline assumptions concerning cooking fuel mix, stove technology use, stove thermal
efficiency, electricity grid, and forest renewability factor. A discussion of notable changes
between results for Phase I and Phase II is provided Appendix B. Table 5-1 displays baseline
LCA results for all investigated current and projected Indian cooking fuel types, currently nearly
70 percent of India's population relies on firewood, dung cake, and crop residue to provide their
cooking energy (Dalberg 2013b), and most households still rely on traditional mud stoves to
consume these fuels (Smith et al. 2000).
The remainder of this chapter focuses on quantifying the environmental impact of
interventions for India's cooking fuel mix through actions such as changing the cooking fuel
types in the country-wide mix, adopting improved stoves, improving overall stove efficiency,
and shifting the fuel type and associated technology used for India's electrical grid. The results
also identify which of the environmental impact categories assessed contribute the most to Indian
economy-wide impacts. Select uncertainty results and sensitivity analyses are presented to
increase understanding of the level of confidence readers should have in the LCA results by
cooking fuel type.
Table 5-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - India
GCCP
CED
FDP
WDP
PMFP
Fuel Type
(kg C02 eq)
(MJ)
(kg oil eq)
(m3)
(kg PM10 eq)
Hard Coal
963
7.21E+3
172
0.397
19.8
Dung Cake
263
1.30E+4
0.152
1.68E-3
24.3
Crop Residue
119
1.01E+4
7.90E-3
8.72E-5
11.4
Firewood
196
6.52E+3
5.94E-3
6.54E-5
5.54
Charcoal from Wood
402
1.09E+4
0.011
1.20E-4
20.5
Kerosene
180
3.09E+3
70.9
0.239
0.171
LPG
157
2.61E+3
58.7
0.193
0.136
Natural Gas
117
2.04E+3
48.7
0.039
0.019
Electricity
457
5.70E+3
122
3.25
1.91
Sugarcane Ethanol
121
1.33E+4
31.0
643
4.38
Biogas from Cattle Dung
11.4
4.06E+3
-
1.02
0.210
Biomass Pellets
141
3.91E+3
13.72
0.357
0.302
5-1
-------
Section 5—Updated LCA Results for India
Table 5-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - India
POFP
FEP
TAP
ODP
BC
Fuel Type
(kg NMVOC)
(kg P
eq)
(kg SO2
eq)
(kg CFC-
11 eq)
(kg BC eq)
Hard Coal
7.87
2.37E-3
1.87
3.01E-8
4.10
Dung Cake
18.8
0.189
0.736
1.40E-9
5.27
Crop Residue
8.22
9.80E-3
0.598
7.28E-11
2.48
Firewood
5.38
7.36E-3
0.377
5.46E-11
1.22
Charcoal from Wood
10.4
0.014
0.209
1.03E-10
4.58
Kerosene
0.481
3.77E-3
0.291
6.20E-8
0.021
LPG
0.341
3.37E-3
0.256
6.56E-8
0.012
Natural Gas
0.046
7.05E-5
0.027
7.25E-8
2.07E-3
Electricity
2.66
3.75E-3
4.54
4.24E-7
-0.016
Sugarcane Ethanol
0.633
0.038
4.35
2.82E-6
0.757
Biogas from Cattle Dung
0.114
-
0.106
-
0.035
Biomass Pellets
1.520
0.006
0.502
5.31E-8
0.026
5.1 Cooking Fuel Mix Scenario Results - India
Cooking fuel mix scenario results provide the most comprehensive perspective on the
options for cookstove sector improvements included in Phase II of this work. Figure 5-1 shows
the effect of various model parameters on climate change impacts per GJ of cooking energy
delivered for India. Figure 5-1 also serves as a model for interpretation of subsequent figures in
this section. At the top of the figure is the baseline (current) fuel mix applying the best available
estimate of current stove technology use and stove thermal efficiency. Results for potential fuel
mix scenarios evaluated are presented according to a series of four technology options, as
described in Table 5-2.
Table 5-2. Cooking Fuel Mix Scenario Technology Options {Figure Key)
Fuel Mix Scenario Axis Labels
Description
Current Tech-Current Eff-Current
Grid1
Assumes current stove technology, current average stove
thermal efficiency values, and 2013 electrical grid mix.
Imp Tech-Current Eff-Current Grid
Assumes improved stove technology use, current average
stove thermal efficiency values, and 2013 electrical grid mix.
Imp Tech-Imp Eff-Current Grid
Assumes improved stove technology use, improved stove
thermal efficiency values, and 2013 electrical grid mix.
Imp Tech-Imp Eff-Clean Grid
Assumes improved stove technology use, improved stove
thermal efficiency values, and the use of clean electricity in
electric cookstoves.
1 Tech = stove technology, Imp = improved, Eff = stove efficiency
The current cooking fuel mix in India yields a GCCP of just below 200 kg of CO2
equivalent emissions per GJ of delivered cooking energy. Baseline cooking fuel mix results for
5-2
-------
Section 5—Updated LCA Results for India
all other countries analyzed in this study exceed 375 kg of CO2 equivalent emissions. The figure
demonstrates that realizing further GCCP impact reductions will be a challenge for India as the
country moves to adopt modern, fossil-based cooking fuels. This result is largely because
firewood and crop residue, which together comprise just under 60 percent of the current cooking
fuel mix, each have among the lowest single fuel GCCP impact scores of the commonly used
fuels. As discussed in Section 4.1 and Section 4.2, 100 percent of CO2 combustion emissions
associated with crop residue and 76 percent of CO2 combustion emissions associated with
firewood are considered to be from renewable biomass and therefore do not contribute to the
overall GCCP impact. The previous assumption for firewood is based on the Phase II baseline
forest renewability factor, which is subject both to uncertainty and the potential to change over
time. Increases in electricity use that do not assume reliance on a cleaner electricity grid lead to
increases in fuel mix GCCP. In general, scenarios that include appreciable quantities of electric
stove use are quite sensitive to the underlying electrical grid mix assumption. A maximum 33
percent reduction in GCCP is realized by the Diverse Modern Fuel mix scenario assuming
adoption of improved stove technology, improved stove thermal efficiency, and a clean
electricity grid.
5-3
-------
Section 5—Updated LCA Results for India
kg C02 eq/GJ Cooking Energy Delivered
0 50 100 150 200 250 300
Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
ci
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
-o
Imp Tech-Imp Eff-Current Grid
Q.
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
-o
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Hard Coal
¦ Dung C ake
¦ Crop Residue
¦ Firewood
¦ Charcoal from Wood
¦ Kerosene
¦ LPG
Natural Gas
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
¦ Biomass Pellets
Figure 5-1. India GCCP cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
Figure 5-2 presents PMFP cooking fuel mix results for India. In the current fuel mix
scenario, the traditional fuels contribute the vast majority of particulate matter emissions. All
four future cooking fuel mix scenarios propose reductions in the use of traditional fuels, which
leads to a minimum 46 percent reduction in PMFP impacts across all scenarios relative to the
baseline. Including the possibility of stove technology upgrades increases the range of impact
reductions to between 58 and 64 percent, depending upon the scenario. Within a given cooking
fuel mix, the adoption of improved stove technology and thermal efficiency is responsible for
between six and 19 percent of the total reduction in impact score. Cooking fuel mixes that rely
more heavily on traditional fuels benefit the most from stove technology and efficiency
upgrades. Figure 5-2 shows that the modest increases in charcoal use explored in the Improved
Biomass, Increased Electricity, and Diverse Modern Fuel mix scenarios contribute
5-4
-------
Section 5—Updated LCA Results for India
disproportionately to fuel mix PMFP impact, over 90 percent of charcoal's PMFP impact is due
to kiln emissions. BC impact results, provided in Appendix A, follow a trend very similar to
those exhibited by PMFP in Figure 5-2.
ro Current Tech-Current Eff-Current Grid
_CQ
Current Tech-Current Eff-Current Grid
o
o Imp Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
I Hard Coal
Firewood
LPG
I Sugarcane Ethanol
I Dung Cake
I Charcoal from Wood
Natural Gas
Biogas from Cattle Dung
I Crop Residue
Kerosene
I Electricity
iBiomass Pellets
Figure 5-2. India PMFP cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
As demonstrated in Figure 5-3, CED is greatest for the baseline scenario. Energy demand
in the baseline scenario is largely driven by the low thermal efficiency of traditional cookstoves
used to burn firewood, dung, and crop residue. The modest increase in reliance on charcoal and
5-5
-------
Section 5—Updated LCA Results for India
sugarcane ethanol present in both the Improved Biomass and Diverse Modern Fuel scenarios is
clearly visible in the figure due to inefficient energy conversion in these fuel types' respective
supply-chains. In the absence of stove technology improvements, the projected fuel mix
scenarios yield CED reductions between 20 and 29 percent, compared to the baseline scenario.
Further CED reductions (between 31 and 44 percent compared to the baseline scenario) can be
realized when the effect of stove technology and efficiency improvements are included. Modern
liquid/gas fuels and biomass pellets demonstrate the lowest CED of the cooking fuel types
considered for India.
MJ/GJ Cooking Energy Delivered
« Current Tech-Current Eff-Current Grid
CQ
Current Tech-Current Eff-Current Grid
1,000 2,000 3,000 4,000 5,000 6,000 7,000
I I
o
^r
o
c>
p
<
CQ
—
g
o
<2
•a
a
E
w
•o
-o
o
>
5
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
^ Current Tech-Current Eff-Current Grid
U.
c
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
•3 Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Hard Coal
¦ Dung Cake
¦ Crop Residue
Firewood
¦ Charcoal from Wood
¦ Kerosene
¦ LPG
Natural Gas
¦ Electricity
¦ Sugarcane Ethanol
CBiogas from Cattle Dung
¦ Biomass Pellets
Figure 5-3. India CED cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
5-6
-------
Section 5—Updated LCA Results for India
5.2 Baseline Normalized Results - India
The concept and methodology behind presenting normalized results are discussed in
Section 4.6. Generally, normalized results show which impact categories are most strongly
linked to the activity under study. If normalization factors were perfectly calibrated, the sum of
personal equivalent emissions for all sectors in the economy would equal the population of a
given country, which for India is approximately 1.3 billion people. However, normalized
emission estimates are uncertain due to lack of geographic granularity and the ever-changing
nature of national or global level estimates for many categories, indicating that the relative
magnitude of results (and not the specific person equivalency value) is of greater importance and
validity. Normalized results only indicate the contribution level of the cookstove sector to
national economy-wide impacts. Normalized results do not imply that impact categories are of
greater or lesser significance.
Normalized results for India are presented in Figure 5-4. PMFP, CED, and BC impact
categories all show a strong dependence on activity in the cooking sector at a national level. The
importance of the cooking sector to these impact categories indicates an opportunity to reduce
national environmental impacts by way of interventions in the cooking sector. Mitigating
combustion emissions associated with traditional biomass fuels through adoption of improved
stoves or decreasing the overall reliance on traditional biomass cooking fuels would lead to the
largest reductions in normalized results in India.
Normalized BC impacts exceed the expected maximum of 1.3 billion person equivalents,
likely due to a number of factors, including the large negative BC impact that is associated with
coal-based electricity production in India. To make up for this negative forcing, the total of other
sectors must be greater than 100 percent of net national characterized BC emissions. The
uncertainty of the BC impact assessment and emission inventories could also account for a
portion of the observed phenomena.
The figure shows very low normalized impacts of the cooking sector on ODP. Other
impact categories show only modest contributions from the cooking sector, indicating that
between two and six percent of national emissions in each category are associated with
household cooking.
5-7
-------
CTQ
C
m o
P
Is5
a-
n
Sg.
O
o
c-i
% a
fu £
£3 CP
-p
O &
CO Tl
•g S'
td o
ST
OJ r>
VI o
t^e.
§ tf>
P O
o 3
c ^
s§
O Q.
td ^
o' 3
I I
° s
as g
» *§
o I
in fcrj
n c
re S
a —¦
a 91
"* &
a 5'
Person Equivalents
H->
t—»
t—'
1—*
~—>
to
4^
oo
o
to
V
ON
"oo
o
o
©
o
o
o
o
o
o
o
©
©
©
©
©
©
©
©
"o
©
©
©
©
©
©
o
©
o
o
o
o
o
o
o
o
o
o
©
©
©
©
©
©
o
o
o
©
©
©
©
©
©
©
©
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modem Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diveise Modern Fuels
Current
BAU 2040
Improved Biomass
Increased Electricity
Diverse Modern Fuels
¦ n
it
ii
ii
in
¦ii
hi
cn
CD
3
&
£2
-------
Section 5—Updated LCA Results for India
5.3 Stove Efficiency Sensitivity - India
Figure 5-5 presents the effect of stove thermal efficiency improvement on PMFP of select
cookstoves in India. Each of the bars is labeled with the associated current or future, improved
stove thermal efficiency value for that stove group. The figure shows that fuel type is the greatest
determinant of PMFP impact among the traditional fuels in India. It is not possible, for example,
for dung cake burned in an improved cookstove to become a competitive option with either crop
residue or firewood regardless of the stove type used to burn these fuels. The figure also shows
that the movement from traditional to improved cookstoves provides more substantial reductions
in PMFP than does seeking the best possible thermal efficiency from traditional stoves.
Promotion of the highest possible thermal efficiency for improved cookstoves can yield
appreciable reductions in environmental impact as evidenced by the best performing crop residue
and firewood improved cookstoves realizing 24 and 17 percent reductions in PMFP,
respectively, relative to the current average thermal efficiency for these two stove groups.
5-9
-------
Section 5—Updated LCA Results for India
30.0
CD
>
15
Q
£p
-------
Section 5—Updated LCA Results for India
Figure 5-6 presents the effect of stove thermal efficiency improvement on CED of select
cookstoves in India. Each of the bars is labeled with the associated current or future, improved
stove thermal efficiency value for that stove group. CED is less strongly determined by fuel type
than is PMFP as is visible from a comparison of Figure 5-5 and Figure 5-6. Results for CED, like
PMFP, show that the movement to improved cookstoves still has more potential for
environmental impact reduction than is possible from stove thermal efficiency improvement
within a given stove type; however, the percent reductions in CED due to upgrades from
improved to traditional stoves are less dramatic than those demonstrated for PMFP because the
overall percent difference in CED between traditional and modern fuels is lower than that
observed for PMFP. Comparatively, the reductions in CED attributable to the assumed stove
thermal efficiency improvements are roughly equivalent to the reductions for PMFP, 24 and 17
percent for improved crop residue and firewood stoves, respectively.
5-11
-------
Section 5—Updated LCA Results for India
14,000
o 12,000
Si
I 10,000
CU
tJD
c
'J2
o
^ 8,000
O
6,000
4,000
2,000
0
Fuel Type:
Stove Type:
Stove
Efficiency:
Crop Residue
Traditional
Current
Crop Residue
Traditional
Future,
Improved
Crop Residue Crop Residue
Improved Improved
Current
Future,
Improved
Dung Cake
Traditional
Current
Dung Cake
Improved
Current
Dung Cake
Improved
Future,
Improved
Firewood
Traditional
Current
Firewood
Traditional
Future,
Improved
Firewood
Improved
Current
Firewood
Improved
Future,
Improved
Figure 5-6. Effect of stove thermal efficiency improvement on CED: traditional cooking fuel results in India.
5-12
-------
Section 5—Updated LCA Results for India
5.4 Electrical Grid Mix Sensitivity - India
Figure 5-7 shows the effect of the underlying electrical grid mix on GCCP of electric
cookstoves in India. Based on the modeled changes to the electrical grid between 2013 and 2021,
the GCCP of electric cookstoves can be reduced by between 16 and 33 percent. As demand for
electricity increases beyond 2021, it is possible that the carbon footprint of electricity production
could again increase as high carbon sources may be required to satisfy the growth in demand.
CED, FDP, WDP, PMFP, POFP, and Terrestrial Acidification Potential (TAP) all follow a trend
similar to GCCP.
I 450 £7 434
Q 400 379 ¦ 397
& 350 - ¦ M2 I
£ ¦296
300 — ¦ _ ¦
£ 250
| 200 1 1 1
1 z ~I I =t = = —76
8 I I I ¦ i
u u
PQ ^
^ ^ ^ ^ ^
Electricity ^ S> ^
Grid; ^ J* # J* # S J*
' ~ ^ J- ^ ^ sP"
Figure 5-7. Effect of electrical grid mix on GCCP impact of electric cookstoves in India.
BC impacts, as shown in Figure 5-8, tend to increase as the electrical grid mix moves
away from coal-based power generation and relies more heavily on other fuels. The TERI
Efficiency-2031 scenario still uses a significant amount of coal, but relies heavily on IGCC
technology that produces far lower sulfur dioxide emissions, resulting in a higher BC impact
score. Sulfur emissions associated with coal combustion generate a short-term cooling effect that
is responsible for the negative impact scores visible in the figure.
FEP and ODP impacts are highest for both the Low Carbon and IEA Blue Map grids.
Increases in ODP potential are contributed by solar, nuclear, and natural gas electricity
generation. Increases in FEP are due to the production of the electronic components in solar
panels.
5-13
-------
Section 5—Updated LCA Results for India
CD
(-
>
*o
a
C
W
M
'J2
o
o
U
a
cr
o
O
CQ
M
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
-0.01
-0.02
-0.03
3
-0.016
Electricity x
Grid: J?
9
y
&
0.018
i
-0.010
0.026
i
i
-0.013
oTV
$
\
0.073
-0.012
T
•N
C>
£
0.003
-0.011
0.029
t
T
4$
-------
Section 5—Updated LCA Results for India
600
T3
!—I
0>
.>
Q
C
W
M
S
3
o
o
O
O
CT
CI
o
o
M
543
500
400
300
200
100
0
Fuel Type:
Stove Type:
Forest
Renewability
Factor:
Firewood
Firewood
Firewood
Firewood
Firewood
Firewood
Biomass
Biomass
Charcoal
Charcoal
Pellets
Pellets
from Wood
from Wood
3-Stone
3-Stone
Traditional
Traditional
Improved
Improved
Improved
Improved
Traditional
Traditional
Low
High
Low
High
Low
High
Low
High
Low
High
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Estimate of
Renewability
Renewability
Renewability
Renewability
Renewability
Renewability
Renewability
Renewability
Renewability
Renewability
(59.2%)
(76.3%)
(59.2%)
(76.3%)
(59.2%)
(76.3%)
(59.2%)
(76.3%)
(59.2%)
(76.3%)
Figure 5-9. Comparative effect of forest product renewability assumption on GCCP in India.
5-15
-------
Section 5—Updated LCA Results for India
5.6 Allocation Approach Sensitivity - India
This section describes the effect of LCA modeling conventions on cookstove
environmental impacts for crop residue, biogas, and sugarcane ethanol. The crop residue
allocation issue centers on the question of how best to distribute environmental impacts of
agricultural production between food crops and other forms of agricultural biomass such as straw
and stover. For biogas, the modeling question concerns how to allocate impacts of anaerobic
digester operation between the biogas and the digestate. Sugarcane ethanol modeling choices
concern both the agricultural impacts of sugarcane production and how best to deal with surplus
electricity generated from bagasse combustion during molasses and ethanol production. Detailed
methodology options applied to address these allocation questions are outlined in Section 3.3.
Figure 5-10 shows the effect of LCA modeling choices on GCCP impact for sugarcane
ethanol and biogas. The sugarcane ethanol GCCP impact is particularly sensitive to the
application of system expansion, which credits the cookstove with avoided electricity production
that results from combustion of bagasse, a co-product of sugarcane cultivation, during molasses
and ethanol production. The magnitude of GCCP associated with avoided electricity production
is enough to make the life cycle impacts of sugarcane ethanol net negative from a climate change
perspective. The GCCP of sugarcane ethanol is not particularly sensitive to the choice between
economic and physical allocation, which yields a six percent difference in impact score. Overall,
the avoided electricity production in the ethanol supply chain has a beneficial effect on all impact
categories assessed.
As is evident in Figure 5-10, the GCCP of biogas is sensitive to all three allocation
modeling options. Economic allocation for biogas yields a GCCP impact score that is 46 percent
lower than the score that is calculated using the cut-off method, which attributes 100 percent of
anaerobic digestion impacts to biogas, thereby leaving digestate burden free. System expansion
produces a net negative impact score due to avoided fertilizer production when land- applying
the digestate. The same is true for FDP, WDP, POFP, and ODP. Impact categories strongly
affected by land application of digestate such as PMFP, FEP, TAP, and BC all yield impact
scores that are higher using the system expansion approach. Both PMFP and TAP are increased
due to ammonia emissions from the field.
Figure 5-11 shows that CED is less sensitive to the choice of modeling technique than
GCCP. The choice of economic allocation versus system expansion to model sugarcane ethanol
production leads to a 14 percent difference in estimated CED. Use of system expansion to model
biogas production yields a 38 percent reduction in estimated CED compared to the baseline, cut-
off method. The effects of LCA allocation approach selection on crop residue impacts are similar
for both India and China. Section 6.6 provides a detailed discussion using results for China to
illustrate trends.
5-16
-------
Section 5—Updated LCA Results for India
•o 140
g 121 ^130
I 120
I 100 l l
i 80 ¦ ¦
I 60
¦ ¦
8" 20
8 o ¦ ¦
^ -20 'H —
U A
O 0
OO
M -20
Fuel Type:
-11
-4
Sugarcane
Ethanol
Sugarcane
Ethanol
Sugarcane
Ethanol
Biogas from
Cattle Dung
Biogas from
Cattle Dung
Biogas from
Cattle Dung
Stove Type:
Modern
Modern
Modern
Modern
Modern
Modern
Allocation
Method:
Allocation,
Physical
(Baseline)
Allocation,
Economic
System
Expansion
System
Expansion
Cut-off
(Baseline)
Allocation,
Economic
Figure 5-10. Effects of allocation methodology choice on sugarcane ethanol and biogas
GCCP impact in India.
-o 16,000
14,000
13,562
13.334
11.700
SB 12>000
10.000
8.000
6.000
4.000
2.000
Sugarcane
Sugarcane
Sugarcane
Biogas from
Bioeas from
Biogas from
Ethanol
Ethanol
Ethanol
Cattle Dung
Cattle Dung
Cattle Dung
Modern
Modern
Modern
Modern
Modern
Modern
Allocation,
Allocation,
System
System
Cut-off
Allocation,
Physical
Economic
Expansion
Expansion
(Baseline)
Economic
(Baseline)
Fuel Type:
Stove Type:
Allocation
Method:
Figure 5-11. Effects of allocation methodology choice on sugarcane ethanol and biogas
CED impact in India.
5-17
-------
Section 5—Updated LCA Results for India
5.7 Stove Group Uncertainty Results - India
Figure 5-12 shows GCCP for traditional biomass fuels in India. All results in the figure
represent the baseline LCA modeling convention. The height of the bar in each figure represents
the Monte Carlo mean around which the error bars are centered. The analysis mean (triangle in
figure) for each impact category is the characterized expected value as it was entered into the
openLCA model. The analysis and Monte Carlo mean deviate from one another depending upon
the distribution used, and in the case of lognormally distributed data, depending upon the
geometric standard deviation.
The uncertainty range associated with GCCP of rice production has a strong effect on the
impacts of cooking with crop residue if the mass ratio of production between food crops and crop
residues is used to allocate the impacts of agricultural production. The height of the uncertainty
band is due to the range of potential CH4 emissions associated with rice production. Methane
emissions per kg of rice are dependent upon yield, duration of inundation, and incorporation of
organic matter. Use of the cut-off method to model cooking with crop residue represents only the
impacts of the cooking process itself and assumes that 100 percent of agricultural impacts are
attributable to food crops, thereby treating crop residues as a waste or a secondary by-product
that does not contribute to the demand for agricultural production. GCCP uncertainty ranges for
other stove and fuel types are generally much more narrow relative to the impact score of a given
stove, which improves the ability to discern true differences in the climate impact of stoves
burning different fuel types, as well as between improved and traditional stoves for a given fuel.
The figure shows, for example, minimal overlap between the uncertainty ranges of improved and
traditional stoves burning dried dung cake, which indicates that justifiable reductions in GCCP
are possible if households upgrade to improved stoves.
Figure 5-13 shows POFP impact scores for the same stoves and fuels displayed in the
previous figure. Generally, the uncertainty ranges associated with the emission of volatile
organic compounds, which contribute to this impact category, are wider than they are for all
other impact categories in this study. This wide range obscures the ability to differentiate
between certain stove-fuel combinations regarding their contribution to POFP impact. Dung cake
appears to have much higher potential to contribute POFP emissions, while improved stoves
burning firewood realize a significant reduction in POFP emissions as compared to other stoves
burning traditional fuels.
Figure 5-14 shows WDP impact scores for the fuels that demonstrate the greatest WDP
potential per GJ of delivered cooking energy. Normalized results in the previous section indicate
that most fuels contribute very marginally to WDP at the national level. Figure 5-14 confirms
that WDP for cooking fuels is predominantly due to the production of agricultural crops.
However not all crops place similar demands on water resources. Water use of sugarcane is
noticeably lower than for both rice and wheat, due to the significantly greater yield potential of
sugarcane per hectare. The share of WDP that is attributed to crop residues is shown to be highly
sensitive to the choice of LCA modeling technique as demonstrated by the disparity in WDP
impacts between the cut-off method and physical allocation. The cut-off method attributes 100
percent of water demand to food crop production, leaving crop residues burden free.
5-18
-------
Section 5—Updated LCA Results for India
A notable contrast from Phase I results regards the contribution of electricity production
to WDP impacts. Electricity was shown to generate very high water use, largely due to turbine
water use necessary for hydropower production. Phase II results are adjusted to reflect the fact
that turbine water is still available for environmental use and is not considered to contribute to
WDP, significantly reducing the WDP of electricity production. Hard coal is included in the
figure to provide a reference point for the WDP of fossil fuels.
5-19
-------
Section 5—Updated LCA Results for India
4,000
3,500
3,000
2,500
"O
5-.
Q
c
W
bD
| 2,000
o
o
U
—i
O
-------
Section 5—Updated LCA Results for India
50
P 20
fP 10
Fuel Type:
Stove Type:
Allocation
Method:
Crop
Crop
Crop
Crop
Crop
Crop
Dung Cake
Dung Cake
Firewood
Firewood
Firewood
Residue
Residue -
Residue -
Residue
Residue -
Residue -
Rice Straw
Wheat
Rice Straw
Wheat
Traditional
Traditional
Traditional
Improved
Improved
Improved
Traditional
Improved
3-Stone
Traditional
Improved
Cut-off
Allocation
Allocation
Cut-off
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Monte Carlo Mean AAnalysis Mean
Figure 5-13. India POFP uncertainty analysis results for traditional cooking fuels modeled with various allocation approaches
and stove technologies.
5-21
-------
Section 5—Updated LCA Results for India
1,200
-s 1,000
§ 800 jg | , J
g
IB
W 600 HI == —
1 l.ll.i II
Crop Crop Crop Crop Crop Crop Crop Hard Coal Electricity Sugarcane Sugarc;
Fuel Type: Residue Residue - Residue - Residue - Residue - Residue - Residue - Ethanol Ethan
Rice Straw Sugarcane Wheat Rice Straw Sugarcane Wheat
Tops Tops
Stove Type:
TVaHi ti r^ncil TYciHi tinnal lTVc»r1i ti r\nc»1 TVqHi tir\nQ 1 Tmr»rr>\r<=»r1 Tmrvrrwr^H Tmrvrr\\7f»H TrQr1itir\r»c» 1 \/T prn \/T /^rn \/fr\H^i
0
A
*
**"
Crop
Crop
Crop
Crop
Crop
Crop
Crop
Hard Coal
Electricity
Sugarcane
Sugarcane
Fuel Type:
Residue
Residue -
Residue -
Residue -
Residue -
Residue -
Residue -
Ethanol
Ethanol
Rice Straw
Sugarcane
Wheat
Rice Straw
Sugarcane
Wheat
Tops
Tops
Stove Type:
Traditional
Traditional
Traditional
Traditional
Improved
Improved
Improved
Traditional
Modern
Modern
Modem
Allocation
Cut-off
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
Allocation
System
Method:
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Expansion
¦ Monte Carlo Mean A Analysis Mean
Figure 5-14. India WDP uncertainty analysis results for select cooking fuels modeled with various allocation approaches and
stove technologies.
5-22
-------
Section 6—Updated LCA Results for China
6. UPDATED LCA RESULTS FOR CHINA
In China, only 27 percent of national cooking energy is provided by traditional biomass
fuels, and China is the only country studied to rely on coal for a notable portion of its cooking
energy. Modern fuels such as LPG, electricity, and natural gas provide nearly 45 percent of
cooking fuel, which is a higher proportion of cooking energy than is observed in other nations
studied (Dalberg 2014). China is characterized by an electrical grid that is associated with high
environmental impacts due to heavy reliance on coal. These aspects of the current cooking fuel
mix have a significant impact on the trends observed in the LCA results for China. China is also
more industrialized than the other countries studied, and therefore has a correspondingly higher
national energy demand (Enerdata 2016), which impacts normalized LCA results presented in
Section 6.2.
Table 6-1 summarizes LCA results for all fuel types and impact categories in China. The
results are representative of baseline assumptions concerning cooking fuel mix, stove technology
use, stove thermal efficiency, electricity grid, and forest renewability factor, as documented in
Sections 2.2.2 and 3.1.2. A discussion of notable changes between results for Phase I and Phase
II is provided in Appendix B. Various forms of coal and electricity demonstrate the greatest
GCCP, significant since all three figure prominently within the current cooking fuel mix. Coal
powder and sugarcane ethanol demonstrate the greatest CED, the former because of low stove
thermal efficiency, and the latter because of energy loss in the supply chain.
Table 6-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - China
GCCP
CED
FDP
WDP
PMFP
Fuel Type
(kg C02 eq)
(MJ)
(kg oil eq)
(m3)
(kg PM10 eq)
Coal Powder
1.16E+3
1.08E+4
254
1.15
21.5
Coal Briquettes
593
5.37E+3
125
0.986
0.989
Coal Honeycomb
527
6.37E+3
149
0.899
1.08
Firewood
190
7.61E+3
3.63E-3
4.00E-5
6.50
Crop Residue
64.1
7.45E+3
0.010
1.18E-4
10.0
Kerosene
225
3.53E+3
76.9
0.480
0.266
Biomass Pellets
140
3.78E+3
12.3
0.417
0.311
Electricity
612
7.22E+3
118
4.07
1.65
LPG
213
3.41E+3
74.4
0.461
0.248
Natural Gas
154
2.37E+3
55.3
0.025
0.048
Coal Gas
254
3.69E+3
82.3
0.576
0.495
Sugarcane Ethanol
113
1.31E+4
24.9
643
4.33
Biogas from Cattle
Dung
11.4
4.06E+3
-
1.02
0.210
1 1
6-1
-------
Section 6—Updated LCA Results for China
Table 6-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - China
POFP
FEP
TAP
ODP
BC
Fuel Type
(kg NMVOC)
(kg P eq)
(kg SO2 eq)
(kg CFC -11
eq)
(kg BC eq)
Coal Powder
3.30
0.116
1.66
1.13E-7
4.45
Coal Briquettes
0.700
0.091
1.20
1.27E-7
0.105
Coal Honeycomb
1.82
0.084
1.05
1.15E-7
0.160
Firewood
2.23
4.50E-3
0.242
3.34E-11
1.42
Crop Residue
5.52
0.013
0.367
9.52E-11
2.20
Kerosene
0.582
0.013
0.960
1.85E-7
-0.032
Biomass Pellets
1.340
0.012
0.53
2.54E-8
0.031
Electricity
2.31
0.078
5.27
1.67E-7
-0.148
LPG
0.500
0.012
0.898
1.81E-7
-0.031
Natural Gas
0.181
6.80E-4
0.143
9.74E-7
-2.04E-3
Coal Gas
1.31
0.042
0.803
1.39E-5
0.038
Sugarcane Ethanol
0.511
0.039
4.33
2.75E-6
0.748
Biogas from Cattle
Dung
0.114
-
0.106
-
0.035
6.1 Cooking Fuel Mix Scenario Results - China
Cooking fuel mix scenario results provide the most comprehensive perspective on the
options for cookstove sector improvements included in the second phase of this project. Table
6-2 provides a guide to interpretation of bar axis labels.
Table 6-2. Cooking Fuel Mix Scenario Technology Options {Figure Key)
Fuel Mix Scenario Parameter Options
Description
Current Tech-Current Eff-Current Grid1
Assumes current stove technology, current average
stove thermal efficiency values, and 2013 electrical grid
mix.
Imp Tech-Current Eff-Current Grid
Assumes improved stove technology use, current
average stove thermal efficiency values, and 2013
electrical grid mix.
Imp Tech-Imp Eff-Current Grid
Assumes improved stove technology use, improved
stove thermal efficiency values, and 2013 electrical grid
mix.
Imp Tech-Imp Eff-Clean Grid
Assumes improved stove technology use, improved
stove thermal efficiency values, and the use of clean
electricity in electric cookstoves.
1 Tech = stove technology, Imp = improved, Eff = stove thermal efficiency
Figure 6-1 shows the effect of various model parameters on GCCP impact per GJ of
cooking energy delivered for China. The baseline (current) cooking fuel mix yields
approximately 420 kg of CO2 equivalent emissions per GJ of cooking energy delivered.
6-2
-------
Section 6—Updated LCA Results for China
Assuming constant use of stove technology and current stove thermal efficiencies, the BAU 2030
cooking fuel mix realizes only a seven percent reduction in GCCP. This cooking fuel mix
assumes that relative reliance on coal reduces slightly, while the use of crop residue and
firewood falls by over 50 percent to provide 11 percent of national cooking energy. LPG and
coal gas use both increase in the BAU 2030 scenario. If improved stove technologies are adopted
and stove thermal efficiencies improve for each fuel type within the scenario, the reduction in
GCCP impact increases to nearly 26 percent below the current baseline cooking fuel mix impact.
Overall, 58 percent of potential reductions, relative to the baseline, for the BAU 2030 scenario
are attributable to stove technology and efficiency upgrades.
The Diverse Modern Fuel mix for China yields the greatest overall reduction in climate
change impact. The cooking fuel mix alone realizes a 31 percent reduction in GCCP impact as
compared to the baseline, which can be improved to 48 percent if stove technology and
efficiency upgrades are achieved. Electric cookstoves are assumed to provide greater than 15
percent of cooking energy in three of the four future cooking fuel mix scenarios. In these
scenarios, electric cookstoves are a significant contributor to GCCP, which makes the results
sensitive to future changes in electrical grid mix. As an example, applying the assumption of a
clean electricity grid to the Increased Electricity scenario improves GCCP impact reductions
from 27 percent to 52 percent of the baseline impact.
50
kg C02 eq/GJ Cooking Energy Delivered
100 150 200 250 300 350
400
450
c3 Current Tech-Current Eff-Current Grid
CQ
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
<
CQ
I £
.2 o
CQ '£
c W
-a
<
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
>
Q
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Coal Powder
¦ Coal Briquettes
¦ Coal Honeycomb
¦ Crop Residue
¦ Firewood
¦ Kerosene
¦ LPG
¦ Natural Gas
Coal Gas
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
Biomass Pellets
Figure 6-1. China GCCP cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
6-3
-------
Section 6—Updated LCA Results for China
Figure 6-2 demonstrates that reductions in PMFP are more sensitive to future cooking
fuel mix upgrades than GCCP. PMFP impacts are also more sensitive, for a given cooking fuel
mix, to stove technology and efficiency upgrades. Depending on the cooking fuel mix, stove
technology and efficiency upgrades are able to reduce PMFP impacts by an additional four to 17
percent beyond those attributable to cooking fuel mix shifts alone. Cooking fuel mixes that rely
more heavily on traditional coal and biomass fuels are the most sensitive to stove technology
improvement. In the absence of stove technology and efficiency changes, reductions in PMFP
impact between 31 and 81 percent are achievable through cooking fuel shifts alone. Movement
away from the use of coal powder, firewood, and crop residue is responsible for most potential
PMFP reductions. Even modest changes in the cooking fuel mix such as those represented by the
BAU 2030 scenario can drastically reduce PMFP emissions.
o.o
1.0
kg PM10 eq/GJ Cooking Knergy Delivered
2.0 3.0 4.0 5.0
6.0
Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
o
m
o
(N
~
<
CQ
-o
o •ti
C/2 o
$ "C
O V
~ 5
I *
.2 o
CQ "C
W
>
-a
<
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
s
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Coal Powder
¦ Coal Briquettes
¦ Coal Honeycomb
¦ Crop Residue
¦ Firewood
¦ Kerosene
¦ LPG
¦ Natural Gas
Coal Gas
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
Biomass Pellets
Figure 6-2. China PMFP cooking fuel mix results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
Figure 6-3 shows that adoption of the future cooking fuel mixes alone can reduce CED
by between 15 and 34 percent. Lower CED is mostly attributable to reduced demand for coal
powder, firewood, and crop residue, which have three of the four highest CED values of all the
6-4
-------
Section 6—Updated LCA Results for China
Chinese cooking fuels included in this study. For the BAU 2030 and Increased Electricity
cooking fuel mix scenarios, the additional adoption of improved stove technologies and
increased thermal efficiencies can double the reductions possible through cooking fuel mix shifts
to 34 and 32 percent, respectively. The Diverse Modern Fuels scenario holds the greatest
potential for CED reduction, realizing a near 50 percent reduction in current energy demand per
GJ of cooking energy assuming the adoption of improved stove technology.
Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
o Imp Tech-Current Eff-Current Grid
|D
<
m
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
S '3 Imp Tech-Current Eff-Current Grid
« -c
O
£ £2
% &
.2 o
m -q
w
<*)
-o
I -a
^ o
o 3
jo Cl.
>
3
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
1,000 2,000
MJ/GJ Cooking Energy Delivered
3,000 4,000 5,000 6,000
7,000
¦ Coal Powder
¦ Coal Briquettes
¦ Coal Honeycomb
¦ Crop Residue
¦ Firewood
¦ Kerosene
¦ LPG
¦ Natural Gas
Coal Gas
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
Biomass Pellets
Figure 6-3. China CED cooking fuel mix results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
6.2 Baseline Normalized Results - China
The concept and methodology behind a normalized presentation of results is included in
detail in Section 4.6. Generally, the results show which impact categories are most strongly
linked to the activity studied. If normalization factors were perfectly calibrated, the sum of
personal equivalent emissions for all sectors would equal the population of a given country,
which for China is approximately 1.4 billion people. However, normalized emission estimates
6-5
-------
Section 6—Updated LCA Results for China
are uncertain due to lack of geographic granularity and the ever-changing nature of national or
global level estimates for many categories, indicating that the relative magnitude of results and
not the specific person equivalency value is of greater importance and validity. Normalized
results indicate only the level of the cookstove sector contribution to the national economy-wide
impacts; they do not imply that impact categories are of greater or lesser significance.
Figure 6-4 indicates that greater than 15 percent of national emissions contributing to
PMFP, FEP, and BC are attributable to the cooking sector. PM and BC emissions in large part
are produced by coal powder, firewood, and crop residue combustion. Other impact categories
show a moderate link to the cooking sector with between five and ten percent of national
emissions attributable to the cooking sector. WDP and ODP are the exception and do not appear
to be significantly linked to the cooking sector in China. Normalized CED is lower for China
than for other countries studied due to China's greater level of industrialization and higher
energy demand per capita.
6-6
-------
crs
=
-!
«
OS
s
as
s
o
"?
3
&
N*
n
a.
n
>
¦I
>v:;
O,
2? ^ o
ru o
a
13
o
5i
Cu
® £ °
K' ^ °
(Sob.
8 w
i-K hi.
o -2
B 5
O 3
pa co
£
rT
u
g
ts z; n
K" B) °
I g &
K £ ®
:o§
Jp 63 O
£L K ^
O O
§->§
O W
ES «
tn ti
rtT
£ §
Is
V? P-
° 3
#,-¦ s
S S
7I I
C/3 ^
2 c
I SL
3. S
® S"
Person Equivalents
to
o
©
"o
o
o
-p.
o
©
"o
o
o
o
©
o
o
o
o
p
"o
o
o
o
o
o
o
(O
o
o
"o
o
o
o
"o
o
o
o
o
o
o
"o
o
o
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modern Fuels
Current
BAU 2030
Increased Electricity
Advanced Biomass & Electricity
Diverse Modem Fuels
I
z/i
cd
o
5'
3
CTn
£
&
»
CD
CL
r
n
>
2?
CD
n
e
3
02
-------
Section 6—Updated LCA Results for China
6.3 Stove Efficiency Sensitivity - China
Figure 6-5 presents GCCP for various coal-based cooking options in China. Each of the
bars are labeled with the associated current or future improved stove thermal efficiency value for
that stove group. Upstream energy losses, particularly associated with electricity production, are
not included in these thermal efficiency values. The figure clearly shows the benefit of increased
coal-based stove thermal efficiency in reducing climate impacts associated with cooking, even as
the underlying feedstock remains the same or similar. The improvement demonstrated in the
figure is not due solely to stove thermal efficiency, however, as fuel form also changes across the
individual stove groups. The potential benefit of increasing stove thermal efficiency within a
given stove group (e.g., coal powder, traditional) as opposed to adoption of improved stoves or
fuel forms varies. The GCCP of coal powder-based cookstoves varies considerably from 840 to
nearly 1400 kg CO2 eq per GJ of delivered cooking energy. Overall, this range represents a
nearly 40 percent potential reduction of GCCP impact when switching from traditional to
improved coal powder stoves. The best performing traditional coal powder stoves can realize a
29 percent reduction in GCCP impact, relative to the current average stove thermal efficiency,
for the same stove type. More advanced forms of coal fuel demonstrate less variation in stove
thermal efficiency within a given stove group (e.g., honeycomb coal, improved) and upgrades in
fuel form (e.g., from powder to honeycomb briquettes) present even more potential to mitigate
impacts.
6-8
-------
Section 6—Updated LCA Results for China
1,600
Q
§
c
w
60
c
5
c
o
U
o
d-
-------
Section 6—Updated LCA Results for China
6.4 Electrical Grid Mix Results - China
Figure 6-6 shows that with the adoption of an improved electrical grid, it is possible to
reduce climate change impacts associated with electric cookstove use by nearly 90 percent
(LBNL AIS-2050 scenario). More modest interventions such as those represented by the LBNL
CIS grid estimate for the year 2050 indicate a potential 43 percent reduction even in the absence
of stove efficiency improvements. Reductions are attributable both to shifts in the electrical grid
fuel mix and to the adoption of more advanced electricity generation technology that reduces fuel
consumption per unit of delivered energy.
The best performing grids represent rapid departures from the current electricity fuel mix
in China. The LBNL AIS 2050 scenario relies heavily on nuclear technology, while the IEA
Blue Map scenario relies on an even mixture of renewables, gas, nuclear, and advanced coal
technology. The affect that these improvements could have on the relative performance between
electricity and other cooking fuels is an important consideration and can be explored in the
results files for each country. While most impact categories follow a downward trend similar to
that exhibited by climate change, ODP and BC impacts are two exceptions. ODP impacts tend to
increase as more natural gas is included in the grid, while BC impacts increase as coal
combustion is reduced. The high relative ODP of natural gas is due to emissions associated with
long distance transport via pipeline. Sulfur dioxide emissions produced during coal combustion
exhibit a short-term cooling effect, which account for the potential increase in BC impact as coal
use drops.
^ 700
o
Vh
o
§ 600
£P
O
W 500
ep
2
o
6 400
O
(N 300
o
o
^ 200
100
00
Electricity
Grid:
Figure 6-6. Effect of electrical grid mix on GCCP impact of electric cookstoves in China.
612
2013 Grid EIA - 2030 LBNL CIS - LBNL AIS - LBNL CIS - LBNL AIS - IEA Blue
(IEA) 2030 2030 2050 2050 Map - 2050
6-10
-------
Section 6—Updated LCA Results for China
6.5 Forest Renewabilitv Factor Sensitivity - China
The use of two separate methodologies to determine the fraction of forestry products that
are renewably produced and are therefore carbon neutral leads to a 20 percent difference in the
estimate of total forest products that are derived from sustainable operations. The assumptions
behind these two methods were discussed previously in Section 4.2. The baseline values for
Phase II of this study are from the WISDOM database (Drigo 2014) and lead to reductions in
GCCP impact between 31 and 40 percent as compared to impacts associated with the low
renewability factor used as the baseline in Phase I (Figure 6-7). The difference between the two
forestry renewability factors affects the relative GCCP impact of the stoves depicted below and
the modern liquid/gas fuels. In specific instances, the choice of renewability factor is enough to
influence whether adoption of modern fuels yields an increase, decrease, or no significant effect
on emissions contributing to climate change.
« 400
2?
! 350
a
& 300
w
250
200
O 150
O
cr
o
-------
Section 6—Updated LCA Results for China
modeling question concerns how to allocate impacts of anaerobic digester operation between the
biogas and the digestate. Sugarcane ethanol modeling questions concern both the agricultural
impacts of sugarcane production and how best to deal with surplus co-produced electricity
generated from bagasse during molasses and ethanol production.
Figure 6-8 (below) and Figure 6-9 demonstrate the effect of LCA allocation modeling
choices on GCCP and PMFP impacts of stoves burning crop residue. When comparing the
figures, it is clear that modeling choice has a variable effect depending upon the impact category
being considered. GCCP impacts are not affected by the choice between a cutoff approach and
system expansion, whereas the opposite is true for particulate emissions. The specific form of
system expansion employed assumes that crop residue utilized in cookstoves avoids the field
burning of those same residues. The biogenic origin of crop residues yields a limited climate
impact, but the more controlled burning of crop residues in cookstoves produces a significant
quantity of avoided particulate emissions from field burning. In general, GCCP is shown to
increase if either physical or economic allocation is used to assign a portion of agricultural
impacts to crop residue, and the specific type of crop modeled also influences GCCP results.
PMFP shows a much lower sensitivity to the use of allocation and to the choice between physical
or economic factors because the majority of PMFP impact is associated with the use phase when
applying a cutoff, economic, or physical allocation approach. Fossil depletion, water depletion,
eutrophication, acidification, and ozone depletion all show noticeable increases if a portion of
agricultural impacts are allocated to the residue. Like PMFP, BC impacts are also strongly
affected if the system expansion approach is used. Photochemical oxidant formation potential is
not shown to be sensitive to any of the LCA modeling choices.
6-12
-------
Section 6—Updated LCA Results for China
700
644
600
-------
Section 6—Updated LCA Results for China
25
> 20
"o
Q
EP
O
[§ 15
op
o
o
O
1—-
g
10
a-
CD
M
-10
-15
Fuel Type:
Allocation
Method:
Crop Residue
Crop Residue
Crop Residue -
Crop Residue -
Crop Residue -
Crop Residue -
Crop Residue -
Crop Residue -
Crop Residue -
Crop Residue -
Maize
Maize
Rice
Rice
Wheat
Wheat
Average
Average
Cut-off
System
Allocation,
Allocation,
Allocation,
Allocation,
Allocation,
Allocation,
Allocation,
Allocation,
(Baseline)
Expansion
Physical
Economic
Physical
Economic
Physical
Economic
Physical
Economic
Figure 6-9. Effects of LCA allocation approach on crop residue PMFP impact in China.
6-14
-------
Section 6—Updated LCA Results for China
Figure 6-10 shows that the TAP impact of biogas and sugarcane ethanol are sensitive to
the system expansion approach, which is used to avoid allocating the burdens of multi-output
processes. Acidification impacts associated with sugarcane ethanol are reduced as a result of the
avoided electricity production generated because of bagasse combustion. Avoided electricity
production also reduces environmental impacts for GCCP, FDP, POFP, and FEP between 60 and
160 percent, relative to the use of physical allocation. Results in other impact categories are less
sensitive to this choice. The consideration of system expansion results for estimating the
environmental impact of sugarcane ethanol use is justified for a number of reasons: (1) sugarcane
is the main agricultural product and all LCA modeling approaches agree that it should be
attributed a majority share of agricultural impacts, and (2) electricity production is a high value
use of by-product bagasse, and the electricity is often used as a direct input within ethanol
production and processing.
System expansion can be used to avoid the need to allocate impacts between biogas and
the solid digestate that exit the digester. The approach credits both the environmental benefits
and burdens of digestate land application to the biogas. For a number of impact categories like
TAP, this approach leads to a marked increase in impacts associated with biogas. System
expansion also leads to a greater than 30 percent decrease in GCCP, CED, WDP, and POFP
impact. Environmental benefits realized are generally due to avoided fertilizer production,
whereas increased environmental impacts are a result of emissions associated with the land
application of digestate.
The system expansion approach is valid for biogas only if the digestion process impacts
the decision to utilize the digestate as a fertilizer and soil amendment or if it has an impact on the
quality of the product destined for use as an agricultural amendment. A review of the literature
indicates a slight and somewhat variable impact of the digestion process on the fertilizer value of
the digestate, when compared to the application of unprocessed manure. If this difference were
more pronounced, then the net effect of the digestion process on avoided fertilizer production
and agricultural emissions would be of greater importance to this analysis. In the absence of this
observation, the cutoff approach is the more justifiable choice for modeling the environmental
impacts of biogas production, which reinforces its choice as the baseline method for this
analysis.
6-15
-------
Section 6—Updated LCA Results for China
6.0
-o
flj
Q 5.0
&
-------
Section 6—Updated LCA Results for China
1,400
1.200
>
O
Q
&
o
W
cp
3
o
o
u
g
u
(N
o
u
ca
1,000
800
600
400
200
0
Fuel Type:
Stove Type:
Allocation
Method:
Crop
Firewood
Biomass
Coal
Coal
Coal
Coal Gas
Kerosene
LPG
Natural Gas
Electricity
Ethanol
Biogas
Residue
Pellets
Powder
Briquettes
Honeycomb
Improved
Improved
Improved
Improved
Improved
Improved
Traditional
Improved
Modern
Modern
Modem
Modern
Modem
Cut-off
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Allocation.
Cut-off
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
Physical
¦ Monte Carlo Mean AAnalysis Mean
Figure 6-11. China GCCP uncertainty analysis results for improved stoves and modern cooking fuels modeled with various
allocation approaches.
6-17
-------
Section 6—Updated LCA Results for China
Figure 6-12 presents uncertainty ranges for PMFP impact scores for a wide range of
cooking fuels and stove types. The uncertainty ranges associated with particulate emissions are
generally wider than those for GCCP, which makes it difficult to distinguish differences between
the stoves burning traditional fuels, based on results for this impact category. According to this
figure, a strategy that promotes an upgrade from traditional to improved stove designs for the
traditional fuels may not prove effective at reducing PM emissions in China. Improved forms of
the traditional fuels such as honeycomb briquettes (coal) and biomass pellets (firewood) do
realize significant reductions in PM emissions. From the perspective of PM emissions, the use of
coal-based electricity as a cooking fuel yields emissions comparable to those possible with
honeycomb briquettes. Modern liquid and gas fuels all yield significantly lower PM emissions
than the traditional fuels when burned in either traditional or improved stove types.
6-18
-------
Section 6—Updated LCA Results for China
90
80
a 70
-------
Section 7—LCA Results for Kenya
7. LCA RESULTS FOR KENYA
Sixty-five percent of Kenyan households rely on firewood for cooking energy (KNBS
2012), and the major portion of this fuel is still consumed in three-stone cookstoves (Githiomi et
al. 2012, SEI 2016). Compared to other nations studied, Kenya is relatively early in the transition
to the use of improved or modern fuels and stoves. Charcoal use is slowly on the rise and now
comprises 17 percent of national cooking energy (CBS 2002, KNBS 2012). Kerosene use,
currently providing less than 12 percent of cooking energy, appears to be beginning to fall in
favor of modern fuels such as LPG (Dalberg 2013a). The results presented here explore how the
environmental impact of the cooking sector will change if Kenya's cooking sector evolves
similarly to the transition that has already been realized in Ghana, or if other pathways provide
greater opportunities.
Table 7-1 presents summarized LCA results for all fuel types and impact categories in
Kenya. The results are representative of baseline assumptions concerning cooking fuel mix,
stove technology use, stove thermal efficiency, electricity grid, and forest renewability factor.
Charcoal and firewood carry significantly greater GCCP impacts than do other cooking fuel
options due to poor efficiency of current stove and kiln technologies and the high percentage of
Kenyan forest products that are harvested using unsustainable practices. As with other nations,
biogas and sugarcane ethanol demonstrate the lowest GCCP. Charcoal and ethanol demonstrate
the highest CED due to energy losses during processing. Significantly greater processing energy
losses are associated with African-produced fossil fuels than are reported for either India or
China, leading to greater potential impacts across most of the reported categories. In Kenya,
firewood produces more PM emissions than do charcoal stoves per GJ of delivered energy, due
to the continued reliance on three stone fires. POFP of charcoal is double the POFP reported in
India, attributable to higher reported values of NMVOC emissions at the kiln. PMFP of charcoal
is roughly half of the PMFP observed in India, with the differences in reported kiln emissions
responsible for the difference. Differences in the underlying electricity mix in each nation help to
explain some of the difference in impact scores between countries. The Kenyan electricity mix
demonstrates environmental impacts competitive with other modern fuels given that 66 percent
of power is generated from hydroelectric and geothermal sources (IE A 2013 c). The relative
impact scores for other cooking fuel types and impact categories largely follow trends similar to
those demonstrated by other nations investigated in this study.
Table 7-1. Summary Table of Single Cooking Fuel Results by Impact Category
(Impact/GJ Delivered Cooking Energy) - Kenya
GCCP
CED
FDP
WDP
PMFP
Fuel Type
(kg C02 eq)
(MJ)
(kg oil eq)
(m3)
(kg PM10 eq)
Firewood
439
9.11E+3
7.20E-3
7.96E-5
15.5
Charcoal from Wood
808
1.22E+4
5.35
0.068
8.40
Biomass Pellets
261
4.21E+3
19.4
0.635
0.152
Kerosene
223
7.96E+3
186
0.856
0.202
LPG
216
7.51E+3
175
0.816
0.196
Electricity
238
7.40E+3
150
5.58
0.331
7-1
-------
Section 7—LCA Results for Kenya
Table 7-1. Summary Table of Single Cooking Fuel Results by Impact Category
(Impact/GJ Delivered Cooking Energy) - Kenya
GCCP
CED
FDP
WDP
PMFP
Fuel Type
(kg C02 eq)
(MJ)
(kg oil eq)
(m3)
(kg PM10 eq)
Biogas from Cattle
Dung
13.4
4.21E+3
-
3.29
0.187
Sugarcane Ethanol
90.0
1.31E+4
26.0
643
4.25
POFP
FEP
TAP
ODP
BC
Fuel Type
(kg NMVOC)
(kg P eq)
(kg SO2
eq)
(kg CFC-11
eq)
(kg BC eq)
Firewood
5.87
8.92E-3
0.226
6.62E-11
3.32
Charcoal from Wood
22.9
0.011
0.208
4.77E-8
2.03
Biomass Pellets
1.47
5.66E-3
0.186
2.31E-8
0.026
Kerosene
0.832
9.96E-3
0.540
1.30E-7
5.70E-4
LPG
0.783
0.011
0.524
1.48E-7
1.70E-4
Electricity
1.46
1.85E-3
1.17
3.08E-8
-0.032
Biogas from Cattle
Dung
0.084
-
5.13E-3
-
0.040
Sugarcane Ethanol
0.431
0.034
4.08
2.74E-6
0.755
7.1 Cooking Fuel Mix Scenario Results - Kenya
Cooking fuel mix scenario results provide the most comprehensive perspective on the
options for cookstove sector improvements included in the second phase of this project. Table
7-2 provides a guide to interpretation of bar axis labels.
Table 7-2. Cooking Fuel Mix Scenario Technology Options {Figure Key)
Fuel Mix Scenario Parameter Options
Description
Current Tech-Current Eff-Current Grid1
Assumes current stove technology, current average stove
thermal efficiency values, and 2013 electrical grid mix.
Imp Tech-Current Eff-Current Grid
Assumes improved stove technology use, current average
stove thermal efficiency values, and 2013 electrical grid
mix.
Imp Tech-Imp Eff-Current Grid
Assumes improved stove technology use, improved stove
thermal efficiency values, and 2013 electrical grid mix.
Imp Tech-Imp Eff-Clean Grid
Assumes improved stove technology use, improved stove
thermal efficiency values, and the use of clean electricity
in electric cookstoves.
1 Tech = stove technology, Imp = improved, Eff = stove thermal efficiency
Figure 7-1 presents GCCP results for each cooking fuel mix scenario and the range of
included stove technology options. Results for the BAU 2030 scenario represent only marginal
changes relative to the current cooking fuel mix and yield roughly equivalent results, although
7-2
-------
Section 7—LCA Results for Kenya
the attribution to individual cooking fuel types changes slightly. Assuming improved stove
technology use for the BAU 2030 fuel mix holds the potential to reduce GCCP impact by
approximately 32 percent. If improvements in the thermal efficiency for each stove type are also
adopted, this reduction can be increased to 44 percent relative to the baseline. Limited electricity
use is assumed in all but the Diverse Modern Fuels mix, which explains the limited sensitivity of
fuel mix results to the possible adoption of a cleaner electricity grid.
In the Ghana Transition (for Kenya) scenario, an increased reliance on Charcoal and LPG
to provide 27 and 24 percent of national cooking energy, respectively, leads to a slight increase
in GCCP impact per GJ of delivered cooking energy. Adoption of improved stove and kiln
technology helps realize a 27 percent reduction in GCCP impact primarily by reducing the
impact associated with firewood and charcoal use and production. Targeting improvements in
stove thermal efficiencies facilitates an additional 11 percent reduction in GCCP. The Slow
Transition scenario produces impacts similar to the Ghana Transition (for Kenya) scenario, but a
larger fraction of impact is attributable to firewood as opposed to LPG and charcoal.
The Diverse Modern Fuels scenario realizes a 25 percent reduction in GCCP impact
based on changes in the cooking fuel mix alone. The adoption of improved stove technology and
advancements in stove thermal efficiency demonstrate a smaller relative effect on this fuel
scenario because the current efficiency of modern stoves tends to be much closer to the future
improved efficiency than is the case for traditional fuels. In other words, the benefits of increased
thermal efficiency are already being considered within the current scenario assumptions for stove
technology and thermal efficiency for the modern fuel options. The introduction of a clean
electricity grid leads to a modest, but noticeable, reduction in GCCP impact of approximately
three percent. The maximum GCCP reduction calculated for the Diverse Modern Fuel Scenario
is 51 percent, relative to the baseline.
7-3
-------
Section 7—LCA Results for Kenya
000
£ Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
o
CI
O
-------
Section 7—LCA Results for Kenya
ca
m
©
n
<
CQ
U
Current Tech-Current EfF-Current Grid
Current Tech-Current EfF-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp EfF-Current Grid
Imp Tech-Imp EfF-Clean Grid
Current Tech-Current EfF-Current Grid
Imp Tech-Current EfF-Current Grid
Imp Tech-Imp EfF-Current Grid
Imp Tech-Imp EfF-Clean Grid
G
O
¦-P
H
I
Current Tech-Current EfF-Current Grid
Imp Tech-Current EfF-Current Grid
Imp Tech-Imp EfF-Current Grid
Imp Tech-Imp EfF-Clean Grid
Current Tech-Current EfF-Current Grid
—
o
3
Imp Tech-Current EfF-Current Grid
Imp Tech-Imp EfF-Current Grid
Imp Tech-Imp EfF-Clean Grid
MJ/GJ Cooking Energy Delivered
1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000
I I I I I
Firewood
ILPG
I Charcoal from Wood
I Electricity
Biomass Pellets ¦ Kerosene
Biogas from Cattle Dung ¦ Sugarcane Ethanol
Figure 7-2. Kenya CED cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
Figure 7-3 shows that PMFP impacts are not particularly sensitive to the BAU 2030,
Ghana Transition (for Kenya), and Slow Transition Fuel mix scenarios in the absence of
assumed technology improvements. This lack of sensitivity is attributable to the fact that
substituting charcoal for firewood has a more limited effect on PMFP emissions as compared to
other possible substitutions, due to the emission of PM at the kiln. Increases in LPG use translate
directly into reductions in relevant particulate emissions. The ability to achieve PMFP reductions
responds more positively to technology improvements. The adoption of improved stove
technology for the BAU 2030 fuel mix scenario leads to a 69 percent reduction in PMFP.
Assuming both improved stove technology and thermal efficiency values under the Ghana
Transition (for Kenya) fuel mix yields a PMFP reduction of 77 percent. Using the same
technology assumptions for the Diverse Modern Fuels scenario reduces PMFP impact by nearly
90 percent relative to the baseline. The introduction of a clean grid has very little effect on PMFP
impact even for the Diverse Modern Fuels scenario, in which it comprises 13 percent of the
cooking fuel mix, due to low PMFP impact of both current and potential electricity grids.
7-5
-------
Section 7—LCA Results for Kenya
CQ
Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
D
<
ffi
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
,e
O
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
o
55
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
2 JS
p
22
CD
>
3
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
kg PM10 eq/GJ Cooking Energy Delivered
0 2.0 4.0 6.0 8.0 10.0 12.0
III
14.0
¦ Firewood
¦ Charcoal from Wood
¦ Biomass Pellets
¦ Kerosene
¦ LPG
¦ Electricity
¦ Biogas from Cattle Dung
¦ Sugarcane Ethanol
Figure 7-3. Kenya PMFP cooking fuel mix scenario results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
7.2 Baseline Normalized Results - Kenya
The concept and methodology behind a normalized presentation of results is included in
detail in Section 4.6. Generally, normalized results show which impact categories are most
strongly linked to the activity studied. If normalization factors were perfectly calibrated, the sum
of personal equivalent emissions for all sectors would equal the population of a given country,
which for Kenya is approximately 44 million people. However, normalized emission estimates
are uncertain due to lack of geographic granularity and the ever-changing nature of national or
global level estimates for many categories, indicating that the relative magnitude of results and
not the specific person equivalency value is of greater importance and validity. Normalized
results indicate only the level of the cookstove sector contribution to national economy-wide
impacts, they do not imply that impact categories are of greater or lesser significance.
Normalized results for Kenya, presented in Figure 7-4, indicate that the cooking sector
contributes significantly to national energy demand and emissions responsible for PMFP, POFP,
and BC impact. As with India and China, the normalized impacts for Kenya are highest for BC
7-6
-------
Section 7—LCA Results for Kenya
and PMFP, in part due to the incomplete inventory of BC pollutant emissions that are accounted
for in the normalization factor, and also due to the presence of negative characterization factors
within the method which allows for BC impacts greater than 100 percent of net national impact.
Regardless of the specific values, the results indicate that cookstove use is a dominant
contributor to BC impacts. Normalized impacts for POFP are noticeably higher in Kenya than
they are for India and China, which is largely due to the use of charcoal. Normalized CED results
indicate that approximately 43 percent of national energy demand in Kenya is attributable to the
cooking sector, which is higher than that realized for both India and China, and is attributable to
lower per capita energy demand in Kenya. Normalized results associated with GCCP indicate
that approximately seven percent of national GHG emissions are associated with household
cooking within the current fuel mix scenario. Other impact categories do not appear to be
particularly dependent upon the current cooking sector.
7-7
-------
£ 5
=.2
21
rS
3
o
a
o
a?
es
o
o
c5
3
o
o
o-
H
3"
Q
CD
3
O)
r
O
m
5
o"
CTQ
£>
C/3
t*
o
O
&3
00
eg
B
O
&3
3
CD
m
f
o
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
O
o
o
"d
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
o
tn
D
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
Tl
O
TJ
Current
^ BAU 2030
0 Ghana T ransition
Slow Transition
Diverse Modern Fuels
Current
^ BAU 2030
Ghana T ransition
- Slow Transition
Diverse Modern Fuels
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
TJ
o
hd
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
•n
M
hd
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
O
a
l~d
Current
BAU 2030
Ghana T ransition
Slow Transition
Diverse Modern Fuels
td
O
-------
Section 7—LCA Results for Kenya
7.3 Stove Efficiency Sensitivity - Kenya
Figure 7-5 shows the PMFP of wood-based stove groupings, highlighting the potential
reductions in particulate emissions attributable to stove and kiln efficiency improvements. Each
of the bars is labeled with the associated current or future improved stove thermal efficiency
value for that stove group. The figure indicates that thermal efficiency is an important indicator
of PMFP emissions within a given fuel type. The potential PMFP reductions evident in this
figure are heightened by the knowledge that cookstoves make a significant contribution to
national PM emissions as is indicated by the normalized results, and also that over 50 percent of
households still burn solid wood fuel in a three-stone fire. An over 99 percent reduction in PMFP
emissions is possible if pelletized biomass is adopted as a replacement for three-stone fires.
Charcoal cookstoves have a lower potential to achieve reduced PMFP emissions as compared to
non-carbonized solid woodfuel due to kiln emissions. Still, a 65 percent reduction in charcoal
cooking emissions is possible if improved charcoal stoves burning fuel from a high-performing
kiln are substituted for traditional cookstoves burning charcoal from an average kiln, the latter of
which constitutes 45 percent of charcoal usage in Kenya today (Clough 2012).
7-9
-------
Section 7—LCA Results for Kenya
20
13%
•73 18
P 16
10
0
Fuel Type:
Stove Type:
Stove
Efficiency:
15°/«
22% 20%
18% 20%
Firewood
Firewood
Firewood
Firewood
Firewood
Firewood
Biomass
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Pellets
from
from
from
from
from
from
from
from
Wood-
Wood-
Wood-
Wood-
Wood-
Wood-
Wood-
Wood-
Avg Kiln
Avg Kiln
Imp Kiln
Imp Kiln
Avg Kiln
Avg Kiln
Iinp Kiln
Imp Kiln
3-Stone
3-Stone
Trad
Trad
Imp
Imp
Imp
Trad
Trad
Trad
Trad
Imp
Imp
Imp
Imp
Current
Future,
Current
Future,
Current
Future,
Current
Current
Future,
Current
Future,
Current
Future,
Current
Future,
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Figure 7-5. Kenya PMFP effect of stove thermal efficiency modeled with various stove technologies.
(Axis abbreviations: Imp = improved, Trad = traditional, Avg = average)
7-10
-------
Section 7—LCA Results for Kenya
7.4 Electricity Grid Mix Sensitivity - Kenya
The current electrical grid mix in Kenya is predominantly fueled by hydropower, oil, and
geothermal energy. All future grid projections, which are forced to tackle the challenge of
rapidly increasing consumer demand, indicate reliance on a greater diversity of fuel types, which
include nuclear, coal, natural gas, wind, and solar. Figure 7-6 shows that the majority of future
grid mixes have the potential to significantly reduce the carbon footprint of electric cookstoves
that utilize their power. The McKinsey grid, however, would yield a negligible change in the
carbon footprint of electric stoves, owing to a dramatic increase in reliance on fossil fuels as
compared to the current grid mix. Even the current Kenyan grid mix provides a relatively clean
source of cooking energy, producing just 240 kg of CO2 equivalent emissions per GJ of delivered
cooking energy in comparison to over 450 kg for both India and China. If any of the four
cleanest electrical grid mix projections can be realized, electricity as a source of cooking energy
will have the optimal climate performance of all the cooking fuels studied, with the exception of
biogas and potentially sugarcane ethanol, depending upon where its true GCCP lands within the
calculated uncertainty range.
Results in all other impact categories, with the exception of FEP and BC, decrease for all
electric cookstoves relying on any of the projected future grid mixes, taking the 2013 grid mix as
baseline.
300
250
I 200
bfi
a
3
o
o
U
150
a
» 100
CI
O
O
8
238
235
50
97
0
Electricity
Grid:
2013 Grid
(IEA)
ROK Low
Demand -
2031
ROK Least
Cost Base
Scenario -
2031
ROK High
Demand -
2031
McKinsey - Low Carbon
2040 2040
Figure 7-6. GCCP of electric cookstove with various electrical grid mix options in Kenya.
7-11
-------
Section 7—LCA Results for Kenya
7.5 Forest Renewabilitv Factor Sensitivity - Kenya
Figure 7-7 shows the effect of forest renewability factors on the GCCP impact of wood-
based fuels produced and consumed in Kenya. Kenya has the highest percentage of forestry
operations that are considered to be non-renewable of the four nations studied. The estimate
based on the WISDOM database (Drigo 2014) indicates that only 36 percent of forestry products
are currently produced via sustainable operations. This scenario is labeled as the high
renewability (high renew.) option in the figure and is taken as the baseline for Phase II of this
study. A method previously used to determine forest renewability factors indicated that 100
percent of forest land is managed unsustainably, and therefore emissions associated with wood
combustion are not considered to be carbon neutral (low renew.). The assumptions behind these
two methods were discussed previously in Section 4.2.
The choice between forest renewability factors yields a 24 to 42 percent difference in
impact scores depending upon fuel and stove type. Solid firewood options are slightly more
sensitive to the choice of renewability factor than is charcoal, likely owing to greater GCCP
contributions for charcoal that are not related to the affected carbon dioxide emissions (e.g.,
CH4). The figure also clearly shows that while the choice of renewability factor is important,
there are other decisions such as fuel form and stove/kiln efficiency that have a greater impact on
GCCP impacts per GJ of delivered heat.
Forest renewability factor is incredibly important in the determination of whether even
the best performing wood-based options, biomass pellets and improved firewood stoves, are able
to compete with the GCCP of modern liquid and gas fuel options. Given the uncertainty ranges
for each fuel and assuming the high renewability factor, the use of firewood in improved stoves
has a climate impact which is roughly equivalent to the modern fossil fuels. If the low
renewability factor is applied, then the climate impact of firewood combustion in improved
stoves exceeds that of all the modern fuel options.
Forest renewability factors are based on current estimates regarding the sustainability of
forestry operations. If significant efforts are made to improve the efficiency of firewood use, for
example through the widespread adoption of pelletized wood stoves, then the renewability factor
could improve over time. General improvements in forestry practices, such as increased effort to
replant following harvest, could also improve the sustainability of national forestry operations.
Alternatively, increased consumer demand for wood products without the adoption of improved
practices will surely cause the sustainability of national forestry operations to deteriorate.
7-12
-------
Section 7—LCA Results for Kenya
"O
1,600
1,400
a<\V
>
<&
&
^ ' ,cA >'
cx^
^ x\C^ AS*-*
-------
Section 7—LCA Results for Kenya
7.6 Stove Group Uncertainty Results - Kenya
Uncertainty results for stoves burning different forms of wood-based fuel are presented
for both GCCP and PMFP in Figure 7-8 and Figure 7-9, respectively. All results in the figures
represent the baseline LCA modeling convention. The height of the bar in each figure represents
the Monte Carlo mean around which the error bars are centered. The analysis mean (triangle in
figure) for each impact category is the characterized expected value as it was entered into
openLC A. The analysis and Monte Carlo mean deviate from one another depending upon the
distribution used, and in the case of lognormally distributed data, depending upon the geometric
standard deviation.
GCCP of charcoal stoves is significantly greater than GCCP of firewood stoves due to
inefficient wood and energy conversion at the kiln. Particularly noticeable in the figure is that
almost no overlap in uncertainty ranges exists between any of the firewood and charcoal stove
options for GCCP. However, differences within a fuel type tend to be obscured by overlap in the
uncertainty ranges. Despite the overlap, it seems reasonable to assume that real reductions in
climate change potential could be realized by adopting improved firewood stoves and improved
charcoal stoves in combination with improved kiln technology, relative to traditional stove
technologies for both fuel sources. GCCP of biomass pellet and improved firewood stoves is
relatively similar, considering the overlap in uncertainty range. Stove emission inventories
register similar CO2 emissions for the two stove types despite a significant difference in stove
thermal efficiency.
1 L400
>
^ 1,200
c
M
bo
G
M
o
o
O
1,000
800
600
O 400
o"
-------
Section 7—LCA Results for Kenya
Upper boundaries of PMFP impact tend to be higher for solid firewood than for charcoal.
In general, the uncertainty range for PMFP impact of firewood options is wider than for other
fuel and stove types. There is a high degree of overlap between the uncertainty ranges of the
firewood and charcoal cookstoves. From the perspective of PMFP impacts, it is challenging to
justify the promotion of one traditional cooking fuel type over the other, although the higher end
of the potential impact range associated with firewood stoves can be avoided by promoting
improved firewood and charcoal stoves. Improved charcoal stoves and kiln technology realize a
significant reduction in PMFP as compared to traditional stoves and average kiln performance.
Biomass pellets present an opportunity to drastically reduce PMFP impacts while still utilizing
firewood resources.
70
13 60
-------
Section 7—LCA Results for Kenya
emissions during agricultural production of sugarcane. Variable rates of nitrogen fertilization and
nitrogen volatilization as ammonia are responsible for the wide uncertainty range projected for
sugarcane ethanol.
T3
"3
Q
(D
c
w
7.0
6.0
5.0
m 4.0
3
o
o
O
3.0
a
It
o
CN
o
M
2.0
1.0
i
i
0.0
Fuel Type:
Stove Type:
¦ Monte Carlo Mean A Analysis Mean
Figure 7-10. Kenya TAP uncertainty analysis results for select cooking fuels modeled with
various allocation approaches and stove technologies.
(Axis abbreviations: Imp = improved, Trad = traditional, Avg = average)
Firewood
Charcoal
Charcoal
Kerosene
LPG
Electricity
*
Biogas
Sugarcane
from
from
from
Ethanol
Wood -
Wood -
Cattle
Avg Kiln
Imp Kiln
Dung
3-Stone
Trad
Imp
Imp
Modern
Modern
Modern
Modern
7-16
-------
Section 8—LCA Results for Ghana
8. LCA RESULTS FOR GHANA
Firewood and charcoal are the predominant cooking fuels used in Ghana today, providing
approximately 42 and 32 percent, respectively, of national cooking energy (GLSS6 2014). The
current cooking fuel mix in Ghana has seen a dramatic drop in the use of firewood since the late
1990s in favor of LPG use. Charcoal use has increased only slowly from 26 percent of the
cooking fuel mix in the mid-1980s to a peak of 34-37 percent in the year 2008 (GLSS1 2008).
Results presented here describe the potential shifts in environmental impact if this trend
continues and help demonstrate if other sources of cooking energy provide a favorable
alternative to LPG. Ghana relies more heavily on charcoal use than other nations studied and for
this reason, the environmental effect of improved charcoal stove and kiln technology is of
particular interest.
Table 8-1 presents summarized LCA results for all fuel types and impact categories
considered for Ghana. The results are representative of baseline assumptions concerning cooking
fuel mix, stove technology use, stove thermal efficiency, electricity grid, and forest renewability
factor. Charcoal from wood has the greatest GCCP impact per GJ of delivered cooking energy.
In general, GCCP impacts for wood-based fuels in Ghana are slightly greater than those reported
for India and China and are approximately 50 percent lower than those reported for Kenya.
Differences observed between nations are primarily due to current technology adoption and
forest renewability factors specific to each country. Like Kenya, energy demand and FDP of
kerosene and LPG is significantly greater than it is for India and China due to inefficient refinery
operations. Other single fuel impact scores show reasonable order of magnitude alignment with
results for other countries.
Table 8-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - Ghana
GCCP
CED
FDP
WDP
PMFP
F ucl lype
(kg CO2 eq)
(MJ)
(kg oil eq)
(m3)
(kg PM10 eq)
Firewood
228
9.20E+3
7.36E-3
8.16E-5
15.5
Crop Residue
120
1.04E+4
7.56E-3
8.32E-5
15.4
Charcoal from Wood
712
1.57E+4
10.6
0.135
10.2
Biomass Pellets
152
4.35E+3
21.2
0.868
0.162
Kerosene
284
8.24E+3
193
0.078
0.104
LPG
274
7.67E+3
179
0.127
0.114
Electricity
259
7.94E+3
150
7.58
0.310
Sugarcane Ethanol
94.3
1.32E+4
27.2
644
4.26
Biogas from Cattle Dung
13.4
4.21E+3
-
3.29
0.187
8-1
-------
Section 8—LCA Results for Ghana
Table 8-1. Summary Table of Single Fuel Results by Impact Category (Impact/GJ
Delivered Cooking Energy) - Ghana
POFP
FEP
TAP
ODP
BC
Fuel Type
(kg
NMVOC)
(kgPeq)
(kg SO2 eq)
(kg CFC-11
eq)
(kgBC eq)
Firewood
5.99
9.12E-3
0.230
6.77E-11
3.33
Crop Residue
9.02
9.37E-3
0.616
6.96E-11
3.36
Charcoal from Wood
30.1
0.016
0.335
9.41E-8
2.49
Biomass Pellets
1.51
6.14E-3
0.206
6.83E-8
0.026
Kerosene
1.48
1.49E-3
0.239
3.09E-8
0.012
LPG
1.39
3.30E-3
0.265
6.42E-8
0.011
Electricity
1.39
1.97E-3
1.10
3.61E-7
-0.030
Sugarcane Ethanol
0.457
0.034
4.10
2.77E-6
0.755
Biogas from Cattle Dung
0.084
-
5.13E-3
-
0.040
8.1 Cooking Fuel Mix Scenario Results - Ghana
Cooking fuel mix scenario results provide the most comprehensive perspective on the
options for cookstove sector improvements included in the second phase of this project. Table
8-2 provides a guide to interpretation of bar axis labels.
Table 8-2. Cooking Fuel Mix Scenario Technology Options {Figure Key)
Fuel Mix Scenario Parameter Options
Description
Current Tech-Current Eff-Current Grid1
Assumes current stove technology, current average
stove thermal efficiency values, and 2013 electrical
grid mix.
Imp Tech-Current Eff-Current Grid
Assumes improved stove technology use, current
average stove thermal efficiency values, and 2013
electrical grid mix.
Imp Tech-Imp Eff-Current Grid
Assumes improved stove technology use, improved
stove thermal efficiency values, and 2013 electrical
grid mix.
Imp Tech-Imp Eff-Clean Grid
Assumes improved stove technology use, improved
stove thermal efficiency values, and the use of clean
electricity in electric cookstoves.
1 Tech = stove technology, Imp = improved, Eff = stove thermal efficiency
Figure 8-1 presents GCCP results for each cooking fuel mix scenario and the range of
technology options included. GCCP impact is more sensitive to improvements in stove
technology than to the future cooking fuel mix changes presented here. The BAU 2030,
Moderated Growth, and Fast Growth scenarios do not deviate more than six percent from current
GCCP impact in the absence of stove technology and efficiency upgrades. Stove technology
upgrades for these three scenarios reduce GCCP impact scores by between 28 and 33 percent
8-2
-------
Section 8—LCA Results for Ghana
relative to the baseline scenario. Stove efficiency upgrades for each stove type have the potential
to reduce impact by a further six percent for each of the three scenarios.
The Diverse Fuel Mix scenario realizes a 14 percent reduction in GCCP relative to the
baseline, based on changing the cooking fuel mix alone. Adoption of improved stove technology
for this scenario increases that reduction to 35 percent with stove thermal efficiency
improvements adding an additional nine percent reduction relative to baseline impacts. The
proposed cooking fuel mix substitutions demonstrate limited effect on GCCP impact as a result
of similar GCCP impact scores for LPG and firewood, which largely replace one another in the
various fuel mix scenarios. Stove technology and efficiency upgrades achieve most of their
reductions for firewood and charcoal fuel types. Scenario results are not greatly affected by the
introduction of a cleaner electrical grid as reliance on electricity in the scenario cooking fuel
mixes is limited.
§ Current Tech-Current Eff-Current Grid
QQ
Current Tech-Current Eff-Current Grid
o
o Imp Tech-Current Eff-Current Grid
s
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
£
o
O
-o
o
Current Tech-Current Eff-Current Grid
c3
PL-
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
g Imp Tech-Current Eff-Current Grid
t-H
o _
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
50
100
kg C02 eq/GJ Cooking Energy Delivered
150 200 250 300" 350 400
450
¦ Crop Residue
¦ Firewood
¦ Charcoal from Wood
¦ Kerosene
¦ LPG
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
¦ Biomass Pellets
Figure 8-1. Ghana GCCP cooking fuel mix results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
8-3
-------
Section 8—LCA Results for Ghana
Figure 8-2 shows cooking fuel mix scenario results for PMFP. PM emissions respond
strongly to fuel mix shifts due to the low PMFP of LPG, which increases as a component of the
cooking fuel mix in all future projections. LPG use increases to comprise 60 percent of the
cooking fuel mix in the Fast Growth scenario, and PMFP impact is reduced to just 45 percent of
its current level. Layering on stove technology and efficiency upgrades yields a 73 percent
reduction in impact relative to the baseline. The Diverse Modern Fuel mix realizes PMFP
reductions within a few percentage points of those described for the Fast Growth scenario.
kg PM10 eq/GJ Cooking Energy Delivered
0.0 2.0 4.0 6.0 8.0 10.0 12.0
Current Tech-Current Eff-Current Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
-o
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
Current Tech-Current Eff-Current Grid
"O
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Crop Residue
¦ Firewood
¦ Charcoal from Wood
¦ Kerosene
¦ LPG
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
¦ Biomass Pellets
Figure 8-2. Ghana PMFP cooking fuel mix results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
Figure 8-3 depicts the CED for cooking fuel mix scenarios in Ghana. Shifts in fuel mix
alone result in a maximum reduction in CED of 16 percent, which is associated with the Diverse
Modern Fuel mix. CED reductions are more sensitive to the proposed technology and efficiency
upgrades than they are to fuel mix shifts. Between 67 and 100 percent of potential CED
reductions are attributable to stove technology and efficiency upgrades. Depending upon the
8-4
-------
Section 8—LCA Results for Ghana
scenario, CED reductions between 38 and 50 percent are possible. Most of the improvements in
energy demand are attributable to a reduced reliance on solid wood fuel and charcoal.
MJ/GJ Cooking Energy Delivered
0 2,000 4,000 6,000 8,000 10,000 12,000
£ Current Tech-Current Eff-Current Grid
PQ
Current Tech-Current Eff-Current Grid
o
o
(N
P
<
CQ
o
O
-d
s
Imp Tech-Current Eff-Current Grid
Imp Tech-Imp Eff-Current Grid
Imp Tech-Imp Eff-Clean Grid
¦ Crop Residue
¦ Firewood
¦ Charcoal from Wood
¦ Kerosene
¦ LPG
¦ Electricity
¦ Sugarcane Ethanol
¦ Biogas from Cattle Dung
1 Biomass Pellets
Figure 8-3. Ghana CED cooking fuel mix results.
(Axis abbreviations: Imp = improved, Tech = stove technology, Eff = stove efficiency)
8.2 Baseline Normalized Results - Ghana
The concept and methodology behind a normalized presentation of results was introduced
in detail in Section 4.6. Generally, the results show which impact categories are most strongly
linked to the activity of study. If normalization factors were perfectly calibrated, the sum of
personal equivalent emissions for all sectors would equal the population of a given country,
which for Ghana is approximately 26 million people. However, normalized emission estimates
8-5
-------
Section 8—LCA Results for Ghana
are uncertain due to lack of geographic granularity and the ever-changing nature of national or
global level estimates for many categories, indicating that the relative magnitude of results and
not the specific person equivalency value is of greater importance and validity. Normalized
results indicate only the level of the cookstove sector contribution to national economy-wide
impacts, they do not imply that impact categories are of greater or lesser significance.
Normalized results for Ghana, displayed in Figure 8-4, indicate that at the national level,
CED, PMFP, POFP, and BC impacts are those most prominently linked to the cooking sector.
Results generally align with the findings for other countries. However, normalized CED is
greater for Ghana than it is for other nations, which is due primarily to two factors. First, Ghana
has the lowest per capita energy demand of any of the nations studied, meaning there are fewer
total economy energy impacts from other sectors. Second, Ghana relies on charcoal for nearly
one-third of national cooking energy, which leads to a greater CED per GJ of delivered cooking
energy than is observed in other nations. Normalized results for POFP are also somewhat greater
for Ghana than for other nations, primarily attributable to charcoal use and production. PMFP
and BC results are primarily associated with the use of charcoal and firewood, and drop
dramatically as LPG use increases as a component of the fuel mix. FDP, WDP and TAP show
the potential for modest increases in normalized impacts if Ghana moves towards
implementation of a fuel and technology mix in line with the options analyzed in this study.
8-6
-------
crs
=
-!
«
os
i.
h
cr
•s
a
»
B
c
N
fSs
a
r
o
>
•t
n>
oo
-• 03
%
o
o
5T
c« ^ n
rrg 8 o
po O Tj
5 ^ HH
P 50
£ B (i
p rs vi
fl> £?-
W td y
B
hi
3-
§
o
55 Q
t£
Co
m
O
3
o
?5
Zt
° s
K s
ere
o
O
O
O
hd
O
W
a
~a
a
T)
d
>Td
*T]
>Td
O
Tl
"d
¦n
tn
hd
O
a
-d
W
n
05 M
2 =
3 £.
3.2
® "
Person Equivalents
o
"o
o
o
o
o
o
o
o
o
o
o
a
o
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
Current
BAU 2030
Moderated Growth
Fast Growth
Diverse Modem Fuels
¦I
t/3
a>
o
o'
s
n
>
2?
CD
9
-------
Section 8—LCA Results for Ghana
8.3 Stove Efficiency Sensitivity - Ghana
Figure 8-5 shows the GCCP of select stove groups, highlighting the potential effect of
improved thermal efficiency on reductions in environmental impact. Each of the bars is labeled
with the associated current or future improved stove thermal efficiency value for that stove
group. The figure shows that traditional fuels with lower starting thermal efficiencies have
significantly more relative potential to reduce GCCP emissions through adoption of both
improved stove technology or use of the best performing traditional stoves. Modern fuels have
already realized high stove thermal efficiencies and therefore, realizing substantial relative gains
is challenging. For example, the best performing traditional firewood stove produces
approximately 38 percent less GHG emissions than does the current average stove. The percent
reduction achieved by upgrading LPG stoves is only 11 percent. Even larger gains are possible
for charcoal users as a result of potential efficiency upgrades at both the stove and the kiln. The
figure shows that, considered in isolation, stove efficiency and kiln improvements have similar
potential to improve the performance of a traditional charcoal stove, both with percent reductions
in GCCP impact between 33 and 38 percent. Combining these strategies yields a total GHG
emission reduction of approximately 58 percent. As the figure shows, thermal efficiency is not
the only, or even the predominant, determinant of GCCP, but it does have a significant effect on
stove performance for a given stove grouping or fuel type.
-------
Section 8—LCA Results for Ghana
900
-d
*a5
Q
14%
800
& 700
c
w
M
c
2
o
o
a
a
"a*
600
500
(N
o
a
bo 400
300
200
100
0
Fuel Type:
Stove
Type:
Stove
Efficiency:
Firewood
Firewood
Firewood
Firewood
Charcoal
Charcoal
Charcoal
Charcoal
LPG
LPG
Electricity
Electricity
from Wood,
from Wood,
from Wood,
from Wood,
Avg. Kiln
Avg. Kiln
Improved
Improved
Kiln
Kiln
Traditional
Traditional
Improved
Improved
Traditional
Traditional
Traditional
Traditional
Modern
Modern
Modern
Modern
Current
Future,
Current
Future,
Current
Future,
Current
Future,
Current
Future,
Current
Future,
Improved
Improved
Improved
Improved
Improved
Improved
Figure 8-5. GCCP effects of stove thermal efficiency in Ghana for various kiln and stove technologies.
8-9
-------
Section 8—LCA Results for Ghana
8.4 Electricity Grid Mix Sensitivity - Ghana
The current electrical grid in Ghana is fueled by hydropower, oil, and natural gas in
descending order of contribution (IEA 2013d). All potential future electrical grids, as described
by Ghana's Energy Commission, indicate future reliance on a more diverse palette of fuel
options. The primary trend projected by Ghana's Energy Commission is decreased reliance on
hydropower as electricity demand grows. The three Ghana EC scenarios represent the full range
of what the Ghana Energy Commission envisions as probable in the near term (GEC 2006),
while the fourth Low Carbon scenario indicates the potential carbon footprint of a clean grid,
which relies on Ghana's considerable solar, wind, and hydropower resources (IRENA 2013).
The renewable and nuclear Ghana EC grid mixes achieve only moderate reductions in GCCP as
compared to the present grid (Figure 8-6). The Ghana EC thermal scenario shows that if coal
power is relied upon to service increasing demand, it is possible that the GCCP of electrical
cookstove use will increase in the future. The Low Carbon electricity scenario produces
approximately 115 kg of CO2 equivalent emissions per GJ of delivered cooking energy, which is
a 55 percent improvement as compared to electric cookstoves relying on the current grid.
T3
-------
Section 8—LCA Results for Ghana
8.5 Forestry Renewabilitv Factor Sensitivity - Ghana
Figure 8-7 shows the effect of forest renewability factors on the GCCP impact of wood-
based fuels produced and consumed in Ghana. The high renewability factor (high renew.), which
is presented as the baseline value for this study, is derived from the WISDOM database (Drigo
2014) and indicates that approximately 71 percent of forest products in Ghana are produced
renewably. A method previously used to determine forest renewability factors indicated that 100
percent of forest land is managed unsustainably and therefore emissions associated with wood
combustion are not considered to be carbon neutral (low renew.). The assumptions behind these
two methods were discussed previously in Section 4.2.
The choice between forest renewability factors yields a 66 to 102 percent difference in
impact score depending upon cooking fuel and stove type. The disparity between high and low
renewability factors is greater in Ghana than it is for other countries studied. While the
magnitude of GCCP impact varies for charcoal by a factor of two depending upon the
renewability factor selected, the choice does not influence its performance relative to other fuels.
Charcoal exhibits the highest GCCP impact regardless of renewability factor. The choice of
renewability factor is, however, critical in determining the relative performance of firewood and
biomass pellet cookstoves. The baseline high renewability factor yields GCCP impact scores that
are lower than the modern liquid and gas fossil fuel options. Assuming the low renewability
factor, even the most efficient option, biomass pellets, is unable to produce GCCP impacts that
are competitive with the modern fossil fuels.
Forest renewability factors are based on current estimates regarding the sustainability of
forestry operations. If significant efforts are made to improve the efficiency of firewood use, for
example through the widespread adoption of pelletized wood stoves, then the renewability factor
could improve over time. General improvements in forestry practices such as increased effort to
replant following harvest, could also improve the sustainability of national forestry operations.
Alternatively, increased consumer demand for wood products without the adoption of improved
practices will surely cause the sustainability of forestry operations to deteriorate.
8-11
-------
Section 8—LCA Results for Ghana
1,800
1,600
Q 1,400
|»
D
c
pa
1,200
S
o
o
O
a 1,000
-------
Section 8—LCA Results for Ghana
8.6 Stove Group Uncertainty Results - Ghana
Figure 8-8 shows CED stove group uncertainty results for a wide range of fuel and stove
technology options in Ghana. All results in the figures incorporate baseline LCA modeling
conventions. The height of the bar in each figure represents the Monte Carlo mean around which
the error bars are centered. The analysis mean (triangle in figure) for each impact category is the
characterized expected value as it was entered into openLCA. The analysis and Monte Carlo
mean deviate from one another depending upon the distribution used, and in the case of
lognormally distributed data, depending upon the geometric standard deviation.
Uncertainty ranges for CED tend to be more narrow, as a percentage of mean impact,
than they are for other impact categories. The reduced energy demand potential of biomass
pellets as compared to three-stone fires is apparent in the figure and indicates one possible
avenue for reducing wood consumption in Ghana. The figure also shows the high-energy
demand of charcoal use when traditional stoves are used to burn charcoal from average kilns.
Significant reductions in CED are possible if improved charcoal stoves and improved kiln
technology are adopted. Despite CED reductions associated with improved charcoal technology,
this fuel type and the charcoal supply chain are unable to compete with the best available
improved firewood cookstoves or biomass pellet cookstoves. Sugarcane ethanol demonstrates
the second highest CED of all cooking options in Ghana. The energy demand of LPG and
electric cookstoves in Ghana is higher than the energy demand associated with the same cooking
options in India or China. Elevated CED for both fuel options is due to petroleum refining, which
also effects electric stove use due to Ghana's heavy reliance on fuel oil as a source of electrical
energy.
8-13
-------
Section 8—LCA Results for Ghana
•g 20,000
| 18,000
| 16,000
& 1 iOOO
| 12,000
m 10,000
bsi '
| 8,000
8 6,000
o
^ 4,000
o
^ 2,000
0
Fuel
Type: s>
a0Y
Stove
Type:
t\y~ jO
\\^
-------
Section 8—LCA Results for Ghana
13
Q
z
bp
Fuel ^
Type: C*
Stove
Type:
80
70
60
50
40
30
20
10
0
o1?
.*SC'
\V*S
^
,0
vV
^,\co9 ^vfsS& a£J*°
c\#~ cJc^C
op
,o*
Trad 3-stone Imp Trad Trad Imp Imp
¦ Monte Carlo Mean A Analysis Mean
Imp Mod
Figure 8-9. Ghana POFP uncertainty analysis results for select cooking fuels modeled with
various stove technologies.
(Axis abbreviations: Imp=improved, Trad=traditional, Mod=modern)
8-15
-------
Section 9—Key Takeaways By Country and Study Conclusions
9. KEY TAKEAWAYS BY COUNTRY AND STUDY CONCLUSIONS
This section highlights several key takeaways for each country studied, followed by
noteworthy conclusions and trends for this phase of work.
9.1 Key Takeawavs
This section describes the key takeaways common to each country, followed by findings
unique to the countries studied.
9.1.1 Findings Common to all Countries
• Normalized results for all countries show that BC and PMFP impact categories are
strongly linked to the cooking sector. Results also show that impacts in these two
categories are sensitive to the projected future fuel mix and stove technology shifts
considered in this study, indicating multiple pathways by which to reduce impacts
attributable to the cooking sector.
• Utilization of modern cooking fuels such as liquefied petroleum gas (LPG), natural
gas, biomass pellets, and ethanol resulted in significant reductions in PMFP and BC,
categories strongly linked to the cooking sector.
• Normalized results confirm that traditional fuels pose a significant risk to human
health (e.g., due to PMFP). The possibility of using renewably sourced wood fuel in
combination with the adoption of improved or pelletized stoves could significantly
reduce hazardous emissions while still allowing the use of traditional biomass
resources.
• The sensitivity analysis shows that a significant range in potential environmental
impact exists between the worst and best performing cookstoves within a given stove
type (e.g. firewood traditional).
• Biogas and biomass pellets hold significant potential to reduce household air
emissions attributable to the cooking sector.
• Updated LCI information for the agricultural production of sugarcane indicates
significant upstream environmental impacts associated with ethanol production.
9.1.2 India
• Normalized BC impacts in India are high relative to other nations and are
disproportionately influenced by the use of dung and crop residues in the current
cooking fuel mix.
• The current, coal heavy electricity mix in India and high electrical grid losses
contribute to the poor performance of electric cookstoves relative to other modern
fuel options.
9-1
-------
Section 9—Key Takeaways By Country and Study Conclusions
• Realizing further GCCP impact reductions will be a challenge for India as the country
moves to adopt modern fossil-based cooking fuels. GCCP of the current Indian
cooking fuel and stove technology mix is at minimum 49 percent lower than that
realized by the other nations studied due to India's continued reliance on biomass
fuels, relatively high baseline forest renewability, and an absence of significant
contributions from stoves that exhibit particularly poor performance such as
traditional coal powder and charcoal cookstoves, all of which drive up GCCP impact
in the other countries.
9.1.3 China
• In China, one-third of cooking fuel energy is produced from coal, which
disproportionately contributes to the country's normalized PMFP and BC impacts.
Potential reductions in environmental impact realized by switching from coal powder
to advanced forms of coal consumption such as honeycomb briquettes or coal gas
provide a robust option for consistently improving performance of the cooking sector
across all impact categories.
• The current coal-heavy electricity mix in China results in poor performance of
electric cookstoves for most impact categories assessed relative to other modern fuel
options. Upgrades to China's electricity sector will be required for electric cookstoves
to achieve environmental impact scores in line with, or better than, other modern
fuels.
9.1.4 Kenya
• Scenario results show that reductions in Kenyan cooking sector emissions, compared
to other countries studied, are more sensitive to adoption of improved stove
technologies and thermal efficiencies, when holding cooking fuel mix constant. This
is because Kenya currently relies heavily on three-stone fires and traditional wood
stoves, which are associated with low thermal efficiencies and notable air emissions
during cookstove use.
• Forest renewability is important in determining if the best performing wood-based
options, biomass pellets and improved firewood stoves, can compete with the GCCP
of modern liquid and gas fuel options. This is especially true for Kenya, which has
the lowest forest renewability among the four study nations.
• Low availability of renewable wood resources in Kenya indicates that following
Ghana's lead in pursuing increased charcoal use as a means of improving urban air
quality could lead to significant pressure on other environmental impact categories
and forest resources. While charcoal may serve to reduce emissions in the household,
it does not reduce cumulative emissions across the supply-chain and serves as an
inefficient use of forest resources.
• The electricity grid in Kenya has the lowest GCCP of all nations studied due to the
prevalence of hydropower and geothermal energy in their electrical grid mix.
9-2
-------
Section 9—Key Takeaways By Country and Study Conclusions
However, Kenya currently has the lowest national electrification rate, 23 percent, of
any of the four countries studied (World Bank 2012). Electricity demand and
generation capacity are expected to rise dramatically in the next 20 years, and the
projected grid mixes yield further improvements in environmental performance.
Electricity may prove a viable alternative to LPG and kerosene in urban Kenya if
electrification and generation grow as expected.
9.1.5 Ghana
• Ghana demonstrates the second highest sensitivity to improvements in stove
technology and efficiency, following Kenya, indicating the potential to reduce
cooking sector emissions even in the absence of cooking fuel mix shifts.
• Of the four study nations, Ghana is most heavily reliant on charcoal energy as a
source of cooking fuel (GLSS6 2014). Significant improvements in environmental
performance are possible through improved charcoal stove and kiln technology
adoption. However, even assuming the most optimistic adoption of charcoal
technology, this fuel demonstrates consistently poor environmental performance
relative to other cooking options and places a heavy burden on forest resources.
• Normalized CED of Ghana's cooking sector is significantly higher than that realized
for other nations, which is due largely to inefficient energy conversion in charcoal
kilns and the LPG refining process, as well as lower overall national per capita energy
consumption for all sectors compared to national per capita energy use in the other
study countries.
9.2 Conclusions
Normalized results across the nations studied agree that, at the national level, the cooking
sector has the greatest potential to contribute to BC and PMFP economy-wide impacts. Cooking
fuel mix results show that BC and PMFP can both be reduced dramatically by strategies that
focus either on changing the cooking fuel mix or through the adoption of improved stove
technologies. The latter strategy is particularly effective for nations and fuels that currently rely
most heavily on traditional technologies. Firewood use in Kenya, for example, is still largely
reliant on the use of three-stone fires, and consequently huge gains in fuel efficiency and
associated emissions reductions are possible via technology improvements alone.
Normalized CED also tends to reveal a significant link between household cooking and
national energy demand, but the results between countries vary. China and Ghana represent the
extreme ends of the spectrum with China demonstrating the lowest normalized CED scores, in
part attributable to real differences in current average CED of the national cooking fuel mix for
each country, as China and Ghana have the lowest and highest CED per GJ of delivered energy,
respectively. National per capita energy demand also plays a significant role in normalized
impacts, and to the extent that this is responsible for differences in normalized impact, this
difference reflects less on the cooking sector than it does on national energy use and the level of
industrialization. The current average cooking fuel mix in India and China, for example, has
9-3
-------
Section 9—Key Takeaways By Country and Study Conclusions
similar CED values per GJ of delivered cooking energy, and yet normalized impacts are greater
for India as a result of lower national energy use.
Reductions in some impact categories such as CED and GCCP are more challenging to
achieve than either PMFP or BC emissions. The dramatic differences in PMFP impact that exists
between traditional and modern fuels are significantly muted for GCCP and CED, thereby
reducing the effectiveness of all but the most dramatic substitutions of fuel type and stove
technology. CED and GCCP impact reductions are more easily achieved in China, Kenya, and
Ghana due to the fact that all three nations currently rely heavily on either coal powder or
charcoal, which are the poorest performing fuels for these impact categories considered in the
respective nations, allowing for beneficial emission reductions regardless of what fuel
substitution is made. Biogas demonstrates notably lower GCCP impact than any other fuel
considered in the study. While GCCP is a critical impact category at the global level, it must be
remembered that normalized results show that GCCP is not necessarily driven by the cooking
sector in each study nation.
As discussed, two separate methodologies were considered over the course of this study
to determine the fraction of forestry products that are renewably produced and are therefore
carbon neutral. Significant differences in GCCP impacts result from the choice of methodology,
and should be considered when evaluating results. For example, in China the choice of
renewability factor is enough to influence whether adoption of modern fuels yields an increase,
decrease, or no significant effect on emissions contributing to climate change. In Ghana, the
choice of renewability factor is critical in determining the relative performance of firewood and
biomass pellet cookstoves. The baseline high renewability factor yields GCCP impact scores that
are lower than the modern liquid and gas fossil fuel options, whereas assuming the low
renewability factor, even the most efficient option, biomass pellets, is unable to produce GCCP
impacts that are competitive with the modern fossil fuels.
Consistent with the Phase I results, it is more challenging to realize cooking fuel mix
level improvements in environmental performance than initial appearances imply. Particularly
when looking at single fuel results where it is obvious that dramatic, often order of magnitude,
differences in impact exist between the worst and best performing fuels in a given impact
category. First, the 100 percent substitution that the differences in single fuel impact scores
imply are not possible at the level of national cooking fuel mix. Additionally, realized reductions
in aggregate impact scores of a fuel mix tend to be muted by canceling factors that occur when
simultaneous shifts occur involving multiple fuel types. The Ghana Transition (for Kenya)
scenario applicable to Kenya provides an example of this phenomenon. The scenario realizes a
significant reduction in firewood use, which is offset by increases in charcoal and LPG
consumption. Despite dramatic shifts in the underlying cooking fuel mix, the average GCCP
increases by a few percentage points assuming stove and kiln technology remain constant.
The results also show that fuel mix substitutions designed to address a single impact
category can lead to the exacerbation of other environmental impacts. For example, increases in
charcoal use in Kenya or Ghana, which are primarily targeted to realize urban air quality
improvements and reductions in PM emissions, have and will continue to lead to increased
demand for firewood, greater GCCP impact, and increased PMFP emissions at the location of the
kiln.
9-4
-------
Section 9—Key Takeaways By Country and Study Conclusions
Results from the second phase of work confirm the single fuel outcomes of Phase I and
provide a deeper understanding of the interplay between cooking fuel mix substitution, stove
technology improvement, and stove thermal efficiency increases in achieving environmental
impact reductions. The results presented in this report are only a subset of the full results
available in the supporting files, and have been selected to highlight key trends while serving as
a guide for interpretation of the full results available. The uncertainty analysis included in the
second phase of work serves to increase confidence in how robust the differences in
environmental performance observed are between cooking fuels. Areas where overlap in
uncertainty ranges obscures clear distinctions between fuels highlight areas for potential future
refinement and study. Normalized impacts help to focus our attention on the impact categories
most strongly influenced by the cooking sector.
Finally, this analysis does not capture many social and economic dimensions that
strongly influence the discussion surrounding appropriate policy options and technology choices
within the cooking sector. The Global Alliance for Clean Cookstoves is furthering additional
research in those areas, the results of which can be used in conjunction with findings in this
study. The results presented here and in accompanying documents will provide the greatest
insight when considered alongside information and indicators aimed at social and economic
understanding of the cooking sectors in India, China, Kenya, and Ghana. Considerations for
future cookstove LCA research include repeating this analysis for other countries such as
countries in South America to expand the geographic relevance of this work.
9-5
-------
Section 10—References
10. REFERENCES
ADP (Accenture Development Partnerships). 2012. Ghana Market Assessment - Sector
Mapping. April, http://www.cleancookstoves.org/resources_files/ghana-market-
assessment-mapping.pdf Accessed 6 December, 2016.
Adelekan, B.A. 2014. Potentials of Selected Tropical Crops and Manure as Sources ofBiofuels.
Institute of Agricultural Research and Training, http ://cdn. intechopen. com/pdfs-
wm/31318.pdf Accessed 22 March, 2017.
AED (Academy for Educational Development). 2007. Fuel Efficient Stove Programs in IDP
Settings - Summary Evaluation Report, Uganda (USAID).
AED (Academy for Educational Development). 2008. Fuel Efficient Stove Programs in IDP
Settings - Summary Evaluation Report Darfur, Sudan (Washington, D.C.: United States
Agency for International Development).
Afrane G., and A. Ntiamoah. 2011. Comparative Life Cycle Assessment of Charcoal, Biogas and
LPG as Cooking Fuels in Ghana. Journal of Industrial Ecology. 15(4): 539-549.
Afrane, G., and A. Ntiamoah. 2012. Analysis of the Life-Cycle Costs and Environmental
Impacts of Cooking Fuels Used in Ghana. Applied Energy 98: 301-306.
http://www.sciencedirect.com/science/article/pii/S030626191200259Q. Accessed 22
March, 2017
Ahmad, R., and N. Jabeen. 2009. Demonstration of Crop Improvement in Sunflower (Gelianthus
annus L.) by the Use of Organic Fertilizers under Saline Conditions. Pakistan Journal of
Botany. 41(3): 1373-1384.
Arthurson, V. 2009. Closing the Global Energy and Nutrient Cycles through Application of
Biogas Residue to Agricultural Land - Potential Benefits and Drawbacks. Energies. 2:
226-242.
ASTAE (Asia Sustainable and Alternative Energy Program). 2013. China: Accelerating
Household Access to Clean Cooking and Heating. September. World Bank, Washington,
DC. https://openknowledge.worldbank.org/bitstream/handle/10986/16662/814950WPQP
12980Box0379837B00PUBLIC0.pdf?sequence=l Accessed 22 March, 2017.
Bailis, R., V. Berrueta, C. Chengappa, K. Dutta, R. Edwards, O. Masera, D. Still, and K.R.
Smith. 2007. 'Performance Testing for Monitoring Improved Biomass Stove
Interventions: Experiences of the Household Energy and Health Project.' Energy for
Sustainable Development 11(2). AED (Academy for Educational Development). 2007.
Fuel Efficient Stove Programs in IDP Settings - Summary Evaluation Report, Uganda
(USAID).
10-1
-------
Section 10—References
Bailis, R., R. Drigo, A. Ghilardi, and O. Masera. 2015. The Carbon Footprint of Traditional
Woodfuels. Nature Climate Change. 5: 266-272.
Baumann, H., and A.M. Tillman. 2004. The Hitch Hiker's Guide to LCA: An Orientation in Life
Cycle Assessment Methodology and Application. Lund, Sweden: Studentlitteratur AB.
Bay Area Air Quality Management District. 2008. Particulate Matter.
http://hank.baaqmd.gov/pln/pm/ Accessed 22 March, 2017.
Beer, J. 2005. High Efficiency Electric Power Generation; The Environmental Role.
https://mitei.mit.edu/svstem/files/beer-combustion.pdf. Accessed 8 June, 2016.
Berglund, M. 2006. Biogas Production from a Systems Analytical Perspective. Ph.D. thesis,
Lund University, Lund, Sweden.
Berick, A. 2006. Heat Losses in a Cook Pot at Constant Temperature. Aprovecho Research
Center. http://aprovecho.org/?pavbox_id=142 Accessed 22 March, 2017.
Bhattacharya, S.C., D.O. Albina, A.M. Khaing. 2002a. Effects of Selected Parameters on
Performance and Emission of Biomass-Fired Cookstoves. Biomass andBioenergy. 23:
387-395.
Bhattacharya, S.C., D.O. Albina, A.P. Salam. 2002b. Emission Factors of Wood and Charcoal-
Fired Cookstoves. Biomass and Bioenergy. 23: 453-469.
Booker, K. 2012. Performance of Charcoal Cookstoves for Haiti Part 1: Results from the Water
Boiling Test. Lawrence Berkeley National Laboratory. LBNL Paper LBNL-5021E.
https://escholarship.org/uc/item/0qm438xv Accessed 25 February, 2016.
Borjesson P., and M. Berglund. 2006. Environmental Systems Analysis of Biogas Systems-Part
1: Fuel-Cycle Emissions. Biomass and Bioenergy 30(5): 469-485.
Carter, E.M., M. Shan, X. Yang, J. Li, and J. Baumgartner. 2014. Pollutant Emissions and
Energy Efficiency of Chinese Gasifier Cooking Stoves and Implications for Future
Intervention Studies. Environmental Science & Technology. 48. 6461-6467.
Cashman, S., M. Rodgers, M. Huff, R. Feraldi, and B. Morelli. 2016. Life-Cycle Assessment of
Cookstove Fuels in India and China. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-15/325. https://cfpub.epa.gov/si/si_public_record
_report.cfm?dirEntrvId=322551 Accessed 23 March, 2017.
Cao, G., X. Zhang, S. Gong, and F, Zheng. 2008. Investigation on Emission Factors of
Particulate Matter and Gaseous Pollutants from Crop Residue Burning. Journal of
Environmental Sciences. 20: 50-55.
10-2
-------
Section 10—References
Castellano, A., A. Kendall, M. Nikomarov, and T. Swemmer. 2015. Electric Power & Natural
Gas, Brighter Africa: The Growth Potential of the Sub-Saharan Electricity Sector.
McKinsey & Company, http://www.mckinsev.com/-/media/mckinsev/dotcom
/client_service/epng/pdfs/brighter_africa-the_growth_potential_of_the_sub-
saharanelectricitvsector.ashx Accessed 5 May, 2016.
CBS (Central Bureau of Statistics). 2002. Kenya 1999 Population and Housing Census. Republic
of Kenya. https://www. google. com/webhp?sourceid=chrome-instant&ion=l&espv
=2&ie=UTF-8#q=Kenva+1999+Population+and+Housing+Census Accessed 15 April,
2016.
CES (Center for Energy Studies). 2001. Efficiency Measurement ofBiogas, Kerosene, and LPG
Stoves. Submitted to: Biogas Support Program. Lilitpur, Nepal.
Clough, L. 2012. The Improved Cookstove Sector in East Africa: Experience from the
Developing Energy Enterprise Programme (DEEP). GVEP International. London, UK.
Collivignarelli, C., A. Mazzu, M. Vaccari, and F. Vitali. 2010. Comparison of Biomass Stoves
Appropriate Technologies for Household Energy in Developing Countries. Third
International Symposium on Energy from Biomass and Waste.
Dalberg (Dalberg Global Development Advisors). 2012. State of Clean Cooking Energy Sector
in Sub-Saharan Africa. Prepared for Africa Clean Cooking Energy Solutions Initiative.
http://siteresources.worldbank.org/EXTAFRREGTOPENERGY/Resources/7173Q5-
1355261747480/ACCES_State_of_the_Clean_Cooking_Energv_Sector_in_SubSaharan
Africa_Report_DRAFT_Dalberg.pdf Accessed 23 March, 2017.
Dalberg (Dalberg Global Development Advisors). 2013a. Kenya Market Assessment. Global
Alliance for Clean Cookstoves. http://cleancookstoves.org/resources/234.html Accessed
22 March, 2017.
Dalberg (Dalberg Global Development Advisors). 2013b. India Cookstoves and Fuels Market
Assessment. Global Alliance for Clean Cookstoves. www.cleancookstoves.org
/resources_files/india-cookstove-and-fuels-market-assessment.pdf Accessed 22 March,
2017.
Dalberg (Dalberg Global Development Advisors). 2014. China Stoves and Fuels Market
Assessment. Global Alliance for Clean Cookstoves. May presentation: preliminary
findings, 19 May 2014.
Dones, R., C., Bauer, and R., Bollinger, et al. 2007. Sachbilanzen von Energiesystemen:
Grundlagen fur den okologischen Vergleich von Energiesystemen und den Einbezug von
Energiesystemen in Okobilanzen fur die Schweiz. [Life cycle of energy systems:
foundations for the ecological comparison of energy systems and the inclusion of energy
systems in life cycle assessment for the Switzerland.'] Final report ecoinvent No. 6-VI,
Paul Scherrer Institut Villigen, Swiss Centre for Life Cycle Inventories, Dubendorf, CH.
10-3
-------
Section 10—References
Drigo, R. 2014. WISDOM Case Studies, http://www.wisdomproiects.net/global/cs.asp. Accessed
22 March, 2017.
EC (European Commission). 2011. Preparatory Studies for Ecodesign Requirements ofEuPs
(III): Lot - 23 Domestic and Commercial Hobs and Grills included when incorporated in
Cookers, Task 4: Technical Analysis of existing Products, http://www.eup-
network.de/fileadmin/user_upload/Produktgruppen/Lots/Final_Documents/Lot
_23_Task_4_Final.pdf Accessed 22 March, 2017.
Ecoinvent Centre. 2010. Ecoinvent data v2.2. Ecoinvent reports No. 1-25, Swiss Centre for Life
Cycle Inventories, Dubendorf, CH.
Enerdata. 2016. Global Energy Statistical Yearbook 2016: Total Energy Consumption - China.
Accessed December 1, 2016.
Energica. 2009. Pre-feasibility study for an improved Cookstove Project in Northern Ghana.
Prepared for Danish Energy Agency, http://www.care.org/sites/default/files/documents
/CC-2009-GHA-Cookstoves_Ghana.pdf. Accessed 3 June, 2016.
Energy Commission. 2010. Draft Bioenergy Policy for Ghana, Executive Summary.
http://cleancookstoves.org/resources/69.html Accessed 2 June, 2016.
ERG (Eastern Research Group). 2014. Quality Assurance Project Plan for Comparative Life
Cycle Assessment of Cooking Fuel Options in China and India.
FAO (Food and Agriculture Organization). 2010. Global Forest Resources Assessment 2010:
Main Report. FAO Forestry Paper 163. Food and Agriculture Organization of the United
Nations, Rome, Italy.
FAO (Food and Agriculture Organization). 2016. AQUASTAT Website. Food and Agriculture
Organization of the United Nations (FAO). Accessed December 1, 2016.
FAOSTAT (Food and Agriculture Organization of the United Nations Statistics Division). 2016.
FAOSTAT Data Download Tool. Statistics Division, http://faostat3.fao.org/home/E
Accessed 14 March, 2016.
GACC (Global Alliance for Clean Cookstoves). 2015. Fuels.
http://cleancookstoves.org/technologv-and-fuels/fuels/. Accessed 22 March, 2017
GACC (Global Alliance for Clean Cookstoves). 2016. The Clean Cooking Catalog.
http://catalog.cleancookstoves.org/. Accessed 9 June, 2016.
GACC (Global Alliance for Clean Cookstoves). 2017a. Comparative Analysis of Fuels for
Cooking: Life Cycle Environmental Impacts and Economic and Social Consideration.
Prepared by Eastern Research Group (ERG), https://cleancookstoves.org/assets-
facit/Comparative-Analvsis-for-Fuels-FullReport.pdf Accessed 22 March, 2017.
10-4
-------
Section 10—References
GACC (Global Alliance for Clean Cookstoves). 2017b. Country Profiles: Focus Countries,
Ghana, http://cleancookstoves.org/countrv-profiles/focus-countries/l-ghana.html
Accessed 29 March, 2017 Accessed 6 December, 2016.
GACC (Global Alliance for Clean Cookstoves). 2017c. Country Profiles: Focus Countries,
Kenya, http://cleancookstoves.org/country-profiles/focus-countries/4-kenya.html
Accessed 29 March, 2017 Accessed 6 December, 2016Gambhir, A., T. Napp, C. Emmott,
L. Vallejo, and G. Anandarajah. 2012. India's C02 Emissions pathway to 2050. The
Grantham Institute, Imperial College. London, EN. https://www.imperial.
ac.uk/media/imperial-college/grantham-institute/public/publications/institute-reports-and-
analvtical-notes/India's-emissions-pathwavs-to-2050—summarv-report.pdf Accessed 10
December, 2016.
Gao, S., P. Xu, F. Zhou, H. Yang, C. Zheng, W. Cao, S. Tao, S. Piao,, Y. Zhao, X. Ji, Z. Shang,
and M. Chen. 2016. Quantifying Nitrogen Leaching Response to Fertilizer Additions in
China's Cropland. Environmental Pollution. 11. 241-251.
GEC (Ghana Energy Commission). 2006. Strategic National Energy Plan 2006-2020.
https://s3.amazonaws.com/ndpc-static/pubication/Strategic+National+Energy+
Plan+2006-2020.pdf Accessed 6 May, 2016.
GEC (Ghana Energy Commission). 2015. National Energy Statistics 2005-2014. Strategic
Planning and Policy Division. Accra, Ghana, http://energvcom. gov. gh/files/Energv%
20Statistics_2015.pdf Accessed 22 March, 2017.
Githiomi, J.K., Mugendi, D.N., and Kung'u, J.B. 2012. Analysis of Household Energy Sources
and Woodfuel Utilisation Technologies in Kiambu, Thika, and Maragwa Districts of
Central Kenya. Journal of Horticulture and Forestry. 4(2): 43-48.
GLSS1 (Ghana Statistical Service). 2008. Ghana Living Standards Survey 1: 1987-1988. Office
of the President. http://www. statsghana. gov, gh/nada/index.php/catalog/7 Accessed 2
December, 2016.
GLSS2 (Ghana Statistical Service). 1993. Ghana Living Standards Survey 2: 1988-1989, Second
Round. Accra, Ghana. http://www. statsghana. gov, gh/nada/index.php/catalog/4 Accessed
23 March, 2017.
GLSS3 (Ghana Statistical Service). 1995. Ghana Living Standards Survey: Report on the Third
Round. Accra, Ghana, http://www. statsghana. gov, gh/nada/index. php/catalog/12
Accessed 2 December, 2016.
GLSS4 (Ghana Statistical Service). 2000. Ghana Living Standards Survey: Report of the Fourth
Round. Accra, Ghana, http://www. statsghana. gov, gh/nada/index.php/catalog/14/study-
description Accessed 2 December, 2016.
10-5
-------
Section 10—References
GLSS5 (Ghana Statistical Service. 2008. Ghana Living Standards Survey: Report of the Fifth
Round. Office of the President. http://www.statsghana.gov.gh/nada/index.php/catalog/5
Accessed 2 December, 2016.
GLSS6 (Ghana Statistical Service). 2014. Ghana Living Standards Survey: Round 6. Accra,
Ghana, http://www.statsghana.gov.gh/glss6.html Accessed 2 December, 2016.
Goedkoop M.J., R., Heijungs, and M., Huijbregts, et al. 2008. A Life Cycle Impact Assessment
Method which comprises Harmonised Category Indicators at the Midpoint and the
Endpoint Level; First Edition Report I: Characterisation, http://www.leidenuniv.nl/
cml/ssp/publications/recipe_characterisation.pdf Accessed 22 March, 2017.
GreenDelta. 2016. OpenLCA 1.5.0. Berlin, Germany, http://www.openlca.org/openlca/ Accessed
23 March, 2017.
Gov. of India [Government of India Ministry of Petroleum and Natural Gas Economics and
Statistics Division (MoPNG)]. 2014. Indian Petroleum and Natural Gas Statistics 2013-
14. Available at http://petroleum.nic.in/docs/pngstat.pdf Accessed August 17, 2015.
GREET (The Greenhouse gases, Regulated Emissions, and Energy use in Transportation Model).
2008. Transportation Fuel Cycle Analysis Model. Version 1.8b. Argonne National
Laboratory, Argonne, IL. https://greet.es.anl.gov/ Accessed 22 March, 2017.
GSF (Gold Standard Foundation). 2015. The Gold Standard: Quantification of Climate related
Emission Reduction of Black Carbon and Co-Emitted Species due to the Replacement of
Less Efficient Cookstoves with Improved Efficiency Cookstoves. The Gold Standard
Foundation, Geneva-Cointrin, Switzerland.
GSS (Ghana Statistical Service). 2012. 2010 Population and Housing Census: Summary Report
of Final Results. Accra, Ghana, http ://www. statsghana. gov, gh/docfiles/201 Ophc
/Census2010_Summarv_report_of_final_results.pdf Accessed 22 March, 2017.
Gurung, J.B. 1997. Review of Literature on Effects of Slurry Use on Crop Production. Prepared
for The Biogas Support Program. Kathmandu, Nepal, http://citeseerx.ist.psu.edu/viewdoc
/download?doi= 10.1.1.458.6298&rep=rep 1 &tvpe=pdf Accessed 23 March, 2016.
GVEP International. 2012a. Kenya Market Assessment - Sector Mapping.
http://www.cleancookstoves.org/resources_files/kenva-market-assessment-mapping.pdf
Accessed 22 March, 2017.
GVEP International. 2012b. The Improved Cookstove Sector in East Africa: Experience from the
Developing Energy Enterprise Programme (DEEP). September. http://www.gvepint
ernational.org/sites/default/files/deep_cookstoves_report_lq_for_web.pdf Accessed 22
March, 2017.
10-6
-------
Section 10—References
GVEP International. 2012c. Briquette Businesses in Uganda. February. http://www.gvepint
ernational.org/sites/default/files/briquette_businesses_in_uganda.pdf Accessed 22 March,
2017.
Habib, G., C. Venkataraman, M. Shrivastava, R. Banerjee, J.W. Stehr, and R.R. Dickerson.
2004. New Methodology for Estimating Biofuel Consumption for Cooking: Atmospheric
Emissions of Black Carbon and Sulfur Dioxide from India. Global Biogeochemical
Cycles 18: GB3007.
Haraldsen, T.K., U. Andersen, T. Krogstad, R. S0rheim. 2011. Liquid Digestate from Anaerobic
Treatment of Source-Separated Household Waste as Fertilizer for Barley. Waste
Management Resources.29: 1271-1276.
Hiemstra-van der Horst, G., and A.J. Hovorka. 2008. Reassessing the "Energy Ladder":
Household Energy Use in Maun, Botswana. Energy Policy. 36(9): 3333-3344.
Hu, Y., Q. Liao, S. Wang, and X. Yan. 2011. Statistical Analysis and Estimation of N Leaching
from Agricultural Fields in China, [in Chinese] Soils 43(1): 19-25.
Huang, J., B.G. Ridout, H. Zhang, C. Xu, and F. Chen. 2013. Water Footprint of Cereals and
Vegetables for the Beijing Market. The Journal of Industrial Ecology. 18(1): 40-48.
IARC (International Agency for Research on Cancer). 2010. Household Use of Solid Fuels and
High-Temperature Frying. Vol 95. http://monographs.iarc.fr/ENG/Monographs/
vol95/mono95-6.pdf Accessed 11 April, 2016.
IARI (Indian Agricultural Research Institute). 2012. Crop Residues Management with
Conservation Agriculture: Potential, Constraints and Policy Needs. Indian Agricultural
Research Institute, New Delhi. http://www.iari.res.in/files/Important_Publications-2012-
13.pdf Accessed 16 March, 2016.
IEA (International Energy Agency). 2006. Energy for Cooking in Developing Countries. In
World Energy Outlook 2006. https://www.iea.org/publications/freepublications/publica
tion/cooking.pdf Accessed 3 June, 2016.
IEA (International Energy Agency). 2007. World Energy Outlook 2007: China and India
Insights. 978-92-64-02730-5. http://www.worldenergvoutlook.org/media/weowebsite/
2008-1994/weo_2007.pdf Accessed 30 March, 2016.
IEA (International Energy Agency). 2010. Energy Technology Perspectives 2010: Scenarios and
Strategies to 2050. 978-92-64-08597-8 https://www.iea.org/publications/freepublications
/publication/etp2010.pdf Accessed 28 March, 2016.
IEA (International Energy Agency). 2013a. India: Electricity and Heat for 2013. http://www.iea.
org/statistics/statisticssearch/report/?countrv=INDIA&product=electricitvandheat&vear=
2013. Accessed 3 June 2016.
10-7
-------
Section 10—References
IEA (International Energy Agency). 2013b. China, People's Republic of: Electricity and Heat for
2013. http://www.iea.org/statistics/statisticssearch/report/?countrv=CHINA&product=
electricitvandheat&vear=2013. Accessed 3 June 2016.
IEA (International Energy Agency). 2013c. Kenya: Electricity and Heat for 2013. http://www.iea
.org/statistics/statisticssearch/report/?countrv=KENYA&product=electricitvandheat&vea
r=201. Accessed 3 June 2016.
IEA (International Energy Agency). 2013d. Ghana: Electricity and Heat for 2013. http://www.iea
.org/statistics/statisticssearch/report/?countrv=GHANA&product=electricitvandheat&vea
r=2013. Accessed 3 June 2016.
IEA (International Energy Agency). 2014. Africa Energy Outlook, https://www.iea.org/publi
cations/freepublications/publication/WEO2014_AfricaEnergvOutlook.pdf Accessed 23
March, 2017.
IEA (International Energy Agency). 2015. India Energy Outlook: World Energy Outlook Special
Report. http://www.worldenergvoutlook.org/media/weowebsite/2015/IndiaEnergvQut
look_WEQ2015.pdf Accessed 1 April, 2016.
IEA (International Energy Agency). 2016. IEA Sankey Diagram: Kenya Balance (2011).
https://www.iea.org/Sankev/#?c=Kenva&s=Balance Accessed 29 March, 2017.
IPCC (Intergovernmental Panel on Climate Change). 2006. 2006 IPCC Guidelines for National
Greenhouse Gas Inventories. Prepared for United Nations Environment Program.
http://www.ipcc-nggip.iges.or.ip/public/2006gl/ Accessed 2 March, 2016.
IPCC (Intergovernmental Panel on Climate Change). 2013. Climate change 2013: The physical
science basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, edited by Stocker, T.F., D. Qin, G.-K.
Plattner, et al. Cambridge University Press, Cambridge, United Kingdom and New York,
NY, USA.
IRENA (International Renewable Energy Agency). 2013. Africa's Renewable Future.
http://www.irena.org/DocumentDownloads/Publications/Africa_renewable_future.pdf.
Accessed 3 June, 2016.
ISO (International Standards Organization). 2010a. ISO 14040:2006, Environmental
Management-Life Cycle Assessment-Principles and Framework.
http://www.iso.org/iso/catalogue_detail?csnumber=37456 Accessed 23 March, 2017.
ISO (International Standards Organization). 2010b. ISO 14044:2006, Environmental
Management-Life Cycle Assessment-Requirements and Guidelines.
http://www.iso.org/iso/catalogue_detail?csnumber=38498 Accessed 23 March, 2017.
10-8
-------
Section 10—References
Jadhav, R. 2015. National Floor Level Minimum Wage Enhanced from Rs. 137 to Rs. 160 per
day. The Times of India. http://timesofindia.indiatimes.com/citv/pune/National-floor-
level-minimum-wage-enhanced-From-Rs-13 7-T o-Rs-160-per-dav/articleshow/47
974965.cms Accessed 8 June, 2016.
Jetter, J. J., and P. Kariher. 2009. Solid-Fuel Household Cook Stoves: Characterization of
Performance and Emissions. Biomass andBioenergy. 33: 294-305.
Jetter, J., Y. Zhao, K.R. Smith, B. Khan, T. Yelverton, P. DeCarlo, and M.D. Hays. 2012.
Pollutant Emissions and Energy Efficiency under Controlled Conditions for Household
Biomass Cookstoves and Implications for Metrics useful in setting International Test
Standards. Environmental Science & Technology 46(19): 10827-10834.
JungbluthN., F. Dinkel, G. Doka, M. Chudacoff, A. Dauriat, M. Spielmann, J. Sutter, N. Kijun,
M. Keller, and K. Schleiss. 2007. Life Cycle Inventories of Bioenergy. Final report
Ecoinvent data v2.0. Volume: 17. Swiss Centre for LCI, ESU. Duebendorf and Uster,
CH.
Kalia, A.K., and S.P. Singh. 1998. Horse Dung as a Partial Substitute for Cattle Dung for
Operating Family-Size Biogas Plants in a Hilly Region. Bioresource Technology. 64: 63-
66.
KNBS (Kenya National Bureau of Statistics). 2012. 2009 Kenya Population and Housing
Census: Analytical Report on Housing Conditions, Amenities and Household Assets.
Volume XI. Republic of Kenya, http ://www.knbs. or.ke/index.php?option=com
_phocadownload&view=categorv&id=109:population-and-housing-census-
2009&Itemid=599 Accessed 11 August, 2016.
Koster, J.R., L.M. Cardenas, R. Bol, D. Lewicka-Szczebak, M. Senbayram, R. Well, A.
Giesemann, and K. Dittert. 2015. Anaerobic Digestates lower N2O Emissions compared
to Cattle Slurry by affecting Rate and Product Stoichiometry of Denitrification - An N2O
Isomer Case Study. Soil Biology and Biochemistry. 81: 65-74.
Kumar, S., L.C. Malav, M.K. Malav, and S.A. Khan. 2015. Biogas Slurry: Sources ofNutrients
for Eco-friendly Agriculture. International Journal of Extension Research. 2: 42-46.
Macedo, I.C., J.E.A. Seabra, and J.E.A.R. Silva. 2008. Greenhouse Gas Emissions in the
Production and use of Ethanol from Sugarcane in Brazil: the 2005/2006 averages and a
prediction for 2020. Biomass and Bioenergy 32(7): 582-595.
MacCarty, N. 2009. Results of Testing of the CleanCook Stove for Fuel Use and Carbon
Emissions. Aprovecho Research Center: Advanced Studies in Appropriate Technology
Laboratory. https://prqiectgaia.eom/wp-content/uploads/2016/08/Results-of-Tests-of-the-
CleanCook-for-Fuel-Use-and-Carbon-Emissions-Aprovecho-2009.pdf Accessed 23
March, 2017.
10-9
-------
Section 10—References
MacCarty, N., D. Still, and D. Ogle. 2010. Fuel use and Emissions Performance of Fifty Cooking
Stoves in the Laboratory and related Benchmarks of Performance. Energy for Sustainable
Development. 14. 161-171.
Mainali, B., S. Pachauri, and Y. Nagai. 2012. Analyzing Cooking Fuel and Stove choices in
China till 2030. Journal of Renewable and Sustainable Energy 4: 1-14.
Makadi, M., A. Tomocsik, and V. Orosz. 2012. Digestate: A New Nutrient Source-Review.
Biogas, edited by S. Kumar, ISBN: 978-953-51-0204-5, InTech, DOI: 10.5772/31254.
http://www.intechopen.com/books/biogas Accessed 4 March, 2016.
Malla, S., and G.R. Timilsina. 2014. Household Cooking Fuel Choice and Adoption of Improved
Cookstoves in Developing Countries. World Bank: Policy Research Working Paper.
http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2014/06/
03/000158349 20140603155910/Rendered/PDF/WPS6903.pdf Accessed 1 April, 2016.
Michael, D.C., S. Zhou, X. Wu, and G. Chen. 2013. China's Energy Future: Reaching for a Clean
World. The Boston Consulting Group, http://www.bcg.com/documents/filel27070.pdf
Accessed 15 February, 2017.
Mikled, C., S. Jiraporncharoen, andN. Potikanond. 2002. Utilization of Fermented Slurry as
Bio-Fertilizer. From Proceedings of the Biodigester Workshop. March 2002.
http://www.mekarn.org/procbiod/choke.htm Accessed 4 March, 2016.
MPNG (Ministry of Petroleum & Natural Gas), Economics and Statistics Division. 2014. 2013-
14 Indian Petroleum and Natural Gas Statistics: Table III. 17. Government of India, New
Delhi, http://www.indiaenvironmentportal.org.in/files/file/pngstat%202013-14.pdf
Accessed 23 March, 2017.
NBSC (National Bureau of Statistics, China). 2008. Communique on Major Data of the Second
National Agricultural Census of China (No.4).
http://www.stats.gov.cn/enGliSH/NewsEvents/200802/t20080229_25997.html Accessed
23 March, 2017.
Nkoa, R. 2014. Agricultural Benefits and Environmental Risks of Soil Fertilization with
Anaerobic Digestates: a Review. Agronomic Sustainable Development. 34: 473-492.
NREL (National Renewable Energy Laboratory). 2012. U.S. LCI database.
https://uslci.lcacommons.gov/uslci/searchAccessed 23 March, 2017.
Odeh, N.A., and T.T. Cockerill. 2008. Life Cycle GHG Assessment of Fossil Fuel Power Plants
with Carbon Capture and Storage. Energy Policy. 36: 367-380.
Paliwal, U. M. Sharma, J.F. Burkhart. 2016. Monthly and Spatially Resolved Black Carbon
Emission Inventory of India: Uncertainty Analysis. Atmospheric Chemistry and Physics.
16: 12457-12476.
10-10
-------
Section 10—References
Pennise, D.M., K.R. Smith, J.P. Kithinji, M.E. Rezende, T.J. Raad, J. Zhang, and C. Fan. 2001.
Emissions of Greenhouse Gases and other Airborne Pollutants from Charcoal making in
Kenya and Brazil. Journal of Geophysical Research, 106(20). 24143-24155.
Plochl, M., and M. Heiermann. 2006. Biogas Farming in Central and Northern Europe: A
Strategy for Developing Countries? Leibniz Institute of Agricultural Engineering.
Potsdam-Bornim, Germany, https://ecommons.cornell.edu/bitstream/handle/1813
/105 5 9/Invited%200verview%20Plochl%20final%2027F eb2006. pdf?sequence= 1.
Accessed 23 March, 2016.
Poeschl, M., S. Ward, and P. Owende. 2012. Environmental Impacts of Biogas Deployment -
Part I: Life Cycle Inventory for Evaluation of Production Process Emissions to Air.
Journal of Cleaner Production. 24: 168-183.
Radcliffe, D.E., D.K. Reid, K. Blomback, C.H. Bolster, A.S. Collick, Z.M. Easton, W.
Francesconi, D.R. Fuka, H. Johnsson, K. King, M. Larsbo, M.A. Youssef, A.S. Mulkey,
N.O. Nelson, K. Persson, J.J. Ramirez-Avila, F. Schmieder, and D.R. Smith. 2015.
Applicability of Models to Predict Phosphorus Losses in Drained Fields: A Review.
Journal of Environmental Quality. 44: 614-628.
Robinson, James. Stove Thermal Efficiency and Emissions Interim Test Report - RocketWorks
Wood Stove, 11-12 February 2013.
ROK (Republic of Kenya). 2011. Updated Least Cost Power Development Plan Study Period:
2011-2031. http://www.renewableenergv.go.ke/downloads/studies/LCPDP-2011-2030-
Studv.pdf Accessed 23 March, 2017.
Saud T., R. Gautam, and T.K. Mandal, et al. 2012. Emission Estimates of Organic and Elemental
Carbon from Household Biomass Fuel used over the Indo-Gangetic Plain (IGP), India.
Atmospheric Environment. 61: 212-220.
Schaetzke, M. 1995. Ermittlung des Energieverbrauchs und des Wirkungsgrades von
Elektrokochstellen, Hauswirtschaft und Wissenschaft. 43(5): 195-199.
SEI (Stockholm Environment Institute). 2016. Country Energy Planning in Kenya: Local
Participation and Local Solutions in Migori County. Working Paper No. 2016-01.
https://www.sei-international.org/mediamanager/documents/Publications/SEI-WP-2016-
01-Kenva-countv-energv-planning.pdf Accessed 23 March, 2017.
Shen, G., W. Wang, Y. Yang, C. Zhu, Y. Min, M. Xue, J. Ding, W. Li, B. Wang, H. Shen, R.
Wang, X. Wang and S, Tao. 2010a. Emission Factors and Particulate Matter Size
Distribution of Poly cyclic Aromatic Hydrocarbons from Residential Coal Combustions in
Rural Northern China. Atmosphere and Environment. 44(39): 5737-5743.
Shen, G., Y. Yang, W. Wang, S. Tao, C. Zhu, Y. Min, M. Xue, J. Ding, B. Wang, R. Wang, H.
Shen, W. Li, X. Wang, and A.G. Russell. 2010b. Emission Factors of Particulate Matter
10-11
-------
Section 10—References
and Elemental Carbon for Crop Residues and Coals Burned in Typical Household Stove
in China. Environmental Science and Technology. 44(18): 7157-7162.
Shen, G., S. Tao, S. Wei, Y. Zhang, R. Wang, B. Wang, W. Li, H. Shen, Y. Huang, Y. Chen, H.
Chen, Y. Yang, W. Wang, X. Wang, W. Liu and S.L.M. Simonich. 2012a. Emissions of
Parent, Nitro, and Oxygenated PAHs from Residential Wood Combustion in Rural China.
Environmental Science and Technology. 46(15): 8123-8130.
Shen, G., S. Wei, W. Wei, Y. Zhang, Y. Min, B. Wang, R. Wang, W. Li, H. Shen, Y. Huang, Y.
Yang, W. Wang, X. Yang, X. Wang and S, Tao. 2012b. Emission Factors, Size
Distributions and Emission Inventories of Carbonaceous Particulate Matter from
Residential Wood Combustion in Rural China. Environmental Science and Technology.
46(7): 4207-4214.
Shen, G., S. Tao, S. Wei, Y. Chen, Y. Zhang, H. Shen, Y. Huang, D. Zhu, C. Yuan, H. Wang, Y.
Wang, L. Pei, Y. Liao, Y. Duan, B. Wang, R. Wang, Y. Lu, W. Li, X. Wang, and X.
Zheng. 2013. Field Measurement of Emission Factors of PM, EC, OC, Parent, Nitro-, and
Oxy-Polycyclic Aromatic Hydrocarbons for Residential Briquette, Coal Cake, and Wood
in Rural Shanxi, China. Environmental Science and Technology. 47(6): 2998-3005.
Singh P, H. Gundimeda, and M. Stucki. 2014a. Environmental Footprint of Cooking Fuels: a
Life Cycle Assessment of Ten Fuel Sources used in Indian Households. International
Journal of Life Cycle Assessment 19: 1036-1048.
Singh, P., and H. Gundimeda. 2014b. Life Cycle Energy Analysis (LCEA) of Cooking Fuel
Sources used in India Households. Energy and Environmental Engineering 2(1): 20-30.
Sloss, L. 2012. Black Carbon Emissions in India. IEA Clean Coal Centre. ISBN 978-92-9029-
529-7.
Smith, K., J. Grylls, P. Metcalfe, B. Jeffrey, and A. Sinclair. 2007. Nutrient Value of Digestate
from Farm-based Biogas Plants in Scotland. Prepared for Scottish Executive
Environment and Rural Affairs Department, http://www.gov.scot/resource/doc/
1057/0053041.pdf Accessed 4 March, 2016.
Smith K.R., R. Uma, V.V.N. Kishore, J. Zhang, V. Joshi, and M.A.K. Khalil. 2000. Greenhouse
Implications of Household Stoves: an Analysis for India. Annual Review of Energy and
the Environment 61: 212-220.
SID (Society for International Development). 2015. Sources of Cooking Fuel.
http://inequalities.sidint.net/kenva/national/sources-of-cooking-fuel/ Accessed 23 March,
2017.
Somasundaram, E., M.M. Amanullahaiyapuri, K. Thirukkumaran, and K. Sathyamoorthi. 2007.
Influence of Organic Sources of Nutrients on the Yield and Economics of Crops under
Maize based Cropping System. Journal of Applied Scientific Research. 3: 1774-1777.
10-12
-------
Section 10—References
Sweeney, D. 2015. Field Evaluation of Alternative and Traditional Cooking Fuels in Haiti.
http://d-lab.mit.edu/. Accessed 3 June, 2016.
TekCarta. 2015. Households Average Size (68 countries). Available at
http://www.generatorresearchnakono.com/tekcarta/databank/households-average-
household-size/ Accessed 10 August, 2015.
TERI (The Energy and Resources Institute). 2006. National Energy Map for India: Technology
Vision 2030. Office of the Principal Scientific Adviser to the Government of India. 81-
7993-099-8. http://www.teriin.org/div/psa-fullreport.pdf Accessed 29 March, 2016.
Thurber, M.C., H. Phadke, S. Nagavarapu, G. Shrimali, and H. Zerriffi. 2014. 'Oorja' in India:
Assessing a Large-Scale Commercial Distribution of Advanced Biomass Stoves to
Households. "Energy for Sustainable Development. 19: 138-150.
Tsiropoulos, I., A.P.C. Faaij, J.E.A. Seabra, L. Lundquist, U. Schenker, J.-F. Briois, and M.K.
Patel. 2014. Life Cycle Assessment of Sugarcane Ethanol Production in India in
comparison to Brazil. International Journal of Life Cycle Assessment 19: 1049-1067.
UN (United Nations). 2007. Bagepalli CDM Project: Project Definition Document.
https://cdm.unfccc.int/filestorage/Y/X/O/YXOEHNRZMVDU24JTSOLG36P19A7I0W/
CDM-PDD.pdf?t=ZEN8b25hMWpsfDAEIASaHO0wmclziHtiJJhl Accessed 23 March,
2017.
UNCCD (United Nations Convention to Combat Desertification). 2015. Combating
desertification in Asia. http://www.unccd.int/en/regional-access/Asia/Pages/default.aspx.
Accessed 23 March, 2017.
USDA (US Department of Agriculture), U.S. EPA (US Environmental Protection Agency).
2015. US Federal LCA Digital Commons Life Cycle Inventory Template, https://data.nal
.usda.gov/dataset/us-federal-lca-commons-life-cvcle-inventorv-unit-process-template
Accessed January 2015.
U.S. EPA (Environmental Protection Agency). 2012. Report to Congress on Black Carbon. EPA-
450/R-12-001. https://www3.epa.gov/airqualitv/blackcarbon/2012report/fullreport.pdf
Accessed 11 December, 2016.
Venkataraman, C., A.D. Sagar, G. Habib, N. Lam, and K.R. Smith. 2010. The Indian National
Initiative for Advanced Biomass Cookstoves: The Benefits of Clean Combustion. Energy
for Sustainable Development 14(2): 63-72.
Wang, S., W. Wei, L. Du, G. Li, and J. Hao. 2009. Characteristics of Gaseous Pollutants from
Biofuel-Stoves in Rural China. Atmospheric Environment. 43: 4148-4154.
Wang C., Yang Y., Zhang Y. 2012. Rural Household Livelihood Change, Fuelwood
Substitution, and Hilly Ecosystem Restoration: Evidence from China. Available at
10-13
-------
Section 10—References
http://www.sciencedirect.com/science/article/pii/S1364032112000822 Accessed 29
March, 2017.
Wang, R. S. Tao, W. Wang, J. Liu, H. Shen, G. Shen, B. Wang, X. Liu, W. Li, Y. Huang, Y. Lu,
H. Chen, B. Li, Y. Chen, C. Wang, D. Zhu, X. Wang, B. Li, W. Liu, and J. Ma. 2014.
Black Carbon Emissions in China from 1949 to 2050. Environmental Science &
Technology. 46: 7595-7603.
Wang, X., L. Yang, Y. Steinberger, Z. Liu, S. Liao, and G. Xie. 2013. Field Crop Residue
estimate and availability for Biofuel Production in China. Renewable and Sustainable
Energy Reviews. 27: 864-875.
Wang, G.L., Y.L. Ye, X.P. Chen, and Z.L. Cui. 2014. Determining the Optimal Nitrogen Rate
for Summer Maize in China by Integrating Agronomic, Economic and Environmental
Aspects. Biogeosciences. 30: 3031-3041.
Wei, S., G. Shen, Y. Zhang, M. Xue, H. Xie, P. Lin, Y. Chen, X. Wang, and S. Tao. 2014. Field
Measurement on the Emissions of PM, OC, EC and PAHs from Indoor Crop Straw
Burning in Rural China. Environmental Pollution. 184: 18-24.
Weidema, B. 2000. Avoiding Co-Product Allocation in Life-Cycle Assessment. Journal of
Industrial Ecology. 4(3): 11-33.
Wilson, D.L., D.R Talancon, R.L. Winslow, X. Linares, and A.J. Gadgil. 2016. Avoided
Emissions of a Fuel-Efficient Biomass Cookstove Dwarf Embodied Emissions.
Development Engineering, http://www. sciencedirect.com/science/article/pii/S23 5
2728515300464 Accessed 24 February, 2016.
Winrock International. 2009. Commercialization of Improved Cookstoves for Reduced Indoor
Air Pollution in Urban Slums of Northwest Bangladesh. Washington, D.C.: United States
Agency for International Development. http://pdf.usaid.gov/pdf_docs/PNAD0851 .pdf
Accessed 9 June, 2016.
World Bank. 2012. Access to electricity. Washington, D.C. http://data.worldbank.org
/indicator/EG.ELC.ACCS.ZS. Accessed 7 June 2016.
World Bank (The World Bank). 2013. China: Accelerating Household Access to Clean Cooking
and Heating. Asia Sustainable and Alternative Energy Program. Washington, DC: World
Bank. http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/
IB/2013/10/01/000356161 20131001104809/Rendered/PDF/814950WP0P12980Box03 7
9837B00PUBLIC0.pdf Accessed 4 April, 2016.
World Bank. 2014a. GDP Growth (Annual %). Washington, D.C.
http://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG Accessed 23 March, 2017.
World Bank. 2014b. Population, total. Washington, D.C.
http://data.worldbank.org/indicator/SP.POP.TOTL Accessed 23 March, 2017.
10-14
-------
Section 10—References
World Bank. 2014c. Population growth (Annual %). Washington, D.C.
http://data.worldbank.org/indicator/SP.POP.GROW Accessed 23 March, 2017.
Xia, Y., and X. Yan. 2011. Life-Cycle Evaluation of Nitrogen-se in Rice-Farming Systems:
Implications for Economically-Optimal Nitrogen Rates. Biogeosciences. 8: 3159-3168.
Zhang, J., K.R. Smith, R. Uma, Y. Ma, V.V.N. Kishor, K. Lata, M.A.K. Khalil, R.A. Rasmussen,
and S.T. Thorneloe. 1999. Carbon Monoxide from Cookstoves in Developing Countries:
Emission Factors. Chemosphere-Global Change Science. 1: 353-366.
Zhang J., K.R. Smith Y. Ma, S. Ye, F. Jiang, W. Qi, P. Liu, M.A.K.Khalil, R.A. Rasmussen, and
S.A. Thorneloe. 2000. Greenhouse Gases and other Airborne Pollutants from Household
Stoves in China: a Database for Emission Factors. Atmospheric Environment 34(26):
4537-4549.
Zhi G., Y. Chen, Y. Feng, S. Xiong, J. Li, G. Zhang, G. Sheng, and J. Fu. 2008. Emission
Characteristics of Carbonaceous Particles from various Residential Coal-Stoves in China.
Environmental Science and Technology 42(9): 3310-3315.
Zhou, N, M.A. McNeil, D. Fridley, J. Lin, L. Price, S. de la Rue du Can, J. Sathaye, and M.
Levine. 2007. Energy use China: Sectoral Trends and Future Outlook. Lawrence
Berkeley National Laboratory, LBNL-61904. https://china.lbl.gov/sites/all/files/lbl-
61904-sectoral-energy-trendian-2007.pdf Accessed 23 March, 2017.
Zhou, N, D. Fridley, M. McNeil, N. Zheng, J. Ke, and M. Levine. 2011. China's Energy and
Carbon Emissions Outlook to 2050. Ernest Orlando Lawrence Berkeley National
Laboratory. LBNL-4472E. https://china.lbl.gov/sites/all/files/lbl-4472e-energv-2050april-
2011.pdf Accessed 28 March, 2016.
10-15
-------
Appendix A
Baseline Single Cooking Fuel Results by Life Cycle Stage
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-l. Single Fuel LCIA Results by Life Cycle Stage for India1
per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Hard Coal
16.2
-
2.17
945
963
Dung Cake
-
-
-
263
263
Crop Residue
-
-
-
119
119
Firewood
-
-
-
196
196
Charcoal from Wood
-
227
00.0
176
402
GCCP100 -
Kerosene
6.04
14.3
15.5
145
180
Climate Change
LPG
5.09
10.6
14.5
127
157
(kg C02 eq)
Natural Gas
0.031
3.86
1.87
112
117
Electricity
-
-
-
457
457
Sugarcane Ethanol
85.9
13.7
20.5
0.953
121
Biogas from Cattle
Dung
-
8.99
-
2.42
11.4
Biomass Pellets
-
50.8
-
90.2
141
Hard Coal
281
-
42.3
6.89E+3
7.21E+3
Dung Cake
-
-
-
1.30E+4
1.30E+4
Crop Residue
-
-
-
1.01E+4
1.01E+4
Firewood
-
-
-
6.52E+3
6.52E+3
Charcoal from Wood
-
5.06E+3
000
5.82E+3
1.09E+4
CED - Energy
Demand (MJ)
Kerosene
120
335
257
2.37E+3
3.09E+3
LPG
104
239
313
1.96E+3
2.61E+3
Natural Gas
13.3
74.0
21.9
1.93E+3
2.04E+3
Electricity
-
-
-
5.70E+3
5.70E+3
Sugarcane Ethanol
2.30E+3
8.62E+3
345
2.08E+3
1.33E+4
Biogas from Cattle
-
2.11E+3
-
1.95E+3
4.06E+3
Dung
Biomass Pellets
-
800
-
3.11E+3
3.91E+3
Hard Coal
6.71
-
0.619
164
172
Dung Cake
-
-
-
0.152
0.152
Crop Residue
-
-
-
7.90E-3
7.90E-3
Firewood
-
-
-
5.94E-3
5.94E-3
Charcoal from Wood
-
5.12E-3
0.00
6.09E-3
0.01
FDP - Fossil
Kerosene
2.83
7.57
5.19
55.3
70.9
Depletion (kg oil
LPG
2.44
5.39
6.48
44.4
58.7
eq)
Natural Gas
0.316
1.74
0.461
46.1
48.7
Electricity
-
-
-
122
122
Sugarcane Ethanol
17.2
6.74
7.02
-
31.0
Biogas from Cattle
-
-
-
-
-
Dung
Biomass Pellets
-
13.72
-
2.10E-4
13.72
Hard Coal
0.312
-
0.020
0.066
0.397
Dung Cake
-
-
-
1.68E-3
1.68E-3
WDP - Water
Depletion (m3)
Crop Residue
-
-
-
8.72E-5
8.72E-5
Firewood
-
-
-
6.54E-5
6.54E-5
Charcoal from Wood
-
5.64E-5
0.000
6.36E-5
0.000
A-l
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-l. Single Fuel LCIA Results by Life Cycle Stage for India1
per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Kerosene
0.058
0.108
0.073
-
0.239
LPG
0.049
0.076
0.069
-
0.193
Natural Gas
3.00E-4
0.037
2.40E-3
-
0.039
Electricity
-
-
-
3.25
3.25
Sugarcane Ethanol
642
0.987
0.082
-1.00E-5
643
Biogas from Cattle
-
1.02
-
-
1.02
Dung
Biomass Pellets
-
0.357
-
-5.28E-18
0.357
Hard Coal
1.66
-
4.94E-3
18.1
19.8
Dung Cake
24.3
24.3
Crop Residue
-
-
-
11.4
11.4
Firewood
-
-
-
5.54
5.54
PMFP -
Charcoal from Wood
-
18.8
0.000
1.73
20.5
Particulate
Kerosene
0.011
0.086
0.034
0.039
0.171
Matter
LPG
9.59E-3
0.068
0.032
0.026
0.136
Formation (kg
Natural Gas
5.88E-5
7.21E-3
1.12E-3
0.011
0.019
PM10 eq)
Electricity
-
-
-
1.91
1.91
Sugarcane Ethanol
4.25
0.080
0.041
8.10E-4
4.38
Biogas from Cattle
-
-
-
0.210
0.210
Dung
Biomass Pellets
-
0.209
-
0.092
0.302
Hard Coal
0.141
-
0.012
7.71
7.87
Dung Cake
-
-
-
18.8
18.8
Crop Residue
-
-
-
8.22
8.22
Firewood
-
-
-
5.38
5.38
POFP -
Charcoal from Wood
-
5.30
0.000
5.05
10.4
Ph otoch emical
Kerosene
0.028
0.187
0.112
0.154
0.481
Oxidant
LPG
0.024
0.135
0.108
0.074
0.341
Formation (kg
Natural Gas
1.40E-4
0.018
4.74E-3
0.023
0.046
NMVOC)
Electricity
-
-
-
2.66
2.66
Sugarcane Ethanol
0.233
0.132
0.150
0.118
0.633
Biogas from Cattle
-
3.59E-3
-
0.110
0.114
Dung
Biomass Pellets
-
0.302
-
1.218
1.520
Hard Coal
8.67E-6
-
1.37E-3
9.91E-4
2.37E-3
Dung Cake
-
-
-
0.189
0.189
Crop Residue
-
-
-
9.80E-3
9.80E-3
FEP -
Firewood
-
-
-
7.36E-3
7.36E-3
Freshwater
Charcoal from Wood
-
6.35E-3
0.00E+0
7.56E-3
0.014
Eutroph ication
Kerosene
1.73E-5
1.07E-3
2.68E-3
-3.94E-6
3.77E-3
(kgPeq)
LPG
1.46E-5
7.37E-4
2.62E-3
-
3.37E-3
Natural Gas
8.92E-8
1.11E-5
5.93E-5
-
7.05E-5
Electricity
-
-
-
3.75E-3
3.75E-3
Sugarcane Ethanol
0.029
4.95E-3
3.48E-3
-1.00E-5
0.038
A-2
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-l. Single Fuel LCIA Results by Life Cycle Stage for India1
per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Biogas from Cattle
-
-
-
-
-
Dung
Biomass Pellets
-
0.005
-
2.50E-4
0.006
Hard Coal
0.076
-
0.011
1.78
1.87
Dung Cake
-
-
-
0.736
0.736
Crop Residue
-
-
-
0.598
0.598
Firewood
-
-
-
0.377
0.377
Charcoal from Wood
-
4.59E-3
0.000
0.205
0.209
TAP-
Terrestrial
Acidification (kg
S02 eq)
Kerosene
0.021
0.168
0.075
0.027
0.291
LPG
0.018
0.153
0.071
0.014
0.256
Natural Gas
1.10E-4
0.013
2.65E-3
0.010
0.027
Electricity
-
-
-
4.54
4.54
Sugarcane Ethanol
3.94
0.322
0.090
-
4.35
Biogas from Cattle
-
-
-
0.106
0.106
Dung
Biomass Pellets
-
0.496
-
0.006
0.502
Hard Coal
1.20E-10
-
1.70E-8
1.30E-8
3.01E-8
Dung Cake
-
-
-
1.40E-9
1.40E-9
Crop Residue
-
-
-
7.28E-11
7.28E-11
Firewood
-
-
-
5.46E-11
5.46E-11
Charcoal from Wood
-
4.71E-11
0.00E+0
5.62E-11
1.03E-10
ODP - Ozone
Kerosene
1.06E-9
1.41E-8
4.68E-8
-1.40E-15
6.20E-8
Depletion (kg
LPG
8.98E-10
2.02E-8
4.45E-8
-6.90E-14
6.56E-8
CFC-11 eq)
Natural Gas
5.48E-12
6.88E-10
7.18E-8
-
7.25E-8
Electricity
-
-
-
4.24E-7
4.24E-7
Sugarcane Ethanol
2.65E-6
1.15E-7
6.22E-8
-5.40E-12
2.82E-6
Biogas from Cattle
-
-
-
-
-
Dung
Biomass Pellets
-
5.30E-8
-
2.00E-12
5.31E-8
Hard Coal
0.345
-
1.10E-4
3.75
4.10
Dung Cake
-
-
-
5.27
5.27
Crop Residue
-
-
-
2.48
2.48
Firewood
-
-
-
1.22
1.22
Black Carbon
and Short-Lived
Charcoal from Wood
-
4.10
0.00E+0
0.479
4.58
Kerosene
7.33E-4
8.93E-3
1.37E-3
0.010
0.021
Climate
LPG
6.13E-4
4.18E-3
1.29E-3
5.92E-3
0.012
Pollutants (kg
BCeq)
Natural Gas
3.79E-6
4.56E-4
4.00E-5
1.57E-3
2.07E-3
Electricity
-
-
-
-0.016
-0.016
Sugarcane Ethanol
0.764
-0.014
2.05E-3
5.34E-3
0.757
Biogas from Cattle
-
-
-
0.035
0.035
Dung
Biomass Pellets
-
-1.58E-3
-
0.028
0.026
1 LC A results presented in this table are calculated as part of this study based on the methodology described in the report body.
A-3
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-2. Single Fuel LCIA Results by Life Cycle Stage for China1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Coal Mix
222
14.3
10.5
616
862
Coal Powder
260
-
17.4
888
1.16E+3
Coal Briquettes
191
29.8
3.79
368
593
Coal Honeycomb
176
27.5
3.49
319
527
GCCP100 -
Firewood
-
-
-
190
190
Climate
Crop Residue
-
-
-
64.1
64.1
Change (kg
Kerosene
27.1
19.9
16.4
161
225
C02 eq)
Biomass Pellets
-
62.3
-
77.5
140
Electricity
-
-
-
612
612
LPG
24.0
24.8
15.9
148
213
Natural Gas
9.25
-
32.8
112
154
Coal Gas
90.8
30.0
41.0
92.5
254
Sugarcane Ethanol
94.4
13.7
4.02
0.956
113
Biogas from Cattle Dung
-
8.99
-
2.42
11.4
Coal Mix
1.34E+3
547
184
6.27E+3
8.34E+3
Coal Powder
1.72E+3
-
307
8.78E+3
1.08E+4
Coal Briquettes
865
1.14E+3
62.7
3.31E+3
5.37E+3
Coal Honeycomb
1.05E+3
1.05E+3
57.8
4.22E+3
6.37E+3
Firewood
-
-
-
7.61E+3
7.61E+3
CED - Energy
Crop Residue
-
-
-
7.45E+3
7.45E+3
Demand (MJ)
Kerosene
461
283
262
2.52E+3
3.53E+3
Biomass Pellets
-
900
-
2.88E+3
3.78E+3
Electricity
-
-
-
7.22E+3
7.22E+3
LPG
409
348
357
2.30E+3
3.41E+3
Natural Gas
75.8
-
360
1.93E+3
2.37E+3
Coal Gas
602
177
523
2.39E+3
3.69E+3
Sugarcane Ethanol
2.38E+3
8.62E+3
55.7
2.08E+3
1.31E+4
Biogas from Cattle Dung
-
2.11E+3
-
1.95E+3
4.06E+3
Coal Mix
31.8
11.5
2.99
150
196
Coal Powder
40.1
-
4.74
210
254
Coal Briquettes
21.2
23.9
1.29
78.9
125
Coal Honeycomb
25.5
22.0
1.19
101
149
Firewood
-
-
-
3.63E-3
3.63E-3
FDP - Fossil
Depletion (kg
oil eq)
Crop Residue
-
-
-
0.010
0.010
Kerosene
8.70
5.76
5.26
57.2
76.9
Biomass Pellets
-
12.26
-
1.90E-4
12.26
Electricity
-
-
-
118
118
LPG
7.70
7.16
7.33
52.2
74.4
Natural Gas
1.50
-
7.64
46.1
55.3
Coal Gas
14.0
2.81
11.2
54.2
82.3
Sugarcane Ethanol
16.9
6.74
1.19
.
24.9
Biogas from Cattle Dung
-
-
-
.
-
Coal Mix
0.860
0.095
0.078
0.016
1.05
A-4
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-2. Single Fuel LCIA Results by Life Cycle Stage for China1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Coal Powder
1.01
-
0.139
8.64E-3
1.15
Coal Briquettes
0.742
0.197
0.017
0.030
0.986
Coal Honeycomb
0.684
0.182
0.016
0.017
0.899
Firewood
-
-
-
4.00E-5
4.00E-5
Crop Residue
-
-
-
1.18E-4
1.18E-4
WDP - Water
Depletion (m3)
Kerosene
0.267
0.134
0.079
-
0.480
Biomass Pellets
-
0.417
-
1.00E-5
0.417
Electricity
-
-
-
4.07
4.07
LPG
0.256
0.125
0.080
-
0.461
Natural Gas
0.012
-
0.013
-
0.025
Coal Gas
0.352
0.208
0.016
-
0.576
Sugarcane Ethanol
642
0.987
0.013
.
643
Biogas from Cattle Dung
-
1.02
-
.
1.02
Coal Mix
0.194
0.139
0.026
10.9
11.2
Coal Powder
0.228
-
0.042
21.2
21.5
Coal Briquettes
0.168
0.289
9.68E-3
0.522
0.989
Coal Honeycomb
0.155
0.267
8.92E-3
0.648
1.08
PMFP -
Firewood
-
-
-
6.50
6.50
Particulate
Matter
Formation (kg
PM10 eq)
Crop Residue
-
-
-
10.0
10.0
Kerosene
0.157
0.058
0.034
0.017
0.266
Biomass Pellets
-
0.168
-
0.143
0.311
Electricity
-
-
-
1.65
1.65
LPG
0.148
0.058
0.033
9.06E-3
0.248
Natural Gas
0.019
-
0.018
0.011
0.048
Coal Gas
0.080
0.346
0.023
0.046
0.495
Sugarcane Ethanol
4.24
0.080
0.013
8.10E-4
4.33
Biogas from Cattle Dung
-
-
-
0.210
0.210
Coal Mix
0.178
0.078
0.070
1.95
2.28
Coal Powder
0.208
-
0.106
2.99
3.30
Coal Briquettes
0.154
0.162
0.036
0.349
0.700
Coal Honeycomb
0.142
0.149
0.033
1.49
1.82
POFP -
Firewood
-
-
-
2.23
2.23
Ph otoch emical
Oxidant
Formation (kg
NMVOC)
Crop Residue
-
-
-
5.52
5.52
Kerosene
0.250
0.098
0.115
0.120
0.582
Biomass Pellets
-
0.244
-
1.096
1.340
Electricity
-
-
-
2.31
2.31
LPG
0.236
0.094
0.123
0.047
0.500
Natural Gas
0.075
-
0.083
0.023
0.181
Coal Gas
0.073
1.04
0.103
0.096
1.31
Sugarcane Ethanol
0.213
0.132
0.049
0.118
0.511
Biogas from Cattle Dung
-
3.59E-3
-
0.110
0.114
FEP -
Coal Mix
0.091
5.30E-3
4.56E-3
2.42E-4
0.102
Freshwater
Coal Powder
0.107
-
8.54E-3
1.32E-4
0.116
A-5
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-2. Single Fuel LCIA Results by Life Cycle Stage for China1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Eutroph ication
Coal Briquettes
0.079
0.011
6.10E-4
4.50E-4
0.091
(kgPeq)
Coal Honeycomb
0.073
0.010
5.68E-4
2.54E-4
0.084
Firewood
-
-
-
4.50E-3
4.50E-3
Crop Residue
-
-
-
0.013
0.013
Kerosene
6.89E-3
3.48E-3
2.44E-3
-
0.013
Biomass Pellets
-
0.012
-
2.40E-4
0.012
Electricity
-
-
-
0.078
0.078
LPG
7.01E-3
2.62E-3
2.79E-3
-1.00E-5
0.012
Natural Gas
2.90E-4
-
3.90E-4
-
6.80E-4
Coal Gas
0.037
4.13E-3
4.90E-4
-
0.042
Sugarcane Ethanol
0.033
4.95E-3
5.00E-4
.
0.039
Biogas from Cattle Dung
-
-
-
.
-
Coal Mix
0.891
0.106
0.063
0.333
1.39
Coal Powder
1.04
-
0.105
0.514
1.66
Coal Briquettes
0.769
0.221
0.022
0.184
1.20
Coal Honeycomb
0.709
0.203
0.020
0.120
1.05
TAP-
Firewood
-
-
-
0.242
0.242
Terrestrial
Crop Residue
-
-
-
0.367
0.367
Acidification
Kerosene
0.644
0.199
0.088
0.029
0.960
1 (kg S02 eq)
Biomass Pellets
-
0.532
-
0.003
0.535
Electricity
-
-
-
5.27
5.27
LPG
0.604
0.200
0.080
0.014
0.898
Natural Gas
0.082
-
0.051
0.010
0.143
Coal Gas
0.365
0.242
0.063
0.133
0.803
Sugarcane Ethanol
3.97
0.322
0.031
.
4.33
Biogas from Cattle Dung
-
-
-
0.106
0.106
Coal Mix
8.13E-9
4.98E-8
5.59E-8
3.14E-9
1.17E-7
Coal Powder
9.52E-9
-
1.02E-7
1.71E-9
1.13E-7
Coal Briquettes
7.01E-9
1.04E-7
1.02E-8
5.83E-9
1.27E-7
Coal Honeycomb
6.47E-9
9.57E-8
9.41E-9
3.32E-9
1.15E-7
Firewood
-
-
-
3.34E-11
3.34E-11
ODP - Ozone
Depletion (kg
CFC-11 eq)
Crop Residue
-
-
-
9.52E-11
9.52E-11
Kerosene
1.00E-7
4.55E-8
3.91E-8
5.22E-13
1.85E-7
Biomass Pellets
-
2.54E-8
-
1.80E-12
2.54E-8
Electricity
-
-
-
1.67E-7
1.67E-7
LPG
1.01E-7
3.42E-8
4.58E-8
3.50E-13
1.81E-7
Natural Gas
4.44E-9
-
9.69E-7
-
9.74E-7
Coal Gas
3.33E-9
1.16E-5
2.29E-6
-
1.39E-5
Sugarcane Ethanol
2.63E-6
1.15E-7
6.22E-9
0
2.75E-6
Biogas from Cattle Dung
-
-
-
.
-
Black Carbon
Coal Mix
-0.047
0.020
2.74E-4
2.32
2.29
and Short-
Coal Powder
-0.055
-
1.63E-4
4.51
4.45
A-6
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-2. Single Fuel LCIA Results by Life Cycle Stage for China1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Lived Climate
Coal Briquettes
-0.040
0.041
4.00E-4
0.103
0.105
Pollutants (kg
BCeq)
Coal Honeycomb
-0.037
0.038
3.67E-4
0.159
0.160
Firewood
-
-
-
1.42
1.42
Crop Residue
_
_
_
2.20
2.20
Kerosene
-0.030
-5.05E-3
2.00E-4
2.62E-3
-0.032
Biomass Pellets
-
-0.014
-
0.045
0.031
Electricity
-
-
-
-0.148
-0.148
LPG
-0.027
-6.54E-3
8.30E-4
1.65E-3
-0.031
Natural Gas
-3.57E-3
-
-3.00E-5
1.56E-3
-2.04E-3
Coal Gas
-0.019
0.059
-4.00E-5
-2.15E-3
0.038
Sugarcane Ethanol
0.756
-0.014
3.40E-4
5.33E-3
0.748
Biogas from Cattle Dung
-
-
-
0.035
0.035
1 LC A results presented in this table are calculated as part of this study based on the methodology described in the report body.
A-7
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-3. Single Fuel LCIA Results by Life Cycle Stage for Kenya1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Feedstock
Fuel
Distribution
Cookstove
Category:
Fuel:
Production
Processing
Use
TOTAL
Firewood
-
-
-
439
439
Charcoal from Wood
-
510
15.4
282
808
Biomass Pellets
-
35.1
-
226
261
GCCP100 -
Climate
Change (kg
C02 eq)
Kerosene
53.5
13.0
3.99
152
223
LPG
48.3
11.7
14.9
141
216
Electricity
-
-
-
238
238
Biogas from Cattle
Dung
-
11.0
-
2.42
13.4
Sugarcane Ethanol
73.8
13.7
1.56
0.953
90.0
Firewood
-
-
-
9.11E+3
9.11E+3
Charcoal from Wood
-
6.28E+3
260
5.63E+3
1.22E+4
Biomass Pellets
-
1104
-
3.11E+3
4.21E+3
CED - Energy
Demand (MJ)
Kerosene
740
4.65E+3
75.7
2.49E+3
7.96E+3
LPG
667
4.31E+3
331
2.20E+3
7.51E+3
Electricity
-
-
-
7.40E+3
7.40E+3
Biogas from Cattle
Dung
-
2.22E+3
-
1.99E+3
4.21E+3
Sugarcane Ethanol
2.39E+3
8.62E+3
26.0
2.08E+3
1.31E+4
Firewood
-
-
-
7.20E-3
7.20E-3
Charcoal from Wood
-
2.00E-3
5.34
5.41E-3
5.35
Biomass Pellets
-
19.4
-
2.10E-4
19.4
FDP - Fossil
Kerosene
14.4
111
1.62
59.5
186
Depletion (kg
LPG
13.0
103
7.12
52.5
175
oil eq)
Electricity
-
-
-
150
150
Biogas from Cattle
Dung
-
-
-
-
-
Sugarcane Ethanol
18.7
6.74
0.565
-
26.0
Firewood
-
-
-
7.96E-5
7.96E-5
Charcoal from Wood
-
2.21E-5
0.068
6.04E-5
0.068
Biomass Pellets
-
0.635
-
1.00E-5
0.635
WDP - Water
Depletion (m3)
Kerosene
0.799
0.021
0.036
-
0.856
LPG
0.721
0.019
0.076
-1.00E-5
0.816
Electricity
-
-
-
5.58
5.58
Biogas from Cattle
Dung
-
3.29
-
-
3.29
Sugarcane Ethanol
642
0.987
6.79E-3
-
643
Firewood
-
-
-
15.5
15.5
Charcoal from Wood
-
6.97
0.039
1.38
8.40
PMFP -
Particulate
Matter
Formation (kg
PM10 eq)
Biomass Pellets
-
0.060
-
0.092
0.152
Kerosene
0.083
0.093
9.11E-3
0.017
0.202
LPG
0.075
0.084
0.029
9.06E-3
0.196
Electricity
-
-
-
0.331
0.331
Biogas from Cattle
Dung
-
-
-
0.187
0.187
A-8
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-3. Single Fuel LCIA Results by Life Cycle Stage for Kenya1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Feedstock
Fuel
Distribution
Cookstove
Category:
Fuel:
Production
Processing
Use
TOTAL
Sugarcane Ethanol
4.17
0.080
3.81E-3
8.10E-4
4.25
Firewood
-
-
-
5.87
5.87
Charcoal from Wood
-
19.5
0.147
3.26
22.9
POFP -
Biomass Pellets
-
0.250
-
1.218
1.469
Ph otoch emical
Kerosene
0.441
0.241
0.038
0.112
0.832
Oxidant
LPG
0.398
0.218
0.118
0.050
0.783
Formation (kg
Electricity
-
-
-
1.46
1.46
NMVOC)
Biogas from Cattle
Dung
-
4.39E-3
-
0.080
0.084
Sugarcane Ethanol
0.166
0.132
0.015
0.118
0.431
Firewood
-
-
-
8.92E-3
8.92E-3
Charcoal from Wood
-
2.49E-3
1.61E-3
6.71E-3
0.011
Biomass Pellets
-
5.40E-3
-
2.60E-4
5.66E-3
FEP -
Freshwater
Eutroph ication
(kg P eq)
Kerosene
8.95E-3
6.80E-4
3.25E-4
5.12E-6
9.96E-3
LPG
8.08E-3
6.10E-4
2.00E-3
-
0.011
Electricity
-
-
-
1.85E-3
1.85E-3
Biogas from Cattle
Dung
-
-
-
-
-
Sugarcane Ethanol
0.029
4.94E-3
1.50E-4
-
0.034
Firewood
-
-
-
0.226
0.226
Charcoal from Wood
-
0.017
0.085
0.107
0.208
Biomass Pellets
-
0.180
-
0.006
0.186
TAP-
Terrestrial
Acidification
(kg S02 eq)
Kerosene
0.191
0.298
0.022
0.029
0.540
LPG
0.173
0.269
0.069
0.014
0.524
Electricity
-
-
-
1.17
1.17
Biogas from Cattle
Dung
-
-
-
5.13E-3
5.13E-3
Sugarcane Ethanol
3.75
0.322
8.61E-3
-
4.08
Firewood
-
-
-
6.62E-11
6.62E-11
Charcoal from Wood
-
1.85E-11
4.76E-8
4.98E-11
4.77E-8
Biomass Pellets
-
2.31E-8
-
1.90E-12
2.31E-8
ODP - Ozone
Kerosene
1.08E-7
1.55E-8
6.71E-9
-2.07E-13
1.30E-7
Depletion (kg
LPG
9.73E-8
1.40E-8
3.68E-8
7.40E-13
1.48E-7
CFC-11 eq)
Electricity
-
-
-
3.08E-8
3.08E-8
Biogas from Cattle
Dung
-
-
-
-
-
Sugarcane Ethanol
2.63E-6
1.15E-7
3.07E-9
1.70E-13
2.74E-6
Firewood
-
-
-
3.32
3.32
Black Carbon
Charcoal from Wood
-
1.68
2.04E-3
0.351
2.03
and Short-
Lived Climate
Pollutants (kg
BCeq)
Biomass Pellets
-
-2.33E-3
-
0.028
0.026
Kerosene
3.60E-3
-6.59E-3
2.95E-4
3.27E-3
5.70E-4
LPG
3.25E-3
-5.95E-3
1.06E-3
1.81E-3
1.70E-4
Electricity
-
-
-
-0.032
-0.032
A-9
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-3. Single Fuel LCIA Results by Life Cycle Stage for Kenya1
1 per GJ delivered heat energy
Life Cycle Stage
TOTAL
Impact
Category:
Fuel:
Feedstock Fuel Cookstove
„ , „ . Distribution TT
Production Processing Use
Biogas from Cattle
Dung
0.040
0.763 -0.014 1.70E-4 5.33E-3
0.040
0.755
Sugarcane Ethanol
1 LC A results presented in this table are calculated as part of this study based on the methodology described in the report body.
A-10
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-4. Single Fuel LCIA Results by Life Cycle Stage for Ghana1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Firewood
-
-
-
228
228
Crop Residue
-
-
-
120
120
Charcoal from Wood
-
521
30.4
160
712
GCCP100 -
Biomass Pellets
-
42.2
-
109
152
Climate
Kerosene
129
1.03
0.921
152
284
Change (kg
LPG
117
0.927
14.9
141
274
C02 eq)
Electricity
-
-
-
259
259
Sugarcane Ethanol
74.9
13.7
4.67
0.953
94.3
Biogas from Cattle
Dung
-
11.0
-
2.42
13.4
Firewood
-
-
-
9.20E+3
9.20E+3
Crop Residue
-
-
-
1.04E+4
1.04E+4
Charcoal from Wood
-
8.77E+3
513
6.40E+3
1.57E+4
Biomass Pellets
-
1241
-
3.11E+3
4.35E+3
IcED - Energy
Kerosene
1.24E+3
4.48E+3
24.6
2.49E+3
8.24E+3
|Demand (MJ)
LPG
1.12E+3
4.02E+3
337
2.20E+3
7.67E+3
Electricity
-
-
-
7.94E+3
7.94E+3
Sugarcane Ethanol
2.42E+3
8.62E+3
78.0
2.08E+3
1.32E+4
Biogas from Cattle
Dung
-
2.22E+3
-
1.99E+3
4.21E+3
Firewood
-
-
-
7.36E-3
7.36E-3
Crop Residue
-
-
-
7.56E-3
7.56E-3
Charcoal from Wood
-
3.07E-3
10.5
7.14E-3
10.6
FDP - Fossil
Depletion (kg
oil eq)
Biomass Pellets
-
21.2
-
2.10E-4
21.2
Kerosene
26.4
107
0.473
59.5
193
LPG
23.8
95.9
7.23
52.5
179
Electricity
-
-
-
150
150
Sugarcane Ethanol
18.7
6.74
1.70
-
27.2
Biogas from Cattle
Dung
-
-
-
-
-
Firewood
-
-
-
8.16E-5
8.16E-5
Crop Residue
-
-
-
8.32E-5
8.32E-5
Charcoal from Wood
-
3.38E-5
0.135
7.84E-5
0.135
Biomass Pellets
-
0.868
-
-5.28E-18
0.868
1 WDP - Water
Kerosene
0.045
6.63E-3
0.027
-
0.078
Depletion (m3)
LPG
0.040
5.99E-3
0.081
-
0.127
Electricity
-
-
-
7.58
7.58
Sugarcane Ethanol
643
0.987
0.020
-
644
Biogas from Cattle
Dung
-
3.29
-
-
3.29
Firewood
-
-
-
15.5
15.5
PMFP -
Particulate
Crop Residue
-
-
-
15.4
15.4
Charcoal from Wood
-
8.35
0.077
1.75
10.2
Biomass Pellets
-
0.070
-
0.092
0.162
A-ll
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-4. Single Fuel LCIA Results by Life Cycle Stage for Ghana1
I per GJ delivered heat energy
Life Cycle Stage
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
TOTAL
Formation (kg
Kerosene
0.078
6.15E-3
2.31E-3
0.017
0.104
PM10 eq)
LPG
0.070
5.55E-3
0.029
9.06E-3
0.114
Electricity
-
-
-
0.310
0.310
Sugarcane Ethanol
4.17
0.080
0.011
8.10E-4
4.26
Biogas from Cattle
Dung
-
-
-
0.187
0.187
Firewood
-
-
-
5.99
5.99
Crop Residue
-
-
-
9.02
9.02
Charcoal from Wood
-
26.0
0.291
3.80
30.1
POFP -
Ph otoch emical
Oxidant
Formation (kg
Biomass Pellets
-
0.289
-
1.218
1.508
Kerosene
1.33
0.027
9.07E-3
0.111
1.48
LPG
1.20
0.024
0.117
0.050
1.39
NMVOC)
Electricity
-
-
-
1.39
1.39
Sugarcane Ethanol
0.162
0.132
0.045
0.118
0.457
Biogas from Cattle
Dung
-
4.39E-3
-
0.080
0.084
Firewood
-
-
-
9.12E-3
9.12E-3
Crop Residue
-
-
-
9.37E-3
9.37E-3
Charcoal from Wood
-
3.81E-3
3.17E-3
8.87E-3
0.016
FEP -
Biomass Pellets
-
5.88E-3
-
2.60E-4
6.14E-3
Freshwater
Kerosene
1.09E-3
3.50E-4
4.64E-5
3.61E-6
1.49E-3
Eutroph ication
LPG
9.90E-4
3.10E-4
2.01E-3
-1.00E-5
3.30E-3
(kgPeq)
Electricity
-
-
-
1.97E-3
1.97E-3
Sugarcane Ethanol
0.029
4.94E-3
4.50E-4
-1.00E-5
0.034
Biogas from Cattle
Dung
-
-
-
-
-
Firewood
-
-
-
0.230
0.230
Crop Residue
-
-
-
0.616
0.616
Charcoal from Wood
-
0.055
0.167
0.113
0.335
TAP-
Biomass Pellets
-
0.200
-
0.006
0.206
Terrestrial
Kerosene
0.183
0.019
7.70E-3
0.029
0.239
Acidification
LPG
0.165
0.017
0.069
0.014
0.265
(kg S02 eq)
Electricity
-
-
-
1.10
1.10
Sugarcane Ethanol
3.75
0.322
0.026
-
4.10
Biogas from Cattle
Dung
-
-
-
5.13E-3
5.13E-3
Firewood
-
-
-
6.77E-11
6.77E-11
Crop Residue
-
-
-
6.96E-11
6.96E-11
Charcoal from Wood
-
2.83E-11
9.40E-8
6.63E-11
9.41E-8
ODP - Ozone
Depletion (kg
CFC-11 eq)
Biomass Pellets
-
6.83E-8
-
1.90E-12
6.83E-8
Kerosene
1.87E-8
1.08E-8
1.42E-9
6.00E-15
3.09E-8
LPG
1.69E-8
9.73E-9
3.76E-8
-6.00E-14
6.42E-8
Electricity
-
-
-
3.61E-7
3.61E-7
Sugarcane Ethanol
2.64E-6
1.15E-7
9.21E-9
5.00E-13
2.77E-6
A-12
-------
Appendix A—Baseline Single Cooking Fuel Results by Life Cycle Stage
Table A-4. Single Fuel LCIA Results by Life Cycle Stage for Ghana1
I per GJ delivered heat energy
Life Cycle Stage
TOTAL
Impact
Category:
Fuel:
Feedstock
Production
Fuel
Processing
Distribution
Cookstove
Use
Biogas from Cattle
Dung
Black Carbon
and Short-
Lived Climate
Pollutants (kg
BCeq)
Firewood
Crop Residue
Charcoal from Wood
Biomass Pellets
Kerosene
LPG
Electricity
Sugarcane Ethanol
Biogas from Cattle
Dung
3.33
3.36
2.05 4.03E-3 0.437
-1.54E-3 - 0.028
9.43E-3 -2.60E-4 -1.79E-4 3.26E-3
8.51E-3 -2.40E-4 1.07E-3 1.81E-3
-0.030
0.763 -0.014 5.20E-4 5.33E-3
0.040
3.33
3.36
2.49
0.026
0.012
0.011
-0.030
0.755
0.040
1 LC A results presented in this table are calculated as part of this study based on the methodology described in the report body.
A-13
-------
Appendix B
Comparison of Results Updates between
Phase I and Phase II Study for India and China
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-l includes a comparison of the Phase I and Phase IILCA results for traditional
fuels in India. Several study assumptions were modified for the second phase of this work based
improved country-specific data, Phase 1 peer review suggestions, and additional stakeholder
input. Notes are provided regarding reasons behind the more dramatic changes in estimated
impact between the two phases of work:
• Baseline forest renewability factor for Phase II estimates that a higher percentage of
forest products are produced using renewable practices, which leads to the
assumption that associated CO2 emissions are carbon neutral, thereby contributing to
the observed change in GCCP impact of firewood.
• CED of hard coal drops in Phase II due to a reduction in the estimated embodied
energy of coal feedstock at the mine that was implemented to align the background
unit process heating value with the country-specific fuel heating value for India
(Singh et al. 2014a).
• WDP impacts for the second phase of work are significantly lower than those in
Phase I due to implementation of the assumption that hydropower turbine water
should not be characterized in estimates of WDP impact because the water is still
available for environmental uses, does not leave the waterway, and is therefore not
depleted.
• The reduction in PMFP, BC, and POFP emissions for kerosene is due to reduced
estimates of use phase emissions as documented in Table 2-1 and SI2.
• The reduction in the charcoal estimated FDP in Phase II is due to the removal of
processing energy associated with the assumption that chunk charcoal (i.e., not
briquettes) are being used.
• FEP of traditional fuels decreased significantly in Phase II due to a change in the
characterization factor used to estimate FEP impact of land applied ash. Only 5%, as
opposed to 100%, of land applied phosphorus is now assumed to make its way to
local water bodies to contribute to FEP. This assumption is in better alignment with
estimates associated with characterization of nutrients contained in land applied
manure (Goedkoop et al. 2008).
• ODP results have dropped significantly in Phase II resulting from the replacement of
HCFC 1211 and Halon 1301 withHCFC-123 and HFC-227ea, respectively.
Table B-l. Comparison on Phase I and Phase II LCA Results, Traditional Fuels, India4
Impact
Category
Value
Description
Hard Coal
Dung
Cake
Crop
Residue
Firewood
Kerosene
Charcoal
GCCP100-
Climate Change
(kg CO2 eq)
Phase II Result
963
263
119
196
181
402
Phase I Result
963
191
132
539
181
572
Percent Change1
0%
28%
-11%
-174%
0%
-42%
B-l
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-l. Comparison on Phase I and Phase IILCA Results, Traditional Fuels, India4
Impact
Category
Value
Description
Hard Coal
Dung
Cake
Crop
Residue
Firewood
Kerosene
Charcoal
Energy Demand
- CED (M.T)
Phase II Result
7.21E+03
1.30E+04
1.01E+04
6.52E+03
3.09E+03
1.09E+04
Phase I Result
1.40E+04
1.30E+04
9.70E+03
7.70E+03
2.60E+03
1.00E+04
Percent Change
-94%
0%
4%
-18%
16%
8%
FDP - Fossil
Phase II Result
172
0.152
7.90E-03
5.90E-03
71
0.011
depletion (kg oil
eq)
Phase I Result
243
0.155
7.60E-03
6.40E-03
65.7
0.117
Percent Change
-41%
-2%
4%
-7%
7%
-944%
WDP - Water
depletion (m3)
Phase II Result
0.397
1.68E-03
8.72E-05
6.54E-05
0.239
1.20E-04
Phase I Result
16.6
1.19
0.058
0.049
36.3
0.629
Percent Change
n.a.2
n.a.2
n.a.2
n.a.2
n.a.2
n.a.2
PMFP -
Particulate
matter formation
(kg PMio eq)
Phase II Result
19.8
24.3
11.4
5.54
0.171
20.5
Phase I Result
19.3
23.6
11.3
4.72
0.308
19.5
Percent Change
2%
3%
1%
15%
-80%
5%
POFP -
Photochemical
oxidant
formation (kg
NMVOC)
Phase II Result
7.87
18.8
8.22
5.38
0.483
10.4
Phase I Result
7.86
18.7
8.75
6.02
1.16
10.5
Percent Change
0%
1%
-7%
-12%
-140%
-1%
FEP -
Freshwater
Phase II Result
2.37E-03
0.188586
9.80E-03
7.36E-03
3.79E-03
0.014
eutrophication
(kg P eq)
Phase I Result
2.10E-03
3.82
0.187
0.157
3.30E-03
0.278
Percent Change
11%
n.a.3
n.a.3
n.a.3
13%
n.a.3
TAP - Terrestrial
acidification (kg
S02 eq)
Phase II Result
1.87
0.736
0.598
0.377
0.292
0.209
Phase I Result
1.87
0.749
0.616
0.4
0.398
0.209
Percent Change
0%
-2%
-3%
-6%
-36%
0%
ODP - Ozone
depletion (kg
CFC-11 eq)
Phase II Result
3.01E-08
1.40E-09
7.28E-11
5.46E-11
6.20E-08
1.03E-10
Phase I Result
8.20E-07
6.20E-08
3.10E-09
2.60E-09
2.40E-06
4.50E-09
Percent Change
-2626%
-4328%
-4161%
-4658%
-3772%
-4257%
Black Carbon
and Short-Lived
Climate
Phase II Result
4.1
5.27
2.48
1.22
0.021
4.58
Phase I Result
3.91
5.01
2.42
1.04
0.045
4.27
Pollutants (kg
BC eq)
Percent Change
4%
5%
2%
14%
-112%
7%
1 Percent change calculated as (Phase II Value-Phase I Value)/Phase II Value.
2 Removal of turbine water changes basis of impact category from water consumption to water depletion.
3 Change in the characterization factor associated with land application of ash residue reduces impact by a factor of 20.
4 LCA results presented in this table are calculated as part of this study based on the methodology described in the report body.
B-2
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-2 includes a comparison of the Phase I and Phase IILCA results for modern
cooking fuels in India. Several study assumptions were modified for the second phase of this
work based improved country-specific data, Phase 1 peer review suggestions, and additional
stakeholder input. Notes are provided regarding reasons behind the more dramatic changes in
estimated impact between the two phases of work:
• The reduction in LPG GCCP and POFP is due to lower estimates of use phase
emissions.
• WDP impacts for the second phase of work are significantly lower than the WDP
impacts in Phase I due to implementation of the assumption that hydropower turbine
water should not be characterized in estimates of WDP impact, because the water is
still available for environmental uses, does not leave the waterway, and is therefore
not depleted.
• The increase in biomass pellet eutrophication potential is attributable to wood ash
(waste disposal) during the pelletization process.
• Increases in impact associated with sugarcane ethanol are generally attributable to
impacts associated with agricultural production of sugarcane. New unit processes
based on the latest research representing Indian sugarcane production have been
developed for Phase II.
• ODP results have dropped significantly in Phase II resulting from the replacement of
HCFC 1211 and Halon 1301 withHCFC-123 and HFC-227ea, respectively.
Table B-2. Comparison of Phase I and Phase II LCA Results, Modern Fuels, India4
Impact
Category
Value
Description
LPG
Natural
Gas
Electricity
Sugarcane
Ethanol
Biogas
Biomass
Pellets
GCCP100-
Climate Change
(kg CO2 eq)
Phase II Result
157
117
457
121
11.4
141
Phase I Result
297
-
415
95.7
10.5
134
Percent Change2
-89%
n.a.1
9%
21%
8%
5%
Energy Demand
- CED (M.T)
Phase II Result
2.61E+03
2.04E+03
5.70E+03
1.33E+04
4.06E+03
3.91E+03
Phase I Result
1.70E+03
-
5.40E+03
6.50E+03
1.80E+03
2.00E+03
Percent Change
35%
n.a.1
5%
51%
56%
49%
FDP - Fossil
depletion (kg oil
eq)
Phase II Result
58.7
48.7
122
31
-
13.7
Phase I Result
44.9
-
91.4
18.3
-
6.25
Percent Change
24%
n.a.1
25%
41%
0%
54%
WDP - Water
depletion (m3)
Phase II Result
0.193
0.039
3.25
643
1.02
0.357
Phase I Result
29.2
-
515
88.6
1.04
35.6
Percent Change
n.a.3
n.a.1
n.a.3
n.a.3
n.a.3
n.a.3
PMFP -
Particulate
matter formation
(kg PMio eq)
Phase II Result
0.136
0.019
1.91
4.38
0.21
0.302
Phase I Result
0.142
-
1.69
0.167
0.077
0.212
Percent Change
-4%
n.a.1
12%
96%
63%
30%
POFP -
Photochemical
oxidant
Phase II Result
0.341
0.046
2.66
0.633
0.114
1.52
Phase I Result
0.687
-
2.01
0.342
0.114
0.237
B-3
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-2. Comparison of Phase I and Phase II LCA Results, Modern Fuels, India4
Impact
Category
Value
Description
LPG
Natural
Gas
Electricity
Sugarcane
Ethanol
Biogas
Biomass
Pellets
formation (kg
NMVOC)
Percent Change
-102%
n.a.1
24%
46%
0%
84%
FEP -
Freshwater
eutrophication
(kg P eq)
Phase II Result
3.37E-03
7.05E-05
3.75E-03
0.0375
0
0.006
Phase I Result
2.50E-03
-
3.40E-03
0.037
0
3.40E-03
Percent Change
26%
n.a.1
9%
1%
0%
38%
TAP - Terrestrial
acidification (kg
SO2 eq)
Phase II Result
0.256
0.027
4.54
4.35
0.106
0.502
Phase I Result
0.316
-
4
0.498
0.106
0.291
Percent Change
-24%
n.a.1
12%
89%
0%
42%
ODP - Ozone
depletion (kg
CFC-11 eq)
Phase II Result
6.56E-08
7.25E-08
4.24E-07
2.82E-06
0.00E+00
5.31E-08
Phase I Result
2.10E-06
0.00E+00
1.40E-06
6.30E-06
0.00E+00
3.20E-07
Percent Change
-3101%
n.a.1
-230%
-123%
0%
-503%
Black Carbon
and Short-Lived
Climate
Pollutants (kg
BC eq)
Phase II Result
0.012
2.10E-03
-0.016
0.757
0.035
0.026
Phase I Result
7.30E-03
_
-0.019
-5.40E-03
6.80E-03
0.02
Percent Change
39%
n.a.1
-15%
101%
81%
24%
1 Not applicable due to the absence of natural gas in Phase I.
2 Percent change calculated as (Phase II Value-Phase I Value)/Phase II Value.
3 Removal of turbine water changes basis of impact category from water consumption to water depletion.
4 LCA results presented in this table are calculated as part of this study based on the methodology described in the report body.
Table B-3 includes a comparison of the Phase I and Phase II LCA results for traditional
fuels in China. Several study assumptions were modified for the second phase of this work based
improved country-specific data, Phase 1 peer review suggestions, and additional stakeholder
input. Notes are provided regarding reasons behind the more dramatic changes in estimated
impact between the two phases of work:
• Baseline forest renewability factor for Phase II estimates that a higher percentage of
forest products are produced using renewable practices, which leads to the
assumption that associated CO2 emissions are carbon neutral, thereby contributing to
the observed change in GCCP impact of firewood.
• CED of coal briquettes drops in Phase II due to a reduction in the estimated embodied
energy of coal feedstock at the mine that was implemented to align the background
unit process heating value with the country specific, fuel heating value for China
(Zhang et al. 2000).
• WDP impacts for the second phase of work are significantly lower than the WDP
impacts in Phase I due to implementation of the assumption that hydropower turbine
water should not be characterized in estimates of WDP impact because the water is
still available for environmental uses, does not leave the waterway, and is therefore
not depleted.
• PMFP and BC impacts of coal feedstock increase significantly due to the inclusion of
PM>2.5 in the Phase II emissions inventory. The value for coal cookstoves is used as
a proxy (Singh et al. 2014a) and is linearly scaled based on the difference in thermal
B-4
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
efficiency between the cookstove from the original source and the stove type in
China.
• The reduction in TAP for coal powder is due to a reduced estimate of use phase
emissions due to the use of the geometric mean as the analysis value in Phase II of the
study as opposed to the arithmetic mean. The difference in the two means is
significant for SO2 emissions due to the high standard deviation of the reported SO2
emissions.
• ODP results have dropped significantly in Phase II resulting from the replacement of
HCFC 1211 and Halon 1301 withHCFC-123 and HFC-227ea, respectively.
Table B-3. Comparison of Phase I and Phase II LCA Results, Traditional Fuels, China4
Impact
Category
Value
Description
Coal Mix
Coal
Powder
Coal
Briquettes
Coal
Honeycomb
Firewood
Crop
Residue
Kerosene
GCCP100 -
Climate
Change (kg
CO2 eq)
Phase II Result
862
1.16E+03
593
527
190
64.1
225
Phase I Result
1.01E+03
1.29E+03
784
695
281
54.7
207
Percent
Change1
-18%
-11%
-32%
-32%
-48%
15%
8%
Energy
Demand - CED
(MJ)
Phase II Result
8.35E+03
1.08E+04
5.37E+03
6.37E+03
7.61E+03
7.45E+03
3.53E+03
Phase I Result
1.10E+04
1.30E+04
8.90E+03
7.60E+03
6.50E+03
7.90E+03
2.90E+03
Percent Change
-32%
-20%
-66%
-19%
15%
-6%
18%
FDP - Fossil
depletion (kg
oil eq)
Phase II Result
195
253
125
149
3.63E-03
0.01
76.9
Phase I Result
179
213
158
134
2.45E-03
0.015
67.7
Percent Change
8%
16%
-26%
11%
32%
-49%
12%
WDP - Water
depletion (m3)
Phase II Result
1.05
1.15
0.986
0.899
4.00E-05
1.20E-04
0.48
Phase I Result
44.5
19.1
76.3
63.7
0.019
0.118
72.3
Percent Change
n.a.2
n.a.2
n.a.2
n.a.2
n.a.2
n.a.2
n.a.2
PMFP -
Particulate
matter
formation (kg
PM10 eq)
Phase II Result
11.2
21.5
0.989
1.08
6.5
10
0.266
Phase I Result
1.81
2.96
0.68
0.631
1.49
3.4
0.232
Percent Change
84%
86%
31%
41%
77%
66%
13%
POFP -
Photochemical
oxidant
formation (kg
NMVOC)
Phase II Result
2.28
3.3
0.7
1.82
2.23
5.52
0.582
Phase I Result
2.33
3.31
1.2
1.5
1.81
2.52
0.425
Percent Change
-2%
0%
-71%
17%
19%
54%
27%
FEP -
Freshwater
Phase II Result
0.102
0.116
0.091
0.084
4.50E-03
0.013
0.013
B-5
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-3. Comparison of Phase I and Phase II LCA Results, Traditional Fuels, China4
Impact
Category
Value
Description
Coal Mix
Coal
Powder
Coal
Briquettes
Coal
Honeycomb
Firewood
Crop
Residue
Kerosene
eutrophication
(kg P eq)
Phase I Result
0.11
0.137
0.089
0.076
0.061
0.38
0.01
Percent Change
-8%
-18%
2%
10%
n.a.3
n.a.3
19%
TAP-
Terrestrial
acidification
(kg S02 eq)
Phase II Result
1.39
1.66
1.2
1.05
0.242
0.367
0.96
Phase I Result
3.72
5.94
1.6
1.42
0.289
0.301
0.867
Percent Change
-167%
-257%
-34%
-35%
-19%
18%
10%
ODP - Ozone
depletion (kg
CFC-11 eq)
Phase II Result
1.17E-07
1.13E-07
1.27E-07
1.15E-07
3.34E-11
9.52E-11
1.85E-07
Phase I Result
6.40E-06
8.40E-07
1.30E-05
1.10E-05
9.90E-10
6.20E-09
3.80E-05
Percent Change
-5369%
-642%
-10154%
-9477%
-2862%
-6412%
-20464%
Black Carbon
and Short-
Li ved Climate
Pollutants (kg
BC eq)
Phase II Result
2.29
4.45
0.105
0.16
1.42
2.2
-0.032
Phase I Result
0.043
0.041
0.047
0.044
0.298
0.693
-0.032
Percent Change
98%
99%
55%
73%
79%
69%
1%
1 Percent change calculated as (Phase II Value-Phase I Value)/Phase II Value.
2 Removal of turbine water changes basis of impact category from water consumption to water depletion.
3 Change in the characterization factor associated with land application of ash residue reduces impact by a factor of 20.
4 LCA results presented in this table are calculated as part of this study based on the methodology described in the report body.
Table B-4 includes a comparison of the Phase I and Phase II LCA results for modern
fuels in China. Several study assumptions were modified for the second phase of this work based
improved country-specific data, Phase 1 peer review suggestions, and additional stakeholder
input. Notes are provided regarding reasons behind the more dramatic changes in estimated
impact between the two phases of work:
• Sugarcane ethanol, biogas, and coal gas were not included in Phase I for China, and
so the percent change has not been calculated.
• WDP impacts for the second phase of work are significantly lower than the WDP
impacts in Phase I due to implementation of the assumption that hydropower turbine
water should not be characterized in estimates of WDP impact, because the water is
still available for environmental uses, does not leave the waterway, and is therefore
not depleted.
• The increase in the biomass pellet eutrophication potential is attributable to wood ash
(waste disposal) during the pelletization process.
• ODP results have dropped significantly in Phase II resulting from the replacement of
HCFC 1211 and Halon 1301 withHCFC-123 and HFC-227ea, respectively.
• The decrease in BC impact score for LPG is associated with crude oil production.
B-6
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
• Coal gas replaced the proposed use of dimethyl ether (DME) in the Phase II study
based on references that demonstrate current use of coal gas (World Bank 2008,
Mainali et al 2012), and the predicted increase in coal gas use predicted by Mainali et
al. (2012). Additionally, the LCI data used for DME in Phase I was adapted from
information initially pertaining to coal gas. Inclusion of coal gas in Phase II prevents
the need for this adaptation.
Table B-4. Comparison of Phase I and Phase II LCA Results, Modern Fuels, China4
Impact
Category
Value
Description
Biomass
Pellets
Electricity
LPG
Natural
Gas
Coal Gas
Sugarcane
Ethanol
Biogas
GCCP100 -
Phase II
Result
140
612
213
154
254
113
11.4
Climate
Change (kg
Phase I
Result
118
496
188
213
-
-
-
CO2 eq)
Percent
Change1
16%
19%
12%
-38%
n.a.2
n.a.2
n.a.2
Energy
Demand -
CED (M.T)
Phase II
Result
3.78E+03
7.22E+03
3.41E+03
2.37E+03
3.69E+03
1.31E+04
4.06E+03
Phase I
Result
2.40E+03
6.10E+03
2.80E+03
2.00E+03
-
-
-
Percent
Change
37%
16%
18%
15%
n.a.2
n.a.2
n.a.2
FDP - Fossil
depletion (kg
oil eq)
Phase II
Result
12.3
118
74.4
55.3
82.3
24.9
-
Phase I
Result
8.12
95.6
64.4
48.6
-
-
-
Percent
Change
34%
19%
13%
12%
n.a.2
n.a.2
n.a.2
Phase II
Result
0.417
4.07
0.461
0.025
0.576
643
1.02
WDP - Water
depletion (m3)
Phase I
Result
49.2
524
57.1
5.77
-
-
-
Percent
Change
n.a.3
n.a.3
n.a.3
n.a.3
n.a.2
n.a.2
n.a.2
PMFP -
Particulate
Phase II
Result
0.311
1.65
0.248
0.048
0.495
4.33
0.21
matter
formation (kg
Phase I
Result
0.215
1.33
0.198
0.057
n.a.1
n.a.1
n.a.1
PM10 eq)
Percent
Change
31%
19%
20%
-19%
n.a.2
n.a.2
n.a.2
POFP -
Photochemical
oxidant
Phase II
Result
1.34
2.31
0.5
0.181
1.31
0.511
0.114
formation (kg
NMVOC)
Phase I
Result
0.26
1.87
0.401
0.226
-
-
-
Percent
Change
81%
19%
20%
-24%
n.a.2
n.a.2
n.a.2
FEP -
Phase II
Result
0.012
0.078
0.012
6.80E-04
0.042
0.039
-
Freshwater
eutrophication
Phase I
Result
0.02
0.063
8.00E-03
6.80E-04
-
-
-
(kg P eq)
Percent
Change
-65%
19%
35%
-1%
n.a.2
n.a.2
n.a.2
B-7
-------
Appendix B—Comparison of Results Updates between
Phase I and Phase II Study for India and China
Table B-4. Comparison of Phase I and Phase II LCA Results, Modern Fuels, China4
Impact
Category
Value
Description
Biomass
Pellets
Electricity
LPG
Natural
Gas
Coal Gas
Sugarcane
Ethanol
Biogas
TAP-
Terrestrial
acidification
(kg S02 eq)
Phase II
Result
0.535
5.27
0.898
0.143
0.803
4.33
0.106
Phase I
Result
0.392
4.27
0.683
0.17
-
-
-
Percent
Change
27%
19%
24%
-19%
n.a.2
n.a.2
n.a.2
ODP - Ozone
depletion (kg
CFC-11 eq)
Phase II
Result
2.54E-08
1.67E-07
1.81E-07
9.74E-07
1.39E-05
2.75E-06
0
Phase I
Result
2.30E-07
2.30E-06
2.90E-05
3.40E-05
-
-
-
Percent
Change
-807%
-1275%
-15898%
-3391%
n.a.2
n.a.2
n.a.2
Black Carbon
and Short-
Li ved Climate
Pollutants (kg
BC eq)
Phase II
Result
0.031
-0.148
-0.031
-2.00E-03
0.038
0.748
0.035
Phase I
Result
0.011
-0.121
-0.018
-2.20E-03
-
-
-
Percent
Change
64%
18%
42%
-7%
n.a.2
n.a.2
n.a.2
1 Percent change calculated as (Phase II Value-Phase I Value)/Phase II Value.
2 Not applicable due to absence of fuel in Phase I work.
3 Removal of turbine water changes basis of impact category from water consumption to water depletion.
4 LCA results presented in this table are calculated as part of this study based on the methodology described in the report body.
B-8
-------
Appendix C—Data Quality
Appendix C
Data Quality
-------
Appendix C—Data Quality
A general introduction to data quality criteria and assessment is presented in Section 1.4
of the main report. Results of the data quality evaluation are catalogued in Table C-l through
Table C-4. Data quality cannot be assessed using the exact same data quality metrics for all types
of data used in this project. The following list of data quality considerations provides specific
interpretations of the generalized Data Quality Rubric presented in Table 1-5 for the main data
sources used in this project.
Additional data quality information is provided in the SI files.
• Cooking Fuel Mix
o Data quality is only assessed for Current and BAU Scenarios. Other scenarios are
included as a form of sensitivity analysis. Actual likelihood of cooking fuel mix
adoption at a future date in line with these scenarios should be assessed based on
the rationale described in the main report and in SI3 for each cooking fuel mix.
o Data Source Reliability - This quality criterion is used to evaluate the
institution/publication and the methods used to estimate the future cooking fuel
mix. Standard rubric criteria and interpretation apply.
o Data Completeness - Assesses how well the information is expected to represent
the national cooking fuel mix.
o Temporal Correlation - The quality of current cooking fuel mix estimates is very
sensitive to the age of the associated data. The temporal correlation of future
cooking fuel mixes is not estimated. While some of the information on which
future cooking fuel mix estimates are made pertain to a specific year (e.g., 2030)
this study does not intend to project the future cooking fuel mix at a given point in
the future. The study rather provides a range of potential cooking fuel mix
estimates as part of the sensitivity analysis.
o Technological Correlation - Standard rubric criteria and interpretation apply. All
information meets the highest quality criterion.
o Geographic Correlation - Geographic correlation is considered on a restricted
scale. Only values of 'High' or 'Medium Low' are assigned. Medium Low
geographic data quality represents data that are "Data from area with slightly
similar production conditions." Given that technology and fuel type are fixed, this
is estimated to be a conservative estimate of data quality that would reflect
differences in crop/wood type or climate that have an impact on stove
performance.
• Stove Technology Mix
o Data Source Reliability - This quality criterion is used to evaluate the institution
and publication and the methods with which they provide information pertaining
to the stove technology mix.
o Data Completeness - Assesses how well the information is expected to represent
the national cookstove technology use.
C-l
-------
Appendix C—Data Quality
o Temporal Correlation - The quality of cookstove technology mix estimates is
very sensitive to the age of the associated data. Standard rubric criteria and
interpretation apply.
o Technological Correlation - All information pertaining to the stove technology
mix is representative of appropriate stove-group.
o Geographic Correlation - Geographic correlation is considered on a restricted
scale. Only values of 'High' or 'Medium Low' are assigned. Medium Low
geographic data quality represents data that are "Data from area with slightly
similar production conditions."
Stove Group LCI Data
o Data Source Reliability - All stove emission data are based on verified
measurements.
o Data Completeness - Data completeness is estimated based on the number of
stoves for which a given stove group LCI is developed.
o High- Stove LCI values based on records for >10 stoves.
¦ Medium High- Stove LCI values based on records for >5 stoves.
¦ Medium- Stove LCI values based on records for 3-4 stoves.
¦ Medium Low - Stove LCI values based on records for 2 stoves.
¦ Low - Stove LCI values based on records for 1 stove.
o Temporal Correlation - Age of stove emissions data is not assumed to be a
critical data quality criterion given that the fuel and stove technology are fixed,
regardless of year. Combustion process for a given stove-fuel combination are not
expected to change over time. Date range of references is noted, but quality
estimate is marked with N/A.
o Technological Correlation - The following stove pollutants are considered
necessary for a complete use-phase emissions inventory: (1) CO, (2) CO2, (3)
CH4, (4) N20, (5) NOx, (6) S02, (7) PM2.5, (8) PM>2.5<10, and (9) NMVOCs.
Technology correlation is assessed based on the number of primary pollutant
emission values reported for each specific stove grouping.
¦ ///^/z-Records for all pollutants available for appropriate stove fuel-technology
combination.
¦ Medium High- Records for six or more pollutants available for appropriate
stove fuel-technology combination. Remaining pollutants use proxy value
based on similar technology scaled to appropriate stove efficiency.
¦ Medium- Records for four or five pollutants available for appropriate stove
fuel-technology combination. Remaining pollutants use proxy value based on
similar technology scaled to appropriate stove efficiency.
¦ Medium Low - Records for three or less pollutants available for appropriate
stove fuel-technology combination. Remaining pollutants use proxy value
based on similar technology scaled to appropriate stove efficiency.
C-2
-------
Appendix C—Data Quality
¦ Low - Less than three emission species available for specific stove group.
Proxy emission value not available for all pollutants (i.e., missing some
pollutant flows).
o Geographic Correlation - Geographic correlation is considered on a restricted
scale. Only values of 'High' or 'Medium Low' are assigned. Medium Low
geographic data quality represents data that are "Data from area with slightly
similar production conditions."
• Electricity Mix
o Data Source Reliability - This quality criterion is used to evaluate the institution,
publication, and methodology used to estimate the electricity fuel and technology
mix.
o Data Completeness - For current electrical mixes standard rubric criteria and
interpretation apply. Data completeness is not evaluated for future electrical
energy mixes.
o Temporal Correlation - Standard rubric criteria and interpretation apply.
o Technological Correlation - Technological correlation is an important quality
criterion for electricity mix information. The sources are largely distinguished on
whether they specify the combustion technology used to produce electricity.
o Geographic Correlation - All electricity mix information is based on information
related to the specified country.
C-3
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
Based primarily on two sources: 1) an unpublished
market report from GACC (a leading organization in
the cookstove sector), 2) a published study with data
based on a national survey
Medium
Results are primarily
based on estimates
adapted from Dalberg et
al. (2013b). Limited
information from on
older publication
(Venkataraman et al.
Cooking
Fuel Mix
Current
India
Data Completeness
Covers all rural and urban households from all
regions of India, but data developed from multiple
sources
Medium
SI3
Temporal
Correlation
Fuel mix estimates primarily from2013, with limited
information from 2007 included, less than 10 years of
difference
Medium
2010) is used to break
out the charcoal value.
Gov of hidia (2014)
provides additional
statistics to determine
the split between LPG
Geographic
Correlation
India
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
produced from natural
gas versus crude oil
Data Source
Reliability
IEA Report; Data verified with many assumptions, or
non-verified but from quality source
Medium
Good source of
information. Future
Data Completeness
Representativeness unknown or incomplete data sets
Low
Cooking
Fuel Mix
BAU 2040
India
Temporal
Correlation
Fuel mix estimate for year 2040
N/A
projection by a
reputable agency. Better
SI3
Geographic
Correlation
India
High
documentation of
methods could lead to a
higher quality estimate
Technological
Correlation
Data from technology, process, or materials being
studied
High
Data Source
Reliability
ERG adaptation of (IEA 2007) 2030 fuel mix
projection
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
documentation of
sources and
Cooking
Fuel Mix
Improved
Biomass
India
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
C-l
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
ERG adaptation of (IEA 2007) 2030 fuel mix
projection
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
Cooking
Fuel Mix
Increased
Electricity
India
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
documentation of
sources and
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
Unqualified estimate based on previous LCA results
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
documentation of
sources and
assumptions
Cooking
Fuel Mix
Diverse
Modem
India
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Fuels
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
Data Source
Reliability
Data largely from a market assessment from GACC
that is no longer publicly available, but the data
source was originally recommended by GACC, a
leading organization in the sector
Medium
Dalberg 2014 is used
Cooking
Fuel Mix
Data Completeness
Covers rural and urban households based on a survey
of a representative sample of the country
Medium
lor the majority ot
cooking fuel mix
assumptions. NBSC
2008 is used to
determine specific crop
Current
China
Temporal
Correlation
Data collection completed in 2013, less than six year
of difference
High
SI3
Geographic
Correlation
Covers six provinces in China to be representative of
the whole country
Medium
types for the crop
residues
Technological
Correlation
Data from technology, process, or materials being
studied
High
Cooking
Fuel Mix
BAU 2030
China
Data Source
Reliability
Mainali et al. 2012; data verified with many
assumptions, or non-verified but from quality source
Medium
SI3
C-2
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Completeness
Sufficient number of sites, but a less adequate period
of time
Medium
Temporal
Correlation
Fuel mix estimate for year 2030
N/A
Geographic
Correlation
China
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
Data Source
Reliability
ERG adaptation of Mainali et al. 2012/2030 fuel mix
N/A
Future cooking fuel
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
Data Completeness
N/A - sensitivity analysis
N/A
Cooking
Fuel Mix
Increased
Electricity
China
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
documentation of
sources and
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
ERG adaptation of Mainali et al. 2012/2030 fuel mix
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
documentation of
sources and
Cooking
Fuel Mix
Advanced
Biomass &
China
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Electricity
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
Unqualified estimate based on previous LCA results
N/A
Future cooking fuel
Cooking
Fuel Mix
Diverse
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
documentation of
Modem
Fuels
China
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
C-3
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
sources and
assumptions
Data Source
Reliability
2009 Kenya Population and Housing Census; data
verified based on measurements
High
Cooking
Fuel Mix
Data Completeness
National Survey; representative data from a sufficient
sample of sites over an adequate period, with records
for all necessary inputs/outputs
High
Current
Kenya
Temporal
Correlation
2009; less than 6 years of difference
Medium High
SI3
Geographic
Correlation
Kenya
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
Data Source
Reliability
ERG; qualified estimate; see SI for assumptions.
Medium Low
Cooking
Fuel Mix
Data Completeness
Representative data from a sufficient sample of sites
over an adequate period of time, with records for all
necessary inputs/outputs
High
BAU 2030
Kenya
Temporal
Correlation
N/A
SI3
Geographic
Correlation
Kenya
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
Data Source
Reliability
ERG adaptation of GLSS2 1993,GLSS3 1995,
GLSS4 2000, GLSS5 2008, GLSS6 2014
N/A
Future cooking fuel
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
Ghana
Transition
Data Completeness
N/A - sensitivity analysis
N/A
Cooking
Fuel Mix
Kenya
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
(for Kenya)
Geographic
Correlation
N/A - sensitivity analysis
N/A
documentation of
sources and
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
C-4
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
ERG adaptation of GLSS2 1993,GLSS3 1995,
GLSS4 2000, GLSS5 2008, GLSS6 2014
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
Cooking
Fuel Mix
Slow
Transition
Kenya
Temporal
Correlation
N/A - sensitivity analysis
N/A
part ol the sensitivity
analysis. See SI3 or
additional detail and
documentation of
sources and
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
Unqualified estimate based on previous LCA results
N/A
Future cooking fuel
Data Completeness
N/A - sensitivity analysis
N/A
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
documentation of
sources and
Cooking
Fuel Mix
Diverse
Modem
Kenya
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Fuels
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
Ghana Living Standards Survey; data verified based
on measurements
High
Data Completeness
National survey; representative data from a sufficient
sample of sites over an adequate period, with records
for all necessary inputs/outputs
High
Cooking
Fuel Mix
Current
Ghana
Temporal
Correlation
2012
High
SI3
Geographic
Correlation
Ghana
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
Cooking
Fuel Mix
BAU 2030
Ghana
Data Source
Reliability
ERG; qualified estimate; see SI for assumptions.
Medium Low
SI3
C-5
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Completeness
Representative data from a sufficient sample of sites
over an adequate period of time, with records for all
necessary inputs/outputs
High
Temporal
Correlation
N/A
Geographic
Correlation
Ghana
High
Technological
Correlation
Data from technology, process, or materials being
studied
High
Data Source
Reliability
ERG scenarios project forward based on historic
trends in fuel mix development using the provided
rationale
N/A
Future cooking fuel
mixes are included as
part of the sensitivity
Data Completeness
N/A - sensitivity analysis
N/A
Cooking
Fuel Mix
Moderated
Growth
Ghana
Temporal
Correlation
N/A - sensitivity analysis
N/A
analysis. See SI3 or
additional detail and
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
documentation of
sources and
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
assumptions
Data Source
Reliability
ERG scenarios project forward based on historic
trends in fuel mix development using the provided
rationale
N/A
Future cooking fuel
mixes are included as
part of the sensitivity
analysis. See SI3 or
additional detail and
Data Completeness
N/A - sensitivity analysis
N/A
Cooking
Fuel Mix
Fast Growth
Ghana
Temporal
Correlation
N/A - sensitivity analysis
N/A
SI3
Geographic
Correlation
N/A - sensitivity analysis
N/A
documentation of
sources and
assumptions
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
Cooking
Fuel Mix
Diverse
Modem
Ghana
Data Source
Reliability
Unqualified estimate based on previous LCA results
N/A
Future cooking fuel
mixes are included as
SI3
Fuels
Data Completeness
N/A - sensitivity analysis
N/A
part of the sensitivity
C-6
-------
Appendix C—Data Quality
Table C-l. Cooking Fuel Mix Data Quality Documentation
Data Type
Fuel Mix
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Temporal
Correlation
N/A - sensitivity analysis
N/A
analysis. See SI3 or
additional detail and
documentation of
sources and
assumptions
Geographic
Correlation
N/A - sensitivity analysis
N/A
Technological
Correlation
Data from technology, process, or materials being
studied
N/A
C-7
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Stove
Group LCI
Dung Cake,
Traditional
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 11 stoves
High
Temporal
Correlation
1999-2014
N/A
Geographic
Correlation
India
High
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Stove
Group LCI
Crop
Residue,
Traditional
Global
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Used as proxy stove use
data for Ghana in the
absence of country
specific data. Crop
residue use is low in
Africa, generally.
SI2
Data Completeness
Represents emissions results of 26 stoves
High
Temporal
Correlation
1999-2014
N/A
Geographic
Correlation
India, China
Medium Low
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Stove
Group LCI
Dung Cake,
Improved
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 4 stoves
Medium
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. NOx,
SOx, ash, PM>2.5<10 andNMVOCs based on
traditional, dung stove (IN)
Medium
High
Stove
Group LCI
Firewood,
three-stone
Global
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 11 stoves
High
Temporal
Correlation
2012-2015
N/A
C-8
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination. N20
emissions based on three-stone tire (IN). S02 and
NMVOC emissions based on firewood, traditional
(IN) scaled to thermal efficiency. NOx based on
firewood, trad, GLO
Medium
Stove
Group LCI
Charcoal,
Improved
Ghana
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 11 stoves
High
Temporal
Correlation
2012
N/A
Geographic
Correlation
Ghana
High
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination. S02,
NOx, N20, PM>2.5<10, and NMVOC based on
charcoal, improved (IN)
Medium
Stove
Group LCI
Crop
Residue,
Traditional
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 12 stoves
High
Temporal
Correlation
1999-2010
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for three or less pollutants available for
appropriate stove fuel-technology combination.
Values for PM>2.5
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Stove
Group LCI
Crop
Residue,
Improved
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 8 stoves
Medium
High
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination,
values for ash, S02, and NOx from improved stove
emissions data for China. Proxy value for PM>2.5<10
from traditional stoves IN, adjusted for thermal
efficiency
Medium
High
Stove
Group LCI
Crop
Residue,
Improved
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 26 stoves
High
Temporal
Correlation
1999-2014
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. N20
from Crop Residue, Imp, IN
Medium
High
Stove
Group LCI
Firewood,
three-stone
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 4 stoves
Medium
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination. Proxy
emissions for NOx, S02, and NMVOC from
firewood; trad stove -IN, adi listed for thermal
Medium
C-10
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
efficiency. Ash value estimated on a typical ash
content of wood. (3%). PM<2.5 from Global three-
stone emission tests, adjusted for thermal efficiency
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 14 stoves
High
Stove
Group LCI
Firewood,
Traditional
India
Temporal
Correlation
1999-2014
N/A
SI2
Geographic
Correlation
India
High
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 16 stoves
High
Stove
Group LCI
Firewood,
Traditional
Global
Temporal
Correlation
2002
N/A
SI2
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination
Medium
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 21 stoves
High
Temporal
Correlation
1999-2012
N/A
Stove
Firewood,
China
Geographic
Correlation
China
High
SI2
Group LCI
Traditional
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for traditional stove use in India
used for S02, N20, PM>2.5<10, adjusted for thermal
efficiency. Ash adjusted for thermal efficiency (544
kg of firewood required for 1 G.T of heat in this UP,
371 kg for the ash proxy unit process)
Medium
High
C-ll
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 16 stoves
High
Temporal
Correlation
1999-2002
N/A
Stove
Firewood,
India
Geographic
Correlation
India
High
SI2
Group LCI
Improved
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination,
values for ash, S02, andNMVOCs from improved
stove emissions data for China, adjusted for thermal
efficiency. Proxy for PM>2.5<10 from trad stove IN,
adjusted for thermal efficiency. PM<2.5 is assumed
equivalent to emissions reported as Total Suspended
Particles
Medium
High
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 35 stoves
High
Stove
Group LCI
Firewood,
Improved
Temporal
Correlation
1999-2012
N/A
China
Geographic
Correlation
China
High
SI2
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination, value
for N20 from improved stove emissions data for
India. Proxy for PM>2.5<10 from trad stove IN,
adjusted for thermal efficiency.
Medium
High
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 28 stoves
High
Stove
Group LCI
Firewood,
Improved
Global
Temporal
Correlation
2002-2012
N/A
SI2
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination.
Medium
C-12
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
values for ash, S02, andNMVOCs from improved
stove emissions data for China. Value for N20 from
improved stove use in India. All adjusted for thermal
efficiency. Proxy for PM>2.5<10 from trad stove IN,
adjusted for thermal efficiency.
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 5 stoves
Medium
High
Temporal
Correlation
2002-2015
N/A
Stove
Group LCI
Charcoal,
Traditional
Geographic
Correlation
Global (no region specified)
Medium Low
Global
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination.
Emissions values for traditional charcoal stove in
India used for S02, N20, PM>2.5<10, and NMVOC,
adjusted for thermal efficiency. Ash waste from
traditional charcoal stoves in India used as proxy for
Ash value, adjusted for stove thermal efficiency.
Proxy value for PM<2.5 taken from Charcoal; Impr;
GH, adjusted for thermal efficiency
Medium
SI2
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 2 stoves
Medium Low
Stove
Group LCI
Charcoal,
Traditional
Temporal
Correlation
1999-2014
N/A
This stove has a very
high thermal efficiency
compared to other
traditional stoves.
India
Geographic
Correlation
India
High
SI2
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for improved charcoal stove use in
GH used for PM<2.5, adjusted for thermal efficiency
Medium
High
Stove
Group LCI
Charcoal,
Improved
Ghana
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 2 stoves
Medium Low
C-13
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Temporal
Correlation
2012
N/A
Geographic
Correlation
Ghana
High
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination.
Emissions values for improved charcoal use in India
used for S02, NOx, N20, PM>2.5<10, andNMVOC,
adjusted for thermal efficiency. Ash waste from
traditional charcoal stoves in India used as proxy for
Ash value, adjusted for thermal efficiency
Medium
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 4 stoves
Medium
Temporal
Correlation
2012
N/A
Stove
Group LCI
Charcoal,
Improved
Kenya
Geographic
Correlation
Kenya
High
SI2
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination.
Emissions values for improved charcoal use in India
used for S02, NOx, N20, PM>2.5<10, andNMVOC,
adjusted for thermal efficiency. Ash waste from
traditional charcoal stoves in India used as proxy for
Ash value, adjusted for thermal efficiency
Medium
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Data Completeness
Represents emissions results of 5 stoves
Medium
High
Stove
Coal
Powder,
Traditional
China
Temporal
Correlation
1999-2000
N/A
All VOCs speciated
SI2
Group LCI
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. N20
and PM>2.5
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Stove
Group LCI
Coal
Powder,
Improved
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
All VOCs speciated
SI2
Data Completeness
Represents emissions results of 3 stoves
Medium
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. Proxy
values for N20 and PM>2.5<10 taken from Heat
from Coal; from Improved Stoves - IN, adjusted for
thermal efficiency
Medium
High
Stove
Group LCI
Coal,
Angethi
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 1 stoves
Low
Temporal
Correlation
2014
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. Proxy
value from PM<2.5 from coal powder, improved, CN,
adi listed for thermal efficiency
Medium
High
Stove
Group LCI
Coal
Briquette,
Improved
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing N20 emissions
SI2
Data Completeness
Represents emissions results of 4 stoves
Medium
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. Proxy
PM>2.5< 10value from honeycomb coal minus
PM<2.5, adi listed for thermal efficiency
Medium
High
Stove
Group LCI
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
All VOCs speciated
SI2
C-15
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Honeycomb
Coal,
Traditional
Data Completeness
Represents emissions results of 8 stoves
Medium
High
Temporal
Correlation
1999-2010
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. PM
<2.5 approximated by measured quantity of total
suspended particles. Proxy value for N20 from Heat
from Coal Trad - IN, adjusted for thermal efficiency
difference Proxy value for PM>2.5<10 from Heat
from Honeycomb Coal; Improved - CN, adjusted for
thermal efficiency difference
Medium
High
Stove
Group LCI
Honeycomb
Coal,
Improved
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 13 stoves
High
Temporal
Correlation
1999-2008
N/A
Geographic
Correlation
China
High
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Stove
Group LCI
Coal Gas
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 2 stoves
Medium Low
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
China
High
Technological
Correlation
All emission profiles represent appropriate
technology and fuel
High
Stove
Group LCI
Kerosene,
Wick
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
SI2
Data Completeness
Represents emissions results of 2 stoves
Medium Low
C-16
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for kerosene stove use in China
used for S02, NMVOCs and NOx, adjusted for
thermal efficiency. Missing PM>2.5<10. Emission
factor for Total Suspended Particles considered to be
PM<2.5
Medium
High
Stove
Group LCI
Kerosene,
Pressure
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
SI2
Data Completeness
Represents emissions results of 2 stoves
Medium Low
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for four or five pollutants available for
appropriate stove fuel-technology combination.
Emissions values for kerosene stove use in China
used for S02, NMVOCs and NOx. Emission factor
for Total Suspended Particles considered to be
PM<2.5
Medium
Stove
Group LCI
Kerosene
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
SI2
Data Completeness
Represents emissions results of 4 stoves
Medium
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for kerosene stove use in India used
for N20
Medium
High
Stove
Group LCI
Kerosene
Global
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
SI2
C-17
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Completeness
Represents emissions results of 8 stoves
Medium
High
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination
Medium
High
Stove
Group LCI
LPG
India
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
SI2
Data Completeness
Represents emissions results of 2 stoves
Medium Low
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for LPG stove use in China used for
NOx, S02, and NMVOCs. Emission factor for Total
Suspended Particles considered to be PM<2.5
Medium
High
Stove
Group LCI
LPG
Global
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10,
All VOCs speciated
SI2
Data Completeness
Represents emissions results of 6 stoves
Medium
High
Temporal
Correlation
1999-2000
N/A
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination
Medium
High
Stove
Group LCI
LPG
China
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Geographic
Correlation
China
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Emissions values for kerosene stove use in India used
for N20
Medium
High
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing PM>2.5<10
Data Completeness
Represents emissions results of 3 stoves
Medium
Stove
Group LCI
Temporal
Correlation
1999-2000
N/A
Use phase emissions
used also for India.
Natural Gas
China
Geographic
Correlation
China
High
SI2
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. Proxy
value for S02 from LPG China, adjusted for thermal
efficiency. Proxy value for N20 from LPG India,
adi listed for thermal efficiency
Medium
High
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Missing values for
NOx, PM>2.5<10,
N20, NMVOC.
Data Completeness
Represents emissions results of 4 stoves
Medium
Stove
Group LCI
Ethanol
India
Temporal
Correlation
2009
N/A
S02 not applicable due
to negligible S content.
SI2
Geographic
Correlation
India
High
Technological
Correlation
All emission profiles represent appropriate
technology and fuel. Only emissions of CO, CH4,
PM2.5, and C02 available in laboratory testing
results
Low
Stove
Group LCI
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Biogas
India
Data Completeness
Represents emissions results of 3 stoves
Medium
SI2
Temporal
Correlation
1999-2014
N/A
C-19
-------
Appendix C—Data Quality
Table C-2. Stove Group LCI Data Quality Documentation
Data Type
Stove
Grouping
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Geographic
Correlation
India
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. Proxy
for PM<2.5 from Biogas - RAF
Medium
High
Stove
Group LCI
Biogas
Africa
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
SI2
Data Completeness
Represents emissions results of 1 stoves
Low
Temporal
Correlation
2011
N/A
Geographic
Correlation
Ghana
High
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination.
Missing Values for N20, CH4, and PM>2.5<10 taken
from 'Biogas, Modem, IN'
Medium
High
Stove
Group LCI
Biomass
Pellets
(wood),
Gasifier
Stove
Global
Data Source
Reliability
Verified measurements from peer reviewed academic
literature
High
Most data from China
SI2
Data Completeness
Represents emissions results of 5 stoves
Medium
High
Temporal
Correlation
2012-2014
N/A
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
Records for six or more pollutants available for
appropriate stove fuel-technology combination. NOx
from firewood, improved GLO adjusted for thermal
efficiency. S02 from firewood, improved CN
adjusted for thermal efficiency. N20, NMVOCs from
firewood, improved IN adjusted for thermal
efficiency
Medium
High
C-20
-------
Appendix C—Data Quality
Table C-3. Electricity Mix Data Quality Documentation
Data Type
Scenario
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
IEA data (IEA 2013a-d); Data verified based on
measurements
High
Generation technology
is not specified.
Regionally specific or
regionally adapted unit
processes from
ecoinvent used to
represent combustion
technology
Electricity
Mix
India,
China,
Kenya,
Ghana
Data Completeness
Representative data from a sufficient sample of sites
over an adequate period, with records for all
necessary inputs/outputs
High
Current Mix
Temporal
Correlation
2013
High
SI4
Geographic
Correlation
Specific to each country
High
Technological
Correlation
Data from a different technology using the same
process and/or materials
Medium High
Data Source
Reliability
The Energy and Resources Institute, India (TERI
2006); Economic modeling approach; Data verified
based on some assumptions and/or standard science
and engineering calculations
Medium High
Generation technology
is specified. LCI of
regionally specific or
Electricity
Mix
Future Mix -
TERIBAU,
Hybrid,
Efficiency
Data Completeness
N/A
N/A
regionally adapted unit
processes from
ecoinvent were adjusted
to reflect the efficiency
and emissions of
reported combustion
technologies
India
Temporal
Correlation
2021 and 2031
High
SI4
Geographic
Correlation
India
High
Technological
Correlation
Specifies combustion technology; Data from
technology, process, or materials being studied
High
Data Source
Reliability
IEA Report (IEA 2010); Data verified based on some
assumptions and/or standard science and engineering
calculations
Medium High
Generation technology
is specified. LCI of
regionally specific or
regionally adapted unit
processes from
ecoinvent were adjusted
to reflect the efficiency
and emissions of
Electricity
Mix
IEA 2050,
IEA Blue
India
Data Completeness
N/A
N/A
SI4
Map
Temporal
Correlation
2050
High
Geographic
Correlation
India
High
reported combustion
technologies
C-21
-------
Appendix C—Data Quality
Table C-3. Electricity Mix Data Quality Documentation
Data Type
Scenario
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Technological
Correlation
Specifies combustion technology; Data from
technology, process, or materials being studied
High
Electricity
Mix
Low Carbon
2050
India
Data Source
Reliability
Gambhir et al. 2012; Economic modeling approach;
data verified based on some assumptions and/or
standard science and engineering calculations
Medium High
Description of modeling
method from Gambhir
et al. 2012: "an
integrated assessment
model combining the
energy-technology
TIMES model with a
climate module to
integrate economic
activity with energy
usage"
SI4
Data Completeness
N/A
N/A
Temporal
Correlation
2050
High
Geographic
Correlation
India
High
Technological
Correlation
Data from a different technology using the same
process and/or materials
Medium High
Electricity
Mix
EIA2030
India
Data Source
Reliability
EIA Report; Data verified based on some assumptions
and/or standard science and engineering calculations
Medium High
Generation technology
is not specified.
Regionally specific or
regionally adapted unit
processes from
ecoinvent used to
represent combustion
technology
SI4
Data Completeness
N/A
N/A
Temporal
Correlation
2030
High
Geographic
Correlation
India
High
Technological
Correlation
Generation technology is not specified; Data from a
different technology using the same process and/or
materials
Medium High
Electricity
Mix
BCG Slow
Shift - 2030,
BCG Base -
2030, BCG
Clean - 2030
China
Data Source
Reliability
Michael et al. 2013; Data verified with many
assumptions, or non-verified but from quality source
Medium
Generation technology
is not specified.
Regionally specific or
regionally adapted unit
processes from
ecoinvent used to
represent combustion
technology
SI4
Data Completeness
N/A
N/A
Temporal
Correlation
2030
High
Geographic
Correlation
China
High
C-22
-------
Appendix C—Data Quality
Table C-3. Electricity Mix Data Quality Documentation
Data Type
Scenario
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Technological
Correlation
Generation technology is not specified; Data from a
different technology using the same process and/or
materials
Medium High
Data Source
Reliability
Zhou et al. 2011; economic modeling considering
technology specific factors such as saturation,
efficiency, or usage; data verified based on some
assumptions and/or standard science and engineering
calculations
Medium High
Generation technology
is specified. LCI of
regionally specific or
Electricity
Mix
LBNL
Scenarios
Data Completeness
N/A
N/A
regionally adapted unit
processes from
ecoinvent were adjusted
to reflect the efficiency
and emissions of
reported combustion
China
Temporal
Correlation
2030-2050
High
SI4
Geographic
Correlation
China
High
Technological
Correlation
Specifies combustion technology; Data from
technology, process, or materials being studied
High
technologies
Data Source
Reliability
IEA Report; Data verified based on some assumptions
and/or standard science and engineering calculations
Medium High
Generation technology
is specified. LCI of
regionally specific or
regionally adapted unit
IEA2050
Data Completeness
N/A
N/A
Electricity
Mix
Baseline,
IEA Blue
China
Temporal
Correlation
2050
High
processes from
ecoinvent were adjusted
SI4
Map
Geographic
Correlation
China
High
to reflect the efficiency
and emissions of
Technological
Correlation
Specifies combustion technology; data from
technology, process, or materials being studied
High
reported combustion
technologies
Electricity
Mix
Republic of
Kenya 2031
Grid
Scenarios
Kenya
Data Source
Reliability
ROK 2011; study uses least cost method of
technology selection; data verified based on some
assumptions and/or standard science and engineering
calculations
Medium High
Generation technology
is not specified.
Regionally specific or
regionally adapted unit
processes from
SI4
Data Completeness
N/A
N/A
Temporal
Correlation
2011-2031
High
ecoinvent used to
C-23
-------
Appendix C—Data Quality
Table C-3. Electricity Mix Data Quality Documentation
Data Type
Scenario
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Geographic
Correlation
Kenya
High
represent combustion
technology
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
Data Source
Reliability
Castellano et al. 2015; demand driven estimates of
future energy demand; data verified based on some
assumptions and/or standard science and engineering
calculations
Medium High
Data Completeness
N/A
N/A
Use empirical data
based on GDP growth
from approximately 20
Electricity
Mix
McKinsey
2040
Kenya
Temporal
Correlation
2040
High
SI4
Geographic
Correlation
West Africa
Medium Low
countries
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
Data Source
Reliability
Unqualified estimate based on previous LCA results
Low
Generation technology
Data Completeness
N/A
N/A
is not specified.
Electricity
Low Carbon
Kenya
Temporal
Correlation
Not associated with a specific time period
N/A
Regionally specific or
regionally adapted unit
SI4
Mix
Geographic
Correlation
Kenya; relies on renewable resources that other
references indicate are available in Kenya
High
processes from
ecoinvent used to
represent combustion
technology
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
Electricity
Mix
Ghana EC
Ghana
Data Source
Reliability
GEC 2006; National Energy Plan; data verified with
many assumptions, or non-verified but from quality
source
Medium
Generation technology
is not specified.
Regionally specific or
regionally adapted unit
SI4
Scenarios
Data Completeness
N/A
N/A
Temporal
Correlation
2020
High
processes from
ecoinvent used to
C-24
-------
Appendix C—Data Quality
Table C-3. Electricity Mix Data Quality Documentation
Data Type
Scenario
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Geographic
Correlation
Ghana
High
represent combustion
technology
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
Data Source
Reliability
Qualified estimates based on 2030 renewable energy
capacity estimates from IRENA 2013
Medium Low
Generation technology
is not specified.
Data Completeness
N/A
N/A
Electricity
Low Carbon
Ghana
Temporal
Correlation
Not associated with a specific time period
N/A
Regionally specific or
regionally adapted unit
SI4
Mix
Geographic
Correlation
Ghana
High
processes from
ecoinvent used to
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
represent combustion
technology
Data Source
Reliability
Unqualified estimate based on previous LCA results
Low
Generation technology
is not specified.
Data Completeness
N/A
N/A
Electricity
Low Carbon
Ghana
Temporal
Correlation
Not associated with a specific time period
N/A
Regionally specific or
regionally adapted unit
SI4
Mix
Geographic
Correlation
Ghana; relies on energy resources that other
references indicate are available in Kenya
High
processes from
ecoinvent used to
represent combustion
technology
Technological
Correlation
Generation technology is not specified; data from a
different technology using the same process and/or
materials
Medium High
C-25
-------
Appendix C—Data Quality
Table C-4. LCI Unit Process Data Quality Documentation
Data Type
Unit
Process
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
Peer reviewed literature; verified measurements
High
LCI Unit
Process
Data
Current,
Average
Kiln
Data Completeness
LCI data based on two publications. Missing estimate
of S02 emissions
Medium
A single quantity
estimate is available for
each flow value
Ghana
Temporal
Correlation
Main emissions data published in 2011, remainder
from 2003
Medium,
Medium Low
SI6
Geographic
Correlation
Ghana, Africa
High
Technological
Correlation
Earthen mound kiln
High
Data Source
Reliability
Data verified with many assumptions, or non-verified
but from quality source
Medium
Data Completeness
Smaller number of sites and shorter periods or
incomplete data from an adequate number of sites or
periods
Medium Low
LCI Unit
Process
Crop
India,
China,
Ghana
Temporal
Correlation
Less than 15 years of difference
Medium Low
Please see SI5 for
specifics regarding data
SI5
Data
Residue
Geographic
Correlation
Input values specific to nation of interest when
possible; Emission values calculated on the basis on
nation specific inputs; average data from larger area or
specific data from a close area
Medium
High
sources, assumptions
and aggregation
Technological
Correlation
Not specific agricultural production method
represented. Intended to cover a wide range of
production practices
N/A
C-26
-------
Appendix C—Data Quality
Table C-4. LCI Unit Process Data Quality Documentation
Data Type
Unit
Process
Name
Country
Data Quality
Criteria
Qualitative Data Quality Discussion
Quality
Estimate
Additional Note on
Quality
SI
File
Data Source
Reliability
LCI values based on: peer reviewed literature; verified
measurements or data verified based on some
assumptions and/or standard science and engineering
calculations
Medium
High
LCI Unit
Process
Biogas &
Bioslurry
Global
Data Completeness
Most flow values based on the average of 3 or more
literature sources with the full range of reported values
being reflected in the uncertainty analysis; smaller
number of sites, but an adequate period of time
Medium
High
Please see SI7 for
specifics regarding data
sources, assumptions
and aggregation
SI7
Data
Land
Application
Temporal
Correlation
For the given biogas production and application
methods LCI inputs/outputs are not expected to be time
sensitive
N/A
Geographic
Correlation
Global (no region specified)
Medium Low
Technological
Correlation
All emission profiles represent appropriate material
High
C-27
-------
*>EPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGES,FEES PAID
EPA
PERMIT NO. G-35
Office of Research and
Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |