Transcript
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
Final Colorado
Greenhouse Gas Inventory and
Reference Case Projections 1990-2020
Center for Climate Strategies
October 2007
Principal Authors: Randy Strait, Steve Roe, Alison Bailie, Holly Lindquist, Alison Jamison, Ezra
Hausman, Alice Napoleon
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
ii Center for Climate Strategies
www.climatestrategies.us
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
iii Center for Climate Strategies
www.climatestrategies.us
Executive Summary
This report presents a summary of Colorado’s anthropogenic greenhouse gas (GHG) emissions
and sinks (carbon storage) from 1990 to 2020. The Center for Climate Strategies (CCS) prepared
a preliminary draft GHG emissions inventory and reference case projection for the Colorado
Department of Public Health and Environment (CDPHE) through an effort of the Western
Regional Air Partnership (WRAP).
1
The preliminary draft report was provided to the Climate
Action Panel (CAP) (and its Policy Work Groups [PWGs]) of the Colorado Climate Project to
assist the CAP in understanding past, current, and possible future GHG emissions in Colorado,
and thereby inform the policy option development process. The CAP and the PWGs provided
comments for improving the reference case projections. This report documents the revised
inventory and reference case projections incorporating comments as approved by the CAP.
2
Colorado’s anthropogenic GHG emissions and anthropogenic/natural sinks (carbon storage)
were estimated for the period from 1990 to 2020. Historical GHG emissions estimates (1990
through 2005)
3
were developed using a set of generally accepted principles and guidelines for
state GHG emissions inventories, relying to the extent possible on Colorado-specific data and
inputs. The reference case projections (2006–2020) are based on a compilation of various
existing Colorado and regional projections of electricity generation, fuel use, and other GHG-
emitting activities, along with a set of simple, transparent assumptions described in Appendixes
A through I of this report.
Table ES-1 provides a summary of historical (1990 to 2005) and reference case projection (2010
and 2020) GHG emissions for Colorado. In 2005, on a gross emissions consumption basis (i.e.,
excluding carbon sinks), Colorado accounted for approximately 116 million metric tons (MMt)
of CO
2
e emissions, an amount equal to 1.6% of total United States (US) gross GHG emissions.
On a net emissions basis (i.e., including carbon sinks), Colorado accounted for approximately 89
MMtCO
2
e of emissions in 2005, an amount equal to 1.4% of total US net GHG emissions.
4
Colorado’s GHG emissions are rising more quickly than those of the nation as a whole.
5
1
Draft Colorado Greenhouse Gas Inventory and Reference Case Projections, 1990–2020, prepared by the Center
for Climate Strategies for the Colorado Department of Public Health and Environment (CDPHE) through an effort
of the Western Regional Air Partnership, January 2007.
2
Final Colorado Greenhouse Gas Inventory and Reference Case Projections, 1990–2020, prepared by the Center
for Climate Strategies for the Climate Action Panel of the Colorado Climate Project, October 2007.
3
The last year of available historical data varies by sector; ranging from 2000 to 2005.
4
National emissions from Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2005, April 2007, US EPA
#430-R-07-002, (http://www.epa.gov/climatechange/emissions/usinventoryreport.html).
5
Gross emissions estimates only include those sources with positive emissions. Carbon sequestration in soils and
vegetation is included in net emissions estimates. All emissions reported in this section for Colorado reflect
consumption-based accounting (including emissions from electricity imports). On a national basis, little difference
exists between production-based and consumption-based accounting for GHG emissions because net electricity
imports are less than 1% of national electricity generation.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
iv Center for Climate Strategies
www.climatestrategies.us
Table ES-1. Colorado Historical and Reference Case GHG Emissions, by Sector
a
(Million Metric Tons CO
2
e) 1990 2000 2005 2010 2020 Explanatory Notes for Projections
Energy 75.4 96.0 102.2 114.1 129.1
Electricity Production 31.6 38.7 39.8 45.3 50.0
Coal 30.9 35.1 34.9 40.0 44.2 See electric sector assumptions
Natural Gas 0.71 3.5 4.9 5.2 5.8 in Appendix A
Oil 0.02 0.08 0.02 0.02 0.02
Wood 0.00 0.00 0.00 0.00 0.01
Net Imported Electricity 1.0 2.2 3.1 2.9 2.6
Electricity Consumption Based 32.7 40.9 42.9 48.2 52.6
Residential/Commercial/Industrial (RCI)
Fuel Use
16.3 20.2 21.2 23.6 27.9
Coal 1.6 1.0 1.2 1.3 1.5 Based on US DOE regional projections
Natural Gas 11.8 15.4 16.5 18.8 23.7 Based on US DOE regional projections
Oil 2.8 3.7 3.5 4.1 5.2 Based on US DOE regional projections
Wood (CH
4
and N
2
O) 0.06 0.07 0.04 0.05 0.05 Based on US DOE regional projections
Avoided emissions from recent building
code and demand-side management
(DSM) initiatives
0.00 0.00 0.00 -0.64 -2.5
Based on analysis of Colorado House Bill (HB) 07-1037
for avoided electricity and natural gas, HB 07-1146 for
avoided natural gas, and Xcel settlement electric DSM
Transportation 19.0 25.5 28.0 30.6 36.2
Motor Gasoline 13.3 17.4 18.1 19.2 22.1 Based on US DOE regional projections
Diesel 2.9 4.8 6.5 7.7 9.8 Based on US DOE regional projections
Natural Gas, LPG, other 0.19 0.22 0.22 0.28 0.39 Based on US DOE regional projections
Jet Fuel and Aviation Gasoline 2.5 3.1 3.2 3.4 3.9 Based on US DOE regional projections
Fossil Fuel Industry 7.5 9.3 10.1 11.8 12.3
Natural Gas Industry 3.1 4.8 5.0 6.5 7.3
Increase based on current trend to 2009, then US DOE to
2020
Oil Industry 0.22 0.15 0.16 0.18 0.20
Increase based on current trend to 2009, then US DOE to
2020
Coal Mining (Methane) 4.2 4.3 4.9 5.1 4.8 Assumes no change after 2003
Industrial Processes 0.76 2.1 2.9 3.8 5.9
Cement Manufacture (CO
2
) 0.32 0.56 0.52 0.55 0.62
Based on state's Nonmetallic Minerals employment
projections (2004-2014)
Lime Manufacture (CO
2
) 0.01 0.01 0.01 0.01 0.01 Ditto
Limestone & Dolomite Use (CO
2
) 0.00 0.03 0.04 0.04 0.04 Ditto
Soda Ash (CO
2
) 0.04 0.04 0.04 0.04 0.05 Based on 2004 and 2009 projections for US production
ODS Substitutes (HFC, PFC, and SF
6
) 0.004 1.2 2.1 3.0 5.1 Based on national projections (US State Dept.)
Semiconductor Manufacturing (HFC,
PFC, and SF
6
)
0.05 0.14 0.08 0.06 0.03 Based on national projections (US EPA)
Electric Power T & D (SF
6
) 0.35 0.20 0.19 0.14 0.08 Based on national projections (US EPA)
Waste Management 1.2 1.9 2.1 2.5 3.5
Solid Waste Management 0.79 1.3 1.5 1.8 2.7 Projections primarily based on population
Wastewater Management 0.39 0.57 0.59 0.66 0.84 Projections based on population
Agriculture (Ag) 8.7 9.6 8.9 8.9 9.1
Enteric Fermentation 3.0 3.2 3.2 3.2 3.2
Projections held constant at 2003 levels except for dairy
cattle (see Appendix F)
Manure Management 0.83 1.2 1.2 1.2 1.3 Ditto
Ag. Soils and Residue Burning 4.9 5.2 4.5 4.5 4.5 Projections held constant at 2005 levels
Total Gross Emissions 86.1 109.6 116.1 129.3 147.5
increase relative to 1990 27% 35% 50% 71%
Forestry and Land Use -24.7 -24.7 -24.7 -24.7 -24.7
Historical and projected emissions held constant at 2004
levels.
Agricultural Soils -2.0 -2.0 -2.0 -2.0 -2.0
Historical and projected emissions held constant at 1997
levels.
Net Emissions (including sinks) 59.4 82.9 89.4 102.6 120.8
a
Totals
may not equal exact sum of subtotals shown in this table due to independent rounding.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
v Center for Climate Strategies
www.climatestrategies.us
From 1990 to 2005, Colorado’s gross GHG emissions were up 35% while national gross
emissions rose by 16% during this period. Much of Colorado’s emissions growth can be
attributed to its population growth. From 1990 to 2005, Colorado’s population grew by 43% as
compared with a national population growth of 19%.
Figure ES-1 illustrates the state’s emissions per capita and per unit of economic output.
Colorado’s per capita emission rate is slightly more than the national average of 25
MtCO
2
e/year. Between 1990 and 2005, per capita emissions in Colorado and national per capita
emissions have changed relatively little. Economic growth exceeded emissions growth in
Colorado throughout the 1990–2005 period. From 1990 to 2005, emissions per unit of gross
product dropped by 40% nationally, and by 54% in Colorado.
6
Electricity use and transportation are the state’s principal GHG emissions sources. Together, the
combustion of fossil fuels for electricity generation and in the transportation sector accounted for
61% of Colorado’s gross GHG emissions in 2005. The remaining use of fossil fuels—natural
gas, oil products, and coal—in the residential, commercial, and industrial (RCI) sectors, plus the
emissions from fossil fuel production, constituted another 27% of total state emissions in 2005.
As illustrated in Figure ES-2 and shown numerically in Table ES-1, under the reference case
projections, Colorado’s gross GHG emissions continue to grow, and are projected to climb to
148 MMtCO
2
e by 2020, reaching 71% above 1990 levels. Overall, the average annual projected
rate of emissions growth in Colorado is 1.6% per year from 2005 to 2020. As shown in Figure
ES-3, demand for electricity is projected to be the largest contributor to future emissions growth
accounting for about 36% of total gross GHG emissions in 2020, followed by emissions
associated with transportation (25%), RCI fossil fuel use (19%), and fossil fuel production (8%)
Some data uncertainties exist in this inventory, and particularly in the reference case projections.
Key tasks for future refinement of the inventory and projections include review and revision of
key drivers (such as electricity, fossil fuel production, and transportation fuel use growth rates)
that will be major determinants of Colorado’s future GHG emissions. These growth rates are
driven by uncertain economic, demographic, and land use trends (including growth patterns and
transportation system impacts), all of which deserve closer review and discussion.
Perhaps the variable with the most important implications for GHG emissions is the type and
number of power plants that will be built in Colorado between now and 2020. The assumptions
on VMT and air travel growth also have large impacts on projected GHG emissions growth in
the state. Finally, uncertainty remains on estimates for historic and projected GHG sinks from
forestry, which can greatly affect the net GHG emissions attributed to Colorado.
6
Based on gross domestic product by state (millions of current dollars), available from the US Bureau of Economic
Analysis, http://www.bea.gov/regional/gsp/. The national emissions used for these comparisons are based on 2005
emissions, http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
vi Center for Climate Strategies
www.climatestrategies.us
Figure ES-1. Historical Colorado and US Gross GHG Emissions, Per Capita and
Per Unit Gross Product
0
5
10
15
20
25
30
1990 1995 2000 2005
US GHG/Capita
(tCO2e)
CO GHG/Capita
(tCO2e)
US GHG/$
(100gCO2e)
CO GHG/$
(100gCO2e)
Figure ES-2. Colorado Gross GHG Emissions by Sector, 1990-2020: Historical and
Projected
0
20
40
60
80
100
120
140
160
1990 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Electricity (Consumption Based) Fossil Fuel Industry
RCI Fuel Use * Transportation Gasoline Use
Transportation Diesel Use Jet Fuel/Other Transportation
Agriculture ODS Substitutes
Other Ind. Process Waste Management
* RCI = direct fuel use in residential, commercial, and industrial sectors; ODS Substitutes = ozone depleting
substances substitutes. Other Industrial Processes include process-related GHG emissions from cement and lime
manufacturing; semiconductor manufacture; soda ash, limestone, and dolomite use; and electricity transmission and
distribution systems.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
vii Center for Climate Strategies
www.climatestrategies.us
Figure ES-3. Sector Contributions to Gross Emissions Growth in Colorado, 1990-2020:
Reference Case Projections (MMtCO
2
e Basis)
-2.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0
Electricity (Consumption Based)
RCI Fuel Use
Fossil Fuel Industry
Transportation
ODS Substitutes (HFCs)
Other Ind. Process
Agriculture
Waste Management
MMtCO2e
2005 - 2020
1990 - 2005
RCI = direct fuel use in residential, commercial, and industrial sectors; ODS Substitutes = ozone depleting
substances substitutes; HFC = hydrofluorocarbons.
Emissions of aerosols, particularly “black carbon” (BC) from fossil fuel combustion, could have
significant climate impacts through their effects on radiative forcing.
7
Estimates of these aerosol
emissions on a CO
2
e basis were developed for Colorado based on 2002 and 2018 data from the
WRAP. The results were a total of 6.75 MMtCO
2
e, which is the mid-point of a range of estimated
emissions (4.3–9.2 MMtCO
2
e) in 2002. Based on an assessment of the primary contributors, it is
estimated that BC emissions will decrease substantially by 2018 after new engine and fuel
standards take effect in the onroad and nonroad diesel engine sectors (decrease of about 4.0
MMtCO
2
e). These estimates are not incorporated into the totals shown in Table 2-1 because a
global warming potential for BC has not yet been assigned by the Intergovernmental Panel on
Climate Change (IPCC). By including BC emission estimates in the inventory, however,
additional opportunities for reducing climate impacts can be identified as the scientific knowledge
related to BC emissions improves.
The following identifies the revisions that the CAP made to the inventory and reference case
projections thus explaining the differences between the information presented in this report and
the preliminary information presented in the January 2007 report:
- Energy Supply: Lowered emissions to account for changes in reference case assumptions
associated with Colorado’s Renewable Portfolio Standard (RPS), which was amended
7
Changes in the atmospheric concentrations of GHGs can alter the balance of energy transfers between the
atmosphere, space, land, and the oceans. A gauge of these changes is called radiative forcing, which is a simple
measure of changes in the energy available to the Earth-atmosphere system (IPCC, 1996). Holding everything else
constant, increases in GHG concentrations in the atmosphere will produce positive radiative forcing (i.e., a net
increase in the absorption of energy by the Earth), http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.htm.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
viii Center for Climate Strategies
www.climatestrategies.us
upward in 2007 by the state legislature’s passage of House Bill (HB) 07-1281
(Renewable Energy Standards):
• Investor-Owned Utilities (IOUs) to provide 20% renewable energy by 2020
• Non-IOUs (e.g., municipal utilities and rural electric cooperatives) to provide 10%
renewable energy by 2020
• Incentives for in-state generation, community-based projects, and solar energy
- RCI: Reduced energy consumption in the reference case projections associated with the
passage of HB 07-1146 (Energy Conservation Building Codes) in 2007. This bill requires
local governments who have building codes to adopt energy efficiency codes for certain
buildings.
8
Reduction in emissions is accounted for under the RPS adjustment to avoid
double counting of emission reductions.
- RCI: Reduced energy consumption in the reference case projections associated with the
passage of HB 07-1037 (legislation recently passed requiring that public electric and gas
utilities implement demand-side management programs)
9
and Xcel’s demand side
management commitments under a recent legal settlement, both of which have the effect
of limiting demand growth relative to what it would have been in the absence of these
factors.
10
- Waste Management: Revisions to municipal solid waste (MSW) to reflect revisions the
US Environmental Protection Agency made to the methods for calculating emissions in
US EPA’s State Greenhouse Gas Inventory Tool (SGIT; i.e., change was from use of
regression equations to LANDGEM model equation):
• 1990 emissions decrease from 1.6 to 0.8 MMtCO
2
e
• 2020 emissions decrease from 5.7 to 2.7 MMtCO
2
e
- Forestry: Removed forest soil organic carbon emissions sink as recommended by the
United States Forest Service (USFS). Relative to the January 2007 report, this change
removed 7.1 MMtCO
2
e of emissions from the forest sink pool for 1990 through 2020.
8
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1146.
9
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1037.
10
Comprehensive Settlement Agreement, docket 04A-214E, 04A-215E, and 04A-216E, issued December 3, 2004,
available at http://www.xcelenergy.com/docs/corpcomm/SettlementAgreementFinalDraftclean20041203.pdf.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
ix Center for Climate Strategies
www.climatestrategies.us
Table of Contents
Executive Summary....................................................................................................................... iii
Acronyms and Key Terms .............................................................................................................. x
Acknowledgements...................................................................................................................... xiv
Summary of Findings...................................................................................................................... 1
Introduction................................................................................................................................. 1
Colorado Greenhouse Gas Emissions: Sources and Trends ........................................................... 2
Historical Emissions ................................................................................................................... 4
Overview................................................................................................................................. 4
A Closer Look at the Two Major Sources: Electricity and Transportation............................ 6
Reference Case Projections......................................................................................................... 7
CAP Revisions............................................................................................................................ 9
Key Uncertainties and Next Steps ............................................................................................ 10
Approach................................................................................................................................... 10
General Methodology ........................................................................................................... 10
General Principles and Guidelines........................................................................................ 11
Appendix A. Electricity Use and Supply................................................................................... A-1
Appendix B. Residential, Commercial, and Industrial (RCI) Fuel Combustion ....................... B-1
Appendix C. Transportation Energy Use................................................................................... C-1
Appendix D. Industrial Processes.............................................................................................. D-1
Appendix E. Fossil Fuel Industries.............................................................................................E-1
Appendix F. Agriculture .............................................................................................................F-1
Appendix G. Waste Management .............................................................................................. G-1
Appendix H. Forestry................................................................................................................. H-1
Appendix I. Inventory and Forecast for Black Carbon................................................................I-1
Appendix J. Greenhouse Gases and Global Warming Potential Values: Excerpts from the
Inventory of U.S. Greenhouse Emissions and Sinks: 1990-2000........................... J-1
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
x Center for Climate Strategies
www.climatestrategies.us
Acronyms and Key Terms
AEO – Annual Energy Outlook, EIA
Ag – Agriculture
bbls – Barrels
BC – Black Carbon*
Bcf – Billion cubic feet
BLM – United States Bureau of Land Management
BOD – Biochemical Oxygen Demand
BTU – British thermal unit
C – Carbon*
CaCO
3
– Calcium Carbonate
CAP – Climate Action Panel
CBM – Coal Bed Methane
CCS – Center for Climate Strategies
CDOT – Colorado Department of Transportation
CDPHE – Colorado Department of Public Health and Environment
CFCs – Chlorofluorocarbons*
CH
4
– Methane*
CO – Carbon monoxide*
CO
2
– Carbon Dioxide*
CO
2
e – Carbon Dioxide equivalent*
CRP – Federal Conservation Reserve Program
DRCOG – Denver Regional Council of Governments
DSM – Demand Side Management
EC – Elemental Carbon*
eGRID – US EPA’s Emissions & Generation Resource Integrated Database
EGU – Electricity Generating Unit
EIA – US DOE Energy Information Administration
EIIP – Emissions Inventory Improvement Program
Eq. – Equivalent
FIA – Forest Inventory and Analysis
Gg – Gigagram
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
xi Center for Climate Strategies
www.climatestrategies.us
GHG – Greenhouse Gases*
GWh – Gigawatt-hour
GWP – Global Warming Potential*
HFCs – Hydrofluorocarbons*
IPCC – Intergovernmental Panel on Climate Change*
IOU – Investor-Owned Utilities
kWh – Kilowatt-hour
LF – Landfill
LFGTE – Landfill Gas Collection System and Landfill-Gas-to-Energy
LMOP – Landfill Methane Outreach Program
LNG – Liquefied Natural Gas
LPG – Liquefied Petroleum Gas
Mt – Metric ton (equivalent to 1.102 short tons)
MMt – Million Metric tons
MPO – Metropolitan Planning Organization
MSW – Municipal Solid Waste
MW – Megawatt
MWh – Megawatt-hour
N – Nitrogen*
N
2
O – Nitrous Oxide*
NO
2
– Nitrogen Dioxide*
NO
x
– Nitrogen Oxides*
NAICS – North American Industry Classification System
NASS – National Agricultural Statistics Service
NFRTAQPC – North Front Range Transportation and Air Quality Planning Council
NMVOCs – Nonmethane Volatile Organic Compounds*
O
3
– Ozone*
ODS – Ozone-Depleting Substances*
OM – Organic Matter*
PADD – Petroleum Administration for Defense Districts
PFCs – Perfluorocarbons*
PM – Particulate Matter*
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
xii Center for Climate Strategies
www.climatestrategies.us
PPACG – Pikes Peak Area Council of Governments
ppb – parts per billion
ppm – parts per million
ppt – parts per trillion
PV – Photovoltaic
PWG – Policy Work Group
RCI – Residential, Commercial, and Industrial
RPA – Resources Planning Act Assessment
RPS – Renewable Portfolio Standard
SAR – Second Assessment Report*
SED – State Energy Data
SF
6
– Sulfur Hexafluoride*
SGIT – State Greenhouse Gas Inventory Tool
Sinks – Removals of carbon from the atmosphere, with the carbon stored in forests, soils,
landfills, wood structures, or other biomass-related products.
TAR – Third Assessment Report*
T&D – Transmission and Distribution
Tg – Teragram
TWh – Terawatt-hours
UNFCCC – United Nations Framework Convention on Climate Change
US EPA – United States Environmental Protection Agency
US DOE – United States Department of Energy
USDA – United States Department of Agriculture
USFS – United States Forest Service
USGS – United States Geological Survey
VMT – Vehicle-Miles Traveled
WAPA – Western Area Power Administration
WECC – Western Electricity Coordinating Council
W/m
2
– Watts per Square Meter
WMO – World Meteorological Organization*
WRAP – Western Regional Air Partnership
WW – Wastewater
* – See Appendix J for more information.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
xiii Center for Climate Strategies
www.climatestrategies.us
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
xiv Center for Climate Strategies
www.climatestrategies.us
Acknowledgements
We appreciate all of the time and assistance provided by numerous contacts throughout
Colorado, as well as in neighboring states, and at federal agencies. Thanks go to in particular the
many staff at several Colorado State Agencies for their inputs, and in particular to Jill Cooper,
Jim DiLeo, and the peer review staff of the Colorado Department of Public Health and
Environment (CDPHE) who provided key guidance in developing the preliminary inventory and
reference case projections. Thanks also go to the members of the Climate Action Panel and the
Policy Work Groups who provided review comments for improving the inventory and reference
case projections.
The authors would also like to express their appreciation to Katie Bickel, Michael Lazarus,
Lewison Lem, Katie Pasko, and David Von Hippel of the Center for Climate Strategies (CCS)
who provided valuable review comments during development of this report. Thanks also to
Michael Gillenwater for directing preparation of Appendix J.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
1 Center for Climate Strategies
www.climatestrategies.us
Summary of Findings
Introduction
This report presents a summary of Colorado’s anthropogenic greenhouse gas (GHG) emissions
and sinks (carbon storage) from 1990 to 2020. The Center for Climate Strategies (CCS) prepared
a preliminary draft GHG emissions inventory and reference case projection for the Colorado
Department of Public Health and Environment (CDPHE) through an effort of the Western
Regional Air Partnership (WRAP).
11
The preliminary draft report was provided to the Climate
Action Panel (CAP) (and its Policy Work Groups [PWGs]) of the Colorado Climate Project to
assist the CAP in understanding past, current, and possible future GHG emissions in Colorado,
and thereby inform the policy option development process. The CAP and the PWGs provided
comments for improving the reference case projections. This report documents the revised
inventory and reference case projections incorporating comments as approved by the CAP.
12
Historical GHG emissions estimates (1990 through 2005)
13
were developed using a set of
generally accepted principles and guidelines for state GHG emissions inventories, as described
the “Approach” section below, relying to the extent possible on Colorado-specific data and
inputs. The reference case projections (2006–2020) are based on a compilation of various
existing Colorado and regional projections of electricity generation, fuel use, and other GHG-
emitting activities, along with a set of simple, transparent assumptions described in Appendixes
A through I of this report.
This report covers the six gases included in the US Greenhouse Gas Inventory: carbon dioxide
(CO
2
), methane (CH
4
), nitrous oxide (N
2
O), hydrofluorocarbons (HFCs), perfluorocarbons
(PFCs), and sulfur hexafluoride (SF
6
). Emissions of these GHGs are presented using a common
metric, CO
2
equivalence (CO
2
e), which indicates the relative contribution of each gas to global
average radiative forcing on a Global Warming Potential- (GWP-) weighted basis.
14
The final
appendix to this report provides a more complete discussion of GHGs and GWPs. Emissions of
black carbon (BC) were also estimated. Black carbon is an aerosol species with a positive
climate forcing potential (that is, the potential to warm the atmosphere, as GHGs do); however,
BC currently does not have a GWP defined by the Intergovernmental Panel on Climate Change
(IPCC) due to uncertainties in both the direct and indirect effects of BC on atmospheric
processes (see Appendices I and J for more details).
It is important to note that the emissions estimates for the electricity sector reflect the GHG
emissions associated with the electricity sources used to meet Colorado’s demands,
11
Draft Colorado Greenhouse Gas Inventory and Reference Case Projections, 1990–2020, prepared by the Center
for Climate Strategies for the Colorado Department of Public Health and Environment (CDPHE) through an effort
of the Western Regional Air Partnership, January 2007.
12
Final Colorado Greenhouse Gas Inventory and Reference Case Projections, 1990–2020, prepared by the Center
for Climate Strategies for the Climate Action Panel of the Colorado Climate Project, October 2007.
13
The last year of available historical data varies by sector; ranging from 2000 to 2005.
14
These gases and the concepts of radiative forcing and GWP are described in Appendix J.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
2 Center for Climate Strategies
www.climatestrategies.us
corresponding to a consumption-based approach to emissions accounting (see “Approach”
section below). Another way to look at electricity emissions is to consider the GHG emissions
produced by electricity generation facilities in Colorado. This report covers both methods of
accounting for emissions, but for consistency, all total results are reported as consumption-based.
Colorado Greenhouse Gas Emissions: Sources and Trends
Table 1 provides a summary of GHG emissions estimated for Colorado by sector for the years
1990, 2000, 2005, 2010, and 2020. Details on the methods and data sources used to construct
these estimates are provided in the appendices to this report. In the sections below, we discuss
GHG emission sources (positive, or gross, emissions) and sinks (negative emissions) separately
in order to identify trends, projections, and uncertainties clearly for each.
This next section of the report provides a summary of the historical emissions (1990 through
2005) followed by a summary of the forecasted reference-case projection-year emissions (2006
through 2020) and key uncertainties. We also provide an overview of the general methodology,
principles, and guidelines followed for preparing the inventories. Appendices A through H
provide the detailed methods, data sources, and assumptions for each GHG sector.
Appendix I provides information on 2002 and 2018 black carbon (BC) estimates for Colorado.
CCS estimated that BC emissions in 2002 ranged from 4.3 – 9.2 million metric tons (MMt) of
carbon dioxide equivalent (CO
2
e) with a mid-point of 6.75 MMtCO
2
e. A range is estimated
based on the uncertainty in the global modeling analyses that serve as the basis for converting
BC mass emissions into their CO
2
e. Emissions are expected to drop by about 4.0 MMtCO
2
e/yr
by 2018 as a result of new engine and fuel standards affecting onroad and nonroad diesel
engines. Appendix I contains a detailed breakdown of 2002 emissions contribution by source
sector. Since the IPCC has not yet assigned a global warming potential for BC, CCS has
excluded these estimates from the GHG summary shown in Table 1.
Appendix J provides background information on GHGs and climate-forcing aerosols.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
3 Center for Climate Strategies
www.climatestrategies.us
Table 1. Colorado Historical and Reference Case GHG Emissions, by Sector
a
(Million Metric Tons CO
2
e) 1990 2000 2005 2010 2020 Explanatory Notes for Projections
Energy 75.4 96.0 102.2 114.1 129.1
Electricity Production 31.6 38.7 39.8 45.3 50.0
Coal 30.9 35.1 34.9 40.0 44.2 See electric sector assumptions
Natural Gas 0.71 3.5 4.9 5.2 5.8 in Appendix A
Oil 0.02 0.08 0.02 0.02 0.02
Wood 0.00 0.00 0.00 0.00 0.01
Net Imported Electricity 1.0 2.2 3.1 2.9 2.6
Electricity Consumption Based 32.7 40.9 42.9 48.2 52.6
Residential/Commercial/Industrial (RCI)
Fuel Use
16.3 20.2 21.2 23.6 27.9
Coal 1.6 1.0 1.2 1.3 1.5 Based on US DOE regional projections
Natural Gas 11.8 15.4 16.5 18.8 23.7 Based on US DOE regional projections
Oil 2.8 3.7 3.5 4.1 5.2 Based on US DOE regional projections
Wood (CH
4
and N
2
O) 0.06 0.07 0.04 0.05 0.05 Based on US DOE regional projections
Avoided emissions from recent building
code and demand-side management
(DSM) initiatives
0 0 0 -0.64 -2.5
Based on analysis of Colorado House Bill (HB) 07-1037
for avoided electricity and natural gas, HB 07-1146 for
avoided natural gas, and Xcel settlement electric DSM
Transportation 19.0 25.5 28.0 30.6 36.2
Motor Gasoline 13.3 17.4 18.1 19.2 22.1 Based on US DOE regional projections
Diesel 2.9 4.8 6.5 7.7 9.8 Based on US DOE regional projections
Natural Gas, LPG, other 0.19 0.22 0.22 0.28 0.39 Based on US DOE regional projections
Jet Fuel and Aviation Gasoline 2.5 3.1 3.2 3.4 3.9 Based on US DOE regional projections
Fossil Fuel Industry 7.5 9.3 10.1 11.8 12.3
Natural Gas Industry 3.1 4.8 5.0 6.5 7.3
Increase based on current trend to 2009, then US DOE to
2020
Oil Industry 0.22 0.15 0.16 0.18 0.20
Increase based on current trend to 2009, then US DOE to
2020
Coal Mining (Methane) 4.2 4.3 4.9 5.1 4.8 Assumes no change after 2003
Industrial Processes 0.76 2.1 2.9 3.8 5.9
Cement Manufacture (CO
2
) 0.32 0.56 0.52 0.55 0.62
Based on state's Nonmetallic Minerals employment
projections (2004-2014)
Lime Manufacture (CO
2
) 0.01 0.01 0.01 0.01 0.01 Ditto
Limestone & Dolomite Use (CO
2
) 0.00 0.03 0.04 0.04 0.04 Ditto
Soda Ash (CO
2
) 0.04 0.04 0.04 0.04 0.05 Based on 2004 and 2009 projections for US production
ODS Substitutes (HFC, PFC, and SF
6
) 0.004 1.2 2.1 3.0 5.1 Based on national projections (US State Dept.)
Semiconductor Manufacturing (HFC,
PFC, and SF
6
)
0.05 0.14 0.08 0.06 0.03 Based on national projections (US EPA)
Electric Power T & D (SF
6
) 0.35 0.20 0.19 0.14 0.08 Based on national projections (US EPA)
Waste Management 1.2 1.9 2.1 2.5 3.5
Solid Waste Management 0.79 1.3 1.5 1.8 2.7 Projections primarily based on population
Wastewater Management 0.39 0.57 0.59 0.66 0.84 Projections based on population
Agriculture (Ag) 8.7 9.6 8.9 8.9 9.1
Enteric Fermentation 3.0 3.2 3.2 3.2 3.2
Projections held constant at 2003 levels except for dairy
cattle (see Appendix F)
Manure Management 0.83 1.2 1.2 1.2 1.3 Ditto
Ag. Soils and Residue Burning 4.9 5.2 4.5 4.5 4.5 Projections held constant at 2005 levels
Total Gross Emissions 86.1 109.6 116.1 129.3 147.5
increase relative to 1990 27% 35% 50% 71%
Forestry and Land Use -24.7 -24.7 -24.7 -24.7 -24.7
Historical and projected emissions held constant at 2004
levels.
Agricultural Soils -2.0 -2.0 -2.0 -2.0 -2.0
Historical and projected emissions held constant at 1997
levels.
Net Emissions (including sinks) 59.4 82.9 89.4 102.6 120.8
a
Totals
may not equal exact sum of subtotals shown in this table due to independent rounding.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
4 Center for Climate Strategies
www.climatestrategies.us
Historical Emissions
Overview
In 2005, on a gross emissions consumption basis (i.e., excluding carbon sinks), Colorado
accounted for approximately 116 million metric tons (MMt) of CO
2
e emissions, an amount equal
to 1.6% of total United States (US) gross GHG emissions. On a net emissions basis (i.e.,
including carbon sinks), Colorado accounted for approximately 89 MMtCO
2
e of emissions in
2005, an amount equal to 1.4% of total US net GHG emissions.
15
Colorado’s GHG emissions are
rising more quickly than those of the nation as a whole.
16
From 1990 to 2005, Colorado’s gross
GHG emissions were up 35% while national gross emissions rose by 16% during this period.
Much of Colorado’s emissions growth can be attributed to its population growth. From 1990 to
2005, Colorado’s population grew by 43% as compared with a national population growth of
19%.
Figure 1 illustrates the state’s emissions per capita and per unit of economic output. Colorado’s
per capita emission rate is slightly more than the national average of 25 MtCO
2
e/year. Between
1990 and 2005, per capita emissions in Colorado and national per capita emissions have changed
relatively little. Economic growth exceeded emissions growth in Colorado throughout the 1990–
2005 period. From 1990 to 2005, emissions per unit of gross product dropped by 40% nationally,
and by 54% in Colorado.
17
Electricity use and transportation are the state’s principal GHG emissions sources. Together, the
combustion of fossil fuels for electricity generation and in the transportation sector accounted for
60% of Colorado’s gross GHG emissions in 2000, as shown in Figure 2. The remaining use of
fossil fuels—natural gas, oil products, and coal—in the residential, commercial, and industrial
(RCI) sectors, plus the emissions from fossil fuel production, constituted another 28% of total
state emissions.
Agricultural activities such as manure management, fertilizer use, and livestock (enteric
fermentation) result in CH
4
and N
2
O emissions that account for another 9% of state GHG
emissions. Industrial process emissions comprise about 2% of state GHG emissions in 2000, and
these emissions are rising rapidly due to the increasing use of HFCs and PFCs as substitutes for
15
National emissions from Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2005, April 2007, US EPA
#430-R-07-002, (http://www.epa.gov/climatechange/emissions/usinventoryreport.html).
16
Gross emissions estimates only include those sources with positive emissions. Carbon sequestration in soils and
vegetation is included in net emissions estimates. All emissions reported in this section for Colorado reflect
consumption-based accounting (including emissions from electricity imports). On a national basis, little difference
exists between production-based and consumption-based accounting for GHG emissions because net electricity
imports are less than 1% of national electricity generation.
17
Based on gross domestic product by state (millions of current dollars), available from the US Bureau of Economic
Analysis, http://www.bea.gov/regional/gsp/. The national emissions used for these comparisons are based on 2005
emissions, http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
5 Center for Climate Strategies
www.climatestrategies.us
ozone-depleting chlorofluorocarbons.
18
Other industrial processes emissions result from cement
and lime manufacturing; PFC use in semiconductor manufacture; CO
2
released during soda ash,
limestone, and dolomite use; and SF
6
released from transformers used in electricity transmission
and distribution systems. Landfills and wastewater management facilities produce CH
4
and N
2
O
emissions accounting for the remaining 2% of the state’s gross GHG emissions in 2000.
Figure 1. Historical Colorado and US Gross GHG Emissions, Per Capita and Per Unit
Gross Product
0
5
10
15
20
25
30
1990 1995 2000 2005
US GHG/Capita
(tCO2e)
CO GHG/Capita
(tCO2e)
US GHG/$
(100gCO2e)
CO GHG/$
(100gCO2e)
Figure 2. Gross GHG Emissions by Sector, 2000, Colorado and US
Colorado
Agriculture
9%
Industrial
Process
2%
Industrial
Fuel Use
9%
Waste
2%
Transport
23%
Fossil Fuel
Industry
(CH4)
9%
Res/Com
Fuel Use
10%
Electricity
Consumpti
on
37%
Agric.
7%
Electricity
33%
Waste
4%
Industrial
Fuel Use
14%
Fossil Fuel
Ind. (CH4)
3%
Res/Com
Fuel Use
9%
Industrial
Process
5%
Transport
26%
US
18
Chlorofluorocarbons (CFCs) are also potent GHGs; however they are not included in GHG estimates because of
concerns related to implementation of the Montreal Protocol. See Appendix J in the Inventory and Projections report
for Colorado.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
6 Center for Climate Strategies
www.climatestrategies.us
Based on data from the early 1980s through 2004, Colorado’s forests are estimated to be net
sinks, accounting for –24.7 MMtCO
2
of GHG emissions (the negative value indicates a net
sequestration of CO
2
from the atmosphere). Also, agricultural soils are estimated to sequester an
additional –2.0 MMtCO
2
. With these GHG sinks, Colorado’s net emissions were 59.4 MMtCO
2
in 1990. Due to a lack of information to estimate future trends, these sinks were estimated to
remain constant throughout the forecast period from 2005 through 2020. Thus, with the increase
in GHG emission sources, by 2020, the net emissions in Colorado are estimated to increase to
about 121 MMtCO
2
e.
A Closer Look at the Two Major Sources: Electricity and Transportation
As shown in Figure 2, electricity consumption accounted for about 37% of Colorado’s gross
GHG emissions in 2000 (about 40.9 MMtCO
2
e), which was higher than the national average
share of emissions from electricity consumption (33%).
19
The GHG emissions associated with
Colorado’s electricity sector increased by 8.2 MMtCO
2
e between 1990 and 2000, accounting for
about 35% of the state’s net growth in gross GHG emissions in this time period.
It is important to note that these GHG emissions estimates reflect the GHG emissions associated
with the electricity sources used to meet Colorado demands, corresponding to a consumption-
based approach to emissions accounting. Another way to look at electricity emissions is to
consider the GHG emissions produced by electricity generation facilities in the state (see
“Approach” section below). While we estimate emissions associated with both electricity
production and consumption, unless otherwise indicated, tables, figures, and totals in this report
reflect electricity consumption-based emissions. In 2000, emissions associated with Colorado’s
electricity consumption (40.9 MMtCO
2
e) were slightly higher than those associated with
electricity production (38.7 MMtCO
2
e) see Table 1. The higher level for consumption-based
emissions reflects GHG emissions associated with net imports of electricity to meet the state’s
electricity demand.
20
The consumption-based approach can better reflect the emissions (and
emissions reductions) associated with activities occurring in the state, particularly with respect to
electricity use (and efficiency improvements), and is particularly useful for policy-making.
Under this approach, emissions associated with electricity imported from other states would need
to be covered in those states’ accounts in order to avoid double-counting or exclusions. (Indeed,
Arizona, California, Oregon, New Mexico, and Washington are currently considering such an
approach.)
Like electricity emissions, GHG emissions from transportation fuel use have risen steadily from
1990 through 2000 at an average rate of slightly under 3% annually. In 2002, onroad gasoline
vehicles accounted for about 66% of transportation GHG emissions. Onroad diesel vehicles
accounted for another 20% of emissions, and air travel for roughly 11%. Rail, marine gasoline,
and other sources (natural gas- and liquefied petroleum gas- (LPG-) fueled-vehicles and used in
transport applications) accounted for the remaining 2% of transportation emissions. As the result
19
For the US as a whole, there is relatively little difference between the emissions from electricity use and emissions
from electricity production, as the US imports only about 1% of its electricity, and exports far less. Colorado’s
situation is different, since it is a net electricity importer.
20
Estimating the emissions associated with electricity use requires an understanding of the electricity sources (both
in-state and out-of-state) used by utilities to meet consumer demand. The current estimate reflects some very simple
assumptions, as described in Appendix A.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
7 Center for Climate Strategies
www.climatestrategies.us
of Colorado’s population and economic growth and an increase in total vehicle miles traveled
(VMT) during the 1990s, onroad gasoline use grew 32% between 1990 and 2002. Meanwhile,
onroad diesel use rose 151% during that period, suggesting an even more rapid growth in freight
movement within or across the state. Aviation fuel use grew by 16% from 1990 to 2002.
Reference Case Projections
Relying on a variety of sources for projections of electricity and fuel use, as noted below and in
the Appendices, a simple reference case projection of GHG emissions through 2020 was
developed. Table 2 shows key annual growth rates used to project emissions for Colorado and
provides historical growth rates for comparison. As illustrated in Figure 3 and shown
numerically in Table 1, under the reference case projections, Colorado gross GHG emissions
continue to grow steadily, climbing to approximately 148 MMtCO
2
e by 2020, 71% above 1990
levels. Overall, the average annual projected rate of emissions growth in Colorado is 1.6% per
year from 2005 to 2020. Demand for electricity is projected to be the largest contributor to future
emissions growth accounting for about 36% of total gross GHG emissions in 2020, followed by
emissions associated with transportation (25%), RCI fossil fuel use (19%), and fossil fuel
production (8%) (see Figure 4).
Table 2. Key Annual Growth Rates for Colorado, Historical and Projected
1990-2005 2005-2020 Sources
Population* 2.4% 1.8% Colorado State Demography Office
Employment*
Goods
Services
1.0%
2.8%
2.7%
2.8%
Colorado Department of Labor and Employment
website, based on analysis by the US Bureau of Labor
Statistics.
Electricity Sales 3.0% 2.1% US DOE Energy Information Administration (EIA) data
for 1990-2004 (3.0% growth is mix of increased
residential and commercial electricity sales countered by
a decrease in industrial sales). The growth rate for 2005-
2020 is based on electricity sales forecasts developed for
the energy supply sector, and includes state legislation
passed in 2007 establishing new requirements for
Colorado’s renewable portfolio standard and for
demand-side management programs (see Appendix A).
Vehicle Miles
Traveled
3.1% 2.1% Federal Highway Administration, Highway Statistic;
Metropolitan Planning Organizations and CDPHE
* For the RCI fuel consumption sectors, population and employment projections for Colorado were used together
with US DOE EIA’s Annual Energy Outlook 2006 (AEO2006) projections of changes in fuel use for the EIA’s
Mountain region on a per capita basis for the residential sector, and on a per employee basis for the commercial and
industrial sectors. For instance, growth in Colorado’s residential natural gas use is calculated as the Colorado
population growth times the change in per capita natural gas use for the Mountain region.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
8 Center for Climate Strategies
www.climatestrategies.us
Figure 3. Colorado Gross GHG Emissions by Sector, 1990-2020: Historical and Projected
0
20
40
60
80
100
120
140
160
1990 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Electricity (Consumption Based) Fossil Fuel Industry
RCI Fuel Use * Transportation Gasoline Use
Transportation Diesel Use Jet Fuel/Other Transportation
Agriculture ODS Substitutes
Other Ind. Process Waste Management
* RCI = direct fuel use in residential, commercial, and industrial sectors; ODS Substitutes = ozone depleting
substances substitutes. Other Industrial Processes include process-related GHG emissions from cement and lime
manufacturing; semiconductor manufacture; soda ash, limestone, and dolomite use; and electricity transmission and
distribution systems.
Figure 4. Sector Contributions to Gross Emissions Growth in Colorado, 1990-2020:
Historic and Reference Case Projections (MMtCO
2
e Basis)
-2.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0
Electricity (Consumption Based)
RCI Fuel Use
Fossil Fuel Industry
Transportation
ODS Substitutes (HFCs)
Other Ind. Process
Agriculture
Waste Management
MMtCO2e
2005 - 2020
1990 - 2005
RCI = direct fuel use in residential, commercial, and industrial sectors; ODS Substitutes = ozone depleting
substances substitutes; HFC = hydrofluorocarbons.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
9 Center for Climate Strategies
www.climatestrategies.us
CAP Revisions
The following identifies the revisions that the CAP made to the inventory and reference case
projections thus explaining the differences between the information presented in this report and
the preliminary information presented in the January 2007 report:
- Energy Supply: Lowered emissions to account for changes in reference case assumptions
associated with Colorado’s Renewable Portfolio Standard (RPS), which was amended
upward in 2007 by the state legislature’s passage of House Bill (HB) 07-1281
(Renewable Energy Standards):
• Investor-Owned Utilities (IOUs) to provide 20% renewable energy by 2020
• Non-IOUs (e.g., municipal utilities and rural electric cooperatives) to provide 10%
renewable energy by 2020
• Incentives for in-state generation, community-based projects, and solar energy
- RCI: Reduced energy consumption in the reference case projections associated with the
passage of HB 07-1146 (Energy Conservation Building Codes) in 2007. This bill requires
local governments who have building codes to adopt energy efficiency codes for certain
buildings.
21
Reduction in emissions is accounted for under the RPS adjustment to avoid
double counting of emission reductions.
- RCI: Reduced energy consumption in the reference case projections associated with the
passage of HB 07-1037 (legislation recently passed requiring that public electric and gas
utilities implement demand-side management programs)
22
and Xcel’s demand side
management commitments under a recent legal settlement, both of which have the effect
of limiting demand growth relative to what it would have been in the absence of these
factors.
23
- Waste Management: Revisions to municipal solid waste (MSW) to reflect revisions the
US Environmental Protection Agency made to the methods for calculating emissions in
US EPA’s State Greenhouse Gas Inventory Tool (SGIT; i.e., change was from use of
regression equations to LANDGEM model equation):
• 1990 emissions decrease from 1.6 to 0.8 MMtCO
2
e
• 2020 emissions decrease from 5.7 to 2.7 MMtCO
2
e
- Forestry: Removed forest soil organic carbon emissions sink as recommended by the
United States Forest Service (USFS). Relative to the January 2007 report, this change
removed 7.1 MMtCO2e of emissions from the forest sink pool for 1990 through 2020.
21
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1146.
22
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1037.
23
Comprehensive Settlement Agreement, docket 04A-214E, 04A-215E, and 04A-216E, issued December 3, 2004,
available at http://www.xcelenergy.com/docs/corpcomm/SettlementAgreementFinalDraftclean20041203.pdf.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
10 Center for Climate Strategies
www.climatestrategies.us
Key Uncertainties and Next Steps
Some data gaps exist in this inventory, and particularly in the reference case projections. Key
tasks for future refinement of this inventory and forecast include review and revision of key
drivers, such as the electricity and transportation fuel use growth rates that will be major
determinants of Colorado’s future GHG emissions (See Table 2). These growth rates are driven
by uncertain economic, demographic and land use trends (including growth patterns and
transportation system impacts), all of which deserve closer review and discussion.
Perhaps the variable with the most important implications for GHG emissions is the type and
number of power plants built in Colorado between now and 2020. The assumptions on VMT and
air travel growth also have large impacts on projected GHG emissions growth in the state.
Finally, uncertainty remains on estimates for historic and projected GHG sinks from forestry,
which can greatly affect the net GHG emissions attributed to Colorado.
Emissions of aerosols, particularly BC from fossil fuel combustion, could have significant
impacts in terms of radiative forcing (i.e., climate impacts). Methodologies for conversion of BC
mass emissions estimates and projections to global warming potential involve significant
uncertainty at present, but CCS has developed and used a recommended approach for estimating
BC emissions based on methods used in other states. Current estimates suggest a relatively small
CO
2
e contribution overall from BC emissions, as compared to the CO
2
e contributed from the
gases (about 4 to 8% BC contribution relative to the other gases in 2002, with the fractions
falling in the 2018 forecast; see Appendix I).
Approach
The principal goal of compiling the inventory and reference case projections presented in this
report is to provide the State of Colorado, the CAP, and the PWGs with a general understanding
of Colorado’s historical, current, and projected (expected) GHG emissions. The following
explains the general methodology and the general principles and guidelines followed during
development of the GHG inventory and reference case projections for Colorado.
General Methodology
We prepared this analysis in close consultation with Colorado agencies, in particular, with the
CDPHE staff, the CAP, and the PWGs. The overall goal of this effort was to provide simple and
straightforward estimates, with an emphasis on robustness, consistency, and transparency. As a
result, we relied on reference forecasts from best available state and regional sources where
possible. Where reliable existing forecasts were lacking, we used straightforward spreadsheet
analysis and constant growth-rate extrapolations of historical trends rather than complex
modeling.
In most cases, we followed the same approach to emissions accounting for historical inventories
used by the US EPA in its national GHG emissions inventory
24
and its guidelines for states.
25
24
National emissions from Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2005, April 2007, US EPA
#430-R-07-002, (http://www.epa.gov/climatechange/emissions/usinventoryreport.html).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
11 Center for Climate Strategies
www.climatestrategies.us
These inventory guidelines were developed based on the guidelines from the IPCC, the
international organization responsible for developing coordinated methods for national GHG
inventories.
26
The inventory methods provide flexibility to account for local conditions. The key
sources of activity and projection data used are shown in Table 3. Table 3 also provides the
descriptions of the data provided by each source and the uses of each data set in this analysis.
General Principles and Guidelines
A key part of this effort involves the establishment and use of a set of generally accepted
accounting principles for evaluation of historical and projected GHG emissions, as follows:
- Transparency: We reported data sources, methods, and key assumptions to allow open
review and opportunities for additional revisions by the CAP and PWGs.
- Consistency: To the extent possible, the inventory and projections will be designed to be
externally consistent with current or likely future systems for state and national GHG
emission reporting. We used US EPA tools for state inventories and projections as a
starting point. These initial estimates were then augmented and/or revised as needed to
conform with state-based inventory and base-case projection needs. For consistency in
making reference case projections
27
, we define reference case actions for the purposes of
projections as those currently in place or reasonably expected over the time period of
analysis.
- Comprehensive Coverage of Gases, Sectors, State Activities, and Time Periods. This
analysis aimed to comprehensively cover GHG emissions associated with activities in
Colorado. It covers all six GHGs covered by US and other national inventories: CO
2
,
CH
4
, N
2
O, SF
6
, HFCs, PFCs, and BC. The inventory estimates are for the year 1990, with
subsequent years included up to most recently available data (typically 2002 to 2005),
with projections to 2010 and 2020.
- Priority of Significant Emissions Sources: In general, activities with relatively small
emissions levels were not reported with the same level of detail as other activities.
- Priority of Existing State and Local Data Sources: In gathering data and in cases
where data sources conflicted, we placed highest priority on local and state data and
analyses, followed by regional sources, with national data or simplified assumptions such
as constant linear extrapolation of trends used as defaults where necessary.
- Use of Consumption-Based Emissions Estimates: To the extent possible, we estimated
emissions that are caused by activities that occur in Colorado. For example, we reported
emissions associated with the electricity consumed in Colorado. The rationale for this
25
http://yosemite.epa.gov/oar/globalwarming.nsf/content/EmissionsStateInventoryGuidance.html.
26
http://www.ipcc-nggip.iges.or.jp/public/gl/invs1.htm.
27
“Reference case” is similar to the term “base year” used in criteria pollutant inventories. However, it also
generally contains both a most current year estimate (e.g., 2002 or 2005), as well as estimates for historical years
(e.g., 1990, 2000). Projections from this reference case are made to future years based on business-as-usual
assumptions of future year source activity.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
12 Center for Climate Strategies
www.climatestrategies.us
method of reporting is that it can more accurately reflect the impact of state-based policy
strategies such as energy efficiency on overall GHG emissions, and it resolves double-
counting and exclusion problems with multi-emissions issues. This approach can differ
from how inventories are compiled, for example, on an in-state production basis, in
particular for electricity.
Table 3. Key Sources for Colorado Data, Inventory Methods, and Growth Rates
Source Information provided Use of Information in this Analysis
US EPA State
Greenhouse Gas
Inventory Tool (SGIT)
US EPA SGIT is a collection of linked
spreadsheets designed to help users develop
state GHG inventories. US EPA SGIT
contains default data for each state for most
of the information required for an inventory.
The SGIT methods are based on the
methods provided in the Volume 8
document series published by the Emissions
Inventory Improvement Program
(http://www.epa.gov/ttn/chief/eiip/techrepor
t/volume08/index.html).
Where not indicated otherwise, SGIT is
used to calculate emissions from
residential/commercial/industrial fuel
combustion, transportation, industrial
processes, agriculture and forestry, and
waste. We use SGIT emission factors
(CO
2
, CH
4
and N
2
O per BTU
consumed) to calculate energy use
emissions.
US DOE Energy
Information
Administration (EIA)
State Energy Data (SED)
EIA SED provides energy use data in each
state, annually to 2003.
EIA SED is the source for most energy
use data. We also use the more recent
data for electricity and natural gas
consumption (including natural gas for
vehicle fuel) from EIA website for
years after 2003. Emission factors from
US EPA SGIT are used to calculate
energy-related emissions.
EIA AEO2006
EIA AEO2006 projects energy supply and
demand for the US from 2003 to 2030.
Energy consumption is estimated on a
regional basis. Colorado is included in the
Mountain Census region (AZ, CO, ID, MT,
NM, NV, UT, and WY).
EIA AEO2006 is used to project
changes in per capita (residential), per
employee (commercial/industrial).
American Gas
Association - Gas Facts
Natural gas transmission and distribution
pipeline mileage.
Pipeline mileage from Gas Facts used
with SGIT to estimate natural gas
transmission and distribution
emissions.
US EPA Landfill
Methane Outreach
Program (LMOP)
LMOP provides landfill waste-in-place
data.
Waste-in-place data used to estimate
annual disposal rate, which was used
with SGIT to estimate emissions from
solid waste.
US Forest Service Data on forest carbon stocks for multiple
years.
Data are used to calculate CO
2
flux
over time (terrestrial CO
2
sequestration
in forested areas).
USDS National
Agricultural Statistics
Service (NASS)
USDA NASS provides data on crops and
livestock.
Crop production data used to estimate
agricultural residue and agricultural
soils emissions; livestock population
data used to estimate manure and
enteric fermentation emissions.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
13 Center for Climate Strategies
www.climatestrategies.us
For electricity, we estimated, in addition to the emissions due to fuels combusted at electricity
plants in the state, the emissions related to electricity consumed in Colorado. This entails
accounting for the electricity sources used by Colorado utilities to meet consumer demands.
In the future, a refinement to the analysis would be to estimate other sectoral emissions on a
consumption basis, such as accounting for emissions from transportation fuel used in Colorado,
but purchased out-of-state. In some cases this can require venturing into the relatively complex
terrain of life-cycle analysis. In general, we recommend considering a consumption-based
approach where it will significantly improve the estimation of the emissions impact of potential
mitigation strategies. For example re-use, recycling, and source reduction can lead to emission
reductions resulting from lower energy requirements for material production (such as paper,
cardboard, and aluminum), even though production of those materials, and emissions associated
with materials production, may not occur within the state.
Details on the methods and data sources used to construct the inventories and forecasts for each
source sector are provided in the following appendices.
• Appendix A. Electricity Use and Supply;
• Appendix B. Residential, Commercial, and Industrial (RCI) Fuel Combustion;
• Appendix C. Transportation Energy Use;
• Appendix D. Industrial Processes;
• Appendix E. Fossil Fuel Industries;
• Appendix F. Agriculture;
• Appendix G. Waste Management; and
• Appendix H. Forestry. .
Appendix I contains a discussion of the inventory and forecast for BC. Appendix J provides
additional background information from the US EPA on GHGs and GWP values.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-1 Center for Climate Strategies
www.climatestrategies.us
Appendix A. Electricity Use and Supply
Overview
Colorado’s electric sector has experienced strong growth in the last 15 years, mostly driven by
population and economic growth in the state. These drivers, and the state’s electric sector, appear
likely to experience continued growth for some time. Greenhouse gas (GHG) emissions
associated with electricity production and consumption accounted for about 36% of Colorado’s
gross GHG emissions in 2005.
As noted in the main report, one of the key questions for the state to consider is how to treat
GHG emissions that result from generation of electricity that is produced outside Colorado to
meet electricity needs in the state. In other words, should the state consider the GHG emissions
associated with the state’s electricity consumption, with its electricity production, or with some
combination of the two? This appendix describes GHG emissions from Colorado’s electricity
sector in terms of emissions from both electricity consumption and production, including the
assumptions used to develop the reference case projections. It then describes Colorado’s
electricity trade and potential approaches for allocating GHG emissions for the purpose of
determining the state’s inventory and reference case projections. In addition, as discussed at the
end of this appendix, the reference case projections were updated to reflect recent legislation that
increased requirements for renewable fuels in Colorado’s Renewable Portfolio Standard (RPS),
which was amended by the Colorado State Legislature in 2007 by House Bill (HB) 07-1281
(Renewable Energy Standards), and requirements for demand-side management programs
(DSM). Finally, key assumptions and results are summarized.
Electricity Consumption
At about 10,000 kilowatt-hour (kWh) per capita (2004 data), Colorado has relatively low
electricity consumption per capita. By way of comparison, the annual per capita consumption for
the US was about 12,000 kWh/capita.
28
Figure A1 shows Colorado’s rank compared to other
western states from 1960-1999; while showing stronger increases during this time period than
most states, Colorado’s per capita consumption has been relatively low (2
nd
lowest, effectively
tied with Utah and New Mexico for much of 1985 to 1999). Many factors influence a state’s per
capita electricity consumption, including the impact of weather on demand for cooling and
heating, the size and type of industries in the state, and the type and efficiency of equipment in
use in the residential, commercial and industrial sectors.
28
Census bureau for U.S. population, Energy Information Administration for electricity sales.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-2 Center for Climate Strategies
www.climatestrategies.us
Figure A1. Electricity Consumption per capita in Western States, 1960-1999
Source: Northwest Power Council, 5
th
Power Plan, Appendix A Note: MWhr is Megawatt-hours.
As shown in Figure A2, electricity sales in the Colorado have generally increased steadily from
1990 through 2004. Overall, total electricity consumption increased at an average annual rate of
3% from 1990 to 2004, which can be compared with population growth at a rate of 2.5% per year
and gross state product increases averaging of 4.3%/yr over the same period.
29
During this
period, residential sector consumption grew by an average of 3.4% per year, commercial sector
use grew by 2.2% per year, and industrial sector consumption increased at 4.2% per year. The
industrial sector electricity sales increases in Colorado have not been uniform over this period –
total industrial sector sales increased by 37% from 1993 to 1994, then by less than 4% from 1994
through 2000.
30
29
Populations from Colorado’s Databook. Gross State Production from Bureau of Economic analysis. Available as
http://bea.gov/bea/newsrelarchive/2006/gsp1006.xls
30
CCS checked this value with EIA who were unable to determine the exact source of the increase. The data are
reported directly by utilities to EIA.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-3 Center for Climate Strategies
www.climatestrategies.us
Figure A2. Electricity Consumption by Sector in Colorado, 1990-2004
31
0
5,000
10,000
15,000
20,000
25,000
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
G
i
g
a
w
a
t
t
-
h
o
u
r
s
(
G
W
h
)
Residential
Commercial
Industrial
Source: EIA State Energy Data (SED) (1990-2002) and EIA Electric Power Annual (2003-2004)
The Colorado Energy Forum recently released a report, Colorado’s Electricity Future.
32
This
report provides projections for electricity sales in Colorado, excluding the impacts of any
additional investments in energy efficiency programs. These projections were developed by RW
Beck by compiling forecasts from the largest utility providers in Colorado and extrapolating
these forecasts to smaller electricity suppliers in similar regions. The RW Beck analysis included
a base case forecast, plus high- and low-case sensitivities. The base case projection was used for
the current analysis. Table A1 reports historic and projected annual average growth rates for
electricity use in Colorado.
31
Note that from 1990-2002, the US Department of Energy (US DOE) Energy Information Administration (EIA)
data includes a category referred to as “other,” which included lighting for public buildings, streets, and highways,
interdepartmental sales, and other sales to public authorities, agricultural and irrigation sales where separately
identified, electrified rail and various urban transit systems (such as automated guideway, trolley, and cable
systems). To report total electricity in Figure A2, the sales from the “other” category are included with commercial
sector sales. The decision to include sales listed as “other” with commercial rather than the residential or industrial
sector sales data was based on a comparison of the trends of electricity sales from 2000-2002 with sales are
categorized in 2003 EIA data.
32
Colorado’s Electricity Future: An Analysis by the Colorado Energy Forum Incorporating Three Separate Reports
by: R.W. Beck Inc., Schmitz Consulting LLC, and The Colorado School of Mines (September 2006)
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-4 Center for Climate Strategies
www.climatestrategies.us
Table A1. Electricity Growth Rates, historic and projected
1990-2000 2000-2004 2004-2010 2010-2020
Residential 3.7% 2.6%
Commercial 2.8% 0.6%
Industrial 4.2% 4.1%
Total 3.4% 2.1% 2.9% 2.1%
Historic Projections
Not Available
Source: Historic from EIA data, projections from Colorado’s
Electricity Future (2006).
Electricity Generation – Colorado’s Power Plants
The following section provides information on GHG emissions and other activity associated with
power plants located in Colorado. Note that GHG emissions are reported in this document as
metric tons of CO
2
equivalents (MTCO
2
) or as million metric tons of CO
2
equivalents
(MMtCO
2
). Since Colorado is part of the interconnected Western Electricity Coordinating
Council (WECC) region – electricity generated in Colorado can be exported to serve needs in
other states, and electricity used in Colorado can be generated by plants outside the state. For this
analysis, we estimate emissions on both a production-basis (emissions associated with electricity
produced in Colorado, regardless of where it is consumed) and a consumption-basis (emissions
associated with electricity consumed in Colorado). The following section describes production-
based emissions while the subsequent section, Electricity trade and the allocation of GHG
emissions, reports consumption-based emissions.
As mentioned the main report and as displayed in Figure A3, coal figures prominently in
electricity generation and accounts for 88% of the GHG emissions from power plants in
Colorado. Table A2 reports the carbon dioxide (CO
2
) emissions from the eight plants in
Colorado with the highest emissions. The plant with the highest emissions, Craig, accounts for
24%-27% of Colorado’s GHG emissions. Craig is a large facility with three generator units
having a combined capacity of over 1,300 megawatts (MW). It runs primarily on coal (over
99.5% of energy consumption) but also consumes small amounts of natural gas and oil. As will
be discussed further in the Electricity Trade and Allocation of GHG emissions section, the Craig
Power Plant is owned by Tri-State (49%), Salt River Project (19%), Pacific-Corp West (13%),
Platte River Power Authority (12%) and Xcel Energy (7%). The contracts associated with these
ownership shares lead to a significant level of electricity from these plants being exported
outside the state – the Salt River Project serves customers in central Arizona; Tri-State provides
power to cooperatives in Wyoming and Nebraska, as well at Colorado; and Pacific-Corp West
serves customers in Oregon, Washington and California. The Hayden power plant is also owned
by a mix of Salt River Project (29%), Pacific-Corp West (18%), and Xcel Energy (53%).
Comanche and Cherokee are 100% owned by Xcel Energy.
33
Electricity trade and its impact on
GHG allocation in Colorado are discussed in the section below.
We considered two sources of data in developing the historic inventory of GHG emissions from
Colorado power plants – EIA State Energy Data (SED), which need to be multiplied by GHG
33
Data from US EPA’s Emissions & Generation Resource Integrated Database (eGRID) database, reflecting
ownership levels in 2000.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-5 Center for Climate Strategies
www.climatestrategies.us
emission factors for each type of fuel consumed, and United States Environmental Protection
Agency (US EPA) data on CO
2
emissions by power plant. For total electric sector GHG
emissions, we used the EIA’s State Energy Data (SED) rather than US EPA data because of the
comprehensiveness of the EIA-based data. The US EPA data are limited to plants over 25 MW
and include only CO
2
emissions (US EPA does not collect data on methane (CH
4
) or nitrous
oxide (N
2
O) emissions). Through discussions with staff at the US EPA we also learned that US
EPA data tend to be conservative (that is, overestimate emissions) because the data are reported
as part of a regulatory program, and that during early years of the data collection program,
missing data points were sometimes assigned a large value as a placeholder. However, the US
EPA provides easily accessible data for each power plant (over 25 MW), which would be much
more difficult to extract from EIA data, and the CO
2
emissions from the two sources differ by
less than 2% in most years. Based on this information, we chose to report information from both
data sources, but rely on the EIA data for the inventory values. For total GHG emissions from
electricity production in Colorado, we applied State Greenhouse Gas Inventory Tool (SGIT)
emission factors
34
to EIA’s SED. For CO
2
emissions from individual plants, we used the EPA
database.
Table A2. CO
2
Emissions from Individual Colorado Power Plants, 2000-2005
(Million metric tons CO2) 2000 2001 2002 2003 2004 2005
Cherokee 4.9 4.8 4.3 5.0 4.9 5.2
Comanche 4.4 4.7 5.2 5.4 4.8 4.8
Craig 9.5 9.7 9.7 9.7 10.4 10.5
Hayden 3.6 3.8 4.0 3.6 3.8 4.1
Martin Drake 2.0 2.1 2.0 2.1 1.8 2.2
Pawnee 4.3 4.8 3.6 4.2 3.8 3.2
Rawhide Energy Station 2.2 2.4 2.3 2.5 2.5 2.1
Ray D Nixon 1.6 1.7 1.7 1.7 1.8 1.6
Other Plants 6.0 6.8 6.7 5.4 5.5 6.0
Total CO2 emissions 38.5 40.7 39.5 39.6 39.3 39.6
Source: US EPA Clean Air Markets database for named plants (http://cfpub.epa.gov/index.cfm). Total
emissions calculated from fuel use data provided by SED (EIA). Note: The emissions reported in the above
table are CO
2
only. CH
4
and N
2
O emissions were not included in the power plant data available from the
US EPA.
Table A3 shows the growth in generation by fuel type for all power plants in Colorado between
1990 and 2004. Overall generation grew by 47% over the 15 years, while electricity consumption
grew by 52%. Natural gas-fired generation has been particularly strong, increasing by more than
8-fold from 1994 through 2004. Renewable generation (biomass, solar and wind) grew by a
similar relative amount over the time period, but as of 2004 these resources accounted for only
0.5% of total generation. Coal generation grew more slowly but remains the dominant source of
electricity in the state. Imports grew from an estimated 1,500 Gigawatt-hour (GWh) (4.6% of
state generation) in 1990 to 3,500 GWh (7.6% of state generation) in 2004.
34
SGIT http://www.epa.gov/climatechange/emissions/state_guidance.html, National GHG inventory
http://www.epa.gov/climatechange/emissions/usinventoryreport.html
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-6 Center for Climate Strategies
www.climatestrategies.us
Figure A3. Electricity Generation and CO
2
Emissions from Colorado Power Plants, 2004
Total Generation
47, 908 GWh
Petroleum,
13 GWh, 0%
wind, solar,
biomass,
waste,
255 GWh, 1%
Natural Gas,
10,597 GWh,
22%
Hydroelectric,
1,195 GWh, 2%
Coal,
35,848 GWh,
75%
Total GHG Emissions
39.5 MMTCO
2
e
Natural Gas,
4.6 MMTCO
2
e,
12%
Coal,
34.9
MMTCO
2
e, 88%
Source: Generation data from EIA Electric Power Annual spreadsheets, GHG emissions figures calculated from EIA
data on consumption and SGIT GHG emission factors.
Table A3. Growth in Electricity Generation in Colorado 1990-2004
Growth
1990 2004
Coal 29,815 35,848 20%
Hydroelectric 1,420 1,195 -16%
Natural Gas 1,238 10,597 756%
biomass, solar, wind 4 255 723%
Petroleum 25 13 -49%
Total 32,502 47,908 47%
Generation (GWh)
Source: EIA Electric Power Annual Data
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-7 Center for Climate Strategies
www.climatestrategies.us
Future Generation and Emissions
Estimating future generation and GHG emissions from Colorado power plants requires
estimation of new power plant additions and production levels from new and existing power
plants. There are, of course, large uncertainties, especially related to the timing and nature of
new power plant construction.
The future mix of generating plants in Colorado remains uncertain, as the trends in type of new
builds are influenced by many factors. Since 1982, new fossil-fueled plants in Colorado have
been natural gas-fired; however, concerns about the cost and availability of natural gas seem to
have led to a trend towards a more coal-dominated mix. Recent announcements by several
utilities indicate that coal-fired units will dominate new power plant builds. Xcel Energy has
started construction on the Pueblo unit, an expansion of the Comanche power plant.
Additionally, according a recent Denver Post article, Xcel has proposed to build a separate new
plant, which would be “the nation's first power plant that converts coal to clean-burning gas and
captures carbon emissions - viewed as an environmental breakthrough that will change coal's
image from a belching polluter to an abundant, clean and relatively affordable resource. The
plant could cost from $500 million to $1 billion or more, with a possible construction start in
2009.”
35
In 2004, Colorado became the first state in the country to have voters directly approve a RPS,
rather than have it processed through a state’s legislature.
36
Colorado voters approved
Amendment 37 and the state has recently begun implementation. The RPS requires utilities with
over 40,000 customers to generate (or purchase) a minimum amount of electricity from
renewable sources. Colorado’s RPS requires minimum annual contributions of renewable
electricity of 3% from 2007 through 2010, 6% from 2011 through 2014; and 10% by 2015 and
thereafter. Of the electricity generated each year from renewable sources, at least 4% must come
from solar electric technologies. At least one-half of this percentage must come from solar
electric systems located on-site at customers’ facilities. Other eligible technologies include wind,
geothermal heat, biomass facilities that burn nontoxic plants, landfill gas, animal waste, small
hydroelectric, and hydrogen fuel cells. Energy generated in Colorado is favored; each kWh of
renewable electricity generated in-state will be counted as 1.25 kWh for the purposes of meeting
this standard. The RPS will likely spur additional new wind and solar projects in the state. Xcel
Energy’s Spring Canyon wind farm came on-line in 2006, and three other wind plants have been
proposed for Colorado. Xcel has also announced a project to build and operate an 8 MW solar
central solar power plant in Alamosa, Colorado, that will house two technologies: concentrating
photovoltaic (PV) and advanced flat-plate solar panel units. The plant is expected to be on-line at
the end of 2007, pending regulatory approval, and will be the largest PV central solar station in
the United States.
37
Table A4 presents data on new and proposed plants in Colorado.
35
http://www.denverpost.com/business/ci_4421583
36
Database of State Incentives for Renewables and Efficiency
http://www.dsireusa.org/library/includes/incentive2.cfm?Incentive_Code=CO24R&state=CO&CurrentPageID=1&R
E=1&EE=1
37
http://www.renewableenergyaccess.com/rea/news/story?id=46072
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-8 Center for Climate Strategies
www.climatestrategies.us
Note that proposals for individual plants cover the period through 2010. Beyond this time period
it is necessary to make assumptions about expected growth. Given the many factors affecting
electricity-related emissions and a diversity of assumptions by stakeholders within the electricity
sector, developing a “reference case” projection for the most likely development of Colorado’s
electricity sector is particularly challenging. Therefore, to develop an initial projection, simple
assumptions were made, relying to the extent possible on widely-reviewed and accepted
modeling assessments.
Table A4. New and Proposed Power Plants in Colorado
Plant Name Fuel Status Capacity Notes
generation Emissions
MW GWh MMTCO2e
Colorado Green wind On-line 2003 162 500 0
All power will be sold to Xcel Energy under
a long term Power Purchase Agreement
Spring Canyon wind On-line 2006 60 190 0
All power will be sold to Xcel Energy under
a long term Power Purchase Agreement
Solar plant in
Alamosa,
Colorado
solar
Proposed -
end of 2007
8 13 0
Xcel Energy selected an affiliate of
SunEdison, LLC, to build, own and operate
this plant, PSCo will purchase the power
and renewable energy credits
Xcel Wind
Plants
wind
Proposed -
end of 2007
775 1697 0
Xcel Energy announced its intent to acquire
775 MW of new wind, to be in service by
end of 2007. Xcel signed contracts with
FPL and Invenergy for 400 MW of capacity.
Blue Spruce
Energy Center
gas On-line 2003 280 255 0.16
Generation and Emissions from US EPA
Clean Air Database for 2005
Rawhide
expansion-
Unit D
gas On-line 2004 74 3
less than
0.005
Generation and Emissions from US EPA
Clean Air Database for 2005
Rocky Montain
Energy Center
gas On-line 2004 478 3,261 1.32
Generation and Emissions from US EPA
Clean Air Database for 2005
Xcel Natural
Gas Plants
gas 2007/2012
608 for 2007
193 for 2012
1050 by 2012 0.54
Plants reported in Xcel Bid Evaluation
report, generation based on 15% capacity
factor (peaking plants)
Lamar
Expansion
coal
Application
Pending -
2008
37 259 0.22
Generation based on 0.80 capacity factor,
GHG emissions based on heatrate of 9000
british thermal unit per kWh (BTU/kWh)
Comanche
Expansion
coal
Under
Construction
- Oct 2009
750 5,256 4.38
Generation based on 0.80 capacity factor,
GHG emissions based on heatrate of 9000
BTU/kWh
Expected Annual
Wind and
Solar Plants
Non-
Renewable
Plants
Sources: Colorado Green – E-mail from Tim Oleary, Shell Renewables, November 10, 2006
Spring Canyon – E-mail from Phil Stiles, InvEnergy, October 18, 2006
Solar Plant – Xcel Energy press release, capacity factor from Wiser and Bolinger powerpoint
presentation 2006, newrules.org
Xcel Wind Plants – Xcel press release, http://www.xcelenergy.com/XLWEB/CDA/0,3080,1-1-
1_15531_26314-28906-0_0_0-0,00.html
Blue Spruce, Rawhide, Rocky Mountain – Colorado’s Electricity Future
Xcel Natural Gas – Xcel Bid Evaluation report
Lamar and Comanche expansion – Colorado’s Electricity Future
The reference case projections are based on CCS’s review of the analyses discussed below and
assume:
- Generation from plants in Colorado grows at 2.8% per year from 2006-2010 – this
growth reflects the estimated generation from the new plants that came on-line in 2006 or
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-9 Center for Climate Strategies
www.climatestrategies.us
are under-construction in the state (as reported in Table A3 above) and additional
renewable generation that is required to meet the RPS.
- Generation from plants in Colorado grows at 2.5% per year from 2010 to 2015 and 2.0%
from 2015 to 2020. This reflects the generation growth rate for the Rocky Mountain
region in EIA’s Annual Energy Outlook 2006 (AEO2006). These assumptions lead to
about 4300 MW of new power plant capacity by 2020 (excluding Comanche expansion).
- Generation from existing non-hydro plants is based on holding generation at 2004 levels.
Generation from existing hydro-electric plants is assumed to be 1,597 GWh per year, the
average generation from the last ten years. New plants and changes to existing plants due
to plant renovations and overhauls that result in higher capacity factors are counted as
new generation.
- The RPS requirements are assumed to be met by Xcel Energy and Aquila, the state’s
investor-owned utilities. From 2007 to 2010, three of the public utilities (Colorado
Spring, Holy Cross and Fort Collins) are expected to meet the RPS minimum
requirements. From 2011 onwards, Longmont is also assumed to meet the RPS
requirements. These utilities are expected to meet the requirements through self-
certification and are assumed to meet the total renewable requirements, but not
necessarily the solar requirements.
38
The 2 investor-owned utilities and the 4 “publics”
are estimated to account for 65% of electricity sales.
39
This analysis assumes that 95% of
the renewables will be located in-state and will receive an additional 25% credit toward
the RPS requirements.
40
- New fossil fuel plants built between 2010 and 2020 will be a mix of 80% coal and 20%
natural gas, based on the mix projected for the Rocky Mountain region of WECC in the
AEO2006.
- Following the definition of reference case that CCS is using – i.e., based on existing or
soon-to-be enacted policies – the projections for the electric sector assume that the state
does not enact rules designed to limit GHG emissions.
Electricity Trade and Allocation of GHG Emissions
Colorado is part of the interconnected WECC region - a vast and diverse area covering 1.8
million square miles and extending from Canada through Mexico, including all or portions of 14
western states. The inter-connected region allows electricity generators and consumers to buy
and sell electricity across regions, taking advantage of the range of resources and markets.
Electricity generated by any single plant enters the interconnected grid and may contribute to
meeting demand throughout much of the region, depending on sufficient transmission capacity.
Thus, it is challenging to define, first, which emissions should be allocated to Colorado, and
secondly, to estimate these allocated emissions both historically and into the future. Some
38
Information on utility plans for meeting RPS based on personal communication, Richard Mignogna, Colorado
Department of Regulatory Agencies, October 23, 2006.
39
Based on utility sales data in 2004, from EIA.
40
CCS assumptions, needs verification
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-10 Center for Climate Strategies
www.climatestrategies.us
utilities track and report electricity sales to meet consumer demand by fuel source and plant type;
however, tracing sales to individual power plants may not be possible.
In 2004, Colorado had 62 entities involved in providing electricity to state customers. The state’s
two investor-owned utilities serve approximately 60% of the customers, and provide 58% of the
electricity sales. The state’s 28 electric cooperatives serve 23% of the customers and provide the
same fraction of sales. One federal and 29 municipal utilities account for the remaining 18.5% of
sales and 17% of customers. The top 5 providers of retail electricity in the state are reported in
Table A5; Xcel Energy provided about 55% of retail electricity sales in 2004.
41
Table A5. Retail Electricity Providers in Colorado (2004)
Entity
Ownership
Type
2004
GWh
Xcel Energy Investor-Owned 25,748
City of Colorado Springs Public 4,312
Intermountain Rural Elec Assn Cooperative 1,784
Aquila Inc Investor-Owned 1,735
City of Fort Collins Public 1,350
Total Sales, Top Five Providers 34,928
Total, All Colorado 46,724
Source: EIA state electricity profiles.
Most of the municipal systems and rural electric cooperatives purchase power from other
utilities, including the Western Area Power Administration (WAPA), Xcel Energy, Tri-State
Generation and Transmission Association (Tri-State), or from a municipal joint-action power
authority. Tri-State, Colorado’s one generation and transmission cooperative, has 1300 MW of
generation capacity and supplies power to rural electricity cooperatives in Colorado, Wyoming
and Nebraska. Three municipal power authorities operate within Colorado – the Arkansas River
Power Authority, the Platte River Power Authority, and the Nebraska Municipal Power Pool.
The Platte River Power Authority is the largest of the three and provides electricity to four cities
(Estes Park, Fort Collins, Longmont, and Loveland) in Colorado with 425 MW of installed
generation – including about 6 MW of wind generation in Wyoming.
42
The largest municipal
generator is Colorado Springs Utilities, which owns and operates 633 MW.
43
In 2004, electricity demand (sales + losses
44
) in Colorado was about 51,500 GWh, while
electricity generation in the state was 47,900 GWh. Net imported electricity from other states
provided the additional 3,400 GWh. Also as mentioned above, 620 MW of the capacity at the
Craig and Hayden power plants is owned by out-of-state utilities. Similarly Colorado utilities
own or have long term contracts for 500 MW of hydro capacity and 340 MW of coal capacity
41
EIA state electricity profiles
42
http://www.awea.org/projects/wyoming.html,
http://www.dora.state.co.us/PUC/projects/euir/FinalRpt/Sctn3Rpt.pdf
43
Colorado Springs Fact Book 2004-2005. http://www.csu.org/about/library/2191.pdf
44
Colorado’s electricity losses are assumed to be 10% of total generation, based on information from eGRID,
http://www.epa.gov/cleanenergy/egrid/index.htm. 10% is the average rate of losses, according to this dataset, over
the period 1994-2000.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-11 Center for Climate Strategies
www.climatestrategies.us
from outside of the state. Thus, electricity trade counts for a significant portion of the electric
power associated with Colorado.
Since almost all states are part of regional trading grids, many states that have developed GHG
inventories have grappled with the problem of how to account for electric sector emissions, when
electricity flows across state borders. Several approaches have been developed to allocate GHG
emissions from the electricity sector to individual states for inventories.
In many ways the simplest approach is production-based – emissions from power plants within
the state are included in the state’s inventory. The data for this estimate are publicly available
and unambiguous. However, this approach is problematic for states that import or export
significant amounts of electricity. Under a production-based approach, characteristics of
Colorado electricity consumption would not be fully captured since only emissions from in-state
generation would be considered.
An alternative is to estimate consumption-based or load-based GHG emissions, corresponding to
the emissions associated with electricity consumed in the state. The load-based approach is
currently being considered by states that import significant amounts of electricity, such as
California, Oregon, and Washington.
45
By accounting for emissions from imported electricity,
states can account for increases or decreases in fossil fuel consumed in power plants outside of
the state, due to demand growth, efficiency programs, and other actions in the state. The
difficulty with this approach is properly accounting for the emissions from imports and exports.
Since the electricity flowing into or out of Colorado is a mix of all plants generating on the inter-
connected grid, it is impossible to physically track the sources of the electrons.
The approach taken in this initial inventory is a simplification of the consumption-based
approach. This approach, which one could term “Net-Consumption-based,” estimates
consumption-based emissions as in-state (production-based) emissions plus the emissions from
the net imports. Emissions for net imports are calculated as net electricity imports (in GWh)
multiplied by the average emission intensity for imports (in MtCO
2
e/GWh). Estimating the mix
of electricity generation for the imports/export of a state is possible and several states are
developing data collection approaches to do this. Washington State has developed regular fuel
disclosure reporting.
46
Colorado enacted legislation in 1999 that requires investor-owned utilities
to disclose information on their fuel mix to retail customers.
47
While this information would be
helpful in estimating the fuel mix of electricity that is imported into Colorado by Xcel and
Aquila, the information was not readily available.
48
As a proxy for estimating the mix of historic
45
See for example, the reports of the Puget Sound Climate Protection Advisory Committee
(http://www.pscleanair.org/specprog/globclim/), the Oregon Governor’s Advisory Group On Global Warming
(http://egov.oregon.gov/ENERGY/GBLWRM/Strategy.shtml), and the California Climate Change Advisory
Committee, Policy Options for Reducing Greenhouse Gas Emissions From Power Imports - Draft Consultant Report
(http://www.energy.ca.gov/2005publications/CEC-600-2005-010/CEC-600-2005-010-D.PDF).
46
http://www.cted.wa.gov/site/539/default.aspx
47
Code of Colorado Regulations Rule 723-3-10(f) et seq. Information from Database of State Incentives for
Renewables and Efficiency http://www.dsireusa.org/documents/Incentives/CO17R.htm
48
The fuel mix provided on Xcel Energy’s bills included a breakdown by fuel for electricity provided by plants in
Colorado but did not include the fuel mix of imported electricity. Information on the legislation is listed at:
http://www.dsireusa.org/library/includes/incentive2.cfm?Incentive_Code=CO17R&state=CO&CurrentPageID=1&R
E=1&EE=0
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-12 Center for Climate Strategies
www.climatestrategies.us
and future GHG for Colorado’s electricity imports, emission intensities that reflect the regional
fuel mix were used. Emissions from the Rocky Mountain region of the WECC (excluding
Colorado’s emissions) were used to calculate GHG emission intensity for imports, with estimates
of future Rocky Mountain emissions provided by the AEO2006. These regional emission factors
were 0.61 MtCO
2
e/MWh in 2004, increasing to 0.68 MtCO
2
e/MWh in 2020, reflecting an
increasing domination of coal generation. To estimate GHG emissions for imports, the amount of
net imports to the state (electricity sales + losses – electricity generation) was multiplied by the
regional emission factors.
This method does not account for differences in the type of electricity that is imported or
exported from the state, and as such, it provides a simple method for reflecting the emissions
impacts of electricity consumption in the state. The calculation also ignores “gross” imports –
since Colorado plants have contracts to out-of-state entities, some of the in-state electricity
generation will be exported and gross imports will be greater than net imports. More
sophisticated methods – for example, based on individual utility information on resources used to
meet loads – can be considered for further improvements to this approach.
Summary of Assumptions and Reference Case Projections
As noted, projecting generation sources, sales, and emissions for the electric sector out to 2020
requires a number of key assumptions, including assumptions regarding future economic and
demographic activity, changes in electricity-using technologies, regional markets for electricity
(and competitiveness of various technologies and locations), access to transmission and
distribution, the retirement of existing generation plants, the response to changing fuel prices,
and the fuel/technology mix of new generation plants. The key assumptions described above are
summarized in Table A6.
Figure A4 shows historical sources of electricity generation in the state by fuel source, along
with projections to the year 2020 based on the assumptions described above.
Based on the above assumptions for new generation, coal continues to dominate new generation
throughout the forecast period (2005-2020). Renewable generation shows the highest relative
growth due to the RPS, growing to 5% of total Colorado generation in 2020.
49
Net imports
increase to a maximum of 6,800 GWh in 2008 (13% of Colorado’s generation) then decrease
sharply as the Comanche coal plant expansion comes on-line (imports are 4,500 GWh or 8.0% in
2010) and continue to decrease, relative to total state generation, to 6% in 2020.
49
This level is lower than the 10% RPS due to assumptions on 1) not all utilities will opt to meet the RPS and 2) in-
state renewable generation receives a credit of 1.25 kWh for each kWh generated so a lower amount of total
renewable generation is required. See assumptions in table A5 for more details.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-13 Center for Climate Strategies
www.climatestrategies.us
Table A6. Key Assumptions and Methods for Electricity Projections for Colorado
Electricity sales Average annual growth of 2.8% from 2005 to 2010 and 2.2% per
year from 2010 to 2020, based on regional growth rates in
Colorado’s Electricity Future, which are based on rates in utilities’
integrated resource plans.
Electricity generation 2.9% per year growth from 2005-2010, based on plants under
construction and RPS requirements and 2.2% per year from 2010 to
2020, based on regional growth rates in AEO2006.
Transmission and
Distribution losses
10% losses are assumed, based on average statewide losses, 1994-
2000, (data from eGRID
50
)
New Renewable
Generation Sources
Colorado’s RPS will be met by 2 investor-owned utilities and 4
public owned utilities (65% of electricity sales), 6% of the utilities’
sales met by renewable generation by 2011, 10% by 2015 and in
subsequent years. 95% of the renewable requirements will be met by
in-state sources. New renewable power plants are assumed to be wind
except for the solar set-aside (4% of the renewable requirements).
New Non-Renewable
Generation Sources
(2006-2010)
New generation in this period assumes the Comanche coal plant
expansion will be on-line by 2010 and new natural gas peaking plants
will be built, following Xcel’s Bid Evaluation. Additional electricity
requirements for Colorado will be met through net electricity
imports.
New Non-Renewable
Generation Sources
(2010-2020)
75% coal
25% natural gas
based on mix of new generation projected in AEO2006 for the Rocky
Mountain region of the WECC.
Heat Rates The assumed heat rates for new gas and coal generation are 7000
BTU/kWh and 9000 BTU/kWh, respectively, based on estimates
used in similar analyses.
51
Operation of Existing
Facilities
Existing facilities are assumed to continue to operate as they were in
2004. Improvements in existing facilities that lead to higher capacity
factors and more generation are captured under the new non-
renewable generation sources.
50
http://www.epa.gov/cleanenergy/egrid/index.htm.
51
See, for instance, the Oregon Governor’s Advisory Group On Global Warming
http://egov.oregon.gov/ENERGY/GBLWRM/Strategy.shtml.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-14 Center for Climate Strategies
www.climatestrategies.us
Figure A4. Electricity Generated by Colorado Power Plants plus Estimated Net Imports,
1990-2020
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
1990 1995 2000 2005 2010 2015 2020
G
W
h
Coal Hydroelectric
Natural Gas Petroleum
Biomass, Wind, Solar Imports
Source: 1990-2004 EIA data, 2005-2020 CCS calculations based on assumptions described above, generation
from petroleum resources is too small to be visible in the chart
Figure A5 illustrates the GHG emissions associated with the mix of electricity generation shown
in Figure A4. From 2005 to 2020, the emissions from Colorado electricity generation are
projected to grow at 2.0% per year, slightly lower than the growth in electricity generation, due
to an increased fraction of generation from renewables. As a result, the average emission
intensity (emissions per MWh) of Colorado’s electricity is expected to decrease from 0.82
MtCO
2
/MWh in 2004 to 0.76 MtCO
2
/MWh in 2020.
Figure A5. Colorado GHG Emissions Associated with Electricity Production (Production-
Basis), excludes Imports
0
10
20
30
40
50
60
1990 1995 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Coal Petroleum
Natural gas
Source: CCS calculations based on approach described in text.
Note: Colorado’s electric generation GHG emissions from petroleum sources are less than 0.1 MMtCO
2
e and
too small to be visible in the chart.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-15 Center for Climate Strategies
www.climatestrategies.us
Figure A6 shows the “net-consumption-based” emissions from 1990 to 2020. Total emissions are
greater than the production-based emissions due to the GHG emissions associated with
electricity imports. These GHG emissions are based on the mix of fuels forecast to generate
electricity in the Rocky Mountain region of the WECC, based on results of the AEO2006. The
estimated regional emission factor is about 0.61 MtCO
2
e/MWh in 2004, increasing to 0.68
MtCO
2
e/MWh in 2020, which is lower than Colorado’s GHG emission rate (see Electricity
Trade section above for further information on this factor).
Figure A6. Colorado GHG Emissions Associated with Electricity Use (Consumption-
Basis), including Imports
0
10
20
30
40
50
60
1990 1995 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Coal Petroleum
Natural gas Imports
Source: CCS calculations based on approach described in text.
Note: GHG emissions from imports are estimated using the mix of fuels in the Rocky Mountain region of
WECC (as defined in the AEO2006). Colorado’s electric generation GHG emissions from petroleum
sources are less than 0.1 MMtCO
2
e and too small to be visible in the chart.
Table A7 summarizes the GHG emissions for Colorado’s electric sector from 1990 to 2020.
During this time period, emissions are projected to increase by 71% on a production-basis and
74% on a consumption-basis.
Comparison to Previous State GHG Inventory
The Colorado Department of Public Health and Environment’s (CDPHE) inventory provided
estimates of production-based electric sector GHG emissions. The production-based GHG
emissions that CCS has estimated for this analysis are about 14% higher than the CDPHE
estimates for 1990 and 12% higher than CDPHE for 1997. These differences appear to result
from differences in energy consumption data, although both analyses relied on EIA data. We
discussed the differences with EIA but were unable to determine the cause for changes in energy
consumption data. However, we verified that the energy consumption values used in this analysis
reflect EIA’s current best estimates. The CDPHE analysis also included projections to 2015,
based on AEO1995 projections for energy consumption in the electric sector. The 2015
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-16 Center for Climate Strategies
www.climatestrategies.us
emissions estimates from the CDPHE analysis were 50.2 MMtCO
2
e, only 1.6% larger than the
estimates from this analysis, 49.5 MMtCO
2
e.
Table A7. Colorado GHG Emissions from Electric Sector, Production and Consumption-
based estimates, 1990-2020 (MMtCO
2
e).
1990 1995 2000 2005 2010 2015 2020
Electricity Production Based 31.6 32.9 38.7 39.8 45.3 49.5 54.0
Coal 30.9 31.6 35.1 34.9 40.0 43.7 47.6
CO2 30.8 31.5 34.9 34.7 39.8 43.5 47.4
CH4 and N2O 0.15 0.15 0.17 0.17 0.20 0.21 0.23
Natural Gas 0.71 1.3 3.5 4.9 5.2 5.8 6.3
CO2 0.71 1.27 3.53 4.88 5.22 5.76 6.34
CH4 and N2O 0.00 0.00 0.00 0.00 0.00 0.01 0.01
Oil 0.02 0.02 0.08 0.02 0.02 0.02 0.02
CO2 0.02 0.02 0.08 0.02 0.02 0.02 0.02
CH4 and N2O 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Wood (CH4 and N2O) 0.000 0.000 0.000 0.001 0.001 0.001 0.001
Net Imported Electricity 1.0 2.5 2.2 3.1 3.0 4.5 3.0
Electricity Consumption Based 32.7 35.4 40.9 42.9 48.2 54.0 57.0
Note: Assumes electricity production from renewable fuels are based on 2004 RPS requirements.
Updates to the Reference Case Projection
At the request of the Climate Action Panel (CAP), the reference case projections for the
Electricity Supply sector were modified to include the effects of Colorado’s 2007 RPS law that
increased requirements (relative to the state’s 2004 RPS) for IOUs to use renewable fuel
resources. The reference case was also modified to incorporate the effects for House Bill (HB)
1146 to increase demand-side management (DSM) requirements for electricity. It was assumed
that the fuel resource mix meeting the RPS is the same as that assumed in the 30% RPS policy
developed by the CAP’s Policy Work Group (PWG). In 2020, wind energy is approximately
85% of RPS generation, and photovoltaic (PV), solar thermal, small hydro, biomass and
geothermal are about 3% each.
In addition, total electricity demand was reduced relative to the original reference case in
response to certain DSM activities as discussed for the RCI sector. The effect of this is a
reduction in the overall emissions associated with energy use, and also a reduction of the direct
impact of the RPS program as this is based on a percentage of total electricity sales in the state.
The RPS law requires Investor-Owned Utilities (IOUs) to provide 20% renewable energy by
2020 and non-IOUs (e.g., municipal utilities and rural electric cooperatives) to provide 10%. The
law also includes incentive clauses for in-state generation, community-based projects and solar
energy. The following assumptions were made for the purpose of revising the emissions
projections under the 2007 RPS law:
- In-state generation constitutes 80% of the RPS generation for IOUs; and this energy
receives 125% credit.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
A-17 Center for Climate Strategies
www.climatestrategies.us
- Community-based projects constitute 10% of non-IOU RPS generation; and this energy
receives 150% credit.
- Pre-2015 solar energy constitutes 5% of non-IOU RPS generation; and this energy
receives 300% credit.
- 10% of energy from existing hydro facilities in Colorado is eligible to meet the RPS.
- New RPS generation and reduced electricity demand displace fossil generation in the
same ratio at which fossil generation is being added to the Colorado system over the
study period: 75% coal and 25% gas.
- While the new law goes into effect in 2008, it has no effect until 2011 as there is
sufficient existing renewable generation to meet the requirement prior to that date.
Table A8 shows the revised contributions to electric sector emissions by fuel after incorporating
the effects of the 2007 RPS requirements and the electricity DSM requirements of HB 1146.
Table A9 shows the total emissions in the initial and the revised reference case projections and
the GHG reductions resulting from the revisions.
Table A8. Colorado GHG Emissions from Electric Sector with 2007 RPS, Production and
Consumption-based estimates, 1990-2020 (MMtCO
2
e).
1990 1995 2000 2005 2010 2015 2020
Electricity Production Based 31.6 32.9 38.7 39.8 45.3 47.6 50.0
Coal 30.9 31.6 35.1 34.9 40.0 42.1 44.2
CO2 30.8 31.5 34.9 34.7 39.8 41.9 43.9
CH4 and N2O 0.15 0.15 0.17 0.17 0.20 0.2 0.2
Natural Gas 0.71 1.3 3.5 4.9 5.2 5.5 5.8
CO2 0.71 1.27 3.53 4.88 5.22 5.5 5.8
CH4 and N2O 0.00 0.00 0.00 0.00 0.00 0.0 0.0
Oil 0.02 0.02 0.08 0.02 0.02 0.02 0.02
CO2 0.02 0.02 0.08 0.02 0.02 0.0 0.0
CH4 and N2O 0.00 0.00 0.00 0.00 0.00 0.0 0.0
Wood (CH4 and N2O) 0.000 0.000 0.000 0.001 0.001 0.004 0.006
Net Imported Electricity 1.0 2.5 2.2 3.1 2.9 2.4 2.6
Electricity Consumption Based 32.7 35.4 40.9 42.9 48.1 50.0 52.6
Note: Assumes electricity production from renewable fuels are based on 2007 RPS requirements.
Table A9. Comparison of Colorado GHG Emissions from Electric Sector for Original and
Revised Forecast (Consumption-Based), 2007-2020
2007 2010 2015 2020
Original GHG Emissions (MMtCO
2
e) 44.4 48.2 52.0 57.0
Revised GHG Emissions (MMtCO
2
e) 44.4 48.2 50.0 52.6
Difference (MMtCO
2
e) - 0.1 2.0 4.3
Difference (%) 0.00 0.15 3.91 7.60
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-1 Center for Climate Strategies
www.climatestrategies.us
Appendix B. Residential, Commercial, and Industrial (RCI) Fuel
Combustion
Overview
Activities in the RCI
52
sectors produce carbon dioxide (CO
2
), methane (CH
4
), and nitrous oxide
(N
2
O) emissions when fuels are combusted to provide space heating, process heating, and other
applications. CO
2
accounts for over 99% of these emissions on a million metric tons (MMt) of
CO
2
equivalent (CO
2
e) basis in Colorado. In addition, since these sectors consume electricity,
one can also attribute emissions associated with electricity generation to these sectors in
proportion to their electricity use.
53
If emissions from the generation of the electricity they
consume are not included, the RCI sectors are between them the third-largest source of gross
greenhouse gas (GHG) emissions in Colorado. Direct use of oil, natural gas, coal, and wood in
the RCI sectors accounted for an estimated 21.2 MMtCO
2
e (18%) of gross GHG emissions in
2005.
54
Emissions and Reference Case Projections
Emissions from direct fuel use were estimated using the United States Environmental Protection
Agency’s (US EPA) State Greenhouse Gas Inventory Tool (SGIT) software and the methods
provided in the Emission Inventory Improvement Program (EIIP) guidance document for RCI
fossil fuel combustion.
55
The default data used in SGIT for Colorado are from United States
Department of Energy (US DOE) Energy Information Administration’s (EIA) State Energy Data
(SED). The SGIT default data for Colorado were revised using the most recent data available,
which includes: (1) 2002 SED information for all fuel types;
56
(2) 2003 SED information for
coal, and for wood and wood waste;
57
(3) 2003 and 2004 SED information for natural gas;
6
(4)
2003 and 2004 SED information for petroleum (distillate oil, kerosene and liquefied petroleum
52
The industrial sector includes emissions associated with agricultural energy use and fuel used by the fossil fuel
production industry.
53
Emissions associated with the electricity supply sector (presented in Appendix A) have been allocated to each of
the RCI sectors for comparison of those emissions to the fuel-consumption-based emissions presented in Appendix
B. Note that this comparison is provided for information purposes and that emissions estimated for the electricity
supply sector are not double-counted in the total emissions for the state. One could similarly allocate GHG
emissions from natural gas transmission and distribution, other fuels production, and transport-related GHG sources
to the RCI sectors based on their direct use of gas and other fuels, but we have not done so here due to the difficulty
of ascribing these emissions to particular end-users. Emissions associated with the transportation sector are provided
in Appendix C and emissions associated with fossil fuel production and distribution are provided in Appendix E.
54
Emissions estimates from wood combustion include only N
2
O and CH
4
. Carbon dioxide emissions from biomass
combustion are assumed to be “net zero”, consistent with US EPA and Intergovernmental Panel on Climate Change
(IPCC) methodologies, and any net loss of carbon stocks due to biomass fuel use should be accounted for in the land
use and forestry analysis.
55
GHG emissions were calculated using SGIT, with reference to EIIP, Volume VIII: Chapter 1 “Methods for
Estimating Carbon Dioxide Emissions from Combustion of Fossil Fuels”, August 2004, and Chapter 2 “Methods for
Estimating Methane and Nitrous Oxide Emissions from Stationary Combustion”, August 2004.
56
EIA State Energy Data 2002, Data through 2002, released June 30, 2006,
(http://www.eia.doe.gov/emeu/states/state.html?q_state_a=co&q_state=COLORADO).
57
EIA State Energy Data 2003 revisions for all fuels, and first release of 2004 information for natural gas and
petroleum, (http://www.eia.doe.gov/emeu/states/_seds_updates.html).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-2 Center for Climate Strategies
www.climatestrategies.us
gas) consumption;
6
(5) 2004 electricity consumption data from the EIA’s State Electricity
Profiles;
58
and (6) 2005 natural gas consumption data from the EIA’s Natural Gas Navigator.
59
Note that the EIIP methods for the industrial sector exclude from CO
2
emission estimates the
amount of carbon that is stored in products produced from fossil fuels for non-energy uses. For
example, the methods account for carbon stored in petrochemical feedstocks, and in liquefied
petroleum gases (LPG) and natural gas used as feedstocks by chemical manufacturing plants
(i.e., not used as fuel), as well as carbon stored in asphalt and road oil produced from petroleum.
The carbon storage assumptions for these products are explained in detail in the EIIP guidance
document.
60
The fossil fuel categories for which the EIIP methods are applied in the SGIT
software to account for carbon storage include the following categories: asphalt and road oil,
coking coal, distillate fuel, feedstocks (naphtha with a boiling range of less than 401 degrees
Fahrenheit), feedstocks (other oils with boiling ranges greater than 401 degrees Fahrenheit),
LPG, lubricants, miscellaneous petroleum products, natural gas, pentanes plus,
61
petroleum coke,
residual fuel, still gas, and waxes. Data on annual consumption of the fuels in these categories as
chemical industry feedstocks were obtained from the EIA SED.
Reference case emissions from direct fuel combustion were estimated based on fuel consumption
forecasts from EIA’s Annual Energy Outlook 2006 (AEO2006),
62
with adjustments for
Colorado’s projected population
63
and employment growth. Colorado employment data for the
manufacturing (goods-producing) and non-manufacturing (commercial or services-providing)
sectors were obtained from the Colorado Department of Labor and Employment.
64
Regional
employment data for the same sectors were obtained from EIA for the EIA’s Mountain region.
65
Finally, as discussed at the end of this appendix, these growth projections were modified to
reflect recent legislation and a settlement agreement leading to increased demand-side
management (DSM) requirements in Colorado.
Table B1 shows historic and projected growth rates for electricity sales by sector. Table B2
shows historic and projected growth rates for energy use by sector and fuel type before
incorporating the effects of recent DSM requirements (see discussion under “Updates to the
Reference Case Projections). For the residential sector, the rate of population growth is expected
to be about 2.5% annually between 2004 and 2020; this demographic trend is reflected in the
growth rates for residential fuel consumption. Based on the Colorado Department of Labor and
Employment’s 10-year forecast (2004 to 2014), commercial and industrial employment are
58
EIA Electric Power Annual 2005 - State Data Tables,
(http://www.eia.doe.gov/cneaf/electricity/epa/epa_sprdshts.html).
59
EIA Natural Gas Navigator (http://tonto.eia.doe.gov/dnav/ng/ng_cons_sum_dcu_SCO_a.htm).
60
EIIP, Volume VIII: Chapter 1 “Methods for Estimating Carbon Dioxide Emissions from Combustion of Fossil
Fuels”, August 2004.
61
A mixture of hydrocarbons, mostly pentanes and heavier fractions, extracted from natural gas.
62
EIA AEO2006 with Projections to 2030, (http://www.eia.doe.gov/oiaf/aeo/index.html).
63
Population data for 1990 are from the Colorado State Demography Office
(http://www.dola.state.co.us/demog/AllHist1.cfm). Population for 2000 and 2005 and forecasts for 2010 and 2025
are taken from the Colorado Data Book (http://www.state.co.us/oed/business-development/colorado-data-
book.cfm).
64
Colorado Department of Labor and Employment, Colorado Industry and Occupational Projections, Released
June 2006 (http://www.coworkforce.com/lmi/oeo/oeo.asp).
65
AEO2006 employment projections for EIA’s Mountain region obtained through special request from EIA (dated
September 27, 2006).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-3 Center for Climate Strategies
www.climatestrategies.us
projected to increase at compound annual rates of 2.8% and 2.7%, respectively, and these growth
rates are reflected in the growth rates in energy use shown in Table B2 for the two sectors. The
2004-to-2014 commercial and industrial employment growth rates were carried forward to 2020
for the purpose of estimating emissions for the reference case projections. These estimates of
growth relative to population and employment reflect expected responses of the economy — as
simulated by the EIA’s National Energy Modeling System — to changing fuel and electricity
prices and changing technologies, as well as to structural changes within each sector (such as
shifts in subsectoral shares and in energy use patterns).
Table B1. Electricity Sales Annual Growth Rates, Historical and Projected
Sector
1990-2004*
2004-2020**
Residential
3.4%
1.9%
Commercial
2.6%
2.9%
Industrial
4.2%
1.9%
Total
3.0%
2.4%
* 1990-2004 compound annual growth rates calculated from Colorado electricity sales by
year from EIA state electricity profiles (Table 8),
http://www.eia.doe.gov/cneaf/electricity/st_profiles/e_profiles_sum.html.
** 2004-2020 compound annual growth rate for total for all three sectors taken from forecast
for the energy supply sector (see Appendix A). Growth rates for individual sectors estimated
based on the proportion of the individual sector’s growth rate to the total growth rate for 1990
to 2004.
Table B2. Historic and Projected Average Annual Growth in Energy Use in Colorado, by
Sector and Fuel, 1990-2020
1990-2004*
2005-2010** 2010-2015** 2015-2020**
Residential
natural gas
1.9%
2.9% 2.3% 2.1%
petroleum
4.9%
1.5% 1.6% 1.2%
wood
-2.8%
1.4% -0.3% 0.1%
coal
9.5%
1.3% -0.9% -0.8%
Commercial
natural gas
-0.5%
2.3% 4.1% 3.4%
petroleum
0.7%
-0.5% 2.4% 1.9%
wood
-1.6%
1.0% 1.6% 1.2%
coal
13.9%
0.9% 1.6% 1.2%
Industrial
natural gas
4.6%
2.8% 1.8% 1.8%
petroleum
3.2%
3.8% 3.2% 2.6%
wood
-14.6%
4.4% 3.6% 3.5%
coal
-6.4%
2.4% 1.5% 1.3%
* Compound annual growth rates calculated from EIA SED historical consumption by sector and fuel type for
Colorado. Latest year for which EIA SED information was available for each fuel type is 2003 for coal and
wood/wood waste, 2004 for petroleum (distillate oil, kerosene, and liquefied petroleum gas), and 2005 for
natural gas. Petroleum includes distillate fuel, kerosene, and liquefied petroleum gases for all sectors plus
residual oil for the commercial and industrial sectors.
** Figures for growth periods starting after 2004 are calculated from AEO2006 projections for EIA’s Mountain
region, adjusted for Colorado’s projected population for the residential sector, projections for non-
manufacturing employment for the commercial sector, and projections for manufacturing employment for the
industrial sector.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-4 Center for Climate Strategies
www.climatestrategies.us
Results
Figures B1, B2, and B3 show historic and projected emissions for the RCI sectors in Colorado
from 1990 through 2020, before incorporating the effects of recent DSM requirements (see
discussion under “Updates to the Reference Case Projections). These figures show the emissions
associated with the direct consumption of fossil fuels and, for comparison purposes, show the
share of emissions associated with the generation of electricity consumed by each sector. During
the period from 1990 through 2020, the residential sector’s share of total RCI emissions from
direct fuel use and electricity use ranges from 32% to 34%, the commercial sector’s share of total
emissions ranges from 33% to 38%, and the industrial sector’s share of total emissions ranges
from 29% to 34% of total RCI emissions. Emissions associated with the generation of electricity
to meet RCI demand accounts for about 65% of the emissions for the residential sector, 81% of
the emissions for the commercial sector, and 52% of the emissions for the industrial sector. From
1990 to 2020, natural gas consumption is the next highest source of emissions for all three
sectors accounting for about 31% of total emissions in the residential sector, 16% for the
commercial sector, and 30% for the industrial sector.
Figure B1. Residential Sector GHG Emissions from Fuel Consumption
0
5
10
15
20
25
30
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Electricity
Wood
Natural Gas
Petroleum
Coal
Source: CCS calculations based on approach described in text.
Note: Emissions associated with wood and coal combustion are too small to be seen on this graph.
For the residential sector, for the 15-year period 2005 through 2020, GHG emissions associated
with the use of electricity and natural gas are expected to increase at average annual rates of
about 1.4% and 2.3%, respectively. Emissions associated with the use of petroleum and wood
fuels are expected to increase annually by about 1.4% and 0.3%, on average, respectively, from
2005 through 2020. Emissions associated with the use of coal are expected to decline slightly by
about 0.3% annually, on the average. Total GHG emissions for this sector increase by an average
of about 1.7% annually over the 15-year period.
For the commercial sector, for 2005 through 2020, emissions associated with the use of
electricity and natural gas are expected to increase at annual average rates of about 2.5% and
3.3%, respectively. Emissions associated with the use of petroleum, wood, and coal fuels are
expected to increase annually by about 1.4%, 1.3%, and 1.2%, on average, respectively, from
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-5 Center for Climate Strategies
www.climatestrategies.us
2005 through 2020. Total GHG emissions for this sector increase on average by about 2.6%
annually over the 15-year period.
For the industrial sector, for 2005 through 2020, emissions associated with the use of electricity
and natural gas are expected to increase by at annual average rates of about 1.4% and 2.1%,
respectively. Emissions associated with the use of petroleum, wood, and coal fuels are expected
to increase annually by about 3.1%, 3.8%, and 1.7%, on average, respectively, from 2005
through 2020. Total GHG emissions for this sector increase by about 1.9% annually, on average,
over the 15-year period.
Figure B2. Commercial Sector GHG Emissions from Fuel Consumption
0
5
10
15
20
25
30
35
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Electricity
Wood
Natural Gas
Petroleum
Coal
Source: CCS calculations based on approach described in text.
Note: Emissions associated with wood combustion are too small to be seen on this graph.
Figure B3. Industrial Sector GHG Emissions from Fuel Consumption
0
5
10
15
20
25
30
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Coal
Electricity
Wood
Natural Gas
Petroleum
Source: CCS calculations based on approach described in text.
Note: Emissions associated with wood combustion are too small to be seen on this graph.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-6 Center for Climate Strategies
www.climatestrategies.us
Updates to the Reference Case Projections
After developing the initial GHG inventory and reference case projections in January 2007, the
State Legislature passed and the Governor signed into law bills to implement DSM programs in
Colorado. In addition, there was a legal settlement agreement with Xcel that contains DSM
requirements. These actions include the following:
- House Bill (HB) 07-1037, requiring public electric and gas utilities implement demand-
side management programs;
66
- A settlement agreement mandating additional DSM in Xcel’s service territory;
67
and
- Improved residential and commercial energy conservation building codes pursuant to
HB07-1146.
68
The effects of these actions were incorporated into the reference case projections at the request
of the Climate Action Panel. The combined emissions impact of HB-1037, HB07-1146, and the
Xcel settlement are shown in Table B3.
Table B3. Impact of Recent Electric and Gas DSM Actions in Colorado
Existing State Actions
a
Avoided GHG Emissions (MMtCO
2
e)
2012 2020 Total 2007-2020
Electric DSM (HB07-
1037 and Xcel Settlement
Agreement)
0.74 1.98 14.8
Gas DSM (HB07-1037) 0.17 0.47 3.2
Building codes (Electric -
HB07-1146)
0.04 0.11 0.73
Building codes (Gas-
HB07-1146)
0.02 0.05 0.34
Total 0.97 2.61 19.1
a
Emission reductions associated with electricity DSM associated with HB07-1146 were incorporated
into the reference case projections to the electricity supply sector to avoid double-counting of
emission reductions.
Key Uncertainties
Key sources of uncertainty underlying the estimates above are as follows:
- Population and economic growth are the principal drivers for electricity and fuel use. The
reference case projections are based on regional fuel consumption projections for EIA’s
Mountain modeling region scaled for Colorado population and employment growth
projections. Consequently, there are significant uncertainties associated with the
66
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1037.
67
Comprehensive Settlement Agreement, docket 04A-214E, 04A-215E, and 04A-216E, issued December 3, 2004,
available at http://www.xcelenergy.com/docs/corpcomm/SettlementAgreementFinalDraftclean20041203.pdf.
68
http://www.statebillinfo.com/sbi/index.cfm?fuseaction=Bills.View&billnum=HB07-1146.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
B-7 Center for Climate Strategies
www.climatestrategies.us
projections. Future work should attempt to base projections of GHG emissions on fuel
consumption estimates specific to Colorado to the extent that such data become available.
- The AEO2006 projections assume no large long-term changes in relative fuel and
electricity prices, relative to current price levels and to US DOE projections for fuel
prices. Price changes would influence consumption levels and, to the extent that price
trends for competing fuels differ, may encourage switching among fuels, and thereby
affect emissions estimates.
- The exception to the AEO2006 assumption of no large changes in prices or fuels
consumption is the AEO2006 reference case projections for industrial coal consumption.
The AEO2006 model’s forecast for the EIA’s Mountain region assumes that new coal-to-
liquids plants would be constructed near active coal mines when low-sulfur distillate
prices reach high enough levels to make coal-to-liquids processing economic. Plants are
assumed to be co-production plants with generation capacity of 758 MW and the
capability of producing 33,200 barrels of liquid fuel per day. The technology assumed is
similar to an integrated gasification combined cycle plant, first converting the coal
feedstock to gas, and then subsequently converting the synthetic gas to liquid
hydrocarbons using the Fisher-Tropsch process. As a result, AEO2006 projections
assume a rather significant increase in coal consumption by the coal-to-liquids industrial
sector starting in 2011. For the EIA’s Mountain region, this sector accounts for 17.5% of
total coal consumption in 2011 and 63% of total coal consumption in 2020, with an
annual growth rate of 26% from 2011 to 2020.
69
This increase in coal consumption,
associated with the installation of coal-to-liquids plants starting in 2011, was excluded
from the industrial coal consumption forecasts for Colorado because it is considered to
represent technology that is beyond the “business-as-usual” assumptions associated with
the reference case projections for the industrial coal consumption sector.
69
Coal Market Module of the National Energy Modeling System 2006, as described in Assumptions to the Annual
Energy Outlook 2006, Coal Market Module, Report #: DOE/EIA-0554(2006), March 2006
(http://www.eia.doe.gov/oiaf/aeo/assumption/index.html).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-1 Center for Climate Strategies
www.climatestrategies.us
Appendix C. Transportation Energy Use
Overview
Fuel use in the transportation sector is the second largest source of greenhouse gas (GHG)
emissions in Colorado – accounting for 24% of Colorado’s gross GHG emissions in 2005.
Carbon dioxide (CO
2
) accounts for about 96% of transportation GHG emissions from fuel use.
Most of the remaining GHG emissions from the transportation sector are due to nitrous oxide
(N
2
O) emissions from gasoline engines.
Emissions and Reference Case Projections
Greenhouse gas emissions for 1990 through 2002 were estimated using United States
Environmental Protection Agency’s (US EPA) State Greenhouse Gas Inventory Tool (SGIT) and
the methods provided in the Emission Inventory Improvement Program (EIIP) guidance
document for the sector.
70,71
For onroad vehicles, the CO
2
emission factors are expressed in units
of pounds per million British Thermal Units (lb/MMBTU), and the methane (CH
4
) and N
2
O
emission factors are both in units of grams per vehicle miles traveled (g/VMT). Key assumptions
in this analysis are listed in Table C1. The default data within SGIT were used to estimate
emissions, with the most recently available fuel consumption data (2002) from United States
Department of Energy (US DOE) Energy Information Administration (EIA) State Energy Data
(SED) added.
72
The default VMT data in SGIT were replaced with state-level annual VMT data
from the Colorado Department of Transportation (CDOT).
73
State-level VMT figures were
allocated to vehicle types using the vehicle mix data provided by the Colorado Department of
Public Health and Environment (CDPHE).
74
Onroad gasoline and diesel emissions were forecast based on VMT projections provided by the
Denver Regional Council of Governments (DRCOG), the North Front Range Transportation and
Air Quality Planning Council (NFRTAQPC), the Pikes Peak Area Council of Governments
(PPACG), and CDPHE.
75,76,77
VMT projections from DRCOG were applied to VMT for Adams,
Arapahoe, Boulder, Douglas, Denver, and Jefferson counties. Projections from NFRTAQPC
were applied to Larimer and Weld counties, and projections from PPACG were applied to El
Paso County. Vehicle miles traveled for all other counties were forecast using the 2002-2012
growth rate (assumed to extend to 2020) from the Colorado State Implementation Plan for
70
CO
2
emissions were calculated using SGIT, with reference to EIIP, Volume VIII: Chapter. 1. “Methods for
Estimating Carbon Dioxide Emissions from Combustion of Fossil Fuels”, August 2004.
71
CH
4
and N
2
O emissions were calculated using SGIT, with reference to EIIP, Volume VIII: Chapter. 3. “Methods
for Estimating Methane and Nitrous Oxide Emissions from Mobile Combustion”, August 2004.
72
EIA, State Energy Consumption, Price, and Expenditure Estimates (SED),
http://www.eia.doe.gov/emeu/states/_seds.html
73
Brad Beckham, Environmental Programs Branch Manager, Colorado Department of Transportation.
74
Barbara MacRae, Air Pollution Control Division Technical Services Program, Colorado Department of Public
Health and Environment.
75
Erik Sabina, Regional Transportation Modeler, Denver Regional Council of Governments.
76
Andres Gomez, Regional Transportation Modeler, North Front Range MPO.
77
2005-2010 and 2007-2012 Transportation Improvement Programs, PPACG,
http://www.ppacg.org/Trans/trans.htm.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-2 Center for Climate Strategies
www.climatestrategies.us
Ozone.
78
These VMT projections suggest that the overall state VMT will grow at an average rate
of 2.1% per year between 2002 and 2020.
79
Table C1. Key Assumptions and Methods for the Transportation Inventory and
Projections
Vehicle Type and Pollutants Methods
Onroad gasoline, diesel,
natural gas, and Liquefied
Petroleum Gas (LPG) vehicles
– CO
2
Inventory (1990 – 2002)
EPA SGIT and fuel consumption from EIA SED
Reference Case Projections (2003 – 2020)
Gasoline and diesel fuel projected using VMT projections provided by
Metropolitan Planning Organizations (MPOs) and CDPHE, adjusted by fuel
efficiency improvement projections from AEO2006. Other onroad fuels
projected using Mountain Region fuel consumption projections from EIA
AEO2006 adjusted using state-to-regional ratio of population growth.
Onroad gasoline and diesel
vehicles – CH
4
and N
2
O
Inventory (1990 – 2002)
EPA SGIT, onroad vehicle CH
4
and N
2
O emission factors by vehicle type and
technology type within SGIT were updated to the latest factors used in the US
EPA’s Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2003.
State total VMT replaced with VMT provided by CDOT, VMT allocated to
vehicle types using data from CDPHE.
Reference Case Projections (2003 – 2020)
VMT projections from MPOs and CDPHE.
Non-highway fuel
consumption (jet aircraft,
gasoline-fueled piston
aircraft, boats, locomotives) –
CO
2
, CH
4
and N
2
O
Inventory (1990 – 2002)
EPA SGIT and fuel consumption from EIA SED.
Reference Case Projections (2003 – 2020)
Aircraft projected using Colorado airport operations projections provided by
CDOT (2005-2025) and EIA prime supplier sales volumes for aviation
gasoline (2002-2005), no growth assumed for rail and marine vessels.
The state-level VMT projections were allocated to vehicle types based on national VMT
forecasts by vehicle type reported in EIA’s Annual Energy Outlook 2006 (AEO2006). The
AEO2006 data were incorporated because they indicate significantly different VMT growth rates
for certain vehicle types (e.g., 34% growth between 2002 and 2020 in heavy-duty gasoline
vehicle VMT versus 284% growth in light-duty diesel truck VMT over this period). The
procedure first applied the AEO2006 vehicle-type-based national growth rates to 2002 Colorado
estimates of VMT by vehicle type. These data were then used to calculate the estimated
proportion of total VMT by vehicle type in each year. Next, these proportions were applied to the
78
Colorado State Implementation Plan for Ozone, Colorado Air Pollution Control Divisions,
http://apcd.state.co.us/documents/eac/ms-TSD.pdf, 2004.
79
CDOT provided a state level VMT estimate for 2020. By using the MPO forecasts, CCS incorporated more detail
for the urban areas. The resulting state-level growth rate was similar to that from CDOT (2.1% compared to 2.2%)
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-3 Center for Climate Strategies
www.climatestrategies.us
estimates for total VMT in the state for each year to yield vehicle-specific VMT estimates and
compound annual average growth rates displayed in Tables C2 and C3, respectively.
Table C2. Colorado Vehicle Miles Traveled Estimates (millions)
Vehicle Type 2002 2005 2010 2015 2020
Heavy-Duty Diesel Vehicle
1,188 1,374 1,644 1,924 2,213
Heavy-Duty Gasoline Vehicle
499 555 617 703 792
Light-Duty Diesel Truck
50 60 85 119 167
Light-Duty Diesel Vehicle
21 25 36 50 70
Light-Duty Gasoline Truck
12,333 13,180 14,701 16,266 17,679
Light-Duty Gasoline Vehicle
11,918 12,736 14,206 15,719 17,084
Motorcycle
91 98 109 120 131
Total
26,098 28,027 31,397 34,903 38,135
Table C3. Colorado Vehicle Miles Traveled Compound Annual Growth Rates
Vehicle Type 2002-2005 2005-2010 2010-2015 2015-2020
Heavy-Duty Diesel Vehicle 4.97% 3.65% 3.20% 2.83%
Heavy-Duty Gasoline Vehicle 3.66% 2.13% 2.65% 2.40%
Light-Duty Diesel Truck 6.47% 7.21% 7.06% 6.93%
Light-Duty Diesel Vehicle 6.47% 7.21% 7.06% 6.93%
Light-Duty Gasoline Truck 2.24% 2.21% 2.04% 1.68%
Light-Duty Gasoline Vehicle 2.24% 2.21% 2.04% 1.68%
Motorcycle 2.24% 2.21% 2.04% 1.68%
Onroad gasoline and diesel fuel consumption was forecast by developing a set of growth factors
that adjusted the VMT projections to account for improvements in fuel efficiency. Fuel
efficiency projections were taken from EIA’s AEO2006. These projections suggest onroad fuel
consumption growth rates of 1.2% per year for gasoline and 3.3% per year for diesel between
2002 and 2020.
Gasoline consumption estimates for 1990-2002 were adjusted by subtracting ethanol
consumption. While the historical ethanol consumption suggests continued growth, projections
for ethanol consumption in Colorado were not available. Therefore, ethanol consumption was
assumed to remain at the 2002 level (2.5% of total gasoline consumption) in the reference case
projections. Biodiesel and other biofuel consumption were not included in this inventory,
because historical and projection data were not available for these fuels.
Emissions for aircraft operations for 1990 to 2002 were based on SGIT methods and fuel
consumption from EIA SED. The consumption of international bunker fuels is included in jet
fuel consumption from EIA. This fuel consumption associated with international air flights
should not be included in the state inventory (as much of it is actually consumed out of state);
however, data were not available to subtract this consumption from total jet fuel estimates. The
2002 estimates were then projected to 2005 in order to apply post-2005 projection data available
from CDOT (as described below). Jet fuel emissions were projected based on 2002 and 2005
total operations for Denver International Airport, provided by CDOT. Aviation gasoline
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-4 Center for Climate Strategies
www.climatestrategies.us
emissions were projected from 2002 to 2005 using EIA data for 2002-2005 aviation gasoline
prime supplier sales volumes in Colorado.
80
Emissions from aviation were projected from 2005 to 2020 using general aviation and
commercial aircraft operations data for 2005 and 2025 as provided by CDOT.
81
General aviation
refers to the operation of civilian aircraft for purposes other than commercial passenger transport.
Jet fuel emissions were projected based on commercial aircraft operations forecasts, and aviation
gasoline emissions were projected using the general aviation forecasts. While military fuel
consumption is included in the historical estimates, projections of military aircraft operations
were not available. Jet fuel projections were adjusted to reflect the projected increase in national
aircraft fuel efficiency (indicated by increased number of seat miles per gallon), as reported in
AEO2006. Because AEO2006 does not estimate fuel efficiency changes for general aviation
aircraft, forecast changes in overall aviation gasoline consumption were based solely on the
projected number of general aviation aircraft operations. These data on aircraft operations project
growth rates of 2.2% per year for general aviation and 2.4% per year for commercial operations
between 2005 and 2020. The resulting compound annual average growth rates are displayed in
Table C4.
Table C4. Colorado Aviation Fuels Use Compound Annual Growth Rates
Fuel 2002-2005 2005-2010 2010-2015 2015-2020
Aviation Gasoline -5.32% 2.36% 2.36% 2.36%
Jet Fuel 2.94% 1.28% 1.28% 1.28%
For the rail and marine sectors, 1990 – 2004 estimates are based on SGIT methods and fuel
consumption from EIA SED. For rail, the historic data show a reduction in fuel consumption in
the mid-1990’s followed by no growth through 2004. Therefore, no growth was assumed for this
sector. The marine sector gasoline consumption data show a growth rate of about 0.7% per year
from 1990 to 2004. This historic growth rate was applied to estimate emissions in the forecast
years.
Fuel consumption data from EIA includes nonroad gasoline and diesel fuel consumption in the
commercial and industrial sectors. Therefore, nonroad emissions are included in the Residential,
Commercial, and Industrial (RCI) fuel combustion sector in this inventory (see Appendix B).
Table C5 shows how EIA divides gasoline and diesel fuel consumption between the
transportation, commercial, and industrial sectors.
80
Colorado Prime Supplier Sales Volumes of Petroleum Products, Energy Information Administration,
http://tonto.eia.doe.gov/dnav/pet/xls/pet_cons_prim_dcu_SCO_a.xls.
81
Chris Pomeroy, Senior Aviation Planner, Colorado Division of Aeronautics.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-5 Center for Climate Strategies
www.climatestrategies.us
Table C5. EIA Classification of Gasoline and Diesel Consumption
Sector Gasoline Consumption Diesel Consumption
Transportation Highway vehicles, marine Vessel bunkering, military use,
railroad, highway vehicles
Commercial Public non-highway, miscellaneous use Commercial use for space heating,
water heating, and cooking
Industrial Agricultural use, construction, industrial
and commercial use
Industrial use, agricultural use, oil
company use, off-highway vehicles
Results
As shown in Figure C1, onroad gasoline consumption accounts for the largest share of
transportation GHG emissions. Emissions from onroad gasoline vehicles increased by about 32%
from 1990-2002, covering almost 66% of total transportation emissions in 2002. GHG emissions
from onroad diesel fuel consumption increased by 151% from 1990 to 2002, and by 2002
accounted for 20% of GHG emissions from the transportation sector. Emissions from aviation
grew by 16% from 1990-2002, and were 11% of transportation emissions in 2002. Emissions
from all other categories combined (boats and ships, locomotives, natural gas and liquid
petroleum gas (LPG), and oxidation of lubricants) contributed only 2% of total transportation
emissions in 2002.
Figure C1. Transportation GHG Emissions by Fuel, 1990-2020
0
5
10
15
20
25
30
35
40
1990 1995 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Onroad Gasoline
Onroad Diesel
Jet Fuel/Av. Gas
Boats and Ships
Rail
Other
Source: CCS calculations based on approach described in text.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
C-6 Center for Climate Strategies
www.climatestrategies.us
Key Uncertainties
Projections of VMT and Biofuels Consumption
One source of uncertainty in the projections of transportation sector GHG emissions presented
above is the future-year vehicle mix, which was calculated based on national growth rates for
specific vehicle types. These growth rates may not reflect vehicle-specific VMT growth rates for
the state. Also, onroad gasoline and diesel growth rates may be slightly overestimated because
increased consumption of biofuels between 2002 and 2020 was not taken into account (due to a
lack of data).
International Bunker Fuels
The consumption of international bunker fuels included in jet fuel consumption from EIA is
another uncertainty. At least the bulk of this fuel consumption associated with international air
flights should not be included in the state inventory (as much of it is actually consumed out of
Colorado airspace); data were not, however, available to allow this consumption to be subtracted
from total jet fuel use estimates. Another uncertainty associated with aviation emissions is the
use of general aviation forecasts to project aviation gasoline consumption. General aviation
aircraft consume both jet fuel and aviation gasoline, but fuel- specific data were not available.
Also, jet fuel consumption includes consumption by military aircraft; projections of military
aircraft operations were not available.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-1 Center for Climate Strategies
www.climatestrategies.us
Appendix D. Industrial Processes
Overview
Emissions in the industrial processes category span a wide range of activities, and reflect non-
combustion sources of greenhouse gas (GHG) emissions from several industrial processes. This
sector accounted for about 3% of Colorado’s gross GHG emissions in 2005. The industrial
processes that exist in Colorado, and for which emissions are estimated in this inventory, include
the following:
- Carbon Dioxide (CO
2
) from:
- Production of cement, lime, and soda ash
- Consumption of limestone, dolomite, and soda ash;
- Hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF
6
)
from semiconductor manufacture;
- SF
6
from transformers used in electric power transmission and distribution (T&D)
systems; and
- HFCs and PFCs from consumption of substitutes for ozone-depleting substances (ODS)
used in cooling and refrigeration equipment.
Other industrial processes that are sources of GHG emissions but are not found in Colorado
include the following:
- Nitrous oxide (N
2
O) from nitric and adipic acid production;
- PFCs from aluminum production;
- HFCs from HCFC-22 production; and
- SF
6
from magnesium production and processing.
Emissions and Reference Case Projections
GHG emissions for 1990 through 2005 were estimated using the State Greenhouse Gas
Inventory Tool (SGIT) and the methods provided in the Emissions Inventory Improvement
Project (EIIP) guidance document for this sector.
82
Table D1 identifies for each emissions source
category the information needed for input into SGIT to calculate emissions, the data sources
used, and the historical years for which emissions were calculated based on the availability of
data. Table D2 lists the data sources used to quantify activities related to industrial process
emissions, the annual compound growth rates implied by the estimates of future activity used,
and the years for which the reference case projections were calculated.
82
GHG emissions were calculated using SGIT, with reference to the Emission Inventory Improvement Program,
Volume VIII: Chapter. 6. “Methods for Estimating Non-Energy Greenhouse Gas Emissions from Industrial
Processes”, August 2004. This document is referred to as “EIIP” below.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-2 Center for Climate Strategies
www.climatestrategies.us
Results
Figures D1 and D2 show historic and projected emissions for the Colorado industrial processes
sector from 1990 to 2020. Total gross GHG emissions were about 2.1 million metric tons (MMt)
of carbon dioxide equivalent (CO
2
e) in 2000 (2% of total emissions), rising to about 5.9
MMtCO
2
e in 2020 (4% of total emissions). Emissions from the overall industrial processes
category are expected to grow rapidly, as shown in Figures D1 and D2, with emissions growth
almost entirely due to the increasing use of HFCs and PFCs in refrigeration and air conditioning
equipment.
Table D1. Approach to Estimating Historical Emissions
Source Category
Time
Period
Required Data for
SGIT Data Source
Cement
Manufacturing -
Clinker Production
1990 -
2002
Metric tons (Mt) of
clinker produced
each year.
US Geological Survey (USGS) in Cement: Annual
Report.
Note: USGS aggregates production for groups of states
for confidentiality purposes. In the SGIT, aggregated
production is divided by the number of states for which
production is aggregated to estimate production for a
given state. The number of states included in an
aggregate total may vary from one year to the next. For
example, the USGS generally aggregates clinker
production from Colorado and Wyoming. SGIT divides
this aggregated production by two to estimate
Colorado’s clinker production. This is a limitation in
SGIT and may result in overestimating or
underestimating production for a given state.
Cement
Manufacturing -
Masonry Cement
Production
1990
and
1996-
2000
Mt of masonry
cement produced
each year.
USGS in Cement: Annual Report.
Note: Data limitations are the same as described for
cement production. Data are not available for some
years; in those cases, data for the closest year to that
for which data were missing was used as a surrogate to
fill in production data for missing years (e.g., 1996
production used for 1995, and 2000 production used for
2001 and 2002.
Lime Manufacture 1990,
1995,
2000,
and
2005
Mt of high-calcium
and dolomitic lime
produced each year.
Colorado Department of Public Health and Environment
(CDPHE), Air Pollution Control Division, Stationary
Sources Program provided production data for two
plants for several but not all years. Data for the closest
year were used as surrogates to fill in production data
for missing years (e.g., production data for 1996 were
used for 1995). Production by type of lime (i.e.,
hydrated lime versus quicklime) produced was
estimated using regional lime production data from
USGS, Minerals Yearbook - Lime for various years
(http://minerals.usgs.gov/
minerals/pubs/commodity/lime/index.html#mis). EIIP
methods applied to remove water from hydrated lime
production estimates. Then, national lime production
data from USGS, Minerals Yearbook – Lime, was used
to estimate amount of high-calcium and dolomitic lime.
Soda Ash
Manufacture
Not available. One plant in Colorado produces soda ash. According to
Chapter 6 of the EIIP guidance document, information is
not available to determine how to estimate CO
2
e for the
process used at this plant. Consequently, CO
2
e are not
estimated for soda ash production.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-3 Center for Climate Strategies
www.climatestrategies.us
Table D1. Approach to Estimating Historical Emissions (Continued)
Source
Category
Time
Period
Required Data for
SGIT Data Source
Limestone and
Dolomite
Consumption
1994 -
2002
Consumption of
limestone and dolomite
by industrial sectors.
For default data, the state's total limestone
consumption (as reported by USGS) is multiplied by
the ratio of national limestone consumption for
industrial uses to total national limestone consumption.
Additional information on these calculations, including a
definition of industrial uses, is available in Chapter 6 of
the EIIP guidance document.
Soda Ash 1990 -
2005
Consumption of soda
ash used in consumer
products such as glass,
soap and detergents,
paper, textiles, and
food. Emissions based
on state’s population
and estimates of
emissions per capita
from the US EPA
national GHG inventory.
USGS Minerals Yearbook, 2004: Volume I, Metals and
Minerals,
(http://minerals.usgs.gov/minerals/pubs/commodity/sod
a_ash/).
For population data, see references for ODS
substitutes.
ODS Substitutes 1990 -
2002
Based on state’s
population and
estimates of emissions
per capita from the US
EPA national GHG
inventory.
-- Population data for 1990 from Colorado State
Demography Office
(http://www.dola.state.co.us/demog/AllHist1.cfm).
Population for 2000 and 2005 and forecasts for 2010
and 2025 from the Colorado Data Book
(http://www.state.co.us/oed/business-
development/colorado-data-book.cfm).
-- US 1990-2000 population from US Census Bureau
(http://www.census.gov/popest/archives/EST90INTER
CENSAL/US-EST90INT-01.html).
-- US 2000-2005 population from US Census Bureau
(http://www.census.gov/population/
projections/SummaryTabA1.xls).
Semiconductor
Manufacturing
1990 -
2002
State and national
value of semiconductor
shipments for NAICS
code 334413
(Semiconductor and
Related Device
Manufacturing). Method
uses ratio of state-to-
national value of
semiconductor
shipments to estimate
state’s proportion of
national emissions for
1990 - 2002.
National emissions from US EPA 2005 Inventory of US
Greenhouse Gas Emissions and Sinks: 1990-2003
(http://www.epa.gov/climatechange/
emissions/usgginv_archive.html).
Value of shipments from U.S Census Bureau's 1997
Economic Census (http://www.census.gov/econ/
census02/). For 1997, value of shipments for state was
1.8% of national total. In the 2002 Economic Census,
value of shipments for sate was 1.9% of national total.
Given the uncertainty of this method and that the
proportion of state-to-national value of shipments in
2002 is very close to the proportion for 1997, the 1997
proportion was used for all years, rather than using the
1997 proportion for some years and 2002 proportion for
others.
Electric Power
T&D Systems
1990 -
2002
Emissions from 1990 to
2003 based on the
national emissions per
kilowatt-hour (kWh) and
state's electricity use.
National emissions per kWh from US EPA 2005
Inventory of US Greenhouse Gas Emissions and Sinks:
1990-2003 (http://www.epa.gov/climatechange/
emissions/usgginv_archive.html).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-4 Center for Climate Strategies
www.climatestrategies.us
Table D2. Approach to Estimating Projections
Annual Growth Rates (%)
Source
Category
Time
Period
Projection
Assumptions Data Source
2000
to
2005
2005
to
2010
2010
to
2015
2015
to
2020
Cement
Manufacturing -
Clinker
Production and
Masonry Cement
Production
2003 -
2020
Compound annual
growth rate from
Colorado Nonmetallic
Minerals sector
employment projections
(2004-2014). Assumed
growth is same for 2015
– 2020 as in previous
periods.
Colorado Department
of Labor and
Employment;
(http://www.coworkfor
ce.com/lmi/oeo/oeo.a
sp).
1.2 1.2 1.2 1.2
Lime
Manufacture
2006 -
2020
Ditto Ditto 1.2 1.2 1.2 1.2
Limestone and
Dolomite
Consumption
2003 -
2020
Ditto Ditto 1.2 1.2 1.2 1.2
Soda Ash
Consumption
2003 -
2020
Growth between 2004
and 2009 is projected to
be about 0.5% per year
for US production.
Assumed growth is
same for 2010 – 2020.
Minerals Yearbook,
2005: Volume I, Soda
Ash,
(http://minerals.usgs.
gov/minerals/pubs/co
mmodity/soda_ash/so
da_myb05.pdf).
0.5 0.5 0.5 0.5
ODS Substitutes 2003 -
2020
Based on national
growth rate for use of
ODS substitutes.
EPA, 2004 ODS
substitutes cost study
report
(http://www.epa.gov/o
zone/snap/emissions/
TMP6si9htnvca.htm).
15.8 7.9 5.8 5.3
Semiconductor
Manufacturing
2003 -
2020
National growth rate
(based on aggregate for
all stewardship program
categories provided in
referenced data source)
US Department of
State, US Climate
Action Report, May
2002, Washington,
D.C., May 2002
(Table 5-7).
(http://yosemite.epa.g
ov/oar/globalwarming.
nsf/UniqueKeyLookup
/SHSU5BNQ76/$File/
ch5.pdf).
3.3 -6.2 -9.0 -2.8
Electric Power
T&/D Systems
2003 -
2020
Ditto ditto 3.3 -6.2 -9.0 -2.8
Substitutes for Ozone-Depleting Substances (ODS)
HFCs and PFCs are used as substitutes for ODS, most notably CFCs (CFCs are also potent
warming gases) in compliance with the Montreal Protocol and the Clean Air Act Amendments of
1990.
83
Even low amounts of HFC and PFC emissions, for example, from leaks and other
83
As noted in EIIP Chapter 6, ODS substitutes are primarily associated with refrigeration and air conditioning but
also have many other uses such as fire extinguishers, solvent cleaning, aerosols, foam blowing, and sterilization.
ODS substitutes depend on technology characteristics in a range of equipment. For the US national inventory, a
detailed stock vintaging model was used, but this modeling approach has not been completed at the state level.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-5 Center for Climate Strategies
www.climatestrategies.us
releases under normal use of the products, can lead to high GHG emissions on a carbon-
equivalent basis. Emissions from the use of ODS substitutes in Colorado have increased from
0.004 MMtCO
2
e in 1990 to about 1.2 MMtCO
2
e in 2000, and are expected to increase at an
average rate of 7.7% per year from 2000 to 2020 due to increased substitutions of these gases for
ODS. The projected rate of increase for these emissions is based on projections for national
emissions from the US EPA report referenced in Table D2.
Figure D1. GHG Emissions from Industrial Processes, 1990-2020
0
1
2
3
4
5
6
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Source: CCS calculations based on approach described in text.
Electricity Distribution
Emissions of SF
6
from electrical equipment have experienced declines since the early-nineties
(see brown line in Figure D2), mostly due to voluntary action by industry. SF
6
is used as an
electrical insulator and interrupter in electricity T&D systems. Emissions for Colorado from
1990 to 2002 were estimated based on the estimates of emissions per kilowatt-hour (kWh) from
the US EPA GHG inventory and on Colorado’s electricity consumption. The US Climate Action
Report shows expected decreases in these emissions at the national level, and the same rate of
decline is assumed for emissions in Colorado. The decline in SF
6
emissions in the future reflects
expectations of future actions by the electric industry to reduce these emissions.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-6 Center for Climate Strategies
www.climatestrategies.us
Figure D2. GHG Emissions from Industrial Processes, 1990-2020, by Source
0
1
2
3
4
5
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Cement Manufacture Lime Manufacture
Limestone & Dolomite Use (CO2) Soda Ash Use (CO2)
ODS Substitutes (HFCs, SF6) Semiconductor Manf. (HFCs, SF6)
Electricity Dist. (SF6)
Source: CCS calculations based on approach described in text.
Semiconductor Manufacture
Emissions of SF
6
and HFCs from the manufacture of semiconductors have experienced declines
since 2000 (see yellow line in Figure D2). Emissions for Colorado from 1990 to 2002 were
estimated based on the default estimates provided in SGIT, which uses the ratio of the state-to-
national value of semiconductor shipments to estimate the state’s proportion of national
emissions from the US EPA GHG inventory (US EPA 2005 Inventory of US Greenhouse Gas
Emissions and Sinks: 1990-2003). The US Climate Action Report shows expected decreases in
these emissions at the national level, and the same rate of decline is assumed for emissions in
Colorado. The decline in emissions in the future reflects expectations of future actions by the
semiconductor industry to reduce these emissions.
Cement Manufacture
Colorado has two cement plants that produce clinker (an intermediate product from which
finished Portland and masonry cement are made). Clinker production releases CO
2
when calcium
carbonate (CaCO
3
) is heated in a cement kiln to form lime (calcium oxide) and CO
2
(see Chapter
6 of the EIIP guidance document). Emissions are calculated by multiplying annual clinker
production and annual production of masonry cement by emission factors for these processes.
However, information on clinker and masonry cement production was not readily available;
therefore, the default data provided in SGIT were used to calculate emissions (see black line in
Figure D2). The growth rate for Colorado’s nonmetallic minerals sector (an annual average of
1.2%) was used to project emissions to 2020.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-7 Center for Climate Strategies
www.climatestrategies.us
Lime Manufacture
Lime is a manufactured product that is used in many chemical, industrial, and environmental
applications including steel making, construction, pulp and paper manufacturing, and water and
sewage treatment. Lime is manufactured by heating limestone (mostly CaCO
3
) in a kiln, creating
calcium oxide and CO
2
. The CO
2
is driven off as a gas and is normally emitted to the
atmosphere, leaving behind a product known as quicklime. Some of this quicklime undergoes
slaking (combining with water), which produces hydrated lime. The consumption of lime for
certain uses, specifically the production of precipitated CaCO
3
and refined sugar, results in the
reabsorption of some airborne CO
2
(see Chapter 6 of the EIIP guidance document). Emissions
are estimated by multiplying the amount of high-calcium and dolomitic lime produced by
emission factors for each product.
The Colorado Department of Public Health and Environment (CDPHE) provided annual lime
production data for two sugar refining plants for several but not all years. Lime production for
years for which data were not available was estimated using the production data for the closest
year as a surrogate (e.g., production data for 1996 used for 1995). Total production for the two
plants was combined and then regional factors from the USGS were applied to total production
to estimate the amount of hydrated lime and quicklime produced. EIIP methods were applied to
remove the mass of water contained in the product from the hydrated lime production estimates.
Then, national lime production data from the USGS was used to estimate the amount of high-
calcium and dolomitic lime produced each year. These production figures were entered into the
SGIT to calculate CO
2
emissions (see dark blue line in Figure D2). The growth rate for
Colorado’s nonmetallic minerals sector (i.e., 1.2% annual) was used to project emissions to
2020.
Limestone and Dolomite Consumption
Limestone and dolomite are basic raw materials used by a wide variety of industries, including
the construction, agriculture, chemical, glass manufacturing, environmental pollution control,
and metallurgical industries such as magnesium production.
84
Recent historical data for Colorado
were not available from the USGS; consequently, the default data provided in SGIT were used to
calculate emissions for Colorado (see orange line in Figure D2). The growth rate for Colorado’s
nonmetallic minerals sector (i.e., 1.2% annual) was used to project emissions to 2020.
Soda Ash Consumption
Commercial soda ash (sodium carbonate) is used in many consumer products such as glass, soap
and detergents, paper, textiles, and food. CO
2
is also released when soda ash is consumed (see
Chapter 6 of the EIIP guidance document). SGIT estimates historical emissions (see dark pink
line in Figure D2) based on the state’s population and national per capita emissions from the US
EPA national GHG inventory. According to the USGS, this industry is expected to grow at an
annual rate of 0.5% from 2004 through 2009 for the US as a whole. Information on growth
84
In accordance with EIIP Chapter 6 methods, emissions associated with the following uses of limestone and
dolomite are not included in this category: (1) crushed limestone consumed for road construction or similar uses
(because these uses do not result in CO
2
emissions), (2) limestone used for agricultural purposes (which is counted
under the methods for the agricultural sector), and (3) limestone used in cement production (which is counted in the
methods for cement production).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-8 Center for Climate Strategies
www.climatestrategies.us
trends for the soda ash industry for years later than 2009 was not available; the same (2004 –
2009) growth rate was therefore applied for estimating emissions through 2020.
Key Uncertainties
Key sources of uncertainty underlying the estimates above are as follows:
- Historical clinker production for the cement industry is uncertain because of the reliance
on the SGIT method that divides aggregated clinker production (obtained from the
USGS) for select states evenly between the states. In SGIT, production for Colorado and
Wyoming is aggregated and divided evenly between the two states. Future work on this
category should focus on obtaining actual clinker production data for 1990 through 2005
from plants located in Colorado.
- Since emissions from industrial processes are determined by the level of production in
and the production processes of a few key industries, and, in some cases, of a few key
plants, there is relatively high uncertainty regarding future emissions from the industrial
processes category as a whole. Future emissions depend on the competitiveness of
Colorado manufacturers in these industries, and the specific nature of the production
processes used in plants in Colorado.
- The projected largest source of future industrial emissions, HFCs and PFCs used in
cooling applications, is subject to several uncertainties as well. First, historical emissions
are based on national estimates; Colorado-specific estimates are currently unavailable.
For example, emissions will be driven by future choices regarding mobile and stationary
air conditioning technologies and the use of refrigerants in commercial applications, for
which several options currently exist.
- Greenhouse gases are emitted from several additional industrial processes that are not
covered in the EIIP guidance documents, due in part to a lack of sufficient state data on
non-energy uses of fossil fuels for these industrial processes. These sources include:
- Iron and Steel Production (CO
2
and CH
4
);
- Ammonia Manufacture and Urea Application (CO
2
, CH
4
, N
2
O);
- Aluminum Production (CO
2
);
- Titanium Dioxide Production (CO
2
);
- Phosphoric Acid Production (CO
2
);
- CO
2
Consumption (CO
2
);
- Ferroalloy Production (CO
2
);
- Petrochemical Production (CH
4
); and
- Silicon Carbide Production (CH
4
).
The CO
2
emissions from the above CO
2
sources (other than CO
2
consumption and
phosphoric acid production) result from the non-energy use of fossil fuels. Although the
US EPA estimates emissions for these industries on a national basis, US EPA has not
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
D-9 Center for Climate Strategies
www.climatestrategies.us
developed methods for estimating the emissions at the state level due to data limitations.
If state-level data on non-energy uses of fuels become available, future work should
include an assessment of emissions for these other categories. The CDPHE’s report,
Greeenhouse Gas Emission Inventory and Forecast, 1990 through 2015 (revised October
2002, Table 2.1.2), identifies the industrial sources that exist in Colorado but for which
state-specific methods are not available to estimate GHG emissions.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-1 Center for Climate Strategies
www.climatestrategies.us
Appendix E. Fossil Fuel Industries
This appendix reports the greenhouse gas (GHG) emissions that are released during the
production, processing, transmission, and distribution of fossil fuels. Known as fugitive
emissions, these are methane (CH
4
) and carbon dioxide (CO
2
) gases released via leakage and
venting at coal mines, oil and gas fields, fossil fuel processing facilities, and gas and oil
pipelines. Nationally, fugitive emissions from natural gas systems, petroleum systems, and coal
mines accounted for 2.8% of total US GHG emissions in 2004.
85
In Colorado, this sector has
somewhat higher emissions than the national average accounting for 9% of Colorado’s gross
GHG emissions in 2005. Emissions associated with energy consumed by these processes are
included in the residential, commercial and industrial sector emissions (see Appendix B).
Oil and Gas Production
Colorado currently ranks 12
th
in crude oil production among US states, totaling 63,000 barrels
(bbls) per day and accounting for about 1% of US production.
86
Proved state crude oil reserves
sit at 225 million bbls, which is similarly about 1% of US totals. Oil production in Colorado
peaked in 1988 at 88,000 bbls per day.
87
Colorado has two petroleum refineries, with a combined
crude oil distillation capacity of 94,000 bbls per day.
88
Colorado currently produces almost two and a half times the amount of natural gas that it
consumes. For example, in 2004, Colorado consumed 440 billion cubic feet (Bcf) and produced
1,079 Bcf. Since the year 2000, total natural gas production (combined conventional and
unconventional sources) has increased by 44%.
89
Unconventional oil and gas resources play an increasingly important role in the state. Colorado is
a significant player in the coal bed methane (CBM) industry, placing first in the nation for CBM
proved reserves and a close second to New Mexico for CBM production.
90
United States
Department of Energy (US DOE) Energy Information Administration (EIA) reports annual CBM
production growth rates in Colorado averaging over 50% throughout the first half of the 1990s.
Average CBM growth rates in the state had slowed to 3% annually between 2000 and 2005.
Between 1999 and 2001, natural gas production from CBM accounted for about 60% of total
natural gas production in state. The share of natural gas production from CBM has subsequently
decreased as the growth in conventional gas has outpaced the growth in CBM production. CBM
accounted for 47% of total Colorado natural gas production in 2005.
85
“The U.S. Inventory of Greenhouse Gas Emissions and Sinks”, United States Environmental Protection Agency
(US EPA), 2005.
86
“Petroleum Profile: Colorado”, US DOE EIA website, October 2006, Accessed at
87
“Petroleum Navigator”, US DOE EIA website, October 2006, Accessed at
88
IBID
89
“Natural Gas Navigator”, US DOE EIA website, September 2006, Accessed at
90
“Natural Gas Navigator”, US DOE EIA website, September 2006, Accessed at
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-2 Center for Climate Strategies
www.climatestrategies.us
In addition, Colorado has the largest, and most commercially viable, oil shale deposits of any US
state. While commercial oil shale production is a number of years away, high oil prices have
brought renewed interest. There are currently 5 research and development proposals for in-situ
technology pending approval by the United States Bureau of Land Management (BLM).
91
A
2005 study projected a 12- to 16-year lag before the pilot tests initiated over the next few years
lead to a production growth phase.
92
Given the large uncertainty surrounding future production
from oil shale in Colorado, especially in the 2006-2020 timeframe, this analysis does not include
a specific estimate for oil shale production or for total GHG emissions from this process. While a
high-level review of oil shale research projects was conducted, meaningful GHG emission
intensity estimates could not be provided within the time constraints of this project.
Oil and Gas Industry Emissions
Emissions of CH
4
and entrained CO
2
can occur at many stages of production, processing,
transmission, and distribution of oil and gas. With over 23,000 gas and oil wells in the state, 43
operational gas processing plants, 2 oil refineries, and over 32,000 miles of gas pipelines
93
, there
are significant uncertainties associated with estimates of Colorado’s GHG emissions from this
sector. This is compounded by the fact that there are no regulatory requirements to track CO
2
or
methane emissions. Therefore, estimates based on emissions measurements in Colorado are not
possible at this time.
The State Greenhouse Gas Inventory Tool (SGIT) developed by the United States Environmental
Protection Agency (US EPA) facilitates the development of rough estimates of state-level GHG
emissions.
94
Methane emission estimates are calculated by multiplying emissions-related activity
levels (e.g., miles of pipeline, number of compressor stations) by aggregate industry-average
emission factors. Key information sources for the activity data are the US DOE EIA
95
and
American Gas Association’s annual publication Gas Facts.
96
CH
4
emissions were estimated using SGIT, with reference to the Emissions Inventory
Improvement Project (EIIP) guidance document. Future projections of CH
4
emissions from oil
and gas systems are calculated based on the following key drivers:
- Consumption – See Appendix A, Electricity, and Appendix B, Residential, Commercial and
Industrial Sector for assumptions used in projecting natural gas consumption in Colorado.
Based on those assumptions, Colorado’s natural gas consumption is projected to decline
slightly until 2010, and then grow at a rate of just over 2.5% annually.
- Production – Continued growth over the next few years in both conventional oil and gas, and
CBM appears likely. Drilling permits reported by the Oil and Gas Conservation Commission
91
Kent Walters, BLM, White River Field Office manager
92
Bartis, James T. et al, Oil Shale development in the United States: prospects and policy issues. 2005. Rand
Corporation. Prepared for the National Energy Technology Laboratory of the US Department of Energy.
93
Data from EIA and Gas Facts.
94
CH
4
emissions were calculated using SGIT, with reference to Emission Inventory Improvement Program, Volume
VIII: Chapter. 5. “Methods for Estimating Methane Emissions from Natural Gas and Oil Systems”, March 2005.
95
“Petroleum Navigator” and “Natural Gas Navigator”, US DOE EIA website, November 2006, Accessed at
http://www.eia.doe.gov
96
American Gas Association “Gas Facts, A Statistical Record of the Gas Industry” Referenced annual publications
from 1992 to 2004.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-3 Center for Climate Strategies
www.climatestrategies.us
increased by 94% between 2003 and 2005.
97
While an increase in drilling permits does not
necessarily translate directly to increased production, it is an indication that continued
growth is likely. As a simple estimate for projections, oil and total natural gas production,
processing, refining and transportation rates are forecast to follow recent production trends
in the state through 2009. Actual production over this period could be significantly higher, as
reflected by the strong increase in drilling permits. From 2010 to 2020, growth rates for oil
and gas production are based on regional results from EIA’s Annual Energy Outlook 2006,
98
where these data are available. Simple assumptions were made for growth rates for natural
gas processing and oil refining.
Table E1 provides an overview of data sources and approach used to project future emissions.
Table E1. Approach to Estimating Historical and Projected CH
4
Emissions from Natural
Gas and Oil Systems
Approach to Estimating Historical Emissions Approach to Estimating Projections
Activity Required Data for SGIT Data Source Projection Assumptions
Number wells EIA
Natural Gas
Drilling and
Field Production
Miles of gathering pipeline Gas Facts
101
Emissions estimated assuming natural gas
production trend continues until 2009 at
7.3% annually,
99
then following US DOE
regional projections until 2020, which
average 0.8% annual growth.
100
Natural Gas
Processing
Number gas processing
plants
EIA
102
Emissions follow trend of natural gas
processing volume, which continues to
grow at 9.8% annually until 2009, then
follow US DOE production trends to 2020,
as above.
103
Miles of transmission
pipeline
Gas Facts
101
Number of gas transmission
compressor stations
EIIP
104
Number of gas storage
compressor stations
EIIP
105
Natural Gas
Transmission
Number of liquefied natural
gas (LNG) storage
compressor stations
Unavailable,
assumed
negligible.
Emissions follow trend of state gas
production, as above.
97
Colorado Oil and Gas Conservation Commission staff report. Drilling permits reported at 4363 in 2005, 2917 in
2004, and 2249 in 2003.
98
EIA Annual Energy Outlook 2006 (AEO2006) with Projections to 2030, Energy Information Administration,
Department of Energy, http://www.eia.doe.gov/oiaf/aeo/index.html.
99
Assumption based on EIA data with an average annual growth rate of 7.3% average annual growth between 2000
and 2005.
100
Based on US DOE AEO2006, natural gas production projection for Rocky Mountain region. Accessed at
http://www.eia.doe.gov/oiaf/aeo/supplement/sup_ogc.xls.
101
No Gas Facts available for 1991 and 1993, so a linear relationship was assumed to extrapolate from the previous
and subsequent year.
102
EIA reported data for 1995 and 2004.
103
Growth assumption based on EIA gas processing data. Average annual growth of 9.8% in gas processing volume
between 1990 and 2004.
104
Number of gas transmission compressor stations = miles of transmission pipeline x 0.006 EIIP. Volume VIII:
Chapter 5. March 2005.
105
Number of gas storage compressor stations = miles of transmission pipeline x 0.0015 EIIP. Volume VIII: Chapter
5. March 2005.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-4 Center for Climate Strategies
www.climatestrategies.us
Table E1. Approach to Estimating Historical and Projected CH
4
Emissions from Natural
Gas and Oil Systems (Continued)
Miles of distribution pipeline Gas Facts
101
Total number of services Gas Facts
Number of unprotected steel
services
Ratio estimated
from 2002 data
107
Natural Gas
Distribution
Number of protected steel
services
Ratio estimated
from 2002 data
107
Distribution emissions follow state gas
consumption trend.
106
Oil Production Annual production EIA
108
Emissions follow trend of state oil
production, which is projected to grow at
1.9% annually until 2009
109
, then follow US
DOE regional projections until 2020, which
average 1.4% annual growth.
110
Oil Refining Annual amount refined EIA
111
Emissions projected to follow trend of 0.5%
annual growth in state oil refining.
112
Oil Transport Annual oil transported
Unavailable,
assumed oil
refined = oil
transported
Emissions follow trend of state oil refining,
as above.
Note that potential improvements to production, processing, and pipeline technologies resulting
in GHG emissions reductions have not been accounted for in this analysis.
A potentially significant source of CO
2
that is not currently included in this inventory is that of
”entrained” CO
2
in raw gas emerging from the ground. In some areas entrained CO
2
can be
significantly above pipeline specifications, and must be separated out at gas processing facilities.
Depending on the level of entrained CO
2
in Colorado coal bed methane produced, emissions of
entrained CO
2
from this source may be significant. Unfortunately, this data could not be obtained
within the time constraints of this project.
As noted above, this analysis also does not include a specific estimate for oil shale production.
Note that any commercial development of oil shale in the region would result in increased
carbon dioxide equivalent (CO
2
e) emissions from oil production, refining and transportation, and
that no emissions from oil shale facilities are included in current forecasts. As production of oil
from oil shale is expected to be energy- (and therefore GHG emissions-) intensive, any future oil
shale development could have significant GHG implications.
106
Based on US DOE regional projections.
107
Gas Facts reported unprotected and protected steel services for 2002, but only total services for other years.
Therefore the ratio of unprotected and protected steel services in 2002 was assumed to be the ratio for all other years
(0.4891 for protected services and 0.0045 for unprotected services). This yields more congruent results than the EIIP
guidance of using multipliers of 0.2841 for protected steel services, and 0.0879 for unprotected steel services.
108
Data extracted from the Petroleum Supply Annual for each year.
109
Based on EIA data, with an annual average oil production growth rate of 1.9% between 2000 and 2005.
110
Based on US DOE AEO2006, oil production projection for Rocky Mountain region. Accessed at
http://www.eia.doe.gov/oiaf/aeo/supplement/sup_ogc.xls.
111
Refining assumed to be equal to the total input of crude oil into Petroleum Administration for Defense Districts
(PADD) IV times the ratio of Colorado’s refining capacity to PADD IV’s total refining capacity. No data for 1995
and 1997, so linear relationship assumed from previous and subsequent years.
112
Based on EIA data, average growth in crude refined annually was 1.7% between 2000 and 2004.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-5 Center for Climate Strategies
www.climatestrategies.us
Coal Production Emissions
CH
4
occurs naturally in coal seams, and is typically vented during mining operations for safety
reasons. Coal mine methane emissions are usually considerably higher, per unit of coal
produced, from underground mining than from surface mining.
Colorado has 13 operational coal mines, which together produced 38.5 million short tons of coal
in 2005.
113
Of Colorado’s 13 coal mines, 8 are underground and 5 are surface mines. In this
inventory, CH
4
emissions from coal mines are as reported by the US EPA, and include emissions
from underground coal mines, surface mines, and post-mining activities.
114
Coal mine CH
4
projections are based on coal production projections provided by a contact at the Colorado
Geological Survey.
115
As a simple assumption for projections, CH
4
emissions per unit coal
mined for the three (underground, surface, and post-mining) categories of mining emission
sources were assumed to hold constant at the 2004 level.
116
Results
Table E2 displays the estimated CH
4
emissions from the fossil fuel industry in Colorado from
1990 to 2005, with projections to 2020. Emissions from this sector grew by 35% from 1990 to
2005, and are projected to increase by a further 22% from 2005 to 2020. The natural gas industry
accounts for the majority of both GHG emissions and emissions growth in the fossil fuel industry
as a whole.
Table E2. Methane Emissions and Projections from the Fossil Fuel Industry
(Million Metric Tons CO2e) 1990 1995 2000 2005 2010 2015 2020
Fossil Fuel Industry 7.5 7.0 9.3 10.1 11.8 12.1 12.3
Natural Gas Industry 3.1 3.3 4.8 5.0 6.5 6.8 7.3
Total Methane Emissions (CH4) 3.1 3.3 4.8 5.0 6.5 6.8 7.3
Total Entrained (CO2) n/a n/a n/a n/a n/a n/a n/a
Production (methane emissions) 0.5 0.6 1.8 1.3 1.8 1.8 1.9
Processing 1.0 1.0 1.1 1.2 1.8 1.9 2.0
Methane Emissions (CH4) 1.0 1.0 1.1 1.2 1.8 1.9 2.0
Entrained Gas (CO2) n/a n/a n/a n/a n/a n/a n/a
Transmission (methane emissions) 1.0 1.0 1.1 1.4 1.9 1.9 2.0
Distribution (methane emissions) 0.6 0.7 0.9 1.1 1.1 1.2 1.4
Oil Industry 0.2 0.2 0.2 0.2 0.2 0.2 0.2
Production (methane emissions) 0.2 0.2 0.1 0.2 0.2 0.2 0.2
Refineries (methane emissions) 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Coal Mining (methane emissions) 4.2 3.5 4.3 4.9 5.1 5.1 4.8
Source: CCS calculations based on approach described in text.
113
EIA
114
Emissions from EPA Inventory of Greenhouse Gas Emissions and Sinks: 1990-2004 (April 2006)
http://yosemite.epa.gov/oar/globalwarming.nsf/content/ResourceCenterPublicationsGHGEmissions
USEmissionsInventory2006.html
115
Personal communication with Chris Carroll of the Colorado Geological Survey, December 2006. Coal production
estimates given in million short tons annually at the following rates: 2006: 35, 2007: 39, 2008-2015: 40, 2016-2020:
decline at 1% per year.
116
Based on 2004 EIA data, CH
4
emission intensity calculated to average 0.127 metric tons CO
2
e per short ton
produced coal.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
E-6 Center for Climate Strategies
www.climatestrategies.us
Figure E1 displays the CH
4
emissions from coal mining and natural gas and oil systems, on a
million metric tons (MMt) CO
2
e basis.
Figure E1. Fossil Fuel Industry Emission Trends (Million metric tonnes CO
2
e)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
1
9
9
0
1
9
9
2
1
9
9
4
1
9
9
6
1
9
9
8
2
0
0
0
2
0
0
2
2
0
0
4
2
0
0
6
2
0
0
8
2
0
1
0
2
0
1
2
2
0
1
4
2
0
1
6
2
0
1
8
2
0
2
0
M
M
t
C
O
2
e
Natural Gas Industry
Coal Mining
Oil Industy
Source: CCS calculations based on approach described in text.
Key Uncertainties
Key sources of uncertainty underlying the estimates above are as follows:
- The largest uncertainty in the GHG emission estimates is the future production of fossil
fuels. These industries are difficult to forecast, as they are affected by a mix of drivers,
including: economics, resource supply, fuels demand, technology development, and the
status of regulations applying to the industry, among others. The assumptions used for the
projections, projecting trends for the near-term and EIA’s Annual Energy Outlook 2006
(AEO2006) growth rates through 2020, do not include any significant changes in energy
prices, relative to today’s prices. Large price swings, resource limitations, or changes in
regulations could significantly change future production and the associated GHG
emissions.
- Other uncertainties with potentially significant GHG implications include the fraction of
entrained CO
2
in current and future CBM production, the presence or absence of any
commercial oil shale production, and potential emissions-reducing improvements in oil
and gas production, processing, and pipeline technologies.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
F-1 Center for Climate Strategies
www.climatestrategies.us
Appendix F. Agriculture
Overview
The emissions discussed in this appendix refer to non-energy methane (CH
4
) and nitrous oxide
(N
2
O) emissions from enteric fermentation, manure management, and agricultural soils.
Emissions and sinks of carbon in agricultural soils are also covered. This sector accounted for
about 8% of Colorado’s gross GHG emissions in 2005. Energy emissions related to agricultural
practices (combustion of fossil fuels to power agricultural equipment) are included in the
residential, commercial, and industrial (RCI) fuel consumption sector estimates.
There are two livestock sources of greenhouse gas (GHG) emissions: enteric fermentation and
manure management. Methane emissions from enteric fermentation are the result of normal
digestive processes in ruminant and non-ruminant livestock. Microbes in the animal digestive
system break down food and emit CH
4
as a by-product. More CH
4
is produced in ruminant
livestock than in other animals because of digestive activity in the large fore-stomach to break
down grasses and other high-fiber feeds. Methane and N
2
O emissions from the storage and
treatment of livestock manure (e.g., in compost piles or anaerobic treatment lagoons) occur as a
result of manure decomposition. The environmental conditions of decomposition drive the
relative magnitude of emissions. In general, the more anaerobic the conditions are, the more CH
4
is produced because decomposition is aided by CH
4
producing bacteria that thrive in oxygen-
limited (or oxygen-free) anaerobic conditions. Under aerobic conditions, N
2
O emissions are the
dominant GHG emissions of concern. Emissions estimates from manure management are based
on estimates of the volumes of manure that are stored and treated in livestock operations.
Emissions from manure that is applied to agricultural soils as an amendment or deposited
directly to pasture and grazing land by grazing animals are accounted for in inventories of
emissions from agricultural soils.
The management of agricultural soils can result in N
2
O emissions and in fluxes of carbon
dioxide (CO
2
) that make soils net emitters or net sinks of carbon. In general, soil amendments
that add nitrogen to soils can also result in N
2
O emissions. Nitrogen additions drive underlying
soil nitrification and de-nitrification cycles, which produce N
2
O as a by-product. The emissions
estimation methodologies used in this inventory account for several sources of N
2
O emissions
from agricultural soils, including decomposition of crop residues, synthetic and organic fertilizer
application, manure and sewage sludge application to soils, nitrogen fixation, and cultivation of
histosols (high organic soils, such as wetlands or peatlands). Both direct and indirect emissions
of N
2
O occur from the application of manure, fertilizer, and sewage sludge to agricultural soils.
Direct emissions occur at the site of application, and indirect emissions occur when nitrogen
leaches to groundwater or in surface runoff and is transported off-site before entering the
nitrification/denitrification cycle. Methane and N
2
O emissions also result when crop residues are
burned. Methane emissions occur during rice cultivation; however rice is not grown in Colorado.
The net flux of CO
2
in or out of agricultural soils depends on the balance of carbon losses from
management practices and gains from organic matter inputs to the soil. Carbon dioxide is
absorbed by plants through photosynthesis and ultimately becomes the carbon source for organic
matter inputs to agricultural soils. When inputs are greater than losses, the soil accumulates
carbon and there is a net sink of CO
2
into agricultural soils. Conversely, soil disturbance from the
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
F-2 Center for Climate Strategies
www.climatestrategies.us
cultivation of histosols releases large stores of carbon from the soil to the atmosphere. Finally,
the practice of adding limestone and dolomite to agricultural soils results in CO
2
emissions.
Emissions and Reference Case Projections
Methane and Nitrous Oxide
GHG emissions for 1990 through 2005 were estimated using the United States Environmental
Protection Agency’s (US EPA) State Greenhouse Gas Inventory Tool (SGIT) and the methods
provided in the Emission Inventory Improvement Program (EIIP) guidance document for the
sector.
117
In general, the SGIT methodology applies emission factors developed for the US to
activity data for the agriculture sector. Activity data include livestock population statistics,
amounts of fertilizer applied to crops, and trends in manure management practices. This
methodology is based on international guidelines developed by sector experts for preparing GHG
emissions inventories.
118
Data on crop production in Colorado from 1990 to 2005 and on the number of animals in the
state from 1990 to 2002 were obtained from the USDA National Agriculture Statistical Service
(NASS) and incorporated as defaults in SGIT.
119
Data on fertilizer usage came from Commercial
Fertilizers, a report from the Fertilizer Institute. The default data in SGIT accounting for the
percentage of each livestock category using each type of manure management system was
revised based on local data.
120
Emissions from enteric fermentation and manure management
were estimated based on the annual growth rate in emissions (million metric ton (MMt) carbon
dioxide equivalent (CO
2
e) basis) associated with historical livestock populations in Colorado for
1990 to 2002. The dairy cattle population was assumed to continue growing at the historical
(1990-2002) growth rate of 1.8% per year. All other livestock populations are assumed to remain
at 2002 levels.
121
Colorado has one hog farm of approximately 5,000 hogs that employs a
manure digester. However, the emission reduction (0.001 MMtCO
2
e) from this operation does
not have a significant effect on the statewide manure management emissions.
Crop production data from the United States Department of Agriculture (USDA) National
Agricultural Statistics Service (NASS) were available through 2005; therefore, N
2
O emissions
from crop residues and crops that use nitrogen (nitrogen fixation) were calculated through 2005.
Emissions for the other agricultural crop production practices categories (synthetic and organic
fertilizers, agricultural residue burning) were calculated through 2002. Data were not available to
117
GHG emissions were calculated using SGIT, with reference to EIIP, Volume VIII: Chapter 8. “Methods for
Estimating Greenhouse Gas Emissions from Livestock Manure Management”, August 2004; Chapter 10. “Methods
for Estimating Greenhouse Gas Emissions from Agricultural Soil Management”, August 2004; and Chapter 11.
“Methods for Estimating Greenhouse Gas Emissions from Field Burning of Agricultural Residues”, August 2004.
118
Revised 1996 1ntergovermental Panel on Climate Change Guidelines for National Greenhouse Gas Inventories,
published by the National Greenhouse Gas Inventory Program of the IPCC, available at http://www.ipcc-
nggip.iges.or.jp/public/gl/invs1.htm; and Good Practice Guidance and Uncertainty Management in National
Greenhouse Gas Inventories, published in 2000 by the National Greenhouse Gas Inventory Program of the IPCC,
available at: http://www.ipcc-nggip.iges.or.jp/public/gp/english/.
119
USDA, NASS (http://www.nass.usda.gov/Statistics_by_State/Colorado/index.asp).
120
Jessica Davis, Department of Soil and Crop Sciences, Colorado State University provided revised values for
waste management system percentages.
121
Renee Picanso, Colorado Agricultural Statistics Service, recommended projecting dairy cattle populations based
on historical data and assuming no growth for beef cattle.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
F-3 Center for Climate Strategies
www.climatestrategies.us
estimate nitrogen released by the cultivation of histosols (i.e., the number of acres of high
organic content soils). SGIT data indicate that agricultural residue burning is not a common
practice in Colorado agriculture. These data indicate a small degree of residue burning for barley,
corn and wheat. Another source of information on this topic is a 2002 Western Regional Air
Partnership (WRAP)-sponsored study on agricultural burning in the western US.
122
This study
indicated that only a small amount of residue (2,000 tons) burned for Spring wheat on average
during the mid-1990’s. The Center for Climate Strategies (CCS) used the SGIT data to estimate
CH
4
and N
2
O emissions from agricultural burning.
Historical emissions from agricultural soils, based on USDA NASS and Fertilizer Institute data,
do not show a significant positive or negative trend. Therefore, total emissions for this source
were held constant from the latest year of historical data to 2020.
Soil Carbon
Carbon dioxide is either emitted or sequestered as a result of agricultural practices. Net carbon
fluxes from agricultural soils have been estimated by researchers at the Natural Resources
Ecology Laboratory at Colorado State University, and are reported in the U.S. Inventory of
Greenhouse Gas Emissions and Sinks
123
and the U.S. Agriculture and Forestry Greenhouse Gas
Inventory. The estimates are based on the IPCC methodology for soil carbon adapted to
conditions in the US Preliminary state-level estimates of CO
2
fluxes from mineral soils and
emissions from the cultivation of organic soils were reported in the U.S. Agriculture and
Forestry Greenhouse Gas Inventory. Currently, these are the best available data at the state-level
for this category. The inventory did not report state-level estimates of CO
2
emissions from
limestone and dolomite applications; hence, this source is not included in this inventory at
present.
Carbon dioxide fluxes resulting from specific management practices were reported. These
practices include: conversions of cropland resulting in either higher or lower soil carbon levels;
additions of manure; participation in the Federal Conservation Reserve Program (CRP); and
cultivation of organic soils (with high organic carbon levels). For Colorado, Table F1 shows a
summary of the latest estimates available from the USDA.
124
The latest data available are for
1997 agricultural practices. These data show that changes in agricultural practices are estimated
to result in a net sink of 2.0 MMtCO
2
e/yr in Colorado. Since data are not yet available from
USDA to make a determination of whether the emissions are increasing or decreasing, the net
sink of 2.0 MMtCO
2
e/yr is assumed to remain constant.
122
Non-Burning Management Alternatives on Agricultural Lands in the Western United States, Volume I:
Agricultural Crop Production and Residue Burning in the Western United States, Eastern Research Group, 2002.
123
U.S. Inventory of Greenhouse Gas Emissions and Sinks: 1990-2004 (and earlier editions), US Environmental
Protection Agency, Report # 430-R-06-002, April 2006. Available at:
http://www.epa.gov/climatechange/emissions/usinventoryreport.html.
124
U.S. Agriculture and Forestry Greenhouse Gas Inventory: 1990-2001. Global Change Program Office, Office of
the Chief Economist, US Department of Agriculture. Technical Bulletin No. 1907. 164 pp. March 2004.
http://www.usda.gov/oce/global_change/gg_inventory.htm; the data are in appendix B table B-11. The table
contains two separate IPCC categories: “carbon stock fluxes in mineral soils” and “cultivation of organic soils.” The
latter is shown in the second to last column of Table F1. The sum of the first nine columns is equivalent to the
mineral soils category.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
F-4 Center for Climate Strategies
www.climatestrategies.us
Results
As shown in Figure F1, gross emissions from agricultural sources range between about 8.7 and
8.9 MMtCO
2
e from 1990 through 2005, but do not show a significant positive or negative trend.
By 2020, emissions are projected to increase to about 9.1 MMtCO
2
e associated with an expected
increase in the dairy cattle population. The historic and projected emissions from the agriculture
sector account for about 6% of total gross GHG emissions in 2020. Including the CO
2
sequestration from soil carbon changes, the historic and projected emissions for the agriculture
sector on a net basis would range between about 6.7 and 7.1 MMtCO
2
e/yr.
Table F1. GHG Emissions from Soil Carbon Changes Due to Cultivation Practices
(MMtCO
2
e)
Changes in cropland Changes in Hayland Other Total
4
Plowout of
grassland
to annual
cropland
1
Cropland
manage-
ment
Other
cropland
2
Cropland
converted
to
hayland
3
Hayland
manage-
ment
Cropland
converted
to grazing
land
3
Grazing
land
manage-
ment CRP
Manure
application
Cultivation
of organic
soils
Net soil
carbon
emissions
0.77 (0.15) 0.00 (0.55) (0.04) (0.26) 0.00 (1.25) (0.53) 0.00 (2.00)
Based on USDA 1997 estimates. Parentheses indicate net sequestration.
1
Losses from annual cropping systems due to plow-out of pastures, rangeland, hayland, set-aside lands, and perennial/horticultural
cropland (annual cropping systems on mineral soils, e.g., corn, soybean, cotton, and wheat).
2
Perennial/horticultural cropland and rice cultivation.
3
Gains in soil carbon sequestration due to land conversions from annual cropland into hay or grazing land.
4
Total does not include change in soil organic carbon storage on federal lands, including those that were previously under private
ownership, and does not include carbon storage due to sewage sludge applications.
Figure F1. Gross GHG Emissions from Agriculture (Ag)
0
2
4
6
8
10
12
1990 1995 2000 2005 2010 2015 2020
Year
M
M
t
C
O
2
e
Enteric Fermentation Manure Management
Ag Soils - Crops Ag Soils - Livestock
Ag Soils - Fertilizer Ag Residue Burning
Source: CCS calculations based on approach described in text.
Notes: Ag Soils – Crops category includes: incorporation of crop residues and nitrogen fixing crops (no cultivation
of histosols estimated in Colorado); emissions for agricultural residue burning are too small to be seen in this chart.
Soil carbon sequestration is not shown (see Table F1).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
F-5 Center for Climate Strategies
www.climatestrategies.us
Agricultural burning emissions were estimated to be very small based on the SGIT activity data
(<0.01 MMtCO
2
e/yr from 1990 to 2002). This agrees with the USDA Inventory which also
reports a low level of residue burning emissions (0.02 MMtCO
2
e).
The only standard IPCC source categories missing from this report are N
2
O emissions from
cultivation of histosols and CO
2
emissions from limestone and dolomite application. Estimates
for Colorado were not available; however the USDA’s national estimate for soil liming is about
9 MMtCO
2
e/yr.
125
Key Uncertainties
Emissions from enteric fermentation and manure management are dependent on the estimates of
animal populations and the various factors used to estimate emissions for each animal type and
manure management system (i.e., emission factors that are dependent on several variables,
including manure production levels, volatile solids contents of manures, and CH
4
formation
potential). Each of these factors has some level of uncertainty. Also, animal populations fluctuate
throughout the year, and thus using point estimates introduces uncertainty into the average
annual estimates of these populations. In addition, there is uncertainty associated with the
original population survey methods employed by USDA. The largest contributors to uncertainty
in emissions from manure management are the emission factors, which are derived from limited
data sets.
As mentioned above, for emissions associated with changes in agricultural soil carbon levels, the
only data currently available are for 1997. When newer data are released by the USDA, these
should be reviewed to represent current conditions as well as to assess trends. In particular, given
the potential for some CRP acreage to retire and possibly return to active cultivation prior to
2020, the current size of the CO
2
sink could be appreciably affected. As mentioned above,
emission estimates for soil liming have not been developed for Colorado.
Another contributor to uncertainty in the emission estimates is the projection assumptions. This
inventory assumes that the dairy cattle population will grow at the historical rate. This population
could, however level off or begin to decline before 2020, due to factors such as competition for
water and feed.
125
U.S. Agriculture and Forestry Greenhouse Gas Inventory: 1990-2001. Global Change Program Office, Office of
the Chief Economist, US Department of Agriculture.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
G-1 Center for Climate Strategies
www.climatestrategies.us
Appendix G. Waste Management
Overview
GHG emissions from waste management include emissions from:
- Solid waste management – methane (CH
4
) emissions from municipal and industrial solid
waste landfills (LFs), accounting for CH
4
that is flared or captured for energy production
(this includes both open and closed landfills);
- Solid waste combustion – CH
4
, carbon dioxide (CO
2
), and nitrous oxide (N
2
O) emissions
from the combustion of solid waste in incinerators or waste to energy plants; and
- Wastewater management – CH
4
and N
2
O from municipal wastewater and CH
4
from
industrial wastewater (WW) treatment facilities.
Waste management activities accounted for about 2% of Colorado’s gross GHG emissions in
2005.
Inventory and Reference Case Projections
Solid Waste Management
For solid waste management, we used the US EPA SGIT software and the US EPA Landfill
Methane Outreach Program (LMOP) landfills database
126
as starting points to estimate
emissions. The LMOP data serve as input data to estimate annual waste emplacement for each
landfill modeled by SGIT. SGIT then estimates CH
4
generation for each landfill site. Additional
post-processing outside of SGIT to account for controls is then performed to estimate final CH
4
emissions.
The LMOP database was shared with CDPHE solid waste staff, and CCS was supplied with
additional data on Colorado landfills. These additional data included information on many sites
that were not present in the LMOP database, as well as updated information on sites that were
present in the database (e.g. waste emplacement data, information on controls).
127
In the
combined LMOP and CDPHE dataset for Colorado, there are over 70 sites represented (both
open and closed landfills). Small uncontrolled landfills in some counties were combined for the
purposes of emissions modeling. Two of these sites collect landfill gas for use in a landfill gas to
energy (LFGTE) plant. Another six sites collect and flare landfill gas. These eight sites are listed
in the Table G1. The rest of the sites were assumed to be uncontrolled.
To obtain the annual waste emplacement rate needed by SGIT for each landfill, the waste-in-
place estimate was divided by the number of years of operation. This average annual disposal
rate for each landfill was assumed for all years that the landfill was operating. For sites where the
years of operation were not available, CCS estimated these years based on the waste-in-place
estimate divided by the latest (2006) annual emplacement rate (note that for 2006, CDPHE stated
126
LMOP database is available at: http://www.epa.gov/lmop/proj/index.htm. Updated version of the database
provided by Rachel Goldstein, Program Manager, EPA Landfill Methane Outreach Program, October 2006.
127
Charles Johnson, CDPHE, Solid Waste Division, personal communication and electronic data provided to S.
Roe, CCS, December 2006.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
G-2 Center for Climate Strategies
www.climatestrategies.us
that these were only partial data; CCS did not, however, make any adjustments to reflect that
some emplacement data might not be for a full year). Data were available to estimate annual
emplacement for all but six sites in the combined LMOP-CDPHE dataset.
128
Table G1. Colorado Landfills with Controls
Site Name County Control
County Line LF Arapahoe Flare
Fountain LF El Paso Flare
Foothills LF Jefferson Flare
Denver Regional North LF Weld Flare
Denver Regional South LF Weld Flare
Tower LF Adams Flare
Denver – Arapahoe Arapahoe LFGTE
Boulder LF Boulder LFGTE
CCS performed three different runs of SGIT to estimate emissions from municipal solid waste
(MSW) landfills: (1) uncontrolled landfills; (2) landfills with a landfill gas collection system and
LFGTE plant; and (3) landfills with landfill gas collection and a flare. SGIT produced annual
estimates through 2005 for each of these landfill categories. CCS then performed some post-
processing of the landfill emissions to account for landfill gas controls (at LFGTE and flared
sites) and to project the emissions through 2020. For the controlled landfills, CCS assumed that
the overall methane collection and control efficiency is 75%.
129
Based on input received from stakeholders, CCS revised the initial draft emission estimates for
landfills by using a new version of EPA’s SGIT tool, which became available after the initial
estimates were made. Also based on stakeholder input, CCS excluded the assumption that 10%
of the CH
4
is oxidized in the surface layers of the landfill. The new version of SGIT uses a
different method for quantifying CH
4
generation, which led to significantly lower landfill CH
4
estimates than the original SGIT version.
Growth rates were estimated by using the historic (1995-2005) growth rates of emissions in both
the controlled and uncontrolled landfill categories. The period from 1995 to 2005 was used since
there were a large number of landfill closures during the period from 1987 to 1995 (which could
have affected waste management practices). Hence, the post-1995 period is thought to be most
representative of waste emplacement rates in the future and subsequent emissions. The annual
growth rates are: 3.9% for uncontrolled sites; 2.1% for flared sites; and 1.2% for LFGTE
landfills. The higher growth rate for uncontrolled sites appears to be driven by the establishment
of many small sites located away from population centers during the past 10-12 years.
CCS used the SGIT default for industrial solid waste landfills. This default is based on national
data indicating that industrial landfilled waste is emplaced at approximately 7% of the rate of
MSW emplacement. We assumed that this additional industrial waste emplacement occurs
beyond that already addressed in the emplacement rates for MSW sites described above. Due to a
128
One of these sites is listed as the Lowry Superfund LF (county unknown). It is listed as a flared site, but no data
were available to estimate methane generation.
129
As per EPA’s AP-42 Section on Municipal Solid Waste Landfills:
http://www.epa.gov/ttn/chief/ap42/ch02/final/c02s04.pdf.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
G-3 Center for Climate Strategies
www.climatestrategies.us
lack of data, no controls were assumed to be used on industrial waste landfills in Colorado. For
industrial landfills, the overall growth rate in MSW emissions from 1995 to 2005 (2.6%/yr) was
used to project emissions to 2010 and 2020 (based on the assumption that industrial waste
landfilling will continue to grow at the same rate as MSW landfilling overall).
Solid Waste Combustion
CDPHE staff indicated that little to no municipal solid waste combustion has occurred during the
1990 – 2005 time-frame.
130
This was true for both combustion in municipal solid waste
combustors and open burning of municipal solid waste by Colorado residents. Hence, emissions
for both the inventory and forecast for this sector are estimated to be negligible.
Wastewater Management
GHG emissions from municipal and industrial wastewater treatment were also estimated. For
municipal wastewater treatment, emissions are calculated in EPA’s SGIT based on state
population, assumed biochemical oxygen demand (BOD) and protein consumption per capita,
and emission factors for N
2
O and CH
4
from wastewater treatment The key SGIT default values
are shown in Table G2.
For industrial wastewater emissions, SGIT provides default assumptions and emission factors for
three industrial sectors: Fruits & Vegetables, Red Meat & Poultry, and Pulp & Paper. CDPHE
provided information on flows for the meat and poultry sector.
131
Emissions from wastewater
treatment for the other two industrial sectors are assumed to be negligible. The emissions were
held constant at the 2005 level for the forecast years. This is because the data showed no
significant growth in activity since the late 1990s and a reduction in activity in 2005.
Table G2. SGIT Key Default Values for Municipal Wastewater Treatment
Variable Value
BOD 0.065 kg /day-person
Amount of BOD anaerobically treated 16.25%
CH
4
emission factor 0.6 kg/kg BOD
Colorado residents not on septic 75%
Water treatment N
2
O emission factor 4.0 g N
2
0/person-yr
Biosolids emission Factor 0.01 kg N
2
O-N/kg sewage-N
Source: US EPA State Inventory Tool – Wastewater Module; methodology and factors taken
from US EPA, Emission Inventory Improvement Program, Volume 8, Chapter 12, October
1999: www.epa.gov/ttn/chief/eiip/techreport/volume08/.
Figure G1 shows the emission estimates for the waste management sector. Overall, the sector
accounts for 2.1 MMtCO
2
e in 2005. By 2020, emissions are expected to grow to 3.5
MMtCO
2
e/yr. The growth in emissions is driven largely by the solid waste management sector,
in particular uncontrolled landfills. In 2005, about 48% of the emissions were contributed by the
130
Matt Burgett and Dale Wells, CDPHE, personal communications with S. Roe, CCS, January 2007.
131
Leslie Simpson, CDPHE, personal communication with S. Roe, CCS, November 2006 and electronic data on
industrial wastewater flows. Data for two plants were provided (Cargill Meat Solutions, and Swift Beef). The total
flows for these facilities in each year were used to estimate the amount of meat processed using the SGIT default of
13 cubic meters/Mt processed. The estimated production values were then used within SGIT to estimate methane
emissions. Process wastewater flow data were available for 1990-2005.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
G-4 Center for Climate Strategies
www.climatestrategies.us
uncontrolled landfills sector. By 2020, the contribution from these sites is expected to be about
54%.
As mentioned above, due to data availability, CCS modeled only emissions from meat and
poultry processors in the industrial wastewater treatment sector (and these emissions were held
constant at 2005 levels for the forecast). Only about 4% of the emissions were contributed by the
industrial wastewater treatment sector in 2005. In 2005, about 24% of the waste management
sector emissions were contributed by municipal wastewater treatment systems. Note that these
estimates are based on the default parameters listed in Table G2 above, and might not adequately
account for existing controls or management practices (e.g. anaerobic digesters served by a flare
or other combustion device). By 2020, municipal wastewater treatment is expected to contribute
about the same amount (21%) of the waste management sector emissions, with the reduction in
the fraction of overall emissions ascribed to the wastewater treatment due to the large increase
projected for the emissions from the solid waste sector.
Figure G1. Colorado GHG Emissions from Waste Management
0
0.5
1
1.5
2
2.5
3
3.5
4
1990 1995 2000 2005 2010 2015 2020
M
M
t
C
O
2
e
Uncontrolled LFs Flared LFs
LFGTE LFs Industrial LFs
MSW Combustion Municipal WW
Industrial WW
Source: CCS calculations based on approach described in text.
Notes: LF – landfill; WW – wastewater; LFGTE – landfill gas to energy; emissions for solid waste
combustion were estimated to be negligible.
Key Uncertainties
The methods used to model landfill gas emissions do not adequately account for the points in
time when controls were applied at individual sites. Hence, for landfills, the historical emissions
are less certain than current emissions and future emissions (since each site that is currently
controlled was modeled as always being controlled, the historic emissions estimates are lower
than they should be as a result). The modeling also does not account for uncontrolled sites that
will need to apply controls during the period of analysis due to triggering the requirements of the
federal New Source Performance Standards/Emission Guidelines. As noted above, the available
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
G-5 Center for Climate Strategies
www.climatestrategies.us
data do not cover all of the open and closed landfills in Colorado. For this reason, emissions
could be slightly underestimated for landfills.
For industrial landfills, emissions were estimated using national defaults (with industrial landfill
wastes buried at 7% of the rate of MSW emplacement). It could be that the available MSW
emplacement data within the combined LMOP data used to model the MSW emissions already
captures industrial LF emplacement. As with overall MSW landfill emissions, industrial landfill
emissions are projected to increase between 2005 and 2020. Hence, the industrial landfill
inventory and forecast has a significant level of uncertainty and should be investigated further.
For example, the existence of active industrial landfills that are not already represented in the
LMOP database should be determined. If these sites do not exist and the existing municipal
waste emplacement data are thought to include industrial wastes, then the separate estimate for
industrial landfill emissions can be excluded from the inventory.
For the wastewater sector, the key uncertainties are associated with the application of SGIT
default values for the parameters listed in Table G2 above (e.g. the fraction of the Colorado
population on septic sewers; and the fraction of BOD that is anaerobically decomposed). The
SGIT defaults for emission factors used to estimate wastewater emissions were derived from
national data.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
H-1 Center for Climate Strategies
www.climatestrategies.us
Appendix H. Forestry
Overview
Forestland emissions refer to the net carbon dioxide (CO
2
) flux
132
from forested lands in
Colorado, which account for about 34% of the state’s land area.
133
Forestlands are net sinks of
CO
2
in Colorado. Through photosynthesis, CO
2
is taken up by trees and plants and converted to
carbon in biomass within the forests. CO
2
emissions occur from respiration in live trees and
decay of dead biomass. In addition, carbon is stored for long time periods when forest biomass is
harvested for use in durable wood products. CO
2
flux is the net balance of CO
2
removals from
and emissions to the atmosphere from the processes described above.
Inventory and Reference Case Projections
For over a decade, the United State Forest Service (USFS) has been developing and refining a
forest carbon modeling system for the purposes of estimating forest carbon inventories. The
methodology is used to develop national forest CO
2
fluxes for the official U.S. Inventory of
Greenhouse Gas Emissions and Sinks.
134
The national estimates are compiled from state-level
data. The Colorado forest CO
2
flux data in this report come from the national analysis and are
provided by the USFS.
The forest CO
2
flux methodology relies on input data in the form of plot-level forest volume
statistics from the Forest Inventory Analysis (FIA). FIA data on forest volumes are converted to
values for ecosystem carbon stocks (i.e., the amount of carbon stored in forest carbon pools)
using the FORCARB2 modeling system. Coefficients from FORCARB2 are applied to the plot
level survey data to give estimates of carbon density (megagrams of C per hectare) for a number
of separate carbon pools.
CO
2
flux is estimated as the change in carbon mass for each carbon pool over a specified time
frame. Forest volume data from at least two points in time are required. The change in carbon
stocks between time intervals is estimated at the plot level for specific carbon pools (Live Tree,
Standing Dead Wood, Under-story, Down & Dead Wood, Forest Floor, and Soil Organic
Carbon) and divided by the number of years between inventory samples. Annual increases in
carbon density reflect carbon sequestration in a specific pool; decreases in carbon density reveal
CO
2
emissions or carbon transfers out of that pool (e.g., death of a standing tree transfers carbon
from the live tree to standing dead wood pool). The amount of carbon in each pool is also
influenced by changes in forest area (e.g., an increase in area could lead to an increase in the
associated forest carbon pools and the estimated flux). The sum of carbon stock changes for all
forest carbon pools yields a total net CO
2
flux for forest ecosystems.
132
“Flux” refers to both emissions of CO
2
to the atmosphere and removal (sinks) of CO
2
from the atmosphere.
133
2001 Report on the Condition of Colorado’s Forests, Colorado Forestry Advisory Board,
http://csfs.colostate.edu/library/pdfs/fhr/01fhr.pdf, reports 22.6 million acres of forested lands. The total land area in
CO is 66.4 million acres (http://www.netstate.com/states/geography/co_geography.htm).
134
U.S. Inventory of Greenhouse Gas Emissions and Sinks: 1990-2004 (and earlier editions), US Environmental
Protection Agency, Report # 430-R-06-002, April 2006. Available at:
http://www.epa.gov/climatechange/emissions/usinventoryreport.html
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
H-2 Center for Climate Strategies
www.climatestrategies.us
In preparing these estimates, USFS estimates the amount of forest carbon in different forest types
and as well as in different carbon pools. The different forests include those in the national forest
system and those that are not federally-owned (private and other public forests). USFS also
provides information on forests categorized as being either woodlands (forests with low
productivity) and non-woodlands (e.g. timberlands or productive forest systems).
Carbon pool data for two periods are used to estimate CO
2
flux for each pool. The data shown in
Table H1 are a summary of the FIA data used to derive the carbon pool and flux estimates that
are summarized in Table H2. As shown in Table H1, the current forest carbon pool estimates are
derived from 2004 FIA data. The previous inventory data came from either a previous FIA cycle
or data from the Resources Planning Act Assessment (RPA). The Resources Planning Act
requires the USFS to report on the state of US forest resources on a regular basis; the USFS
publishes the RPA assessment every five years. FIA is a key contributor to RPA. RPA data,
which are generally lower in resolution, are sometimes used in place of FIA cycles. The FIA has
transitioned from a periodic to annual sampling design, which has created some forest inventory
data sets that are not comparable over time, in which case the RPA data are better suited for
estimating carbon densities.
135
Except for the National Forests – Woodlands category, the
interval between the current and previous surveys is around 22 years (early 1980s to 2004; data
from the 1997 RPA report are from the early 1980s (the report contains a time series dating back
to the 1950s).
Table H1. Forestry Data Used to Estimate Forest CO
2
Flux
Forest
Past Inventory
Source
1
Avg.
Year
2
Interval
3
(yr)
Current
Forest Area
(10
3
hectares)
Previous
Forest Area
(10
3
hectares)
National Forests –
Non-Woodlands
RPAdata_CO_1997 2004.2 22.9 4,219 4,020
National Forests –
Woodlands
FISDB21_CO_01_1984 2004.3 6.9 366 174
Non-National Forests –
Non-Woodlands
FIADBWW_CO_1983 2004.2 23.7 2,067 1,975
Non-National Forests –
Woodlands
FIADBWW_CO_1983 2004.2 20.9 2,774 2,351
Totals 9,425 8,519
1
Current Inventory Source for all groups was FISDB21_CO_02_2005.
2
Average year for the current FIA inventory data.
3
The number of years between the current inventory source and the past inventory source.
The data in Table H1 show an increase of 906 kilo-hectares (2.2 million acres) in forested area
from the early-1980s through 2004. About two-thirds of this increase occurred in woodland
forests (as mentioned under key uncertainties below, some of this difference is likely driven by
methodological differences in survey methods).
135
Jim Smith, USFS, personal communication with K. Bickel, CCS, November 7, 2006.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
H-3 Center for Climate Strategies
www.climatestrategies.us
Table H2 provides a summary of the size of the forest carbon pools for the final survey period
and the resultant flux estimates (in units of carbon and CO
2
) developed by the USFS.
3
A total of
31 million metric tons (MMt) CO
2
is estimated to be sequestered in Colorado forests each year,
with about half of this accumulating in the live tree carbon pool. The soil organic carbon and
forest floor pools had the next largest accumulations of carbon at 7 and 5 MMtCO
2
/yr,
respectively. Note that this analysis averages out annual fluctuations in carbon sequestration
rates over an approximately 20-year time interval.
In addition to the forest carbon pools, additional carbon stored as biomass is removed from the
forest for the production of durable wood products; carbon remains stored in the products pool or
is transferred to landfills where much of the carbon remains stored over long period of time. An
estimated 0.8 MMt carbon dioxide equivalent (CO
2
e) is sequestered annually in wood products;
these data are based on a USFS study from 1987 to 1997.
136
.Additional details on all of the forest
carbon inventory methods can be found in Annex 3 to United States Environmental Protection
Agency’s (EPA) 2006 GHG inventory for the US.
137
Recent discussions with the USFS have indicated that there is considerable uncertainty with the
soil organic carbon flux estimates. Due to this uncertainty, their recommendation is to leave this
flux out of the statewide totals for carbon flux. In Table H2, a total forest flux which excludes the
soil organic carbon pool has been provided (-24.7 MMtCO
2
), and this estimate has been used in
the summary tables at the front of this report.
For the 1990 and 2000 historic emission estimates, as well as the reference case projections, the
forest area and carbon densities of forestlands were assumed to be at the same levels as those
shown in Table H2. Hence, there is no change in the estimated future sinks for 2010 and 2020.
These assumptions could change based on feedback from project stakeholders and other state
forestry experts.
In order to provide a more comprehensive understanding of GHG sources/sinks from the forestry
sector, the Center for Climate Strategies (CCS) also developed some rough estimates of state-
wide emissions for methane (CH
4
) and nitrous oxide (N
2
O) from wildfires and prescribed burns.
A study published earlier this year in Science indicated an increasing frequency of wildfire
activity in the western US driven by a longer fire season and higher temperatures.
138
136
http://www.fs.fed.us/ne/global/pubs/books/epa/states.
137
Annex 3 to US EPA’s 2006 report can be downloaded at:
http://yosemite.epa.gov/oar/globalwarming.nsf/UniqueKeyLookup/RAMR6MBLNQ/$File/06_annex_Chapter3.pdf.
138
Westerling, A.L. et al, “Warming and Earlier Spring Increases Western US Forest Wildfire Activity”,
Sciencexpress, July 6, 2006.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
H-4 Center for Climate Strategies
www.climatestrategies.us
Table H2. Forestry CO
2
Flux Estimates for Colorado
Carbon Pool (MMt Carbon)
Forest Live Tree
Standing
Dead
Under-
story
Down &
Dead
Forest
Floor
Soil
Organic
Carbon
National Forests –
Non-Woodlands 346.1 46.0 9.2 27.0 137.0 152.6
National Forests – Woodlands 9.0 0.4 1.5 0.4 9.0 11.4
Non-National Forests –
Non-Woodlands 107.3 13.1 5.5 8.0 55.3 63.8
Non-National Forests –
Woodlands 73.3 1.0 9.2 2.6 61.5 68.7
Totals
535.7 60.5 25.3 38.0 262.8 296.5
Carbon Pool Flux (MMt carbon/year)
Forest Live Tree
Standing
Dead
Under-
story
Down &
Dead
Forest
Floor
Soil
Organic
Carbon
National Forests –
Non-Woodlands -2.72 -0.32 -0.02 -0.20 -0.30 -0.43
National Forests – Woodlands -0.70 -0.03 -0.11 -0.03 -0.71 -0.88
Non-National Forests –
Non-Woodlands -0.03 -0.02 -0.01 0.00 0.05 0.21
Non-National Forests –
Woodlands -0.73 -0.02 -0.10 -0.03 -0.47 -0.82
Totals -4.18 -0.40 -0.24 -0.27 -1.42 -1.92
Carbon Pool Flux (MMt CO
2
/yr)
Forest
Live Tree
Standing
Dead
Under-
story
Down &
Dead
Forest
Floor
Soil
Organic
Carbon
National Forests –
Non-Woodlands -10.0 -1.2 -0.1 -0.7 -1.1 -1.6
National Forests – Woodlands -2.6 -0.1 -0.4 -0.1 -2.6 -3.2
Non-National Forests –
Non-Woodlands -0.1 -0.1 0.0 0.0 0.2 0.8
Non-National Forests –
Woodlands -2.7 -0.1 -0.4 -0.1 -1.7 -3.0
Totals -15.3 -1.5 -0.9 -1.0 -5.2 -7.1
Total Forest Flux = -30.9
Harvested Wood Products
1
= -0.8
Total Statewide Flux = -31.8
Total Excluding Soil Organic
Carbon =
-24.7
NOTE: Totals may not add exactly due to rounding.
1
Source: http://www.fs.fed.us/ne/global/pubs/books/epa/states; For Colorado, HWP are estimated to sequester
0.22 MMtC during the period 1987-1997).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
H-5 Center for Climate Strategies
www.climatestrategies.us
CCS used 2002 emissions data developed by the Western Regional Air Partnership (WRAP) to
estimate CO
2
e emissions for wildfires and prescribed burns.
139
The CO
2
e from CH
4
emissions
from this study were added to an estimate of CO
2
e for N
2
O to estimate a total CO
2
e for fires (the
CO
2
emissions from fires are captured within the carbon pool accounting methods described
above). The nitrous oxide estimate was made assuming that N
2
O was 1% of the emissions of
nitrogen oxides (NO
x
) from the WRAP study. The 1% estimate is a common rule of thumb for
the N
2
O content of NO
x
from combustion sources. CCS is soliciting feedback on this
assumption. The results for 2002 are that fires contributed over 1.2 MMtCO
2
e of CH
4
and NO
x
from about 511,000 acres burned (495,000 acres by wildfires). About 94% of the CO
2
e was
contributed by CH
4
. Note that this level of activity compares to less than 90,000 acres burned in
Colorado in 1996.
140
Given the large swings in fire activity from year to year and the current lack
of data for multiple years, CCS will discuss this issue with project stakeholders before including
the fire emission estimates in the GHG inventory.
Key Uncertainties
It is important to note that there were methodological differences in the two FIA cycles that can
produce different estimates of forested area and carbon density. In the Rocky Mountain Region,
the FIA program modified the definition of forest cover for the woodlands class of forestland.
Earlier FIA cycles defined woodlands as having a tree cover of at least 10%, while the newer
sampling methods used a woodlands definition of tree cover of at least 5% (leading to more area
being defined as woodland). In woodland areas, the earlier FIA surveys might not have
inventoried trees of certain species or with certain tree form characteristics (leading to
differences in both carbon density and forested acreage). Also, surveys since 1999 include all
dead trees on the plots, but data prior to that are variable in terms of these data. The
modifications to FIA surveys are a result of an expanded focus in the FIA program, which
historically was only concerned with timber resources, while more recent surveys have aimed at
a more comprehensive gathering of forest biomass data.
The effect of these changes in survey methods has not been estimated by the USFS. In states like
Colorado with significant areas of woodlands, the change in definition could contribute
significantly to the increases seen in forested area in Colorado and the associated CO
2
flux. For
these reasons, the USFS provided flux estimates separately for woodlands and non-woodlands
(i.e., all other forest classes), so that the relative influence of these classes on total net CO
2
fluxes
in Colorado could be discerned. As shown in Table H2, the contribution from the woodland
areas drives a significant fraction of the flux estimate statewide. Given the differences in FIA
survey methods, the forest flux estimates should be viewed as conservatively high for Colorado
forests (i.e., more likely to have a higher than lower magnitude of CO
2
flux into forest lands).
As mentioned above, CCS included the forestry estimates without the soil carbon pool in the
emissions summary tables (see Tables ES-1 and Table 1) for this report, since the USFS has
indicated a high level of uncertainty for this carbon pool. These uncertainties are likely to remain
until additional data from measurements and potentially improved modeling methods are
developed.
139
2002 Fire Emission Inventory for the WRAP Region Phase I – Essential Documentation, prepared by Air
Sciences, Inc., June 2004.
140
1996 Fire Emission Inventory, Draft Final Report, prepared by Air Sciences, Inc., December 2002.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
I-1 Center for Climate Strategies
www.climatestrategies.us
Appendix I. Inventory and Forecast for Black Carbon
This appendix summarizes the methods, data sources, and results of the development of an
inventory and forecast for black carbon (BC) emissions in Colorado. Black carbon is an aerosol
(particulate matter or PM) species with positive climate forcing potential but currently without a
global warming potential defined by the Intergovernmental Panel on Climate Change (IPCC; see
Appendix J for more information on black carbon and other aerosol species). BC is synonymous
with elemental carbon (EC), which is a term common to regional haze analysis. An inventory for
2002 was developed based on inventory data from the Western Regional Air Partnership
(WRAP) regional planning organization and other sources. This appendix describes these data
and methods for estimating mass emissions of BC and then transforming the mass emission
estimates into carbon dioxide (CO
2
) equivalents (CO
2
e) in order to present the emissions within a
greenhouse gas (GHG) context.
In addition to the PM inventory data from WRAP, PM speciation data from the United States
Environmental Protection Agency’s (US EPA) SPECIATE database were also used: these data
include PM fractions of EC (also known as BC) and primary organic aerosols (also known as
organic material or OM). These data come from ongoing work being conducted by E.H. Pechan
& Associates, Inc. (Pechan) for US EPA on updating the SPECIATE database.
141
These new
profiles have just recently been released by US EPA. As will be further described below, both
BC and OM emission estimates are needed to assess the CO
2
e of BC emissions. While BC and
OM emissions data are available from the WRAP regional haze inventories, the Center for
Climate Strategies (CCS) favored the newer speciation data available from US EPA for the
purposes of estimating BC and OM for most source sectors (BC and OM data from the WRAP
were used only for the nonroad engines sector). In particular, better speciation data are now
available from EPA for important BC emissions sources (e.g., most fossil fuel combustion
sources).
After assembling the BC and OM emission estimates, the mass emission rates were transformed
into their CO
2
e estimates using information from recent global climate modeling. This
transformation is described in later sections below.
Development of BC and OM Mass Emission Estimates
The BC and OM mass emission estimates were derived by multiplying the emissions estimates
for PM with an aerodynamic diameter of less than 2.5 micrometers (PM
2.5
) by the appropriate
aerosol fraction for BC and OM. The aerosol fractions were taken from Pechan’s ongoing work
to update US EPA’s SPECIATE database as approved by US EPA’s SPECIATE Workgroup
members.
After estimating both BC and OM emissions for each source category, we used the BC estimate
as described below to estimate the CO
2
e emissions. Also, as described further below, the OM
141
Version 4.0 of the SPCIATE database and report is expected to be finalized during the Fall of 2006 and will be
provided via EPA’s web site (http://www.epa.gov/ttn/chief/emch/speciation/index.html).
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
I-2 Center for Climate Strategies
www.climatestrategies.us
emission estimate was used to determine whether the source was likely to have positive climate
forcing potential. The mass emission results for 2002 are shown in Table I1.
Development of CO
2
e for BC+OM Emissions
We used similar methods to those applied previously in Maine and Connecticut for converting
BC mass emissions to CO
2
e.
142
These methods are based on the modeling of Jacobson (2002)
143
and his updates to this work (Jacobson, 2005a).
144
Jacobson (2005a) estimated a range of 90:1 to
190:1 for the climate response effects of BC+OM emissions as compared to CO
2
carbon
emissions (depending on either a 30-year or 95-year atmospheric lifetime for CO
2
). It is
important to note that the BC+OM emissions used by Jacobson were based on a 2:1 ratio of
OM:BC (his work in these papers focused on fossil fuel BC+OM; primarily diesel combustion,
which has an OM:BC ratio of 2:1 or less).
For Maine and Connecticut, ENE (2004) applied climate response factors from the earlier
Jacobson work (220 and 500) to the estimated BC mass to estimate the range of CO
2
e associated
with BC emissions. Note that the analysis in the northeast was limited to BC emissions from
onroad diesel exhaust. An important oversight from this work is that the climate response factors
developed by Jacobson (2002, 2005a) are on the basis of CO
2
carbon (not CO
2
). Therefore, in
order to express the BC emissions as CO
2
e, the climate response factors should have been
adjusted upward by a factor of 3.67 to account for the molecular weight of CO
2
to carbon
(44/12).
For this inventory, we started with the 90 and 190 climate response factors adjusted to CO
2
e
factors of 330 and 697 to obtain a low and high estimate of CO
2
e for each sector. An example
calculation of the CO
2
e emissions for 10 tons of PM less than 2.5 microns (PM
2.5
) from onroad
diesel exhaust follows:
BC mass = (10 short tons PM
2.5
) x (0.613 ton EC/ton PM
2.5
) = 6.13 short tons BC
Low estimate CO
2
e = (6.13 tons BC) (330 tons CO
2
e/ton BC+OM) (3 tons BC+OM/ton BC)
(0.907 metric ton/ton) = 5,504 metric tons CO
2
e
High estimate CO
2
e = (6.13 tons BC) (697 tons CO
2
e/ton BC+OM) (3 tons BC+OM/ton BC)
(0.907 metric ton/ton) = 11,626 metric tons CO
2
e
NOTE: The factor 3 tons BC+OM/ton BC comes directly from the global modeling inputs used
by Jacobson (2002, 2005a; i.e., 2 tons of OM/ton of BC).
142
ENE, 2004. Memorandum: “Diesel Black Carbon Calculations – Reductions and Baseline” from Michael
Stoddard, Environment Northeast, prepared for the Connecticut Stakeholder Dialog, Transportation Work Group,
October 23, 2003.
143
Jacobson, 2002. Jacobson, M.Z., “Control of fossil-fuel particulate black carbon and organic matter, possibly the
most effective method of slowing global warming”, Journal of Geophysical Physical Research, volume 107, No.
D19, 4410, 2002.
144
Jacobson, 2005a. Jacobson, M.Z., “Updates to ‘Control of fossil-fuel particulate black carbon and organic matter,
possibly the most effective method of slowing global warming”, Journal of Geophysical Research Atmospheres,
February 15, 2005.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
I-3 Center for Climate Strategies
www.climatestrategies.us
For source categories that had an OM:BC mass emissions ratio >4.0, we zeroed out these
emission estimates from the CO
2
e estimates. The reason for this is that the net heating effects of
OM are not currently well understood (overall OM is thought to have a negative climate forcing
effect or a net cooling effect). Therefore, for source categories where the PM is dominated by
OM (e.g., biomass burning), the net climate response associated with these emissions is highly
uncertain and could potentially produce a net negative climate forcing potential. Further, OM:BC
ratios of 4 or more are well beyond the 2:1 ratio used by Jacobson in his work.
Results and Discussion
We estimate that BC mass emissions in Colorado total about 6.8 million metric tons (MMt) CO
2
e
in 2002. This is the mid-point of the estimated range of emissions. The estimated range is 4.3 –
9.2 MMtCO
2
e (see Table I1), which is roughly 4 to 8% of the estimated emissions for the six
IPCC GHGs. The primary contributing sectors in 2002 were nonroad diesel (54%), onroad diesel
(26%) and rail (7%). The nonroad diesel sector includes exhaust emissions from construction,
industrial and agricultural engines. Agricultural engines contributed about 43% of the nonroad
diesel emissions in 2002, while construction and mining equipment contributed about 32%.
Another significant contributing source sector to BC emissions in Colorado is nonroad gasoline
engines, at over 5% of the total BC CO
2
e. Of this amount, lawn and garden engines contributed
about one-third of the emissions and recreational marine/other equipment contributed another
third. Wildfires and miscellaneous sources such as fugitive dust from paved and unpaved roads
contributed a significant amount of PM and subsequent BC and OM mass emissions (see Table
I1); however the OM:BC ratio is >4 for these sources, so the BC emissions were not converted to
CO
2
e.
Based on 2018 projected emission estimates from the WRAP
145
, there will be a drop in the future
BC emissions for the onroad and nonoad diesel sectors. In 2018, the nonroad sector will
contribute between 0.7 and 1.4 MMtCO
2
e/yr compared to the range of 2.3 to 5.0 MMtCO2e
estimated for 2002. For the onroad diesel sector, the emissions drop to 0.2 to 0.5 MMtCO2e in
2018 from a range of 1.1 to 2.3 MMtCO
2
e in 2002. These reductions are due to new Federal
engine and fuels standards that are currently being phased in to reduce particulate matter
emissions.
While the state of science in aerosol climate forcing is still developing, there is a good body of
evidence supporting the net warming impacts of BC. Aerosols have a direct radiative forcing
because they scatter and absorb solar and infrared radiation in the atmosphere. Aerosols also
alter the formation and precipitation efficiency of liquid water, ice and mixed-phase clouds,
thereby causing an indirect radiative forcing associated with these changes in cloud properties
(IPCC, 2001).
146
There are also a number of other indirect radiative effects that have been
modeled (see, for example, Jacobson, 2002, as noted in footnote on the previous page).
145
Tom Moore, Western Regional Air Partnership, personal communication and data files provided to S. Roe, CCS,
January 2007.
146
IPCC, 2001. Climate Change 2001: The Scientific Basis, Intergovernmental Panel on Climate Change, 2001.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
I-4 Center for Climate Strategies
www.climatestrategies.us
The quantification of aerosol radiative forcing is more complex than the quantification of
radiative forcing by GHGs because of the direct and indirect radiative forcing effects, and the
fact that aerosol mass and particle number concentrations are highly variable in space and time.
This variability is largely due to the much shorter atmospheric lifetime of aerosols compared
with the important GHGs (i.e., CO
2
). Spatially- and temporally-resolved information on the
atmospheric concentration and radiative properties of aerosols is needed to estimate radiative
forcing.
The quantification of indirect radiative forcing by aerosols is especially difficult. In addition to
the variability in aerosol concentrations, some complicated aerosol influences on cloud processes
must be accurately modeled. For example, the warm (liquid water) cloud indirect forcing may be
divided into two components. The first indirect forcing is associated with the change in droplet
concentration caused by increases in aerosol cloud condensation nuclei. The second indirect
forcing is associated with the change in precipitation efficiency that results from a change in
droplet number concentration. Quantification of the latter forcing necessitates understanding of a
change in cloud liquid-water content. In addition to warm clouds, ice clouds may also be affected
by aerosols.
To put the radiative forcing potential of BC in context with CO
2,
the IPCC estimated the radiative
forcing for a doubling of the earth’s CO
2
concentration to be 3.7 watts per square meter (W/m
2
).
For BC, various estimates of current radiative forcing have ranged from 0.16 to 0.42 W/m
2
(IPCC, 2001). These BC estimates are for direct radiative effects only. There is a higher level of
uncertainty associated with the direct radiative forcing estimates of BC compared to those of
CO
2
and other GHGs. There are even higher uncertainties associated with the assessment of the
indirect radiative forcing of aerosols.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
I-5 Center for Climate Strategies
www.climatestrategies.us
Table I1. 2002 BC Emission Estimates
Mass Emissions CO
2
Equivalents
BC OM BC + OM Low High
Contribution
to CO
2
e Sector Subsector
Metric Tons Metric Tons %
Electricity Generating Units (EGUs) Coal 66 102 168 66,469 140,392 1.5%
Oil 0 0 0 17,983 37,983 0.4%
Gas 0 7 7 0 0 0.0%
Other 19 243 262 503 1,063 0.0%
Non-EGU Fuel Combustion (Residential, Commercial, and Industrial)
Coal 11 239 250 43,678 92,253 1.0%
Oil 25 13 39 32,620 68,898 0.8%
Gas 0 300 300 0 0 0.0%
Other
a
1,563 8,337 9,900 2,391 5,049 0.1%
Onroad Gasoline (Exhaust, Brake Wear, & Tire Wear) 168 664 832 63,315 133,728 1.5%
Onroad Diesel (Exhaust, Brake Wear, & Tire Wear) 1,250 522 1,772 1,113,198 2,351,209 25.6%
Aircraft 93 156 249 92,119 194,567 2.1%
Railroad
b
320 105 426 317,249 670,068 7.3%
Other Energy Use Nonroad Gas 212 596 807 209,402 442,283 4.8%
Nonroad Diesel 2,375 779 3,154 2,350,832 4,965,242 54.1%
Other Combustion
c
3 0 3 0 0 0.0%
Industrial Processes 87 0 87 32,917 69,526 0.8%
Agriculture
d
324 0 324 0 0 0.0%
Waste Management Landfills 1 473 474 0 0 0.0%
Incineration 1 0 1 692 1,462 0.0%
Open Burning 0 0 1 190 402 0.0%
Other 4 80 84 0 0 0.0%
Wildfires/Prescribed Burns 6,652 66,543 73,196 0 0 0.0%
Miscellaneous
e
321 5,317 5,638 0 0 0.0%
Totals 13,496 84,478 97,974 4,343,559 9,174,123 100%
a
Primarily wood-fired commercial/industrial boilers with some large diesel engines.
b
Railroad includes Locomotives and Railroad Equipment Emissions.
c
Other Combustion includes Motor Vehicle Fire, Structure Fire, and Aircraft/Rocket Engine Fire & Testing Emissions.
d
Agriculture includes Agricultural Burning, Agriculture/Forestry and Agriculture, Food, & Kindred Spirits Emissions.
e
Miscellaneous includes Paved/Unpaved Roads and Catastrophic/Accidental Release Emissions.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-1 Center for Climate Strategies
www.climatestrategies.us
Appendix J. Greenhouse Gases and Global Warming Potential
Values: Excerpts from the Inventory of U.S. Greenhouse Emissions
and Sinks: 1990-2000
Original Reference: Material for this Appendix is taken from the Inventory of U.S. Greenhouse
Gas Emissions and Sinks: 1990 - 2000, U.S. Environmental Protection Agency, Office of
Atmospheric Programs, EPA 430-R-02-003, April 2002 www.epa.gov/globalwarming/
publications/emissions Michael Gillenwater directed the preparation of this appendix.
Introduction
The Inventory of U.S. Greenhouse Gas Emissions and Sinks presents estimates by the United
States government of U.S. anthropogenic greenhouse gas emissions and removals for the years
1990 through 2000. The estimates are presented on both a full molecular mass basis and on a
Global Warming Potential (GWP) weighted basis in order to show the relative contribution of
each gas to global average radiative forcing.
The Intergovernmental Panel on Climate Change (IPCC) has recently updated the specific global
warming potentials for most greenhouse gases in their Third Assessment Report (TAR, IPCC
2001). Although the GWPs have been updated, estimates of emissions presented in the U.S.
Inventory continue to use the GWPs from the Second Assessment Report (SAR). The guidelines
under which the Inventory is developed, the Revised 1996 IPCC Guidelines for National
Greenhouse Gas Inventories (IPCC/UNEP/OECD/IEA 1997) and the United Nations
Framework Convention on Climate Change (UNFCCC) reporting guidelines for national
inventories
147
were developed prior to the publication of the TAR. Therefore, to comply with
international reporting standards under the UNFCCC, official emission estimates are reported by
the United States using SAR GWP values. This excerpt of the U.S. Inventory addresses in detail
the differences between emission estimates using these two sets of GWPs. Overall, these
revisions to GWP values do not have a significant effect on U.S. emission trends.
Additional discussion on emission trends for the United States can be found in the complete
Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2000.
What is Climate Change?
Climate change refers to long-term fluctuations in temperature, precipitation, wind, and other
elements of the Earth’s climate system. Natural processes such as solar-irradiance variations,
variations in the Earth’s orbital parameters, and volcanic activity can produce variations in
climate. The climate system can also be influenced by changes in the concentration of various
gases in the atmosphere, which affect the Earth’s absorption of radiation.
The Earth naturally absorbs and reflects incoming solar radiation and emits longer wavelength
terrestrial (thermal) radiation back into space. On average, the absorbed solar radiation is
balanced by the outgoing terrestrial radiation emitted to space. A portion of this terrestrial
radiation, though, is itself absorbed by gases in the atmosphere. The energy from this absorbed
terrestrial radiation warms the Earth's surface and atmosphere, creating what is known as the
“natural greenhouse effect.” Without the natural heat-trapping properties of these atmospheric
gases, the average surface temperature of the Earth would be about 33
o
C lower (IPCC 2001).
147
See FCCC/CP/1999/7 at www.unfccc.de
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-2 Center for Climate Strategies
www.climatestrategies.us
Under the UNFCCC, the definition of climate change is “a change of climate which is attributed
directly or indirectly to human activity that alters the composition of the global atmosphere and
which is in addition to natural climate variability observed over comparable time periods.” Given
that definition, in its Second Assessment Report of the science of climate change, the IPCC
concluded that:
Human activities are changing the atmospheric concentrations and distributions of
greenhouse gases and aerosols. These changes can produce a radiative forcing by changing
either the reflection or absorption of solar radiation, or the emission and absorption of
terrestrial radiation (IPCC 1996).
Building on that conclusion, the more recent IPCC Third Assessment Report asserts that
“[c]oncentrations of atmospheric greenhouse gases and their radiative forcing have continued to
increase as a result of human activities” (IPCC 2001).
The IPCC went on to report that the global average surface temperature of the Earth has
increased by between 0.6 ± 0.2°C over the 20th century (IPCC 2001). This value is about 0.15°C
larger than that estimated by the Second Assessment Report, which reported for the period up to
1994, “owing to the relatively high temperatures of the additional years (1995 to 2000) and
improved methods of processing the data” (IPCC 2001).
While the Second Assessment Report concluded, “the balance of evidence suggests that there is
a discernible human influence on global climate,” the Third Assessment Report states the
influence of human activities on climate in even starker terms. It concludes that, “[I]n light of
new evidence and taking into account the remaining uncertainties, most of the observed warming
over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations”
(IPCC 2001).
Greenhouse Gases
Although the Earth’s atmosphere consists mainly of oxygen and nitrogen, neither plays a
significant role in enhancing the greenhouse effect because both are essentially transparent to
terrestrial radiation. The greenhouse effect is primarily a function of the concentration of water
vapor, carbon dioxide, and other trace gases in the atmosphere that absorb the terrestrial radiation
leaving the surface of the Earth (IPCC 1996). Changes in the atmospheric concentrations of these
greenhouse gases can alter the balance of energy transfers between the atmosphere, space, land,
and the oceans. A gauge of these changes is called radiative forcing, which is a simple measure
of changes in the energy available to the Earth-atmosphere system (IPCC 1996). Holding
everything else constant, increases in greenhouse gas concentrations in the atmosphere will
produce positive radiative forcing (i.e., a net increase in the absorption of energy by the Earth).
Climate change can be driven by changes in the atmospheric concentrations of a number of
radiatively active gases and aerosols. We have clear evidence that human activities have affected
concentrations, distributions and life cycles of these gases (IPCC 1996).
Naturally occurring greenhouse gases include water vapor, carbon dioxide (CO
2
), methane
(CH
4
), nitrous oxide (N
2
O), and ozone (O
3
). Several classes of halogenated substances that
contain fluorine, chlorine, or bromine are also greenhouse gases, but they are, for the most part,
solely a product of industrial activities. Chlorofluorocarbons (CFCs) and
hydrochlorofluorocarbons (HCFCs) are halocarbons that contain chlorine, while halocarbons that
contain bromine are referred to as bromofluorocarbons (i.e., halons). Because CFCs, HCFCs, and
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-3 Center for Climate Strategies
www.climatestrategies.us
halons are stratospheric ozone depleting substances, they are covered under the Montreal
Protocol on Substances that Deplete the Ozone Layer. The UNFCCC defers to this earlier
international treaty; consequently these gases are not included in national greenhouse gas
inventories. Some other fluorine containing halogenated substances—hydrofluorocarbons
(HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF
6
)—do not deplete stratospheric
ozone but are potent greenhouse gases. These latter substances are addressed by the UNFCCC
and accounted for in national greenhouse gas inventories.
There are also several gases that, although they do not have a commonly agreed upon direct
radiative forcing effect, do influence the global radiation budget. These tropospheric gases—
referred to as ambient air pollutants—include carbon monoxide (CO), nitrogen dioxide (NO
2
),
sulfur dioxide (SO
2
), and tropospheric (ground level) ozone (O
3
). Tropospheric ozone is formed
by two precursor pollutants, volatile organic compounds (VOCs) and nitrogen oxides (NO
x
) in
the presence of ultraviolet light (sunlight). Aerosols—extremely small particles or liquid
droplets—often composed of sulfur compounds, carbonaceous combustion products, crustal
materials and other human induced pollutants—can affect the absorptive characteristics of the
atmosphere. However, the level of scientific understanding of aerosols is still very low (IPCC
2001).
Carbon dioxide, methane, and nitrous oxide are continuously emitted to and removed from the
atmosphere by natural processes on Earth. Anthropogenic activities, however, can cause
additional quantities of these and other greenhouse gases to be emitted or sequestered, thereby
changing their global average atmospheric concentrations. Natural activities such as respiration
by plants or animals and seasonal cycles of plant growth and decay are examples of processes
that only cycle carbon or nitrogen between the atmosphere and organic biomass. Such
processes—except when directly or indirectly perturbed out of equilibrium by anthropogenic
activities—generally do not alter average atmospheric greenhouse gas concentrations over
decadal timeframes. Climatic changes resulting from anthropogenic activities, however, could
have positive or negative feedback effects on these natural systems. Atmospheric concentrations
of these gases, along with their rates of growth and atmospheric lifetimes, are presented in
Table 10.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-4 Center for Climate Strategies
www.climatestrategies.us
Table 10. Global Atmospheric Concentration (ppm Unless Otherwise Specified), Rate of
Concentration Change (ppb/year) and Atmospheric Lifetime (Years) of Selected
Greenhouse Gases
Atmospheric Variable CO
2
CH
4
N
2
O SF
6
a
CF
4
a
Pre-industrial atmospheric concentration 278 0.700 0.270 0 40
Atmospheric concentration (1998) 365 1.745 0.314 4.2 80
Rate of concentration change
b
1.5
c
0.007
c
0.0008 0.24 1.0
Atmospheric Lifetime 50-200
d
12
e
114
e
3,200 >50,000
Source: IPCC (2001)
a
Concentrations in parts per trillion (ppt) and rate of concentration change in ppt/year.
b
Rate is calculated over the period 1990 to 1999.
c
Rate has fluctuated between 0.9 and 2.8 ppm per year for CO
2
and between 0 and 0.013 ppm per year
for CH4 over the period 1990 to 1999.
d
No single lifetime can be defined for CO
2
because of the different rates of uptake by different
removal processes.
e
This lifetime has been defined as an “adjustment time” that takes into account the indirect effect of
the gas on its own residence time.
A brief description of each greenhouse gas, its sources, and its role in the atmosphere is given
below. The following section then explains the concept of Global Warming Potentials (GWPs),
which are assigned to individual gases as a measure of their relative average global radiative
forcing effect.
Water Vapor (H
2
O). Overall, the most abundant and dominant greenhouse gas in the
atmosphere is water vapor. Water vapor is neither long-lived nor well mixed in the atmosphere,
varying spatially from 0 to 2 percent (IPCC 1996). In addition, atmospheric water can exist in
several physical states including gaseous, liquid, and solid. Human activities are not believed to
directly affect the average global concentration of water vapor; however, the radiative forcing
produced by the increased concentrations of other greenhouse gases may indirectly affect the
hydrologic cycle. A warmer atmosphere has an increased water holding capacity; yet, increased
concentrations of water vapor affects the formation of clouds, which can both absorb and reflect
solar and terrestrial radiation. Aircraft contrails, which consist of water vapor and other aircraft
emittants, are similar to clouds in their radiative forcing effects (IPCC 1999).
Carbon Dioxide (CO
2
). In nature, carbon is cycled between various atmospheric, oceanic, land
biotic, marine biotic, and mineral reservoirs. The largest fluxes occur between the atmosphere
and terrestrial biota, and between the atmosphere and surface water of the oceans. In the
atmosphere, carbon predominantly exists in its oxidized form as CO
2
. Atmospheric carbon
dioxide is part of this global carbon cycle, and therefore its fate is a complex function of
geochemical and biological processes. Carbon dioxide concentrations in the atmosphere
increased from approximately 280 parts per million by volume (ppmv) in pre-industrial times to
367 ppmv in 1999, a 31 percent increase (IPCC 2001). The IPCC notes that “[t]his concentration
has not been exceeded during the past 420,000 years, and likely not during the past 20 million
years. The rate of increase over the past century is unprecedented, at least during the past 20,000
years.” The IPCC definitively states that “the present atmospheric CO
2
increase is caused by
anthropogenic emissions of CO
2
” (IPCC 2001). Forest clearing, other biomass burning, and
some non-energy production processes (e.g., cement production) also emit notable quantities of
carbon dioxide.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-5 Center for Climate Strategies
www.climatestrategies.us
In its second assessment, the IPCC also stated that “[t]he increased amount of carbon dioxide [in
the atmosphere] is leading to climate change and will produce, on average, a global warming of
the Earth’s surface because of its enhanced greenhouse effect—although the magnitude and
significance of the effects are not fully resolved” (IPCC 1996).
Methane (CH
4
). Methane is primarily produced through anaerobic decomposition of organic
matter in biological systems. Agricultural processes such as wetland rice cultivation, enteric
fermentation in animals, and the decomposition of animal wastes emit CH
4
, as does the
decomposition of municipal solid wastes. Methane is also emitted during the production and
distribution of natural gas and petroleum, and is released as a by-product of coal mining and
incomplete fossil fuel combustion. Atmospheric concentrations of methane have increased by
about 150 percent since pre-industrial times, although the rate of increase has been declining.
The IPCC has estimated that slightly more than half of the current CH
4
flux to the atmosphere is
anthropogenic, from human activities such as agriculture, fossil fuel use and waste disposal
(IPCC 2001).
Methane is removed from the atmosphere by reacting with the hydroxyl radical (OH) and is
ultimately converted to CO
2
. Minor removal processes also include reaction with Cl in the
marine boundary layer, a soil sink, and stratospheric reactions. Increasing emissions of methane
reduce the concentration of OH, a feedback which may increase methane’s atmospheric lifetime
(IPCC 2001).
Nitrous Oxide (N
2
O). Anthropogenic sources of N
2
O emissions include agricultural soils,
especially the use of synthetic and manure fertilizers; fossil fuel combustion, especially from
mobile combustion; adipic (nylon) and nitric acid production; wastewater treatment and waste
combustion; and biomass burning. The atmospheric concentration of nitrous oxide (N
2
O) has
increased by 16 percent since 1750, from a pre industrial value of about 270 ppb to 314 ppb in
1998, a concentration that has not been exceeded during the last thousand years. Nitrous oxide is
primarily removed from the atmosphere by the photolytic action of sunlight in the stratosphere.
Ozone (O
3
). Ozone is present in both the upper stratosphere, where it shields the Earth from
harmful levels of ultraviolet radiation, and at lower concentrations in the troposphere, where it is
the main component of anthropogenic photochemical “smog.” During the last two decades,
emissions of anthropogenic chlorine and bromine-containing halocarbons, such as
chlorofluorocarbons (CFCs), have depleted stratospheric ozone concentrations. This loss of
ozone in the stratosphere has resulted in negative radiative forcing, representing an indirect effect
of anthropogenic emissions of chlorine and bromine compounds (IPCC 1996). The depletion of
stratospheric ozone and its radiative forcing was expected to reach a maximum in about 2000
before starting to recover, with detection of such recovery not expected to occur much before
2010 (IPCC 2001).
The past increase in tropospheric ozone, which is also a greenhouse gas, is estimated to provide
the third largest increase in direct radiative forcing since the pre-industrial era, behind CO
2
and
CH
4
. Tropospheric ozone is produced from complex chemical reactions of volatile organic
compounds mixing with nitrogen oxides (NO
x
) in the presence of sunlight. Ozone, carbon
monoxide (CO), sulfur dioxide (SO
2
), nitrogen dioxide (NO
2
) and particulate matter are included
in the category referred to as “criteria pollutants” in the United States under the Clean Air Act
and its subsequent amendments. The tropospheric concentrations of ozone and these other
pollutants are short-lived and, therefore, spatially variable.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-6 Center for Climate Strategies
www.climatestrategies.us
Halocarbons, Perfluorocarbons, and Sulfur Hexafluoride (SF
6
). Halocarbons are, for the
most part, man-made chemicals that have both direct and indirect radiative forcing effects.
Halocarbons that contain chlorine—chlorofluorocarbons (CFCs), hydrochlorofluorocarbons
(HCFCs), methyl chloroform, and carbon tetrachloride—and bromine—halons, methyl bromide,
and hydrobromofluorocarbons (HBFCs)—result in stratospheric ozone depletion and are
therefore controlled under the Montreal Protocol on Substances that Deplete the Ozone Layer.
Although CFCs and HCFCs include potent global warming gases, their net radiative forcing
effect on the atmosphere is reduced because they cause stratospheric ozone depletion, which is
itself an important greenhouse gas in addition to shielding the Earth from harmful levels of
ultraviolet radiation. Under the Montreal Protocol, the United States phased out the production
and importation of halons by 1994 and of CFCs by 1996. Under the Copenhagen Amendments to
the Protocol, a cap was placed on the production and importation of HCFCs by non-Article 5
countries beginning in 1996, and then followed by a complete phase-out by the year 2030. The
ozone depleting gases covered under the Montreal Protocol and its Amendments are not covered
by the UNFCCC.
Hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF
6
) are not
ozone depleting substances, and therefore are not covered under the Montreal Protocol. They are,
however, powerful greenhouse gases. HFCs—primarily used as replacements for ozone
depleting substances but also emitted as a by-product of the HCFC-22 manufacturing process—
currently have a small aggregate radiative forcing impact; however, it is anticipated that their
contribution to overall radiative forcing will increase (IPCC 2001). PFCs and SF
6
are
predominantly emitted from various industrial processes including aluminum smelting,
semiconductor manufacturing, electric power transmission and distribution, and magnesium
casting. Currently, the radiative forcing impact of PFCs and SF
6
is also small; however, they
have a significant growth rate, extremely long atmospheric lifetimes, and are strong absorbers of
infrared radiation, and therefore have the potential to influence climate far into the future (IPCC
2001).
Carbon Monoxide (CO). Carbon monoxide has an indirect radiative forcing effect by elevating
concentrations of CH
4
and tropospheric ozone through chemical reactions with other
atmospheric constituents (e.g., the hydroxyl radical, OH) that would otherwise assist in
destroying CH
4
and tropospheric ozone. Carbon monoxide is created when carbon-containing
fuels are burned incompletely. Through natural processes in the atmosphere, it is eventually
oxidized to CO
2
. Carbon monoxide concentrations are both short-lived in the atmosphere and
spatially variable.
Nitrogen Oxides (NO
x
). The primary climate change effects of nitrogen oxides (i.e., NO and
NO
2
) are indirect and result from their role in promoting the formation of ozone in the
troposphere and, to a lesser degree, lower stratosphere, where it has positive radiative forcing
effects. Additionally, NO
x
emissions from aircraft are also likely to decrease methane
concentrations, thus having a negative radiative forcing effect (IPCC 1999). Nitrogen oxides are
created from lightning, soil microbial activity, biomass burning – both natural and anthropogenic
fires – fuel combustion, and, in the stratosphere, from the photo-degradation of nitrous oxide
(N
2
O). Concentrations of NO
x
are both relatively short-lived in the atmosphere and spatially
variable.
Nonmethane Volatile Organic Compounds (NMVOCs). Nonmethane volatile organic
compounds include compounds such as propane, butane, and ethane. These compounds
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-7 Center for Climate Strategies
www.climatestrategies.us
participate, along with NO
x
, in the formation of tropospheric ozone and other photochemical
oxidants. NMVOCs are emitted primarily from transportation and industrial processes, as well as
biomass burning and non-industrial consumption of organic solvents. Concentrations of
NMVOCs tend to be both short-lived in the atmosphere and spatially variable.
Aerosols. Aerosols are extremely small particles or liquid droplets found in the atmosphere.
They can be produced by natural events such as dust storms and volcanic activity, or by
anthropogenic processes such as fuel combustion and biomass burning. They affect radiative
forcing in both direct and indirect ways: directly by scattering and absorbing solar and thermal
infrared radiation; and indirectly by increasing droplet counts that modify the formation,
precipitation efficiency, and radiative properties of clouds. Aerosols are removed from the
atmosphere relatively rapidly by precipitation. Because aerosols generally have short
atmospheric lifetimes, and have concentrations and compositions that vary regionally, spatially,
and temporally, their contributions to radiative forcing are difficult to quantify (IPCC 2001).
The indirect radiative forcing from aerosols is typically divided into two effects. The first effect
involves decreased droplet size and increased droplet concentration resulting from an increase in
airborne aerosols. The second effect involves an increase in the water content and lifetime of
clouds due to the effect of reduced droplet size on precipitation efficiency (IPCC 2001). Recent
research has placed a greater focus on the second indirect radiative forcing effect of aerosols.
Various categories of aerosols exist, including naturally produced aerosols such as soil dust, sea
salt, biogenic aerosols, sulphates, and volcanic aerosols, and anthropogenically manufactured
aerosols such as industrial dust and carbonaceous aerosols (e.g., black carbon, organic carbon)
from transportation, coal combustion, cement manufacturing, waste incineration, and biomass
burning.
The net effect of aerosols is believed to produce a negative radiative forcing effect (i.e., net
cooling effect on the climate), although because they are short-lived in the atmosphere—lasting
days to weeks—their concentrations respond rapidly to changes in emissions. Locally, the
negative radiative forcing effects of aerosols can offset the positive forcing of greenhouse gases
(IPCC 1996). “However, the aerosol effects do not cancel the global-scale effects of the much
longer-lived greenhouse gases, and significant climate changes can still result” (IPCC 1996).
The IPCC’s Third Assessment Report notes that “the indirect radiative effect of aerosols is now
understood to also encompass effects on ice and mixed-phase clouds, but the magnitude of any
such indirect effect is not known, although it is likely to be positive” (IPCC 2001). Additionally,
current research suggests that another constituent of aerosols, elemental carbon, may have a
positive radiative forcing (Jacobson 2001). The primary anthropogenic emission sources of
elemental carbon include diesel exhaust, coal combustion, and biomass burning.
Global Warming Potentials
Global Warming Potentials (GWPs) are intended as a quantified measure of the globally
averaged relative radiative forcing impacts of a particular greenhouse gas. It is defined as the
cumulative radiative forcing÷both direct and indirect effects÷integrated over a period of time
from the emission of a unit mass of gas relative to some reference gas (IPCC 1996). Carbon
dioxide (CO
2
) was chosen as this reference gas. Direct effects occur when the gas itself is a
greenhouse gas. Indirect radiative forcing occurs when chemical transformations involving the
original gas produce a gas or gases that are greenhouse gases, or when a gas influences other
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-8 Center for Climate Strategies
www.climatestrategies.us
radiatively important processes such as the atmospheric lifetimes of other gases. The relationship
between gigagrams (Gg) of a gas and Tg CO
2
Eq. can be expressed as follows:
( ) ( )
|
|
.
|
\
|
× × =
Gg 1,000
Tg
GWP gas of Gg Eq CO Tg
2
where,
Tg CO
2
Eq. = Teragrams of Carbon Dioxide Equivalents
Gg = Gigagrams (equivalent to a thousand metric tons)
GWP = Global Warming Potential
Tg = Teragrams
GWP values allow policy makers to compare the impacts of emissions and reductions of
different gases. According to the IPCC, GWPs typically have an uncertainty of roughly ±35
percent, though some GWPs have larger uncertainty than others, especially those in which
lifetimes have not yet been ascertained. In the following decision, the parties to the UNFCCC
have agreed to use consistent GWPs from the IPCC Second Assessment Report (SAR), based
upon a 100 year time horizon, although other time horizon values are available (see Table 11).
In addition to communicating emissions in units of mass, Parties may choose also to use
global warming potentials (GWPs) to reflect their inventories and projections in carbon
dioxide-equivalent terms, using information provided by the Intergovernmental Panel on
Climate Change (IPCC) in its Second Assessment Report. Any use of GWPs should be based
on the effects of the greenhouse gases over a 100-year time horizon. In addition, Parties may
also use other time horizons. (FCCC/CP/1996/15/Add.1)
Greenhouse gases with relatively long atmospheric lifetimes (e.g., CO
2
, CH
4
, N
2
O, HFCs, PFCs,
and SF
6
) tend to be evenly distributed throughout the atmosphere, and consequently global
average concentrations can be determined. The short-lived gases such as water vapor, carbon
monoxide, tropospheric ozone, other ambient air pollutants (e.g., NO
x
, and NMVOCs), and
tropospheric aerosols (e.g., SO
2
products and black carbon), however, vary spatially, and
consequently it is difficult to quantify their global radiative forcing impacts. GWP values are
generally not attributed to these gases that are short-lived and spatially inhomogeneous in the
atmosphere.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-9 Center for Climate Strategies
www.climatestrategies.us
Table 11. Global Warming Potentials (GWP) and Atmospheric Lifetimes (Years)
Used in the Inventory
Gas Atmospheric Lifetime 100-year GWP
a
20-year GWP 500-year GWP
Carbon dioxide (CO
2
) 50-200 1 1 1
Methane (CH
4
)
b
12±3 21 56 6.5
Nitrous oxide (N
2
O) 120 310 280 170
HFC-23 264 11,700 9,100 9,800
HFC-125 32.6 2,800 4,600 920
HFC-134a 14.6 1,300 3,400 420
HFC-143a 48.3 3,800 5,000 1,400
HFC-152a 1.5 140 460 42
HFC-227ea 36.5 2,900 4,300 950
HFC-236fa 209 6,300 5,100 4,700
HFC-4310mee 17.1 1,300 3,000 400
CF
4
50,000 6,500 4,400 10,000
C
2
F
6
10,000 9,200 6,200 14,000
C
4
F
10
2,600 7,000 4,800 10,100
C
6
F
14
3,200 7,400 5,000 10,700
SF
6
3,200 23,900 16,300 34,900
Source: IPCC (1996)
a
GWPs used here are calculated over 100 year time horizon
b
The methane GWP includes the direct effects and those indirect effects due to the production of tropospheric
ozone and stratospheric water vapor. The indirect effect due to the production of CO
2
is not included.
Table 12 presents direct and net (i.e., direct and indirect) GWPs for ozone-depleting
substances (ODSs). Ozone-depleting substances directly absorb infrared radiation and
contribute to positive radiative forcing; however, their effect as ozone-depleters also
leads to a negative radiative forcing because ozone itself is a potent greenhouse gas.
There is considerable uncertainty regarding this indirect effect; therefore, a range of net
GWPs is provided for ozone depleting substances.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-10 Center for Climate Strategies
www.climatestrategies.us
Table 12. Net 100-year Global Warming Potentials for Select Ozone Depleting
Substances*
Gas Direct Net
min
Net
max
CFC-11 4,600 (600) 3,600
CFC-12 10,600 7,300 9,900
CFC-113 6,000 2,200 5,200
HCFC-22 1,700 1,400 1,700
HCFC-123 120 20 100
HCFC-124 620 480 590
HCFC-141b 700 (5) 570
HCFC-142b 2,400 1,900 2,300
CHCl
3
140 (560) 0
CCl
4
1,800 (3,900) 660
CH
3
Br 5 (2,600) (500)
Halon-1211 1,300 (24,000) (3,600)
Halon-1301 6,900 (76,000) (9,300)
Source: IPCC (2001)
* Because these compounds have been shown to deplete stratospheric ozone, they are typically referred to
as ozone depleting substances (ODSs). However, they are also potent greenhouse gases. Recognizing the
harmful effects of these compounds on the ozone layer, in 1987 many governments signed the Montreal
Protocol on Substances that Deplete the Ozone Layer to limit the production and importation of a number
of CFCs and other halogenated compounds. The United States furthered its commitment to phase-out
ODSs by signing and ratifying the Copenhagen Amendments to the Montreal Protocol in 1992. Under
these amendments, the United States committed to ending the production and importation of halons by
1994, and CFCs by 1996. The IPCC Guidelines and the UNFCCC do not include reporting instructions
for estimating emissions of ODSs because their use is being phased-out under the Montreal Protocol. The
effects of these compounds on radiative forcing are not addressed here.
The IPCC recently published its Third Assessment Report (TAR), providing the most current and
comprehensive scientific assessment of climate change (IPCC 2001). Within that report, the
GWPs of several gases were revised relative to the IPCC’s Second Assessment Report (SAR)
(IPCC 1996), and new GWPs have been calculated for an expanded set of gases. Since the SAR,
the IPCC has applied an improved calculation of CO
2
radiative forcing and an improved CO
2
response function (presented in WMO 1999). The GWPs are drawn from WMO (1999) and the
SAR, with updates for those cases where new laboratory or radiative transfer results have been
published. Additionally, the atmospheric lifetimes of some gases have been recalculated.
Because the revised radiative forcing of CO
2
is about 12 percent lower than that in the SAR, the
GWPs of the other gases relative to CO
2
tend to be larger, taking into account revisions in
lifetimes. However, there were some instances in which other variables, such as the radiative
efficiency or the chemical lifetime, were altered that resulted in further increases or decreases in
particular GWP values. In addition, the values for radiative forcing and lifetimes have been
calculated for a variety of halocarbons, which were not presented in the SAR. The changes are
described in the TAR as follows:
New categories of gases include fluorinated organic molecules, many of which are ethers that
are proposed as halocarbon substitutes. Some of the GWPs have larger uncertainties than that of
others, particularly for those gases where detailed laboratory data on lifetimes are not yet
available. The direct GWPs have been calculated relative to CO
2
using an improved calculation
of the CO
2
radiative forcing, the SAR response function for a CO
2
pulse, and new values for the
radiative forcing and lifetimes for a number of halocarbons.
Colorado GHG Inventory and Reference Case Projection
CCS, October 2007
J-11 Center for Climate Strategies
www.climatestrategies.us
References
FCCC (1996) Framework Convention on Climate Change; FCCC/CP/1996/15/Add.1; 29
October 1996; Report of the Conference of the Parties at its second session. Revised Guidelines
for the Preparation of National Communications by Parties Included in Annex I to the
Convention, p18. Geneva 1996.
IPCC (2001) Climate Change 2001: A Scientific Basis, Intergovernmental Panel on Climate
Change; J.T. Houghton, Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, C.A.
Johnson, and K. Maskell, eds.; Cambridge University Press. Cambridge, U.K.
IPCC (2000) Good Practice Guidance and Uncertainty Management in National Greenhouse
Gas Inventories. IPCC National Greenhouse Gas Inventories Programme Technical Support
Unit, Kanagawa, Japan. Available online at .
IPCC (1999) Aviation and the Global Atmosphere. Intergovernmental Panel on Climate Change;
Penner, J.E., et al., eds.; Cambridge University Press. Cambridge, U.K.
IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel
on Climate Change; J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg,
and K. Maskell, eds.; Cambridge University Press. Cambridge, U.K.
IPCC/UNEP/OECD/IEA (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas
Inventories. Paris: Intergovernmental Panel on Climate Change, United Nations Environment
Programme, Organization for Economic Co-Operation and Development, International Energy
Agency.
Jacobson, M.Z. (2001) Strong Radiative Heating Due to the Mixing State ofBlack Carbon in
Atmospheric Aerosols. Nature. In press.
UNEP/WMO (2000) Information Unit on Climate Change. Framework Convention on Climate
Change (Available on the internet at .)
WMO (1999) Scientific Assessment of Ozone Depletion, Global Ozone Research and
Monitoring Project-Report No. 44, World Meteorological Organization, Geneva, Switzerland.