EPA 600/B-13/203 | September 2013 | www.epa.gov/ord
United States
Environmental Protection
Agency
EPA U.S. Nine-region MARKAL
Database
Database Documentation
Office of Research and Development
-------
EPA/600/B-13/203
September 2013
EPA U.S. Nine-region MARKAL
DATABASE
Database Documentation
By
Carol Lenox, Rebecca Dodder, Cynthia Gage, Ozge
Kaplan, Dan Loughlin, Will Yelverton
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
Cincinnati, OH 45268
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency
policy and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation of use.
-------
Contents
List of Acronyms and Abbreviations 7
1. Introduction 9
2. MARKAL 9
2.1 Model Description 9
2.2 MARKAL Data Needs 11
2.2.1 Time Horizon 11
2.2.2 System-wide Parameters 11
2.2.3 End-Use Demands 11
2.2.4 Energy Carriers 12
2.2.5 Resource Technologies 12
2.2.6 Process, Conversion, and Demand Technologies 12
2.2.7 Environmental Emissions 13
2.2.8 User-defined Constraints 13
2.3 MARKAL Set Definitions and Naming Conventions 13
3. The EPA Nine-region Database for MARKAL 16
3.1 Structure Overview 16
3.2 Organization of Data 17
3.3 Units 18
3.4 Important Assumptions 18
3.5 Emissions Tracking 19
3.6 Solving a model run 19
3.7 Sectoral Descriptions 20
3.7.1 Resource Supply - Natural Gas 20
3.7.2 Resource Supply - Oil and Refined Products (Including Refineries) 22
3.7.3 Resource Supply - Coal 26
3.7.4 Resource Supply - Biomass and Biofuels 27
3.7.5 Resource Supply - Municipal Solid Waste 32
3.7.6 Resource Supply - Hydrogen 32
3.7.7 Electric Supply Sector 34
3.7.8 Residential Sector 41
-------
3.7.9 Commercial Sector 45
3.7.10 Industrial Sector 50
3.7.11 Transportation Sector 53
3.7.12 Air Quality Regulations and CAIR 60
4. Database Quality Control Process 61
5. Description of the National Database (EPANMD) 62
References 65
Appendices 70
A: MARKAL Parameter Descriptions 70
B: Sector Workbook Description - Oil Resource Supply 76
C: Sector Workbook Description - Natural Gas Resource Supply 80
D: Sector Workbook Description - Coal 84
E: Sector Workbook Description - Unconventional Fuels 88
F: Sector Workbook Description - Refineries 92
G: Sector Workbook Description - Biofuels 95
H: Sector Workbook Description - Biomass 101
I: Sector Workbook Description - Municipal Solid Waste 108
J: Sector Workbook Description - Electric Sector 110
K: Sector Workbook Description - Residential Sector 118
L: Sector Workbook Description - Commercial Sector 124
M: Sector Workbook Description - Industrial Sector 130
N: Sector Workbook Description - Industrial Biofuels 137
O: Sector Workbook Description - Light Duty Transportation 138
P: Sector Workbook Description - Heavy Duty Transportation 144
Q: Sector Workbook Description - Off-Highway Transportation 149
List of Figures
Figure 2.1: Example of a Simple Reference Energy System 10
Figure 3.1: EPAUS9r Regions 16
Figure 3.2 Natural Gas Supply Flow 21
-------
Figure 3.3 PADD Breakout 23
Figure 3.4 Crude Oil Flow 23
Figure 3.5: The Hydrogen Supply Representation in the EPAUS9R 32
Figure 3.6: Nuclear RES 37
Figure 3.7: Total Residential Energy Use by Region in PJ 41
Figure 3.8: Residential Energy Demand by End-Use Type 42
Figure 3.9: Residential CDD coefficient and HDD coefficient 43
Figure 3.10: Commercial Energy Use by Region in pJ 46
Figure 3.11: Commercial Energy Demand by End-Use Type 47
Figure 3.12 Commercial CDD coefficient and HDD coefficient 48
Figure 3.13: Industrial Energy Use by Region in PJ 50
Figure 3.14: Industrial Energy Demand by Sub-Sector 51
Figure 3.15: Light-duty Vehicle Energy Use by Region in PJ 54
Figure 3.16: Distribution of Car Classes 56
Figure 3.17: Heavy Duty Vehicle Energy Use by Type in PJ 57
List of Tables
Table 2.1: Set Definitions in MARKAL 14
Table 2.2: Technology Naming Conventions in MARKAL 15
Table 2.3: Energy Carrier Names in the EPAUS9R 15
Table 3.1 EPAUS9r Workbooks 17
Table 3.2: Fractions Used for QHR(Z)(Y) Values in the EPAUS9R 19
Table 3.3: Oil Naming Conventions 24
Table 3.4: Refined Fuels 24
Table 3.5 Refinery Fuel Outputs 26
Table 3.6: Coal Naming Convention 26
Table 3.7: Biomass Supply Comparison 28
Table 3.8: Existing Electricity Conversion Technologies 35
Table 3.9: New Electricity Conversion Technologies 36
Table 3.10: Residential Demands 42
-------
Table 3.11: Residential Technology and Fuel Combinations 44
Table 3.12: Commercial Demands 46
Table 3.13: Commercial Technology and Fuel Combinations 48
Table 3.14: Industrial Demands 51
Table 3.15: Industrial Energy Carriers 52
Table 3.16: Light Duty Vehicle Fuel and Technology Combinations 54
Table 3.17: Heavy Duty Transportation Demands 57
Table 3.18: Heavy Duty Vehicle Demand Types, Fuel, and Technology Combinations 58
Table 4.1 Data Source Quality Rankings 61
-------
List of Acronyms and Abbreviations
AEO
APB
APPCD
B20
BC
bn-lum-yr
bn-pass-miles
bn-t-miles
bn-vmt
CAFE
CAIR
CAMD
CARD
CBECS
CCS
CDD
CFL
CHIEF
CHP
CNG
DOE
E85
ECAT
EIA
EISA
EPA
ERG
ETSAP
FAME
FAPRI
FGD
GHG
GW
HDD
ICLUS
IGCC
LDV
LED
LFG
LNB
LPG
MARKAL
MATS
MECS
Annual Energy Outlook
Atmospheric Protection Branch
Air Pollution Prevention Control Division
20% Biofuel Blend
Black Carbon
billion lumens per year
billion passenger miles
billion ton miles
billion vehicle miles traveled
Corporate Average Fuel Economy
Clean Air Interstate Rule
Clean Air Markets Division
Center for Agriculture and Rural Development
Commercial Buildings Energy Consumption Survey
Carbon Capture and Sequestration
Cooling Degree Day
Compact Fluorescent Light
Clearinghouse for Inventories and Emissions Factors
Combined Heat and Power
Compressed Natural Gas
Department of Energy
85% Ethanol Fuel Blend
Energy and Climate Assessment Team
Energy Information Administration
Energy Independence and Security Act
Environmental Protection Agency
Eastern Research Group, Inc
Energy Technology Systems Analysis Program
Fatty Acid Methyl Esters
Food and Agricultural Policy Research Institute
Flue Gas Desulfurization
Greenhouse Gases
Gigawatt
Heating Degree Day
Integrated and Climate Land Use Scenarios
Integrated Gasification Combined Cycle
Light Duty Vehicles
Light Emitting Diode
Landfill Gas
Low NOX Burners
Liquid Petroleum Gas
MARket Allocation model
Mercury Air Toxics Standard
Manufacturing Energy Consumption Survey
-------
MOVES
MSW
MSW-DST
Mt
mu
NCDC
NEEDS
NEI
NEMS
NGA
NGCC
NGL
NOAA
NOX
NRC
NREL
NRMRL
O&M
OC
ORD
ORNL
OTAQ
PADD
PJ
PM
PSD
PV
RA
RECS
RES
RES
RPS
RTI
SAGE
SCR
SEDS
SMR
SNCR
SOX
SRI
tcfm-hr
UA
UC
USDA
VMT
Motor Vehicle Emissions Simulator model
Municipal Solid Waste
Municipal Solid Waste Decision Support Tool
million tons
million units
National Climatic Data Center
National Electric Energy Data System
National Emissions Inventory
National Energy Modeling System
Natural Gas
Natural Gas Combined Cycle
Natural Gas Liquids
National Oceanic and Atmospheric Administration
Nitrogen Oxide
National Research Council
National Renewable Energy Lab
National Risk Management Research Lab
Operating and Maintenance
Organic Carbon
Office of Research and Development
Oak Ridge National Laboratory
Office of Transportation Air Quality
Petroleum for Administration Defense Districts
Petajoules
Particulate Matter
Prevention of Significant Deterioration
Photovoltaic
Rural Area
Residential Energy Consumption Survey
Reference Energy System
Reference Energy System
Renewable Portfolio Standard
Research Triangle Institute
Systems for the Analysis of Global Energy model
Selective Catalytic Reduction
State Energy Data Systems
Steam Methane Reform
Selective Non-Catalytic Reduction
Sulfur Oxide
Southern Research Institute
thousand cubic feet per minute per hour
Urbanized Area
Urban Cluster
United States Department of Agriculture
Vehicle Miles Traveled
-------
1. Introduction
The evolution of the energy system in the United States is an important factor in future
environmental outcomes including air quality and climate change. Given this, decision
makers need to understand how a changing energy landscape will impact future air quality
and contribute to meeting mitigation targets and adaptation goals. Energy scenario analyses,
incorporating drivers of emissions such as technological advances, population growth, fuel
availability and utilization, and consumer choice, give important insights into the
environmental effects of the changing energy system. To perform such scenario analyses, a
detailed representation of the energy system is needed. To address this need, the EPA's
Energy and Climate Assessment Team (EC AT) developed a nine-region representation of the
U.S. energy system for use in scenario analysis within the MARKAL modeling framework.
ECAT is part of the Office of Research and Development (ORD), located in the National
Risk Management Research Laboratory (NRMRL), Air Pollution Prevention and Control
Division's (APPCD) Atmospheric Protection Branch (APB). The purpose of this document
is to describe in detail the database, hereafter referred to as the EPA U.S. Nine-region
MARKAL Database (EPAUS9r). The EPAUS9r was originally developed to aid in
technology assessment as part of a larger Air Quality Assessment being performed by EPA
ORD (see "Demonstration of a Scenario Approach for Technology Assessment:
Transportation Sector;" EPA-600/R-04/135, January 2004). In recent years, the MARKAL
modeling framework has been used to research such questions as: What are the impacts of
future energy and technology options on air quality and climate change? What energy
technology future pathways most effectively mitigate climate change and minimize
unintended consequences, such as pollutant emissions? What are the effects of human choice
on the demand side of the energy system?
2. MARKAL
2.1 Model Description
The MARKet ALlocation (MARKAL) model is a data-driven, bottom-up energy systems
economic optimization model. The initial version of the model was developed in the late
1970s at Brookhaven National Laboratory. In 1978, the International Energy Agency adopted
MARKAL and created the Energy Technology and Systems Analysis Program (ETSAP).
ETSAP is a group of modelers and developers that meets every six months to discuss model
developments, extensions, and applications. MARKAL, therefore, benefits from an active
and interactive group of users and developers. For a detailed description of MARKAL, see
the ETSAP MARKAL users manual at http://www.etsap.org/documentation.asp .
The basis of the MARKAL model structure is a network diagram called a Reference Energy
System (RES), which depicts an energy system from resource supply to end-use demand
(Figure 2.1). The RES divides an energy system up into technology stages, energy carriers,
and user demands. The four technology stages represented are resource, process, conversion,
and demand technologies. Resource technologies represent the extraction cost and
availability of resources such as coal, oil, and natural gas. Conversion technologies represent
9
-------
the conversion of fuel inputs into electricity. Process technologies represent other means of
converting resources into end-use fuels including refineries and coal-to-liquid processes.
Demand technologies represent the technologies that meet specific user demands, such as
vehicles, air conditioners, and water heaters. These technologies feed into a final stage
consisting of end-use demands for useful energy services. End-use demands include items
such as residential lighting, commercial air conditioning, and automobile passenger miles
traveled. The stages are connected by the various forms of energy, called energy carriers,
produced and consumed by the system.
Figure 2.1: Example of a Simple Reference Energy System
Resource
Technologies
Process
Technologies
Conversion
Technologies
Demand
Technologies
End-Use
Demand
Gas
Collector
Emissions
Tracking
Electricity
The EPAUS9r is a distinct representation of the U.S. energy system designed to be used
within the MARKAL model structure. The database characterizes the flow of energy
associated with the extraction or import of resources, the conversion of these resources into
useful energy, and the use of the energy in meeting end-use demands within and between the
nine census regions of the United States. A MARKAL model run optimizes technology
penetrations and fuel use within this representation over the specified time horizon using
linear programming techniques to minimize the net present value of the energy system while
satisfying the specified demands, subject to any constraints a user wishes to impose. Outputs
of the model include the technological mix at time intervals into the future, the total system
cost, criteria and greenhouse gas (GHG) emissions, and estimates of energy commodity
prices. A single MARKAL model run generates a least cost pathway to satisfy energy
demands. Using scenario analyses, the model can also be used to explore how the least cost
pathway changes in response to various input changes, such as the introduction of new
energy efficient technologies or a new policy to stimulate CC>2 reductions.
10
-------
2.2 MARKAL Data Needs
A MARKAL database uses a variety of data parameters to describe each element of the RES.
The general categories of data required for a MARKAL model are:
• Time horizon
• System-wide global parameters
• Energy service demands
• Energy carriers
• Resource technology profiles
• Process and demand technology profiles
• Environmental emission factors
• User-defined constraints
2.2.1 Time Horizon
The time horizon constitutes a user-defined number of time periods with each period having
the same number of years. For the EPAUS9r, the time horizon extends from 2005 to 2055
divided into 5-year time periods.
2.2.2 System-wide Parameters
System-wide, otherwise known as global, parameters are assumptions that apply to the entire
model. Two important system-wide parameters of the model are:
• Cost discounting - All costs must be entered in the same monetary unit and discounted to
a common year; 2005 U.S. dollars for the EPAUS9r.
• Subdivision of the year into load fractions - MARKAL subdivides the year into three
seasons Z (summer, winter, intermediate) and 4 times of day Y (day am, day pm, night,
and peak).
2.2.3 End-Use Demands
End-use demands describe the specific energy services to be delivered to individuals or
commercial entities in the economy. Examples of end-use demands include residential space
cooling, personal automotive transport, and industrial process heat. The demand for an
energy service does not refer to the consumption of a particular energy commodity, but rather
to the provision of services such as manufacturing steel, transportation, lighting offices, and
heating homes. These energy services are measured in units of useful energy, which may
vary with sector. For example, in the EPAUS9r, demand for the majority of transport
services is specified in miles traveled, demand for lighting is specified in billion lumens per
year, and demand for industrial process energy is specified in petajoules (PJ). Key demand
related data include:
• projections for useful energy demand services by sector, and
• the load shape of the demand profile by season/day-night-peak (for end use demands that
use electricity).
11
-------
2.2.4 Energy Carriers
Energy carriers are the various forms of energy produced and consumed in the RES. Energy
carriers can include fossil fuels, such as coal with different sulfur content, crude oil, refined
oil products, natural gas, electricity, synthetic fuels produced by model processes, and
renewable energy (e.g., biomass, solar, wind, geothermal, and hydro). Energy carriers
provide the interconnections between the various technologies in the reference energy system
by flowing out of one or more technologies and into others. The model requires that the total
amount of each energy carrier produced in any time period is greater than or equal to the total
amount consumed. Key energy carrier related data include:
• transmission efficiency
• investment and operation and maintenance cost for electricity transmission and
distribution systems
• reserve margin or amount of installed electricity production capacity above the highest
average annual demand
2.2.5 Resource Technologies
Resource technologies are the entry points for raw fuels into and out of the energy system,
including imports and exports, mining and extraction, and renewable energy. These
technologies are generally characterized using stepwise supply curves that indicate how
much of a resource can be obtained at a given price during each model period. Key resource
technology data include:
• bounds indicating the size of each step on each resource supply curve
• a corresponding resource supply cost for each supply step
• cumulative resources limits indicating the total amount of a resource at a particular
supply step that can be delivered over the entire modeling horizon (e.g., total proven size
of a petroleum reservoir)
• cost of transporting resources, either within a region or from region to region.
2.2.6 Process, Conversion, and Demand Technologies
Process technologies are those technologies that change the form, characteristics, or location
of energy carriers. Examples of process technologies in the U.S. model include oil refineries
and hydrogen production technologies. A sub-category of the process technologies is
conversion technologies, which model electricity production (e.g. conversion of one form of
energy to another, as in coal to electricity). Conversion plants are distinguished from other
types of technologies by the fact that they operate on a seasonal/day-night basis. Demand
technologies are those devices that are used to directly satisfy end-use service demands,
including vehicles, furnaces, and electrical devices. These technologies are characterized
using parameters that describe technology costs, fuel consumption and efficiency, and
availability. Key process and demand technology data include:
• cost of investing in new capacity
• fixed operating and maintenance (O&M) costs for installed capacity
12
-------
variable O&M costs according to the operation of installed capacity
fuel delivery costs corresponding to any sectoral difference in the price of an energy
carrier
technical efficiency (usually defined as the ratio between the sum of energy carrier or
useful energy service outputs to the sum of energy carrier inputs)
model year in which the technology first becomes available for investment
availability factors (for process technologies) and capacity utilization factors (for demand
technologies) that describe the maximum percent annual (or season/day-night-peak)
availability for operation or a fixed percent annual (or season/day-night-peak) capacity
utilization per unit of installed capacity
existing installed capacity at the start of the model time horizon
limits on capacity in the form of incremental new investment (absolute or growth rate) or
total installed capacity
"hurdle" rates, or technology specific discount rates, that can be used to represent non-
economic, behavioral aspects of investment choices (e.g., consumer preferences,
expectation of very rapid rates of return, or information gaps).
2.2.7 Environmental Emissions
The EPAUS9r database tracks the production of emissions according to the activity, installed
capacity, or new investment in capacity of a resource or technology. Key environmental
variable related data (expressed in terms of pollutant emissions) include:
• emissions per unit of technology activity, installed capacity, or new investment
• emission constraints, which can take the form of a cap on total emissions in a year, or a
cumulative cap on emissions over the entire modeling horizon.
2.2.8 User-defined Constraints
User-defined constraints are used in the model to set upper, lower, and fixed limits on the use
of fuel types or technology groups. These constraints have three components:
• right hand side value, which specifies the constant value that the constraint is to adhere to
in each period,
• relationship type (upper, lower, fixed), and
• left hand side value, which defines a numeric coefficient (positive or negative) for each
decision variable for each period.
2.3 MARKAL Set Definitions and Naming Conventions
The MARKAL structure uses a pre-defined set of definitions and naming conventions to
organize the RES. Each set represents technologies, energy carriers, or constraints of a
similar type. Within any given set, MARKAL has a number of mandatory parameters that
need to be specified in the model. The main set memberships are listed in Table 2.1.
13
-------
Table 2.1: Set Definitions in MARKAL
Set Name
TCH
SRCENCP
SEP_EXP
SEPJMP
SEP_MIN
SEP_RNW
SEP_STK
PRC
PRE
PRW
CON
ELE
BAS
NBN
STG
DMD
ENV
Set Definition
Technologies
Resource Technology
Export
Import
Extraction
Renewable
Stockpile
Process Technology
Energy
Material (weight)
Conversion
Technology
Electric Conversion
Baseload
Non-baseload
Storage
Demand Technology
Emissions
Set Name
ENT
ENC
ECV
EPS
ENU
ERN
ESY
ELC
LTH
FEQ
DM
DM_COM
DMJND
DM_RES
DM_TRN
Set Definition
Energy Carrier
Standard
Conversion
Fossil
Nuclear
Renewable
Synthetic
Electric
District Heat
Fossil Equivalent
Demand
Commercial
Industrial
Residential
Transportation
ADRATIO User Defined Constraints
REG_ADR Regional Constraint
XARAT Cross Region Constraint
In addition to pre-defined sets, standard naming conventions are used to name the
technologies used in the model. For example, domestically mined fossil fuel step curves start
with MUST (a standard convention) followed by the energy type and the supply step (i.e.
MINNGAD6 is the name for the 6* step in the supply curve for domestically mined natural
gas). The technology naming conventions are listed in Table 2.2.
14
-------
Table 2.2: Technology Naming Conventions in MARKAL
Resource
Technologies
(SRCENCP)
MIN = Fossil Fuels
RNW = Renewables
IMP = Imports
EXP = Exports
STK = Stockpiles
Process
Technologies
(PRC)
P = Process
SC = Collectors
SE = Emissions Tracking
X = Transportation
Tracking
ZZ = Dummy
Conversion
Technologies
(CON)
E = Electric Conversion
Demand
Technologies
(DMD)
COM = Commercial
IND = Industrial
RES = Residential
TRN =
Transportation
The general rule for energy carrier naming conventions is to use a three- to four-character
core name for each principal energy carrier. The specific names for energy carriers would
then add on a two- or three-character descriptor to the core name. The core names are listed
in Table 2.3.
Table 2.3: Energy Carrier Names in the EPAUS9R
Resource
Agricultural Residues
Asphalt
Biodiesel
Biomass - corn stover
Biomass - primary mill
residue
Biomass - soybean oil
Biomass - timber
Biomass - ag residues
Biomass - corn grain
Biomass Energy Crop -
grasses
Biomass Energy Crop -
woody
Biomass Forest Residues
Biomass - municipal solid
waste
Biomass - urban wood waste
Coal
Coke
Compressed Natural Gas
Conventional Gasoline
Distillate Heating Oil
Electricity
Core
Name
AGR
ASP
BDL
BSTV
BPMR
BSYO
BTIM
BAGR
BCRN
BECG
BECW
BFSR
BMSW
BUWW
COA
COK
CNG
GSC
DSH
ELC
Resource
Ethanol
Geothermal
Highway Diesel
Hydrogen
Hydropower
Jet Fuel
Kerosene
Landfill Gas
Liquid Petroleum Gas
Natural Gas
Natural Gas Liquids
Nuclear
Oil
Petrochemical Feedstocks
Petroleum Coke
Reformulated Gasoline
Residual Fuel Oil
Solar
Ultra-low Sulfur Diesel
Wind
Core
Name
ETH
GEO
DSL
H2
HYD
JTF
KER
LFG
LPG
NGA
NGL
NUC
OIL
PFS
PTC
GSR
RFH
SOL
DSU
WND
15
-------
3. The EPA Nine-region Database for MARKAL
3.1 Structure Overview
The EPAUS9r was developed around the nine U.S. Census divisions and covers a modeling
horizon from 2005 to 2055. The nine divisions, considered regions in the database, were
chosen based on the fact that the primary source for data populating the database, the U.S.
Energy Information Administration (EIA) Annual Energy Outlook (AEO) (EIA, 2012c), uses
these same nine divisions.
The technologies chosen for inclusion in the database are commercially available
technologies with historical data for investment costs, efficiencies, and operating and
maintenance costs. Most of these technologies were drawn from the AEO. Other
technologies not yet represented in the AEO are included based on development team
expertise. Data for these technologies were derived from other widely recognized
authoritative sources. Emerging technologies are analyzed by the research team using a
scenario analysis approach in which a range of costs and efficiencies for a new technology
are modeled. Those technologies are not included in the base database.
Essentially, each of the nine-regions in the database has its own conventional RES. The nine
RES structures are then interconnected through a series of trade technology links, so that, for
example, petroleum products refined in Region 2 can be traded to be used in Region 3. The
naming conventions used are essentially the same from one region to another, facilitating
regional and cross-region analysis. The regions are identified by the letter R and the number
of the region, as given in the Figure 3.1:
Figure 3.1: EPAUS9r Regions
EPAUS9r Regions
Rl New England
R2 Middle Atlantic
R3 East North Central
R4 West North Central
R5 South Atlantic
R6 East South Central
R7 West South Central
R8 Mountain
R9 Pacific
In addition to the nine interconnected RES structures, there is a "dummy" supply region
named RO, for coal and imported fuels. For each of the nine-regions, there will be an export
option in RO linked to an associated import option in the region for each commodity. Along
the import path for each commodity entering a region, a transportation cost is added and
16
-------
existing limitations on the supply of each resource are specified, either directly on the
"supply link" or by means of regional infrastructure and transportation process technologies
where the expansion of the supply infrastructure capacity requires investment.
3.2 Organization of Data
The EPAUS9r was developed using ANSWER, which is a Windows interface to MARKAL
developed using MS Visual Basic, MS Access, MS Excel, and requiring the GAMS
mathematical modeling language software. For a complete description of ANSWER see,
"ANSWER MARKAL, An Energy Optimization Tool version 5" (available from Ken Noble of
Nob el-Soft noblesoft@netspeed.com.au).
Software
All data and results referenced in this document were based on the following software
versions:
ANSWER Version 6.4.22 and Gams ScrPRD version 5.9e - available from Noble-
SoftSystem (www.noblesoft.com.au/)
GAMS version 22.7 and XPRESS/GAMS solver - available from GAMS Development
Corporation (www.gams.com)
EPAUS9r_2012 version 1.0 - available from the EPA
All data for the EPAUS9r is organized and transformed from raw data to MARKAL ready
data in Excel workbooks. There are 21 workbooks that make up the database listed in Table
3.1. The appendices of this document contain detailed descriptions of each of the
workbooks.
Table 3.1 EPAUS9r Workbooks
ANSWER Scenario
Name
COAL
OIL
NATGAS
BIOFUEL
BIOMASS
MSW
ELC
ELC_TRDX
REF
UNCNV
H2
Excel Workbook Name
EPAUS9R_[YR]_Coal_vx.x
EPAUS9R_[YR]_Oil_vx.x
EPAUS9R_[YR]_NatGas_vx.x
EPAUS9R_[YR]_Biofuel_vx.x
EPAUS9R_[YR]_Biomass_vx.x
EPAUS9R_[YR]_MSW_vx.x
EPAUS9R_[YR]_ELC_vx.x
EPAUS9R_TRD_ELC_vx.x
EPAUS9R_[YR]_REF_vx.x
EPAUS9R_[YR]_UNCNV_vx.x
EPAUS9R_[YR]_H2_vx.x
Description
Coal resource supply
Domestic and imported oil and imported refined
products
Domestic and imported natural gas and natural gas
liquids
Biofuel production technologies
Biomass resource supply
Municipal solid waste resource supply
Electric sector technologies
Electric sector regional trading
Refineries
Unconventional fuel processes
Hydrogen production and distribution
17
-------
H2TRD
RES
COM
INDUS!
INBIO
TRNJHDV
TRN_LDV
TRN_OH
FUELS
AQREG
CSAPR
CAIRMATS
EPAUS9R_[YR]_TRD_H2_vx.x
EPAUS9R_[YR]_RES_vx.x
EPAUS9R_[YR]_COM_vx.x
EPAUS9R_[YR]_IND_vx.x
EPAUS9R_[YR]_INDBIO_vx.x
EPAUS9R_[YR]_TRN_HDV_vx.x
EPAUS9R_[YR]_TRN_LDV_vx.x
EPAUS9R_[YR]_TRN_OH_vx.x
EPAUS9R_[YR]_FUELS_vx.x
EPAUS9R_[YR]_AQREG_vx.x
EPAUS9R_[YR]_CSAPR-MATS_vx.x
EPAUS9R_[YR]_CAIR-MATS_vx.x
Hydrogen regional trading
Residential sector
Commercial sector
Industrial sector
Industrial biofuels
Transportation sector - heavy duty vehicles
Transportation sector - light duty vehicles
Transportation sector - off-highway vehicles
Energy carrier connections for emissions tracking
Air quality regulations
Cross state air pollution rule and Mercury and air toxics
standard
Clean air interstate rule and Mercury and air toxics
standard
3.3 Units
All costs in the database are given in units of year 2005 million U.S. dollars. Energy carrier
use is given in terms of petajoules (PJ). Most end use demands are also given in terms of PJ,
with the following exceptions:
• Commercial and residential lighting - billion lumens per year (bn-lum-yr)
• Commercial ventilation - thousand cubic feet per minute per hour (tcfm-hr)
• Residential refrigerators and freezers - million units (mu)
• Transportation cars and trucks - billion vehicle miles (bn-vmt)
• Transportation air and passenger rail - billion passenger miles (bn-pass-miles)
• Transportation shipping and freight rail - billion ton miles (bn-t-miles)
3.4 Important Assumptions
There are a number of important assumptions that are used throughout the model database
(sector specific assumptions are discussed in the sector explanations).
• The long-term annual discount rate (DISCOUNT) applied to the economy as a whole is
5%. This discount rate is overridden anytime there is a discount rate (DISCRATE)
applied to a specific technology.
• The fraction of the year (QHR(Z)(Y)) specifies the year by season (Z) and time-of-day
(Y) that best describes the electrical load through the typical year. Table 3.2 lists the
seasonal fractions used in the database.
18
-------
Table 3.2: Fractions Used for QHR(Z)(Y) Values in the EPAUS9R
I-DAM
I-DPM
I-N
I-P
S-DAM
S-DPM
S-N
S-P
W-DAM
W-DPM
W-N
W-P
Intermediate day - AM
Intermediate night - PM
Intermediate night
Summer peak
Summer day - AM
Summer day - PM
Summer night
Summer peak
Winter day - AM
Winter day - PM
Winter night
Winter peak
0.0822
0.0957
0.1532
0.0032
0.0975
0.1087
0.1253
0.0027
0.0815
0.1087
0.1381
0.0032
• The average transmission efficiency (TE(ENT)) of each energy carrier is assumed to
100% unless otherwise stated. This value is based on the fact that, for most energy
carriers, there is no loss of fuel in transporting from one technology to another (i.e. a
rail car transporting coal from the mine to a power plant). For electricity, losses do
occur across transmission lines. In the model, these losses are represented with a
transmission efficiency of 93.5%. This value is based on EIA data found in the State
Electricity Profiles.
• The reserve capacity ((E)RESERVE) for electricity is 0.15.
3.5 Emissions Tracking
The EPAUS9r tracks sectoral emissions for: CO2, NOX, PMio, PM2.5, SO2, VOC, CH4, CO,
Organic Carbon (OC), and Black Carbon (BC). For the residential, commercial, electricity
production, and industrial sectors, emissions are tracked on the fuels coming into the sector.
In the transportation sector, emissions are tracked on the transportation technologies. In
addition to emissions, the database tracks water use in the electric sector.
3.6 Solving a model run
A MARKAL model run optimizes by finding the least cost pathway for meeting the specified
energy service demands utilizing available resources and technologies while satisfying any
pre-defined constraints.
The available fuel resources are characterized using supply curves. The costs and cumulative
amounts for a given supply curve step come from several different resources including the
19
-------
AEO reference case and the U.S. Billion Ton Update: Biomass Supply for a Bioenergy and
Bioproducts Industry (DOE, 201 la).
Technologies which convert resource supplies into end-use technology fuels and the end-use
technologies themselves are chosen by the model based on four variables: technology costs
(investment, variable O&M, and fixed O&M costs), technology efficiency, hurdle rates, and
user-defined constraints.
Technology costs and efficiencies are largely drawn from the AEO.
Technology specific hurdle rates (DISCRATE) are applied differentially to base and advanced
technologies and are chosen to best simulate the consumer's reluctance to purchase certain
technologies or use certain fuels.
User-defined constraints include both fuel and technology shares. Fuel shares are the
percentage of a given demand that must be met by a certain fuel type, and technology shares
are the percentage of a given demand that must be met by a certain technology type. In
MARKAL, these shares are set for the year 2010 based on historical data in the AEO. After
2010, shares are relaxed in order to give the model freedom to switch to different fuels and/or
different technologies. Other user-defined constraints with the EPAUS9r include renewable
portfolio standards (RPS) and regulation induced emissions limits.
A scenario run taking all of this data into account will typically take thirty minutes to an hour
depending upon the computer hardware being used.
3.7 Sectoral Descriptions
The descriptions below give a general overview of each of the sectors within the database.
Detailed calculations for the MARKAL parameters used in the database can be found in the
appendices at the end of this report.
3.7.1 Resource Supply - Natural Gas
There are three sources for natural gas in the EPAUS9r: domestically mined natural gas,
imported liquid natural gas, and pipeline imported Canadian natural gas. These supplies are
characterized in the database using a series of supply curves.
Schematic of Natural Gas Supply
Figure 3.2 shows a schematic of the flow of natural gas in the database within a given region.
20
-------
Figure 3.2 Natural Gas Supply Flow
Refineries
Commercial
Sector
Natural Gas
mined within
the region
Pipeline
natural gas
imported
from other
regions
Residential
Sector
Electric Sector
Pipe
natur
impc
from C
ine
al gas
rted
anada
Imported
natural gas
liquids
Natural
Gas
Industrial
Sector
Transportation
Sector (after
compression)
Exports to Exports to
other regions Mexico
Natural Gas Supply Curves
The naming convention for each supply step begins with three letters, either MUST for
domestically mined resources or IMP for imported resources, followed by NGA, a letter
describing the type (D for domestic, C for Canadian, and L for liquid), and the step number.
Within the database there are a total of nine supply curves used to represent all of the natural
gas available to the model. These include a unique domestic natural gas supply curve in each
of the regions 2 through 8 and two supply curves in Region 0 (RO), one for imported liquid
natural gas and one for Canadian natural gas. The resources in RO are apportioned to the
census regions based on historical data via export and import technologies.
Each of the supply curves has six steps characterized by five parameters: cost (COST), upper
bound (BOUND(BD)Or), cumulative available resource (CUM), annual supply growth rate
(GROWTHr), and annual supply decrease rate (DECAYr). The costs and upper bounds for
step 3 are based on the AEO reference case data for regional and imported price and expected
production or import level. The costs and bounds for the other steps are calculated based on
step 3 using price and quantity elasticities from the EIA National Energy Modeling System
(NEMS) Natural Gas Transmission and Distribution module (EIA, 2012a). The cumulative
resource amount is determined using proved natural gas reserve data from the EIA (EIA,
2012b). Upper bounds on natural gas use are set in the years 2005 and 2010. Beyond those
years the supply from a specific step on the curve is allowed to grow or decline based on
historical natural gas production and imports as reported by EIA. It is important to note that
imported gas supply data are based on net imports and, as of the year 2020, the United States
is set to become a net exporter of liquid natural gas. Therefore, the upper bound on
availability of imported liquid natural gas becomes zero after the 2015 time period.
Natural Gas Transport
During a MARKAL run, domestic natural gas is collected as needed from the various supply
steps and sent through a collector technology where transportation costs are applied and wet
domestic natural gas is separated into dry natural gas (NGA) and natural gas plant liquids
(NGL). In this process it is assumed that for every PJ of wet natural gas, 89.89% is
converted into NGA and 10.11% is converted into NGL. NGL goes into refineries as energy
21
-------
inputs to the refinery process. NGA is delivered to sectors within the region or traded
between regions through trade technologies. The RES has two pipeline technologies for
regional trading of natural gas: existing pipelines with an upper bound on capacity and new
pipeline technology that can add to the existing capacity for an additional cost.
NGA from Canada is imported via pipelines into regions 1,2,4,8, and 9. Similar to regional
trading, imported Canadian gas has two pipeline technologies, existing and new, through
which to transport the gas. Imported liquid natural gas is shipped into existing terminals in
regions 1, 2, 5, and 7. Like the natural gas pipelines, the existing terminals have an upper
bound on capacity. Additional terminal capacity can be purchased at a given cost.
Natural Gas Export to Mexico
In the RES, the U.S. is also able to export NGA to Mexico. This is represented in the
database RES by trading NGA from regions 7, 8 and 9 into RO. Historical data are used to
set the value for how much exported natural gas comes from each region.
3.7.2 Resource Supply - Oil and Refined Products (Including Refineries)
The EPAUS9r represents three different sources for refined petroleum products: domestic
and imported crude oil supply processed through a refinery and imported products that are
already refined. Similar to natural gas, these supplies are characterized in the database using
a series of supply curves.
Crude Oil
Both imported and domestic crude oil are represented in the model by stepwise supply
curves. There are seven supply curves for domestic oil including a unique curve in each of
the regions 2 through 8. For imported oil, there are five crude grades, each of which has a
separate supply curve for each Petroleum Administration for Defense District (PADD), for a
total of 25 unique curves located in Region 0. Imported oil is apportioned to the regions in
which the PADD resides. For example, oil imported into PADD 2 is sent to regions 3 and 4.
Figure 3.3 shows the states that fall in each district.
22
-------
Figure 3.3 PADD Breakout
Petroleum Administration for Defense Districts
PADD 5:
West Coast,
AK, HI
PADD 4:
Rocky
Mountain
PADD 3: Gulf Coast
PADD 1:
East
Coast
Schematic of Crude Oil Supply
Figure 3.4 shows a schematic of the flow of crude oil in the database within a given region.
Figure 3.4 Crude Oil Flow
Crude Oil
mined within
the region
Crude Oil
imported
from other
regions
Imported
Crude Oil into
the PADD
Crude Oil
Refineries
Exports to
other regions
Crude Oil Supply Curve
There are five steps for each of the supply curves. The curves are characterized using five
parameters: cost (COST), upper bound (BOUND(BD)Or), cumulative available resource (CUM),
annual supply growth rate (GROWTHr), and annual supply decrease rate (DECAYr). The
costs and upper bounds for step 3 are based on the AEO reference case data for price and
expected production or import level. The costs and bounds for the other steps are calculated
based on step 3 using price and quantity elasticities from the NEMS Natural Gas
23
-------
Transmission and Distribution module. The cumulative resource amount is determined using
proved crude oil reserve data from the EIA (EIA, 2012b). Upper bounds on crude oil
production or import are set in the years 2005 and 2010. Beyond those years the supply from
a specific step on the curve is allowed to grow or decline based on historical production and
imports as reported by EIA.
Domestic oil uses the naming convention "MINOILD" followed by the step number for the
specific step on the curve. For imported oil, the naming convention is given in Table 3.3.
Table 3.3: Oil Naming Conventions
Resource
IMP
Type
OIL
Description
HH
HL
HV
LL
MH
High sulfur content, High gravity
High sulfur content, Low gravity
High sulfur content, very high gravity
Low sulfur content, Low gravity
Medium sulfur content, High gravity
PADD
1-5
Step
1-5
Oil Transport
During a MARKAL run, domestic oil is collected as needed from the various supply steps
and sent through a collector technology where transportation costs are applied. After that
technology, oil is either delivered to the refinery technology within the region or traded
between regions through trade technologies. All trading between regions carries an
additional cost.
Imported Refined Products
There are thirteen imported refined products represented in the EPAUS9r. These products
are listed in Table 3.4.
Table 3.4: Refined Fuels
MARKAL name
ASP
DSH
DSL
DSD
GSR
GSC
JTF
KER
LPG
MTH
PFS
RFH
RFL
Description
Asphalt
Distillate - Heating Oil No. 2
Distillate - Low Sulfur Highway Diesel
(500 ppm)
Distillate - Ultra-low Sulfur Highway Diesel (15 ppm)
Reformulated Gasoline
Conventional Gasoline
Jet Fuel
Kerosene
Liquid Petroleum gas
Methanol
Petrochemical Feedstocks
High Sulfur Residual Fuel Oil
Low Sulfur Residual Fuel Oil
24
-------
Imported refined products have a separate supply curve for each PADD, for a total of 65
unique curves located in Region 0. There are five steps for each of the supply curves with
the exception of asphalt, which only has one step. The curves are characterized using four
parameters: cost (COST), upper bound (BOUND(BD)Or), annual supply growth rate
(GROWTHr), and annual supply decrease rate (DECAYr). The costs and upper bounds for
step 3 are based on the AEO reference case data for price and expected import level. The
costs and bounds for the other steps are calculated based on step 3 using price and quantity
elasticities from the NEMS Natural Gas Transmission and Distribution module. Upper
bounds on imports are set in the years 2005 through 2020. Beyond those years the supply
from a specific step on the curve is allowed to grow or decline based on historical imports as
reported by EIA.
Like imported oil, the imported refined products are apportioned to the regions in which the
PADD for each supply curve resides. Once the imported refined products are in their
destination regions, these refined products are then supplied to the different end-use sectors
(industrial, commercial, and residential, for example) using transportation technologies in the
RES. Delivery charges are applied to these technologies.
Refineries
The refinery representation in the database consists of three refinery technologies: existing,
new, and high limit. Existing refineries and their output (designated with an "E" at the end
of every name) are characterized based on AEO data for current refinery capacities and
yields. New refinery capacity (designated with an "N") can be built in some of the regions
starting in 2015 and have yields that are similar to the existing refineries. Both existing and
new refineries can output a dummy liquid intermediate (called DLG) that feeds into a high
limit refinery (designated with an "L") which produces higher proportions of gasoline and
diesel. The high limit refineries have a lower cost than new refineries and are not meant to
represent a stand-alone refinery. Instead the high limit refinery represents the ability to
modify the process of an existing refinery to obtain different fuel yields. The existing and
new refineries can produce up to 20% of DLG. This cascaded implementation of the refinery
processes permits more flexibility in fuel production.
There are four key inputs into the refineries: crude oil, natural gas, natural gas plant liquids,
and electricity. Crude oil and natural gas plant liquids serve as feedstock inputs, while
natural gas and electricity are used to meet the refinery energy consumption.
Twelve fuels listed in Table 3.5 are produced by the refineries. The refineries can produce
up to a maximum yield, designated in the database, for each fuel. The sum of the maximum
yields for the twelve fuels is always greater than 1. A limiting parameter (LIMIT) is used to
set the total sum of all fuel produced by the refinery per unit of activity equal to 1. This
difference between the sum of the individual yields and the maximum total yield allows the
refineries to vary the output to meet demands. In this way, the blend and output of the
refinery can be optimized within a specific range.
25
-------
Table 3.5 Refinery Fuel Outputs
MARKAL name
ASP
DSH
DSU
GSR
JTF
KER
LPG
PFS
PTC
RFH
RFL
DIG
Description
Asphalt
Distillate - Heating Oil No. 2
Distillate - Ultra-low Sulfur Highway Diesel (15 ppm)
Reformulated Gasoline
Jet Fuel
Kerosene
Liquid Petroleum gas
Petrochemical Feedstocks
Petroleum Coke
High Sulfur Residual Fuel Oil
Low Sulfur Residual Fuel Oil
Dummy Liquid Fuel
Each region, with the exception of Rl, has its own set of refinery technologies that produces
products used both within the region and by other regions. Trade between regions can take
place either by pipeline, river barges, or ocean vessels. Each of these trade links are
represented in the RES by trade technologies with trading costs attached to them.
3.7.3 Resource Supply - Coal
Forty-one domestically mined coal resources supply curves are represented in the EPAUS9r,
broken out by region of origin, coal type, sulfur content, and mine type. Fourteen coal
regions, five coal types, three sulfur content ranks, and two types of mines are described.
The naming convention for each type of coal is based on these breakouts and is described in
Table 3.6.
Table 3.6: Coal Naming Convention
c
description
Coal
Region
of Origin
NA
CA
SA
El
Wl
GL
DL
WM
WN
SW
WW
RM
ZN
PC
description
Northern Appalachia
Central Appalachia
Southern Appalachia
East Interior
West Interior
Gulf Lignite
Dakota Lignite
Western Montana
Wyoming Northern PRB
Wyoming Southern PRB
Western Wyoming
Rocky Mountains
Southwest
Northwest
Coal
Type
P
B
L
S
G
description
Premium
Bituminous
Lignite
Sub-bituminous
GOB
Sulfur
Content
H
M
L
description
High
Medium
Low
Mine
Type
U
S
description
Underground
Surface
26
-------
Each supply curve step name begins with the three letters "MUST," identifying the supply as
domestically mined, followed by the coal naming convention given above and the step. The
supply curves are represented in Region 0 and then exported into different regions using
export and import technologies. Historical AEO transportation data is used to apply the costs
for sending specific coal types to different regions. Not all types of coal are available in
every region, nor are all types of coal traded between regions.
Coal Supply Curve
Each of the supply curves has eleven steps. The curves are characterized using five
parameters: cost (COST), upper bound (BOUND(BD)Or), cumulative available resource (CUM),
annual supply growth rate (GROWTHr), and annual supply decrease rate (DECAYr). The
costs and upper bounds to 2035 are based on the AEO reference case data for coal supply
curves. Beyond 2035, the costs and bounds grow or decrease in proportion to the years 2030
and 2035. The cumulative resource amount is calculated based on the upper bounds of the
supply curve at each step.
Coal Transport
Coal is delivered to the end-use sectors of the economy (industrial, commercial, electricity,
for example) in each region via transportation technologies.
Coke Imports
Imported coke is represented by a nine-step supply curve in Region 0. Step costs and
cumulative step values are given based on AEO reference case data. Imported coke can be
exported into any of the regions as needed and is delivered to the industrial sector.
Coal to Liquids and Natural Gas
Several process technologies are available in the database that transform coal supply into
liquids or natural gas. The coal to liquids technology outputs diesel fuel and petrochemical
feedstocks. It is first available to the model in 2015. A more efficient and less expensive
technology becomes available in 2030. The coal to natural gas technology outputs synthetic
natural gas as well as petrochemical feedstocks. Region 4 represents a gasification plant with
residual capacity that is already up and running in North Dakota. Other regions have the
ability to build a coal to natural gas plant starting in 2015. Like the coal to liquids, a more
efficient and less expensive technology becomes available in 2030.
3.7.4 Resource Supply - Biomass and Biofuels
The EPAUS9r database characterizes a number of biomass supply chains covering biomass
for use in electricity production, for use in the industrial sector, and for use in the production
of ethanol for transportation. In each region, different biomass feedstock supplies are made
available to the model, with transportation technologies needed to move the biomass and
conversion technologies for ethanol production.
27
-------
Biomass
Data for biomass supply were taken from the Billion Ton Update (DOE, 201 la) which
includes a number of feedstock supply estimates organized by the following major
groupings: (1) forest biomass and wood waste resources, (2) agricultural biomass and waste
resources, and (3) biomass energy crops. Within each of those groupings, there are a number
of biomass feedstock types. For example, energy crops can be divided into perennial grasses,
woody crops and annual energy crops. Table 3.7 below shows the mapping between the
Billion Ton Update Categories and the biomass supply curves incorporated in the nine-region
MARKAL database.
Table 3.7: Biomass Supply Comparison
Billion Ton Update
EPAUS9r Resources
FOREST BIOMASS AND WOOD WASTE RESOURCES
Forest Residues
• Other removal residue
• Conventional wood
• Composite operations (with Federal
Lands)
• Other Forestland Treatment Thinnings
(with Federal Lands)
Forest Residues (FSR)
Mill Residues:
• Unused secondary
• Unused primary
Urban Wood Waste:
• Construction and demolition
• Municipal solid waste
Primary Mill Residues (PMR)
Urban Wood Waste (UWW)
AGRICULTURAL BIOMASS AND WASTE RESOURCES
Primary Agricultural Residues:
• Corn stover
• Barley straw
• Oat straw
• Sorghum stubble
• Wheat straw
Stover (STV)
Agricultural Residues (AGR)
28
-------
Secondary Agricultural Residues and Wastes:
• Cotton gin trash
• Cotton residue
• Manure
• Rice hulls
• Sugarcane trash
• Orchard and vineyard prunings
• Rice straw
• Wheat dust
Not currently used in EPAUS9r.
DEDICATED ENERGY CROPS
Annual energy crops
TT- i • u i
• High-yield sorghum
Perennial grasses
• Switchgrass
„. , ,. .
• Giant Miscanthus
• Sugarcane
Coppice and non-coppice woody crops
• Poplar
• Willow
• Eucalyptus
• Southern Pines
_, _ . ,^^4,
Energy Crops: Annual (ECA)
&j F v /
_, _ _ ,T^^
Energy Crops: Grasses (ECG)
6J F v '
Energy Crops: Woody (ECW)
Agricultural residues reflect the quantity of "straw and stubble" collected from agricultural
lands, including: wheat straw, barley straw, oats straw and sorghum stubble. Quantities of
straw and stubble for specific crops are small relative to corn stover, and are therefore
aggregated to a single supply curve. Forest residues include a variety of forest biomass
resources, including residues from logging and thinning and other removal residues. For
these supply curves, all lands, including federal lands, are included. Mill residues are either
used or unused. Used mill residues are reflected as a single price point, and are available
only to the mills themselves in the model. Urban wood waste includes two categories of
wood waste: construction and demolition (C&D) waste, and wood from municipal solid
waste (MSW). Three types of perennial grasses (switchgrass, Giant Miscanthus, and
sugarcane) are included in the supply curves for Energy Crops - Grasses (ECG), and three
types of woody energy crops (poplar, willow, eucalyptus, southern pines) are included in the
supply curves for Energy Crops - Woody (ECW).
Each of the supply curves has a number of steps, ranging from five to twenty, that give the
cost and upper bound for the given feedstock. Forest residues have 20 supply steps —
substantially more than the other biomass categories. The forest residue prices reflect forest
"roadside prices" that a buyer would pay, and therefore do not include transportation or
preprocessing costs. All of the biomass feedstocks supply curves take into account the cost
29
-------
and energy associated with the production and collection of biomass (and therefore costs are
reflective of "farm gate" or "roadside" prices. The one exception is the inclusion of a factor
accounting for some feedstock degradation and loss for stover and other agricultural residues
(with a slightly higher factor for INP(ENT)p than OUT(ENC)p).
Feedstock transportation includes the diesel use for the truck transport of biomass. A
variable operating and maintenance cost represents the non-fuel costs associated with the
transport of cellulosic feedstocks in particular.
For electricity production, feedstocks are converted from million tons (Mt) to PJ based on
their energy content. However, there are two different collectors: biomass to integrated
gasification and combined cycle (IGCC) and biomass to combustion. The reason for the
separation is that the emissions for the CO2 from biomass feedstock going to combustion are
accounted for on the input fuel, whereas for IGCC, the emissions are accounted for using a
separate emissions accounting technology.
Electric sector biomass-related CO2 and SO2 emissions associated with combustion in the
electric sector are tracked, but it should be noted that the CC>2 emissions are tracked as a
separate CO2 biomass source for the electric sector (e.g., CO2BE) but are not included in the
total CC>2 accounting and are thus assumed to be carbon neutral.
Biofuels
The EPAUS9r characterizes biochemical ethanol production processes from both cellulosic
feedstocks and corn feedstocks and thermo-chemical production processes for ethanol from
cellulosic feedstocks.
For the corn-based biochemical processes, there are four variations of ethanol production:
existing wet mills, existing dry mills, new dry mills, and new dry mills with combined heat
and power. All of the corn-based ethanol production technologies utilize a number of fossil
inputs (including electricity, natural gas, gasoline, and coal for the wet mills), and output
corn-based ethanol. These technologies also produce a number of co-products ranging from
high-fructose corn syrup to dried distiller's grain. These non-energy co-products have been
combined and included via a discount (DELIV(ENT)) on the corn-ethanol price. There is also
a subsidy for corn-based ethanol applied for the first two time period (2005, 2010). This
subsidy is set to zero for the rest of the time periods under the assumption that it will not be
reintroduced. The majority of current U.S. ethanol production is dry mill corn-grain ethanol.
Although most of the growth in ethanol production is dry mill, the residual capacity includes
a number of wet mill facilities.
Cellulosic-based ethanol production via the biochemical platform is set up to use up to five
different feedstocks, although for the purposes of the base model run, only stover and other
agricultural residues are used. Additional feedstocks for biochemical ethanol can be added
for scenario runs.
30
-------
The thermochemical production of ethanol and a gasoline-blendstock is set up to pull from
several different biomass feedstock sources, primarily, woody biomass. For each of these
technologies, there is an investment cost (INVCOST), fixed and variable O&M (FIXOM and
VAROM), two inputs (a biomass feedstock and the gasoline denaturant), and two outputs
(ethanol and mixed higher alcohols that are blended into gasoline). A hurdle rate
(DISCRATE) reflects the high level of investment uncertainty regarding the technology itself,
as well as the uncertainty regarding the availability and price of the biomass feedstock (note
that there is no hurdle rate associated with the biomass feedstocks production and logistics).
Ethanol Regional Trading
The model structure for the inter-regional trading, transport and blending of fuels includes
individual transport modes (truck, barge, and rail) that move the ethanol from one region to
another. Within the importing regions, there are additional technologies that designate where
that ethanol was imported via barge, truck or rail. For each combination of mode, there are
also energy inputs, variable O&M costs (VAROM), and upper bounds (BOUND(BD)O) on the
capacity for transport. Currently the transportation bounds are set to an upper limit for rail
and barge, with much greater flexibility to expand transportation by trucking. However, with
corn-based and corn-stover based ethanol being the key feedstocks for much of the ethanol
production, trucking is limited to go from Regions 3 and 4 to the other Regions.
Ethanol Blending
The model captures the blending of denatured ethanol and gasoline to an E10 blend and an
E85 blend, including the costs associated with conversion of dispensers and storage tanks to
handle E85 blends, as well as a blending of biodiesel into a B20 blend.
Biofuels Constraints
The majority of the biofuels constraints are for specifying lower, upper or fixed bounds on
the production of particular fuels, such as corn-based ethanol or biodiesel, or categories of
fuels, such as advanced biofuels or all exported biodiesel. These constraints are global,
cross-region constraints, which means that the sum of the activity in all nine-regions has to
meet the constraint. Some constraints are included as placeholders, labeled as non-binding
constraints, which can be used in scenarios analyses. There are two share constraints: the
share of new corn-based ethanol production that can come from combined heat and power
(CFtP) facilities, and the share of E85 in the total gasoline/ElO pool. The E85 share
constraint is applied to the blending technology that represents gasoline station retrofits to
dispense E85, and restricts the total dispensing of E85 to 1/3 of the total gasoline/ElO pool.
Biodiesel
Biodiesel production for use in transportation is also available in the model. Biodiesel is
defined as fatty acid methyl esters (FAME) derived from soybean and waste oil. This supply
chain includes the oil pre-processing steps (soybean crushing), the actual biodiesel
production process, and credits for the co-products (e.g., glycerin), although that price credit
drops substantially due to the glycerin glut and assumes no upgrading to refined glycerin.
31
-------
3.7.5 Resource Supply - Municipal Solid Waste
The Municipal Solid Waste (MSW) resource supply is represented in the database by a three
step supply curve in each of the nine-regions. Supply upper bounds for MSW come from the
2006 report The State of Garbage (Simmons et al.). These bounds are given in Mt, which are
then converted to PJ using biomass energy conversion factors taken from the U.S.
Department of Agriculture (USDA). Cost data are derived from average garbage collection
costs for U.S. cities taken from the EPA Municipal Solid Waste Decision Support Tool
(MSW-DST) (Kaplan et al.). MSW supply is either turned into electricity by burning the
waste or from the decomposition of the waste over time into landfill gas.
3.7.6 Resource Supply - Hydrogen
Hydrogen, available in the EPAUS9r for use in the transportation sector, is produced in the
model by a variety of different technologies: coal gasification, natural gas steam methane
reforming, biomass gasification, or electrolysis. The model can also bring in imported
hydrogen in liquid form by truck and by gas pipeline. Much of the data for hydrogen
production come from The Hydrogen Economy: Opportunities, Costs, Barriers, andR&D
Needs (NRC, 2004). Figure 3.5 illustrates the RES for hydrogen as implemented in each of
the nine-regions. The rightmost side of the RES, Hydrogen at Pump, represents the hydrogen
as fuel delivered into vehicle tanks. Hydrogen at Pump represents hydrogen fuel either
delivered to refueling stations via pipeline or truck, or produced on site by any of the six
distributed production technologies. While distributed production technologies combine
production and refueling capabilities, The Refueling Station must receive hydrogen via
delivery and transmission technologies from centralized production technologies.
Figure 3.5: The Hydrogen Supply Representation in the EPAUS9R
Central: Coal Gasification
Central: Coil Gasificatiinw/ CCS
Central: Natural Gas SMR
/i::(::-:
Midsize: Eiiomass Gasificatio:
« Y-+
Midsize: Biomass Gasificatmnv.'.'''^1-S
Midsize: Water E
Midsize: Natural Gas SMR
Midsis: Natural Gas SMRw/CCS
Pipeline or Track
Inter-region Transmission
Pipeline or Track
Inter-region Transmission
CCS = carbon capture and sequestration
SMR = steam methane reforming
Refueling Station
Distributed: Natural Gas SME
Distributed: Water Electron-sis
w/ only gnd electricity
Distribute J: Water Electro lysis
w/ 20% Solar Electniiy
Distribute J: Water Electrolysis
w/30% Wind Electrify
Distributed: Water Electro lysis
w/100% wind elactrcity
Distributed: Water Electro^sis
w/ 100% solar electricity
Hydrogen
At
Pump
32
-------
The transportation costs for hydrogen from centralized production technologies to refueling
stations are characterized differently according to the type of area where the refueling station
is located: Urbanized Area (UA), Urban Cluster (UC) and Rural Area (RA). Generally
speaking, a UA refers to a densely settled territory of 50,000 or more people, a UC to at least
2,500 people but fewer than 50,000 people, and any area outside UA or UC is considered a
RR.
Central and Midsize Production
Both Central and Midsize technologies are centralized production facilities located away
from Refueling Stations, and are therefore connected to the hydrogen distribution
infrastructure. The term Central represents a generic 1,200 ton per day hydrogen production
facility, while Midsize represents a smaller 24 ton per day plant. Both Central and Midsize
plants are dedicated to hydrogen production with no electricity co-generation. All
technologies become available in 2015.
CC>2 capture is available for centralized production technologies including natural gas steam
methane reform, coal gasification, and biomass gasification systems. Criteria pollutant
emissions of these technologies with CCS are assumed to be the same as the corresponding
technologies without the CCS option, except for 862 emissions, which are 50% lower in
those technologies with the CCS option; the amount of CC>2 sequestered is equal to the
difference in CO2 emissions between a CCS technology and its conventional equivalent.
Distributed Production
A distributed production technology is an onsite facility that not only produces hydrogen but
also dispenses hydrogen directly to vehicles. Therefore, a distributed production technology
could be considered a technological combination of small-scale production and refueling
technologies. The distributed production technology in the model represents a 480 kg per day
based hydrogen production capacity.
Intra-region Distribution
Intra-region distribution technologies transport Hydrogen at Plant Gate from Central or
Midsize plants to Refueling Station, and are modeled separately for UA, UC and RR, with
the layouts based on the H^A Delivery Analysis model (Mintz et al.).
For UA, Hydrogen at Plant Gate is transported by pipeline or truck transmission lines and
arrives at the "city gate" of the UA as Hydrogen after Transmission. Hydrogen after
Transmission is then transported by pipeline or truck delivery technologies and arrives at
Refueling Station as Hydrogen Delivered. Three combinations of intra-region distribution
technologies are allowed: pipeline transmission followed by pipeline delivery, pipeline
transmission followed by truck delivery, and truck transmission followed by truck delivery.
The UC intra-region distribution is similar. Although it is reasonable for a transmission route
to be dedicated to a single UA, dedicating a transmission route to a single UC would result in
an unrealistically high cost estimate due to lower demand. As an approximation, the model
combines ten UCs into a single urban unit that shares the same transmission route. Hydrogen
at Plant Gate is transported by pipeline or truck transmission lines and arrives at the "city
33
-------
gate" of the aggregated cluster of UCs as Hydrogen after Transmission. Hydrogen after
Transmission is then transported by pipeline or truck delivery technologies and arrives at
Refueling Station as Hydrogen Delivered. Similarly, three combinations of intra-region
distribution technologies are allowed: pipeline transmission followed by pipeline delivery,
pipeline transmission followed by truck delivery, and truck transmission followed by truck
delivery.
For RR, there is a four-segment highway layout approach. Pipelines or trucks transport and
deliver Hydrogen at Plant Gate along each segment to Refueling Station technologies. The
model allows two technology combinations: pipeline followed by pipeline, and truck
followed by truck.
Inter-Region Transportation
The term Inter-Region Transportation includes both Export and Import technologies. The
Export technology transports Hydrogen at Plant Gate in one census region to the region's
border as Hydrogen Exported, which is then transported by the Import technology of an
adjacent destination region and delivered as Hydrogen at Plant Gate in the destination region.
Vehicle Refueling Stations
Hydrogen from centralized production is dispensed to vehicles at Refueling Station
technologies. Parameters are based on a hypothetical facility with a 2,740 kg per day
hydrogen dispensing capacity.
Constraints Added to Address H2 Transition Issues, Decentralized Hydrogen Production, and
Production by Geographic Area Type
Due to the substantial transportation and distribution infrastructure barriers faced by
centralized and midsize plants, some experts predict that distributed generation at refueling
stations will likely be used to meet initial demands before a full-scale H2 infrastructure is
built. These stations could serve remote, less populated areas where weak economies-of-scale
are justified by high hydrogen delivery costs and low demand. To capture this expectation,
the model contains constraints that require distributed (onsite) technologies to meet a given
fraction of total per-period hydrogen demand. These minimum share constraints gradually
relax and approach zero by 2030. The model also contains logical constraints on pipeline
versus truck delivery within UA and UC area types.
3.7.7 Electric Sector
The Electric sector consists of conversion technologies that take in fuel resources and convert
them to electricity for use in the end-use sectors. Power plant capacity is modeled as
gigawatts (GW), and power plant costs are given in terms of dollars per GW. As electricity
is produced, the output is converted to PJ of electricity through a conversion factor of 31.536
PJ/GW. The technologies represented range from fossil fuel conversion technologies to
nuclear and renewable technologies. In addition to the regular emissions tracking
(ENV_ACT), water consumption is characterized for all power plants using ENV_ACT in terms
of million gallons per PJ of output electricity.
34
-------
The naming convention for electric conversion technologies starts with 'E' for Electricity
followed by three or four characters representing the fuel type and a number of characters
used to represent the technology type. For all technologies with residual capacity, the last
letter is "R." The names of existing non-coal technologies are listed below in Table 3.8.
Table 3.8: Existing Electricity Conversion Technologies
MARKAL
Technology Name
EBIOSTMR
EDSLCCR
EDSLCTR
EHYDCONR
EHYDREVR
ELFGGTRR
ELFGICER
ELFGSTRR
EMSWSTMR
ENGACCRD
ENGACCRO
ENGACCRR
ENGACTR
ENGASTMRO
ENGASTMRR
ERFLSTMR
ESOLPVR
EURNALWRO
EURNALWRR
EWNDR
Description
Wood/Biomass Steam
Diesel Oil Combined-Cycle
Diesel Oil Combustion Turbine
Hydroelectric, Conventional
Hydroelectric, Reversible
Landfill gas to energy: Gas Turbines
Landfill gas to energy: Engines
Landfill gas to energy: Steam Turbines
Municipal Solid Waste Steam
Natural Gas Combined-Cycle; Dry Cooling
Natural Gas Combined-Cycle; Open Loop Cooling
Natural Gas Combined-Cycle; Recirculating Cooling
Natural Gas Combustion Turbine
Natural Gas Steam; Open Loop Cooling
Natural Gas Steam; Recirculating Cooling
Oil Steam (Resid Fuel Oil LS)
Solar Photovoltaic
Pre-Existing Nuclear LWRs; Open Loop Cooling
Pre-Existing Nuclear LWRs; Recirculating Cooling
Wind
The existing technologies are characterized by residual capacity (RESID), fixed O&M
(FKOM), variable O&M (VAROM), plant lifetime (LIFE), availability (AF or AF(Z)(Y)), and
efficiency (lNP(ENT)c).
Coal Plant Retrofits
Residual capacity and costs for new installation for a number of air pollution control retrofits
available to existing coal powered plants are available in the database. For NOX reductions
the model can choose between Low NOX Burner (LNB), Selective Catalytic Reduction
(SCR), Selective Non-Catalytic Reduction (SNCR), or a combination set-up. For SC>2
reductions the model has can build flue gas desulfurization (FGD). PMio retrofits include
fabric filters (FFR), cyclones (CYC), ESP, and ESP upgrades.
35
-------
New Electricity Conversion Technologies
Table 3.9 lists the new electric power production conversion technology options. In addition
to the parameters specified for existing technologies, new technologies have an investment
cost (INVCOST) for new construction. With the exception of the availability factors and
growth constraints for renewables, the specifications for new generating technologies are the
same across the nine MARKAL regions (solar thermal generation is available only in
Regions 4, 7, 8, and 9).
The input for the nuclear technologies is INP(MAT)c instead of INP(ENT)c. In the EPAUS9R
database, nuclear power plants take in metric tons of enriched uranium and produce
electricity in PJ. Whereas all other conversion technologies have a specified efficiency in
terms of INP(ENT)c in (PJ/PJ), nuclear conversion efficiency is specified in terms of
INP(MAT)c in (tons U / PJ ELC). Investment costs represent the cost in the first year the
technology is available.
Table 3.9: New Electricity Conversion Technologies
MARKAL
Technology Name
EBIOIGCC
ECOALIGCC
ECOALIGCCS
ECOALSTM
EGEOBCFS
EGEOR
ELFGGTR
ELFGICE
ELFGSTR
ENGACC05
ENGAACC
ENGAACT
ENGACCCCS
ENGACT05
ESOLPVCEN
ESOLSTCEN
ESOLPVCOM
ESOLPVRES
EURNALWR15
EWNDCL4A
EWNDCL4B
EWNDCL4C
EWNDCL4D
EWNDCL4E
EWNDCL5A
Description
Biomass Integrated Gasification Combined-Cycle
Integrated Coal Gasif. Combined Cycle
Integrated Coal Gasif. Combined Cycle - CO2 Capt.
Pulverized Coal Steam - 2010
Geothermal - Binary Cycle and Flashed Steam
Geothermal, Residual
Landfill gas to energy: Gas Turbines
Landfill gas to energy: Engines
Landfill gas to energy: Steam Turbines
Natural Gas - Combined-Cycle (Turbine)
Natural Gas - Advanced Combined-Cycle (Turbine)
Natural Gas - Advanced Combustion Turbine
Natural Gas Combined Cycle - CO2 Capture
Natural Gas - Combustion Turbine
Solar PV Centralized Generation
Solar Thermal Centralized Generation
Solar PV Distributed Commercial Generation
Solar PV Distributed Residential Generation
Nuclear LWRs, Available in 2015
Wind Generation Class 4 Cost Category A
Wind Generation Class 4 Cost Category B
Wind Generation Class 4 Cost Category C
Wind Generation Class 4 Cost Category D
Wind Generation Class 4 Cost Category E
Wind Generation Class 5 Cost Category A
36
-------
EWNDCL5B
EWNDCL5C
EWNDCL5D
EWNDCL5E
EWNDCL6A
EWNDCL6B
EWNDCL6C
EWNDCL6D
EWNDCL6E
Wind
Wind
Wind
Wind
Wind
Wind
Wind
Wind
Wind
Generation
Generation
Generation
Generation
Generation
Generation
Generation
Generation
Generation
Class
Class
Class
Class
Class
Class
Class
Class
Class
5
5
5
5
6
6
6
6
6
Cost
Cost
Cost
Cost
Cost
Cost
Cost
Cost
Cost
Category
Category
Category
Category
Category
Category
Category
Category
Category
B
C
D
E
A
B
C
D
E
New Solar and Wind
The economics of wind and solar depend strongly on the quality of the available resources,
which vary by region. Regionally-specific availability factors (AF) differentiate the cost-
effectiveness of wind and solar across regions. Two broad solar technology types are
modeled: solar photovoltaic (PV) and concentrating solar thermal (ST). Three types of solar
PV technology are modeled: central electricity generation plants, distributed generation for
residential application and distributed generation for commercial application. The
technology representations for commercial and residential PV are found in the commercial
and residential end-use sector workbooks. One concentrating central solar thermal
technology is modeled.
Three wind technology types are available based on the class of wind resource (Class 4-6)
and five cost categories (A-E). The cost categories are based on the ease of access to wind
resources for each wind class. The difference in capital cost from one category to another
takes into account the cost of transmission interconnection. To constrain the total amount of
wind development, total installed wind capacity (summed across wind classes) was
constrained by region. In addition, inter-regional constraints on maximum installed wind
capacity by cost category and wind class were applied at the national level using the NEMS
input data from the AEO.
Nuclear Technology
The nuclear conversion technology RES consists of mined and imported uranium which feed
materials (instead of energy carriers) into process technologies that enrich the uranium.
These processes create the materials needed for the reactors. The reactors then output
electricity and spent materials to be stockpiled. The RES is pictured in Figure 3.6.
37
-------
Figure 3.6: Nuclear RES
MINURN mined
uranium
IMPURN imported
uranium
PURNU45 Conversion, Enrichment,
Fabrication
^depleted uranium
EURNALWRO Existing Nuclear Plant
open loop
EURNALWRR, Existing Nuclear Plant
recirculatingloop
EURNALWR15 New Nuclear Plant
•spent uranium
A three step supply curve is available for extraction of uranium and a two step supply curve
is available for imported uranium in RO. The uranium goes through a process technology for
conversion, enrichment, and fabrication of uranium to U-235. At this point processed
uranium is sent to the reactors via export technologies out of RO and import technologies into
each region. Mined raw uranium does not have an implicit energy content (like coal)
because it depends on the ultimate level of enrichment. Different nuclear technologies
require uranium enriched to different levels but draw on the same supply of global raw
uranium. As a result, mined uranium, and the other nuclear resources, must be defined as a
material with a cost per unit mass rather than per unit energy.
Carbon Capture and Sequestration (CCS)
CCS, if successfully implemented on a large scale, would allow the continued use of fossil
fuels (especially coal and natural gas) for electric power generation with low atmospheric
emissions of CC>2. The EPAUS9r CCS technology representation focuses on CC>2 capture,
while incorporating CC>2 sequestration (underground injection) as a single cost term. This
emphasis is in keeping with current thinking that capture would account for the largest share
of CCS-related costs (IPCC, 2005) as well as the fact that the feasibility and economics of
sequestration would likely be driven by policy and geologic factors that lie outside the
domain of an energy systems model like MARKAL (Wilson et al.).
CO2 Capture in EPA US9r
CC>2 capture from electric power plants may take place along one of two generic technology
pathways:
Post-Combustion Capture: Equivalent to traditional "smoke stack" controls for 862 and NOX
emissions, this pathway involves separation of CC>2 from the remaining flue gases.
Applicable to both coal-steam and natural gas combustion turbines, the post-combustion
approach is currently the most mature means of retrofitting existing power generation units
(short of a complete repowering).
38
-------
Pre-Combustion Capture: In this approach the carbon is separated from the fuel stream prior
to combustion. The approach to separation most likely to shape the design of new power
plants with CC>2 control, pre-combustion capture is a mature process that the hydrogen,
synthetic fuel, and chemical industries use routinely. The process begins with either steam
reforming or partial oxidation of natural gas, or gasification of coal, to produce hydrogen and
CO (other byproducts need to be removed). A water gas shift reaction then produces
additional hydrogen while converting CO into a high pressure CO2 stream. The higher
pressure simplifies the CO2 capture process (which is typically accomplished via physical
absorption) and reduces its energy requirements, improving overall system efficiency. The
hydrogen is available for use in a combined turbine and steam cycle power generation unit.
Coal-based IGCC plants with CO2 capture are the most frequently mentioned pre-combustion
CCS technology in the literature (IPCC, 2005, MIT, 2007).
EPAUS9r represents each of these CCS technology pathways as part of its electric sector
module. (Note that the model only includes new post-combustion amine capture as a retrofit
option; new coal plants with amine-based CO2 capture are not likely to be competitive with
IGCC or oxyfuel alternatives.) In each representation, the additional power needed to run the
CCS technologies is represented as an energy penalty. This shows up in the model as a
decrease in the efficiencies of the technologies as compared to conventional power plants
without CCS.
CO2 capture effects parameters related to two sets of emissions: economy-wide CO2; CO2E,
which is CO2 from the electric sector; and CO2S, which tracks the amount of CO2
sequestered. These values depend on two quantities: assumed capture efficiency and the
underlying conversion technology emissions rate. Relative to their non-CCS counterparts,
new CCS generating plants and retrofits reduce CO2 and CO2E (the negative emission rates,
in appropriate units) by an amount equal to CO2S. These values account for the fact that the
EPAUS9r electric sector models emissions as a process technology on the fuel chain by
factoring in the underlying conversion plant efficiency; i.e., as described earlier, emissions
are in terms of energy input, while emissions rates tied directly to conversion plants—new
generating units—are expressed in terms of electricity output.
C02 Capture Retrofits
EPAUS9r includes CO2 capture retrofit options for all new coal steam technologies, residual
(existing) coal plants, as well as new IGCC and new NGCC capacity. These retrofits sit as
process technologies on the fuel chain upstream from their corresponding generating
(conversion) technologies. As required by amine-based CO2 scrubbers, the residual coal
retrofits are in line with the flue gas desulfurization (FGD) retrofits (see below), requiring
installation of both if CCS is pursued (CCS pass-through's allow FGD installation
independent of CCS).
As with the new CCS plants, the retrofit power requirements are interpreted here as an
energy penalty (i.e., a base plant output de-rating). The inverse of the retrofit efficiency
(iNP(ENT)p) is the increase in input energy required per unit of retrofit energy output. For a
given retrofit energy penalty, this increase in input energy is related to the assumed energy
penalty through the following relationship:
39
-------
Input energy increase =!/(!+ energy penalty)
The efficiency for new conversion technologies with CCS, i.e., the inverse of the MARKAL
INP(ENT)c, includes the CC>2 capture energy requirements. Hence, these conversion
technologies, as opposed to the retrofit process technologies, do not include an explicit
energy penalty.
CO2 Sequestration in EPAUS9r
In keeping with this aggregate technology representation and the analytical focus on supply-
side CC>2 abatement options, EPAUS9r uses a single figure to represent the cost of CC>2
transport, injection, and long-term monitoring. Geological sequestration enters EPAUS9r as
the ENV_COST parameter on CO2S. EPAUS9r controls CCS market penetration through
region-specific upper bounds on CO2S (ENV_BOUND(UP)), the aggregate amount of CO2
sequestered per model time period.
Electricity Trade
Electricity trade limits in the EPAUS9r represent the non-simultaneous transfer capability of
the transmission network to transfer electricity from one area to another for a single demand
and generation pattern. Trade in electricity is broken into domestic inter-regional trade,
where currently existing, and international transfers between the Canada and Mexico and the
United States. The Canadian transfers are identified by province. For the international
transfers there is a simple series of three supply steps available to identify a price for the
imported electricity.
As the EPAUS9r model simulates the growth in the energy market on a regional basis, the
model assumes the placement of the new generating facilities within each region with the
interconnection of these facilities to the existing transmission grid. It is only appropriate to
assume that these facilities will be integrated into the grid with an eye towards increasing
regional reliability. These facilities could in fact by themselves contribute to increasing the
transfer capability between two neighboring regions.
CHP Electricity Conversion Technologies
Several combined heat and power (CHP) systems are available that are considered a utility
because they produce and sell electricity. These include boiler steam turbines, combustion
turbines, microturbines, fuel cells, and reciprocating engines that use biomass, coal, natural
gas, oil, or waste.
Electric Sector Constraints
Technology-specific constraints are implemented in the electric sector in the early years of
the model time horizon, out to 2020, to help the model to follow historical and predicted
electric production capacity.
40
-------
Renewable Portfolio Standards (RPS)
Thirty-eight states now have defined targets for a percentage of total electricity supply to be
met by designated renewable technologies. No two of these standards are the same and
many of the states have multiple tiers to their standards. Given these considerations,
constraints were developed for the EPAUS9r that try to replicate the state standards at the
regional level. Data was drawn from DOE's Database of State Incentives for Renewables
and Efficiency (DSIRE) (NCSU 2010). Up to three tiers per region were established to
incorporate the state level breakouts and filters were developed in the model to specify which
renewable technologies fell into each tier. RPS constraints start in the year 2010 and go out
until 2055. For most states, once the established standard is met (for example, many states
have RPS goals by 2020), the constraint level remains the same for the rest of the model time
horizon.
3.7.8 Residential Sector
The residential sector representation in the EPAUS9r covers energy service demands for
space heating, space cooling, lighting, water heating, refrigeration, freezing, and other
household uses. These demands make up about 16% of the total energy used in the demand
sectors of the database. The first six demands can be met in a model run by choosing from a
number of detailed technologies. For example, space cooling demand can be met by central
air conditioning, room air conditioning, electric heat pump, or geothermal heat pump. These
six demands represent 80% of the energy use in the residential sector in 2010. The other
20% of energy demand, which come from appliances such as personal computers, TVs,
clothes dryers, ovens, and dishwashers, are met with "Residential Other" technologies that
use electricity, natural gas, or LPG. These "other" technologies do not have efficiency
improvements over time.
In the reference run total energy consumption in the residential sector grows from 14,311 PJ
in 2005 to 16,006 PJ in 2055, an average 1.13% increase per 5-year time period. The total
residential energy use by region is shown in Figure 3.7. Much of the increase in energy use
in Regions 5, 7, and 9 is due to large population increases in those regions.
-Rl
-R2
-R3
-R4
-R5
R6
R7
R8
R9
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055
Figure 3.7: Total Residential Energy Use by Region in PJ
41
-------
Residential Energy Demands
The residential sector is characterized by nine end-use energy demands. The MARKAL
names and units are listed in Table 3.10.
Table 3.10: Residential Demands
Residential Demands
Demand
RSC
RSH
RWH
RLT
RRF
RFZ
ROE
ROG
ROL
Units
PJ/yr
PJ/yr
PPJ
billion lumens/yr
million units
million units
PJ/yr
PJ/yr
PJ/yr
Descriptor
Space Cooling
Space Heating
Water Heating
Lighting
Refrigerators
Freezers
Other - Electricity
Other - Natural Gas
Other - LPG
The percent of total residential energy demand met by end-use type in 2010 and in 2055 is
shown in Figure 3.8.
12055
12010
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
Figure 3.8: Residential Energy Demand by End-Use Type
Demands are first calculated at the national level using the data for energy consumption by
end-use demand and fuel and the average stock equipment efficiency from the AEO
reference case. Regional demands are then determined using calculations based on either
projected population or projected number of households and coefficients calculated for
square footage of heated or cooled space, average heating degree days (HDD) and cooling
degree days (CDD), or number of units from AEO equipment stock data. The calculations
for the individual demands are given below:
42
-------
RSC:
RSH:
RWH:
RLT:
RRF:
RFZ:
ROE:
ROG:
ROL:
cooling
heating
national
national
number
number
national
national
national
coefficient * square footage of air conditioned space * CDD
coefficient * square footage of heated space * HDD
water heating * regional percent of households
lighting demand * regional percent of households
of households * refrigerators per household
of households * freezers per household
other electric * regional percent of households
other natural gas * regional percent of households
other LPG * regional percent of households
National heating and cooling coefficients go down over time based on AEO assumptions that
building shell improvements reduce the energy demands for new and existing buildings. The
coefficients are calculated using AEO heating and cooling demands over time and regional
values for CDD and HDD, population, household air conditioning use, and household square
footage. The CDD and HDD values were taken from the National Climate Data Center
Historical Climatological Series (NCDC, 2009).
16
14
12
10
8
6
4
2
0
•Housing cooling
coefficient (BTU per
sqftperCDD)
-Housing heating
coefficient (BTU per
sqftperHDD)
Figure 3.9: Residential CDD coefficient and HDD coefficient
Residential Technology Choice:
Twenty-nine different technology and fuel combinations listed in Table 3.11 meet the six
main end use demands. Within each of these technology and fuel combinations there are a
number of different available technologies based on vintage year and efficiency.
43
-------
Table 3.11: Residential Technology and Fuel Combinations
End Use
Demand
Space Heating
Space Cooling
Water Heating
Refrigeration
Freezing
Lighting
Technology
Type
Radiant
Heat Pump
Furnace
Wood
Room AC
Central AC
Heat Pump
Incandescent
CFL
LED
Halogen
Linear
Fluorescent
Reflector
Fuel
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Distillate
Geothermal
Distillate
Geothermal
Distillate
Kerosene
LPG
LPG
Solar
Technology costs and efficiencies for space cooling, space heating, water heating,
refrigeration, and freezers are taken from the AEO Residential Technology Equipment Type
Description File (AEO, 2011). Lighting costs and efficiencies were taken from a report
prepared for the EIA on residential and commercial building technologies (EIA, 2007).
Fuel shares are given for space cooling, space heating, and water heating starting in 2010
based on the fuel use in 2005. From 2015 to 2055, the given shares for electricity and natural
gas are relaxed 3% per time period. Diesel, LPG, and kerosene fuel shares are relaxed 5%
per time period out to 2055. Technology shares are given for space heating, space cooling,
water heating, and lighting starting in 2010. For space heating, shares are given for furnaces
and radiant heat. For space cooling, shares are given for heat pumps, central air conditioners,
and room air conditioners. For water heating, shares are given for instantaneous water
heaters. As with electricity and natural gas fuel shares, the shares for all of these
technologies are relaxed 3% per time period. For the lighting technologies, incandescent
lighting shares are reduced 50% by 2020 in keeping with the Energy Independence and
Security Act (EISA) of 2007 and drop off to only 5% of the technologies used in 2055.
Compact fluorescent and LED lighting technologies increase in shares by a minimum of 20%
by 2055. Halogen and linear fluorescent lighting technologies are relaxed 3% per time
period.
44
-------
Residual technologies and new vintages of technologies that use electricity or natural gas and
that are already saturated in the market carry a hurdle rate of 18%. Higher hurdle rates are
assigned to technologies as follows:
28% Compact fluorescent lights
45% LED and linear fluorescent lights
45% Technologies that use diesel, kerosene, or LPG
45% Instantaneous and solar water heaters
45% Electric heat pumps for space heating and cooling
60% Geothermal heat pumps for space heating and cooling
60% Room/window air conditioners
Residential Emissions Accounting:
All fuels coming into the residential sector pass through a "dummy" process technology
which tracks emissions from a particular fuel type. The technology names start with an
"SERES" to indicate emissions tracking for the residential sector, and ends with the three
letter name for the fuel type. For every PJ of natural gas, diesel, kerosene, or LPG that flows
through these dummy technologies to a specific residential technology, such as a natural gas
water heater, the emissions from that fuel are counted. No costs are associated with these
technologies. Emissions for electricity production are handled in the electric sector.
Electricity passes through its own "dummy" technology called "SCRESELC" to get to the
residential sector.
3.7.9 Commercial Sector
The commercial sector representation in the EPAUS9r database covers energy service
demands for space heating, space cooling, lighting, water heating, refrigeration, cooking,
ventilation, office equipment, and other commercial uses. These demands make up about
13.5% of the total energy used in the demand sectors of the database. The first seven
demands can be met in a model run by choosing from a number of detailed technologies. For
example, water heating demand can be met by electric, natural gas, or solar water heaters or
an electric heat pump. These seven demands represent 64% of the energy use in the
commercial sector in 2010 in the AEO. The other 36% of energy demand comes from other
equipment such as office computers and printers, automated teller machines,
telecommunications equipment, medical equipment, and emergency generators. These
demands are met with "Commercial Miscellaneous" and "Commercial Office Equipment"
technologies that use electricity, natural gas, diesel, fuel oil, or LPG. In the base
representation, these "other" technologies do not have efficiency improvements over time.
In the reference run, total energy consumption in the commercial sector grows from 9,420 PJ
in 2005 to 13,172 PJ in 2055, an average 3.42% increase per 5-year time period. The total
commercial energy use by Region is shown in Figure 3.10.
45
-------
3500
3000
2500
•Rl
•R2
R3
•R4
•R5
R6
R7
R8
R9
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055
Figure 3.10: Commercial Energy Use by Region in pj
Commercial Energy Demands
The commercial sector is characterized by thirteen end-use energy demands. The MARKAL
names and units are listed in Table 3.12.
Table 3.12: Commercial Demands
Commercial Demands
Demand
CSH
CSC
CWH
COF
CCK
CLT
CMD
CME
CMN
CML
CMR
CRF
CVT
Units
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
billion lumens/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
tcfm-hr
Descriptor
Space Heating
Space Cooling
Water Heating
Office Equipment
Cooking
Lighting
Misc - DSL
Misc - ELC
Misc - NG
Misc - LPG
Misc - RFL
Refrigeration
Ventilation
The percent of total commercial energy demand met by end-use type in 2010 and in 2055 is
shown in Figure 3.11.
46
-------
120 55
12010
0.0%
5.0%
10.0% 15.0% 20.0% 25.0%
30.0%
Figure 3.11: Commercial Energy Demand by End-Use Type
Demands are calculated by determining the energy intensity per square foot for each end-use
demand from the average stock equipment efficiency in the AEO reference case and
multiplying those intensities by the regional square footage. Heating and cooling demands
are determined using calculations based on either projected population or projected number
of households and coefficients calculated for square footage of heated or cooled space,
average HDDs and cooling degree days CDDs from AEO equipment stock data. The
calculations for the individual demands are given below:
CSC: cooling coefficient * square footage of air conditioned space * CDD
CSH: heating coefficient * square footage of heated space * HDD
CWH: water heating intensity * regional square footage
CLT: lighting intensity * regional square footage
CRF: refrigeration intensity * regional square footage
CVT: ventilation intensity * regional square footage
CCK: cooking intensity * regional square footage
COF: national demand for office equipment per square foot * regional square footage
CME: national demand for "other" electricity per square foot * regional square footage
CMN: national demand for "other" natural gas per square foot * regional square footage
CMD: national demand for "other" diesel per square foot * regional square footage
CML: national demand for "other" LPG per square foot * regional square footage
CMR: national demand for "other" residual fuel per square foot * regional square footage
National heating and cooling coefficients go down over time based on AEO assumptions that
building shell improvements reduce the energy demands for new and existing buildings.
Figure 3.12 shows the change in CDD and HDD coefficient values over time. The
coefficients are calculated using AEO heating and cooling demands over time and regional
values for CDD and HDD, population, household air conditioning use, and household square
footage.
47
-------
25
20
15
10
5 -
•Building heating
coefficient BTU/ft2-
DD-yr
•Building cooling
coefficient BTU/ft2-
DD-yr
Figure 3.12 Commercial CDD coefficient and HDD coefficient
Commercial Technology Choice:
Forty-three different technology and fuel combinations are available to meet the seven main
end use demands are listed in Table 3.13. Within each of these technology and fuel
combinations there are a number of different technology representations based on vintage
year and efficiency.
Table 3.13: Commercial Technology and Fuel Combinations
End Use
Demand
Space Heating
Space Cooling
Water Heating
Ventilation
Cooking
Refrigeration
Technology Type
Heat Pump
Boiler
Furnace
Heat Pump
Centrifugal Chiller
Reciprocating Chiller
Scroll Chiller
Screw Chiller
Rooftop A/C
Window/Wall A/C
Central A/C
CAV
VAV
Central
Walk-in Refrigerator
Walk-in Freezer
Fuel
Air Source
Electric
Air Source
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Natural Gas
Ground Source
Diesel
Diesel
Ground Source
Diesel
Solar
48
-------
Lighting
Reach-in
Refrigerator
Reach-in Freezer
Ice Machine
Beverage Machine
Vending Machine
Incandescent
CFL
LED
Halogen
Linear fluorescent
Mercury Vapor
Metal Halide
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Technology costs and efficiencies for space cooling, space heating, water heating, lighting,
refrigeration, and freezers are taken from the AEO Commercial Technology Equipment Type
Description File (AEO, 201 la).
Fuel shares are given for space cooling, space heating, cooking, and water heating starting in
2010 based on fuel use in 2005. From 2015 to 2055, the given shares for electricity and
natural gas are relaxed 3% per time period. Diesel, LPG, and kerosene fuel shares are
relaxed over 5% per time period out to 2055. Technology shares are given for space heating,
space cooling, water heating, lighting, ventilation, and refrigeration. For space heating,
shares are given for furnaces, boilers, and other electric technologies. For space cooling,
shares are given for rooftop, central, and wall/window air conditioners, air-source and
ground-source heat pumps, and chillers. For water heating, shares are given for solar
systems and heat pumps. As with electricity and natural gas fuel shares, the technology
shares are relaxed 3% per time period. Incandescent lighting shares are reduced 60% by
2020 in keeping with the 2007 EISA and drop off to only 5% of the technologies used in
2055. All other lighting technologies are relaxed 3% per time period. Ventilation and
refrigeration technology splits are held constant throughout the model time horizon.
Commercial technologies and new vintages of technologies that use electricity or natural gas
and that are already saturated in the market carry a hurdle rate of 18%. Higher hurdle rates
are assigned to technologies as follows:
24% All high efficiency technologies (except otherwise noted).
24% Ground source heat pumps, standard efficiency
45% Ground source heat pumps, high efficiency
45% Solar water heaters
60% All high efficiency natural gas technologies
75% Diesel technologies
125% Diesel boilers
49
-------
Commercial Emissions Accounting:
All fuels coming into the commercial sector pass through a "dummy" process technology
which tracks emissions from a particular fuel type. The technology names start with an
"SECOM" to indicate emissions tracking for the residential sector, and ends with the three
letter name for the fuel type. For every PJ of natural gas, diesel, kerosene, or LPG that flows
through these dummy technologies to a specific residential technology, such as a natural gas
water heater, the emissions from that fuel are counted. No costs are associated with these
technologies. Emissions for electricity production are handled in the electric sector.
Electricity passes through its own "dummy" technology called "SCCOMELC" to get to the
commercial sector.
3.7.10 Industrial Sector
The industrial sector representation in the EPAUS9r database covers energy demands for six
main industrial sub-sectors: food, paper, chemicals, nonmetallic mineral products, primary
metals, and transportation equipment. Two additional smaller demands are represented:
other manufacturing, including wood and plastics, and non-manufacturing, including
agriculture and construction. These demands make up about 32.5% of the total energy used
in the demand sectors of the database. The six main industrial subsectors account for 60% of
energy demand in the industrial sector. Another 10% is covered by the other manufacturing
and non-manufacturing demands. Much of the remaining 30%, which includes petroleum
and coal products, is covered in other parts of the database including the refinery sector and
the various resource supply sectors.
In the reference run total energy consumption in the residential sector grows from 42,250 PJ
in 2005 to 57,979 PJ in 2055, an average 3.25% increase per 5-year time period. The total
industrial energy use by region is shown in Figure 3.13.
18000
16000
14000
-Rl
•R2
Ri
•R4
R5
R6
R7
R8
P.9
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055
Figure 3.13: Industrial Energy Use by Region in PJ
50
-------
Industrial Energy Demands
The industrial sector is characterized by eight end-use energy demands. The MARKAL
names and units are listed in Table 3.14.
Table 3.14: Industrial Demands
Industrial Demands
Demands
1C
IF
IM
IN
IP
IT
IO
IXNONM
Units
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
PJ/yr
Description
Chemicals
Food
Primary Metals
Non-metallic Minerals
Paper
Transportation Equipment
Other Manufacturing
Non-manufacturing
The percent of total industrial energy demand met by sub-sector in 2010 and in 2055 is
shown in Figure 3.14.
12055
12010
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%
Figure 3.14: Industrial Energy Demand by Sub-Sector
National demands are calculated for each sub-sector using AEO reference case data for the
value of shipments in dollars and the total energy consumption per dollar shipment. The
energy consumption per dollar shipment is held constant at the 2009 value for all years.
Energy efficiency improvements, which cause the AEO reference case values for energy
consumption per dollar shipment to decrease over time, are ignored in the demand
calculations. Instead, energy efficiency improvements are handled in the process
technologies used to meet the demands. National level demands are then regionalized using
EIA Manufacturing Energy Consumption Survey (MECS) (EIA, 2006) data for fuel
51
-------
consumption by sub-sector by region. With these data, the percent of each sub-sector
demand met by each Region in 2006 was calculated. These percentages were then applied to
the total sub-sector demand at a national level to obtain regional demands. The sub-sector
breakdowns by region in 2006 are assumed to remain consistent throughout the model time
horizon.
Industrial Technology Choice:
The industrial sector is modeled a little differently than the other end-use sectors. The
industrial sector has a three layer structure consisting of a demand layer, an end-use demand
technology layer, and a process technology layer.
Every demand has one end-use demand technology used to represent the sum total of the
energy needs of the different process technologies. These technologies are named in
MARKAL with the first two letters equal to the demand name, followed by "TECHEXT," an
abbreviation for "existing technologies." For example, the demand technology for the food
sub-sector is IFTECHEXT.
Each of the demand technologies is characterized by input energy carriers from up to eight
different process technologies. Each of those energy carriers is given an input value equal to
the percent of total fuel use in that sub-sector by the process technology it represents. Table
3.15 shows the energy carriers and their MARKAL naming conventions.
Table 3.15: Industrial Energy Carriers
Energy Carriers to End-Use Demand Technologies
First two letters
represent the demand
1C, IF, IM, IN, IP, IT, orlO
Final letters
represent the
technology
STM
BOL
PRH
MDR
FAC
EC
FEED
HEAT
Description
Steam
Boilers
Process Heat
Machine Drives
Facility
Electrochemical
Feedstock
Other Heat
Process technologies are represented that output the different demand energy carriers. For
example, non-metallic mineral sector machine drive needs can be met by diesel, electric,
natural gas, or coal powered drives. The model chooses which process technologies to use
based on fuel type, costs, efficiencies, and constraints.
Investment and fixed O&M costs are given for boilers, machine drives, electrochemical, and
process heat technologies. Feedstock, facilities, steam, and other heat technologies do not
52
-------
have any costs associated with them. Changes in technology efficiency are calculated using
the change in energy consumption per dollar shipment values over time from the AEO
reference case run and are characterized in MARKAL using the input energy parameter.
Constraints are placed on process technologies to control the fuel shares used by each sub-
sector. The constraints are based on the percentage a specific energy carrier accounts for the
total energy use of a technology based on MECS. When MECS data are not available, AEO
data are used. Constraints apply over all regions unless the technology does not exist in a
particular region. Because of the end-use driven nature of the industrial sector, energy
carriers and process technologies are tightly constrained. Fuel shares add up to 100% in
2010. From 2015 through 2050, some shares are reduced to allow for fuel switching, with
natural gas and coal lower constraints being reduced 10% to 32% over the course of the
model time horizon.
Industrial Emissions Accounting:
Industrial emissions are accounted for in the process technologies and parameterized as
quantity of emission per PJ of fuel used in a given technology. Only emissions factors
associated with fuel combustion are tracked. Therefore, no emissions are associated with
feedstocks. Emissions from electricity generation are tracked in the electric sector at the
power production plant.
Industrial Pulp and Paper Black Liquor Production
Within the industrial sector, the paper sub-sector creates a bi-product called black liquor.
This black liquor can be gasified to create electricity and steam. The supply of black liquor,
called MININDBL, is characterized in the database with a black liquor supply curve for each
region. No costs are associated with the by-product, but there is an upper bound. The
gasification process technology is called IPBLGCC. Black liquor and mill residue can also
be used to power a boiler. This process technology is called IPBORBIOOO. Both processes
are characterized by investment and fixed O&M costs, efficiencies, and availability factors.
3.7.11 Transportation Sector
The transportation sector in the EPAUS9r database covers energy service demands for the
following sub-categories: light duty vehicles, heavy duty vehicles, and off-highway vehicles.
These demands make up about 38% of the total energy used in the demand sectors of the
database.
Light Duty Vehicles (TL)
The light duty vehicle sub-sector, which accounts for about 55% of the total transportation
demand, represents fuel use for personal vehicle miles traveled. The vehicle technologies
available to the model range in fuel type and efficiency. In addition, there are seven different
class sizes: mini-compact, compact, full size, minivan, pick-up truck, small SUV, and large
SUV.
53
-------
In the reference run, energy consumption for the fleet of personal vehicles decreases at an
average of 2.4% per time period from 2005 to 2030 resulting from increasingly stringent
Corporate Average Fleet Efficiency (CAFE) targets. From 2035, however, increases in
vehicle miles traveled more than offsets the efficiency targets, and energy demands increased
at an average of 2.0% per time period out to 2055. The total personal vehicle energy use by
region is shown in Figure 3.15.
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055
•Rl
•R2
•R3
•R4
•R5
•R6
R7
R8
R9
Figure 3.15: Light-duty Vehicle Energy Use by Region in PJ
Light Duty Vehicle Demands
There is one demand for the light duty vehicles, given the name TL. The demand is given in
billion vehicle miles traveled.
Demands are calculated using the national vehicle miles traveled (VMT) reported in the
AEO. Using population data, VMT per person can be calculated. The data are then
regionalized using the Transportation Energy Consumption Survey regional percentage of
vehicle miles traveled (EIA, 2001). An important assumption made in the database is that
beyond 2035, the VMT per person is held constant.
Light Duty Vehicle Technology Choice:
Nineteen different fuel and technology combinations are available in one or more of the
seven car classes, listed in Table 3.16.
Table 3.16: Light Duty Vehicle Fuel and Technology Combinations
Gasoline
Conventional
Advanced
Hybrid
Plug-in Hybrid (20 miles per
charge)
Plug-in Hybrid (40 miles per
Car Class
Mini-
Compact
X
X
Compact
X
X
X
X
X
Full-size
X
X
X
X
X
Minivan
X
X
X
X
X
Pickup
X
X
X
X
X
Small
SUV
X
X
X
X
X
Large
SUV
X
X
X
X
X
54
-------
Diesel
E85
CNG
LPG
Hydrogen
Electric
charge)
Conventional
Hybrid
Flexfuel
Advanced
Hybrid
Plug-in Hybrid (20 miles per
charge)
Plug-in Hybrid (40 miles per
charge)
Conventional
Flexfuel
Conventional
Flexfuel
Fuel Cell
100 mile range
200 mile range
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Technology costs and efficiencies are calculated by layering EPA Office of Transportation
Air Quality's (OTAQ) vehicle assumptions for advanced technologies on AEO assumptions
regarding conventional vehicle technologies. Data from 2040 through 2055 are held constant
to the 2035 value. AEO adjustment factors are used to reduce efficiencies to account for
real-world driving conditions and vehicle degradation over time. In the transportation sector,
instead of having a new technology created for each change in investment cost and
efficiency, there is one technology with changing investment costs over time and vintaged
efficiencies.
Car class constraints are introduced into the database in the 2010 time period to mimic the
current distribution of cars across the country. From 2010 to 2035 the class splits are
adjusted, with a greater percentage of light-duty transportation demand being met by smaller,
more efficient vehicles. From 2035 on, the splits remain the same. Figure 3.16 shows the
distribution of car classes in 2010 and 2035.
55
-------
12035
12010
Compact
Mini-compact
0%
10%
20%
30%
40%
Figure 3.16: Distribution of Car Classes
In addition to class constraints, a number of other technology or fuel specific constraints are
implemented in the database:
• 100 mile and 200 mile electric vehicles are limited to a maximum penetration equal to
the number of consumers whose needs can be met with those ranges.
• Diesel-powered vehicles are given a regionally specific fixed amount of investment in
each time period.
• Fuel cell vehicles, hybrid, hybrid electric, and advanced gasoline fueled vehicles are
limited to a certain market penetration that increases over time. Most can penetrate
the market 100% by 2025 if the model run finds the 100% market penetration to be
the least cost solution.
• All technology types have a fixed investment constraint in 2010 that mimics the
conditions in the actual market in 2010.
• A global constraint is implemented that forces a reduction in the total fuels going to
light duty vehicles over time. This mimics the national CAFE standard that forces an
improvement in the overall efficiency of light duty vehicles over time. No credit is
given towards the standard for alternative fuel vehicles, and the constraint does not
represent different efficiency targets for cars and trucks.
Conventional vehicles carry a hurdle rate of 40%. All other vehicles carry a hurdle rate of
44%.
Light Duty Vehicle Emissions Accounting:
Passenger vehicle emissions are tracked in two places. Carbon dioxide emissions are
accounted for on the fuel collector coming into the transportation sector to allow for the
differences in emissions when different blends of gasoline and biofuels are used. The
remaining emissions are accounted for on the demand technology itself. The emission
factors are derived from runs of the Motor Vehicle Emission Simulator (MOVES) (EPA,
2010) model. Emission factors for existing vehicles include consideration of pre-2005
56
-------
vintages that leave the fleet, as well as degradation of vehicle emission controls over time.
Emission factors for 2010 and later vintages represent lifetime average emissions and do not
otherwise incorporate degradation.
Heavy Duty Vehicle Use
The heavy duty vehicle sector representation in the EPAUS9r database, which accounts for
43% of the total transportation demand, covers energy service demands for the following
sub-categories: air, bus, commercial trucks, medium and heavy duty trucks, passenger and
freight rail and shipping.
In the reference run total energy consumption in the heavy duty transportation sector grows
from 9,427 PJ in 2005 to 12,964 PJ in 2055, an average 3.2% increase per 5-year time period.
The breakdown by sub-sector is shown in Figure 3.17.
4500
•Air
•Bus
•Commercial Trucks
•Medium duty trucks
•Long Haul heavy duty
trucks
Short Haul heavy duty
trucks
Figure 3.17: Heavy Duty Vehicle Energy Use by Type in PJ
Heavy Duty Transportation Energy Demands
The heavy duty transportation sector is characterized by nine end-use energy demands. The
MARKAL names and units are listed in Table 3.17.
Table 3.17: Heavy Duty Transportation Demands
Heavy Duty Transportation Demands
Name
TA
TB
TC
TM
Description
Domestic Air Transport
Bus
Commercial Trucks (Class 2b)
Medium Duty Trucks (Class 3-6)
Units
bn-pass-miles
bn-vmt
bn-vmt
bn-vmt
Unit Description
billion passenger miles
billion vehicle miles traveled
billion vehicle miles traveled
billion vehicle miles traveled
57
-------
THL
THL
TRF
TRP
TS
Short Haul Heavy Duty Trucks (Class 7-8)
Long Haul Heavy Duty Trucks (Class 7-8)
Freight Rail
Passenger Rail
Shipping (Marine)
bn-vmt
bn-vmt
bn-t-m
bn-pass-miles
bn-t-m
billion vehicle miles traveled
billion vehicle miles traveled
billion ton miles
billion passenger miles
billion ton miles
Demands are calculated by using national level energy consumption from the AEO which are
regionalized using data collected from a number of different sources.
Heavy Duty Transportation Technology Choice:
A number of different technology choices that vary in fuel use and efficiency improvements
are available. Table 3.18 lists these choices. Within some of these combinations there are
different vintage years available.
Table 3.18: Heavy Duty Vehicle Demand Types, Fuel, and Technology Combinations
End Use
Demand
Air
Bus
Commercial
Medium
Fuel Type
Jet Fuel
Gasoline
Gasoline
Diesel
Biodiesel
(20%)
CNG
Hydrogen
Fuel Cell
Gasoline
Diesel
Biodiesel
(20%)
CNG
E85
LPG
Hydrogen
Fuel Cell
Gasoline
Efficiency Improvements
Wing tip design
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Hybrid
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Improved
Aerodynamics
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Auxiliary
Power Unit
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Blended
Wing Design
Geared
Engine
58
-------
and Heavy
Duty Short
Haul
Heavy Duty
Long Haul
Passenger
Rail-
Commuter
Passenger
Rail-
Subways
and
Streetcars
Passenger
Rail-
Intercity
Freight Rail
Shipping
Diesel
Biodiesel
(20%)
CNG
LPG
Diesel
Biodiesel
(20%)
LNG
Diesel
Electricity
Electricity
Diesel
Diesel
Biodiesel
(20%)
LNG
Residual Oil
Diesel
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Improved
Efficiency
Fuel Injection
Fuel Injection
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Advanced
Technology
Auxiliary Power
Unit
Bubble
Lubrication
Hybrid
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Advanced
Hybrid
Smartway
Smartway
Smartway
Smartway
Hybrid
Smartway
Hybrid
Smartway
Hybrid
User defined constraints are given for fuel shares in each of the sub-sectors. These shares are
set for the year 2010 based on historical data in the years leading up to 2010 in the AEO.
After 2010, shares are relaxed to give the model freedom to switch to different fuels and/or
different technologies. In most cases, constraints are relaxed 1-3% per period. For some
fuels, AEO projects increased usage in the future (for example, CNG usage in buses). For
these fuels, the share is constrained to match AEO in 2035.
The base efficiency technologies carry a hurdle rate of 18%. Higher hurdle rates are assigned
to technologies as follows:
59
-------
20%, 24%, 26% Improved efficiency (modelers choice between the three)
24% Hydrogen fuel cell technologies
28% Advanced technology improvements
Off-Highway Diesel and Gasoline Use
Off-road diesel and gasoline used for construction, agriculture, industrial, commercial, and
recreational purposes is tracked in the model using a simple system. Off-highway demand
accounts for about 2% of the total transportation demand. There are two demands, one for
diesel and one for gasoline, and two technologies that feed fuel into those demands. No costs
are associated with the technologies, so essentially, the system is set-up to track the fuel use
by off-road technologies and their corresponding emissions. Any reductions due to
technology efficiency improvements are captured in the associated demands.
National demand for off-road fuel use at the start of the model time horizon is calculated
using data from Oak Ridge National Laboratory (ORNL) Off-Highway Transportation-
Related Fuel Use (Davis and Truett, 2004). Future year demands are calculated using a
growth rate developed from AEO projections of agriculture, construction, and recreational
transportation gasoline and diesel fuel use. National demands are then regionalized.
3.7.12 Air Quality Regulations and CAIR
Air quality regulations affecting the electric sector are represented in the EPAUS9r as both
regional upper bounds (ENV_BOUND)BD) and global limits (GEMLIMT) on NOX and SO2
emissions. The purpose of the air quality regulation representation is to approximate limits
on NOX and SO2 emissions from the electric sector that result from regulations that pre-date
the Clean Air Interstate Rule (CAIR).
Emission limits for 2010 and 2015 were obtained from U.S. EPA Clean Air Market Division
(CAMD) analysis of CAIR (EPA, 2004). The technical support document for that analysis
provides state-level emissions of NOX and SO2 for both CAIR and non-CAIR modeling. We
aggregate the non-CAIR state-level totals to the Census Division to obtain regional
constraints. These constraints are then imposed upon the regional emissions. National
emissions of both species from the electric sector are capped at 2015 levels through 2055.
CAIR emissions limits are represented in a similar manner, although CAMD's CAIR
modeling results were used to develop state-level emission totals. The CAIR emissions limits
also represent the effects of the Mercury Air Toxics Standards (MATS) toxics rule. To
approximate the requirements of that rule, we obtained national totals for SO2 and NOX from
CAMD's analysis of MATS (EPA, 2011) for 2015, 2020 and 2030. Constraint values for
2025 were interpolated.
We expect that CAIR, MATS, and Prevention of Significant Deterioration (PSD) rules will
also impact the utilization of control technologies. Therefore, we represent the impacts of
these rules by applying bounds (BOUND(BD)O) on the use of control technologies. For SO2
controls, we assume that, after 2020, FGD will be used at all coal plants that utilize non-low-
60
-------
sulfur coal. Further, for those plants that do not use low-sulfur coal, but have FGD in place in
2020, FGD must continue to be used (e.g., controlled plants cannot "back off" their use of
controls). For NOX controls, we assume that all coal plants use some form of NOX control
from 2020 through the end of the time horizon. For particulate matter (PM) controls,
upgrades beyond cyclones and electrostatic precipitators (ESP), whether ESP or fabric filters,
are required for all coal units.
4. Database Quality Control Process
To ensure an accurate representation the EPAUS9r has been constructed and evaluated in the
following ways:
• Data were chosen using the established quality guidelines outlined in the Quality
Assurance Project Plan developed for this project.
• Data are fully documented and have been run through quality control checks to
ensure accurate transmission of raw data into the MARKAL database
• The original database was subject to a full model peer review
• The results of a reference case EPANUS9r MARKAL run were assessed against the
results of AEO for the year correlating to the most recent update of the database and
found to be within stated ranges for measures of fuel use system-wide and within
each sector.
The majority of the data were taken from NEMS (EIA, 2009) input data underlying the AEO.
AEO data were selected because the AEO is a nationally recognized source of technology
data, widely used where reference or default data are required. In some cases, AEO data were
not available in a form that could be utilized for the EPAUS9r or better data were determined
to exist. Where better data could be found, data were chosen and ranked using Table 4.1,
given below. The rankings were determined based on the desire to use widely accepted,
quality controlled and/or peer reviewed, data and to minimize the bias in information that is
used in the database.
Table 4.1 Data Source Quality Rankings
Rank
A
B
C
D
E
Quality
Highest
Second
Third
Fourth
Lowest
Source
Federal and state agencies and laboratories
Independent journal articles, academic studies, and
manufacturer product literature
NGO studies, trade journal articles, and conference
proceedings: peer-reviewed
Conference proceedings and other trade literature: non
reviewed
peer-
Individual estimates
The goal in developing the database was that 90% of the data would be from sources in the
top three tiers of data quality.
Another goal of the EPAUS9r is to ensure that the data are fully documented, are entered
properly into the database with required units properly calculated, and are an accurate
61
-------
representation of the information provided in the reference source. EC AT is committed to
using conversion methodologies that are consistent with generally accepted professional
standards. In all work to transform original data into the units and form needed for the
MARKAL model, the conversion factors used are available in the supporting documentation.
A comprehensive peer review of the original EPA database was done to ensure that its
performance is reasonable and that disaggregation to nine-regions has been well represented.
The questions asked of the peer reviewers were as follows:
• Are the results plausible for a tested scenario?
• Are influences between sectors reasonable?
• Are influences and trades between regions reasonable?
• Are there critical regional constraints or ad ratios which are missing from the
database?
• Are some constraints or bounds "over constraining" the model and limiting
flexibility?
• Are there critical weaknesses in the database that significantly influence the results?
The peer reviews were exhaustively documented, and the peer reviewers found no major
errors in the database. Peer review comments and any necessary changes were incorporated
into the database.
Finally, before any release, a reference MARKAL run is performed and calibrated to ensure
that the model is producing reasonable results and providing a plausible, consistent
representation of the key features of the U.S. energy system. The results for the total system
energy consumption and sectoral energy consumption are compared to the AEO. Broad
trends (upward, downward, or changing over the time horizon) are also compared to see if
the EPAUS9r results track with the AEO trends. Finally, the degree of quantitative match
between the EPAUS9r results and AEO are compared. Constraints are added where there is
an underlying feature of the energy system that an unconstrained MARKAL run does not
represent.
Examples of these constraints include
• Transportation LDV class splits are implemented to keep the model run from
choosing all small compact cars, and
• Some level of residential advanced lighting, such as Compact Fluorescent Lamps
(CFL's), is forced in through constraints to overcome the models desire to always
choose inexpensive incandescent lights.
5. Description of the National Database (EPANMD)
In addition to the EPAUS9r, which represents the U.S. energy system at the nine Census
region level, EC AT has a corresponding aggregated national U.S. energy system
representation called the EPANMD (EPA National MARKAL Database). The EPANMD
contains one region (USEPA) which is a summation of the EPAUS9r's nine-regions. This
section describes how parameters were developed for the EPANMD from EPAUS9r data.
62
-------
Resource Supply Curves
In the EPAUS9r, crude oil (domestic and imported), imported refined products, and natural
gas (domestic and imported) are characterized using a series of regional or PADD level
stepped supply curves. In the EPANMD, these supply curves are weighted to become
national curves. Since multiple regions are not present in the EPANMD, the trade
technologies between regions for oil, gas, refined products, and coal are eliminated. Instead
of allowing the products to be distributed to regions from PADDs, each product is collected
by a dummy technology and then delivered to the end-use sectors. These dummy
technologies are new technologies added to the EPANMD. Delivery charges are applied for
delivery to sectors and refineries. The charges for each EPAUS9r region are averaged for a
national level average value.
Biomass and Biofuels
Biomass feedstocks are produced by a supply curve that varies by region. For the EPANMD,
the availability of feedstocks is summed while the costs are averaged for a national level
supply curve. After biomass production, collection and transportation, delivered biomass is
used either for biofuels production for use in the transportation sector, or sent to other sectors
for other forms of energy production (electricity, heat, steam, etc.).
Biofuels are produced by eight technologies: six corn-based for ethanol production and two
cellulosic ethanol production technologies. The input, output, and costs of these technologies
are the same across the EPAUS9r regions and, therefore, remain the same for the EPANMD.
Residual capacity for existing production technologies varies across regions, and is summed
up for a national level RESID.
Municipal Solid Waste
Municipal solid waste (MSW) is used for electricity generation via direct combustion,
utilization of landfill gas, combined heat and power (CHP), and gasification. MSW follows a
similar path as biomass. Production is based on a supply curve and availability is
apportioned by Region. In the EPANMD, the availability is nationalized by summing up the
Regional values.
Refineries
In the EPAUS9r, the regional refineries produce petroleum products that will generally be
used within the demand region, but each region can trade products with other regions.
However, these trade technologies are eliminated in the EPANMD since multiple regions do
not exist. In the EPANMD, three refinery technologies produce all the petroleum products
for use within demand sectors. The start year for existing and high level refineries remain at
2000 and 2010 respectively. However, the start year for new conversion refineries, which
varies across regions in the EPAUS9r, is averaged for a start year of 2015. Cost data for
refineries are equal to the values for Region 7. Residual capacity and lower bounds on
activity for existing refineries are added up across the regions for a national level value.
63
-------
Electricity Generation
The EPANMD contains all of the electricity generating technologies present in the
EPAUS9r. Primarily, parameters are either summed or averaged for national level values.
Availability factors like LIFE, AF, AF(Z)(Y), and PEAK(CON), vary across regions and are
averaged for national level values in the EPANMD. Cost parameters for existing
technologies, including INVCOST, VAROM, and FIXOM, vary by region and are averaged
for the EPANMD. For new technologies, these costs do not vary by region and remain at the
same value for the EPANMD. Input and output parameters vary by region for existing
technologies and are averaged for the national model. These parameters do not vary for new
technologies by region and remain at the same level in the EPANMD. Residual capacity for
existing technologies varies by region and is summed up for the national model. The fuel
constraints, limiting the amount of electricity generated by each fuel (coal, natural gas,
petroleum, and wind), varies by region in the EPAUS9r, and are averaged for a national level
constraint value. In the EPAUS9r, global constraints are used to control the amount of
electricity generated by each wind class. In the EPANMD, these constraints are converted to
an upper bound on each specific wind technology.
Residential, Commercial, Transportation, and Industrial Sectors
End use demands are summed to a national level. Availability and utilization parameters,
including CF, IBOND(BD), LIFE, and START vary by technology type, but are the same
across regions in the EPAUS9r. Thus, these parameters remain the same in the EPANMD.
Efficiency and cost parameters, including EFF and INVCOST and input/output parameters,
including MA(ENT) and OUT(DM), do not vary across regions in the EPAUS9r, and
therefore remain at the same values in the EPANMD. The residual capacity (RESID) for
technology stock already existing at the beginning of the modeling horizon (2000) in the
EPAUS9r is distributed based on regional demand. Therefore, for a national level RESID for
each existing technology, the RESID was summed up across the regions. Weighted averages
across the nine-regions are calculated for the fuel and technology constraints.
Emissions Data
Emission factors do not vary across the regions in the EPAUS9r, and thus remain at the same
values for national emission factors.
64
-------
References
Aden, et al., 2002 Lignocellulosic Biomass to Ethanol Process Design and Economics Utilizing
Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn Stover. NREL/TP-
510-32438.
AEO (Annual Energy Outlook). 2011. Residential Technology Equipment Description File.
October 2011.
—. 201 la. Commercial Technology Equipment Description File (KTECH). October 2011.
Provided by Erin Boedecker.
Argonne National Laboratory. 2007. The Greenhouse Gases, Regulated Emissions, and Energy
Use in Transportation (GREET) Model. Available at: http://greet.es.anl.gov/
Battye, W.; Boyer, K.; Pace, T.G. 2002. Methods for improving global inventories of black
carbon and organic carbonparticulates. (68-D-98-046). Prepared for Environmental
Protection Agency.
Bond, T.C.; Streets, D.G.; Yarber, K.F.; Nelson, S.M.; Woo, J.-H.; Klimont, Z. A Technology-
Based Global Inventory of Black and Organic Carbon Emissions from Combustion. J. of
Geophys. Res. 2004, 109, D14203.
Bond, T.C.; Bhardwaj, E.; Dong, R.; Jogani, R.; Jung, S.; Roden, C.; Streets, D.G.; Trautmann,
N. Historical Emissions of Black and Organic Carbon Aerosols from Energy-related
Combustion, 1850-2000. GlobalBiogeochemical Cycles. 2007, v. 21, GB2018.
Bonskowski, Richard F. 2002. U.S. Metallurgical Coal and Coke Supplies - Prices,
Availability, and the Emerging Future Markets. Iron and Steel Society Annual Conference.
March 2002. Available at: http://www.eia.gov/neic/speeches/isspaper.pdf
Elobeid, A.; Carriquiry, M.; Fabiosa, J.; Mulik, K.; Hayes, D.; Babcock, B.; Dumortier, J.;
Rosas, F. "Greenhouse Gas and Nitrogen Fertilizer Scenarios for U.S. Agriculture and
Global Biofuels" June 2011 Working Paper: 11-WP524 Available at:
http://www.card. iastate.edu/publications/synopsis.aspx?id=l 16
Denholm, P. and W. Short. An Evaluation of Utility System Impacts and Benefits of Optimally
Dispatched Plug-In Hybrid Electric Vehicles. October 2006. (NREL7TP-620-40293)
Available at: http://www.nrel.gov/docs/fy07osti/40293.pdf
DOC (Department of Commerce). 2012. Implicit Price Deflators for Gross Domestic Product.
Available at: http://www.bea.gov/itable/errorNIPA.cfm
DOE (United States Department of Energy). 2008. Documentation for Emissions of Greenhouse
Gases in the United States 2006 (DOE/EIA-0638). October 2008. Available at:
http://www.eia.gov/oiaf/1605/ggrpt/documentation/pdf/0638(2006).pdf
65
-------
—. 2010. Transportation Energy Data Book. Available at: http://cta.ornl.gov/data/index.shtml
—. 201 la. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and Bioproducts Industry
(ORNL/TM-2011/224). December 2011. Available at:
http ://wwwl. eere. energy. gov/biomass/pdfs/billion_ton_update.pdf
Dutta, A., & Phillips, S. (2009). Thermochemical ethanol via direct gasification and mixed
alcohol synthesis of lignocellulosic biomass (NREL/TP-510-45913). Golden, CO: National
Renewable Energy Laboratory: http://www.nrel.gov/docs/fy09osti/45913 .pdf.
Dutta, A., Talmadge, M., Hensley, I, Worley, M., Dudgeon, D., Barton, D., et al. (2011).
Process design and economics for conversion of lignocellulosic biomass to ethanol:
thermochemicalpathway by indirect gasification and mixed alcohol synthesis (NREL/TP-
5100-51400). Golden, CO: National Renewable Energy Laboratory:
http://purl.fdlp.gov/GPO/gpo9356.
EIA (United States Energy Information Administration). 1999. US Coal Reserves: 1997 Update.
(DOE/EIA-0529(97). February 1999. Available at: ftp://ftp.eia.doe.gov/coal/052997.pdf
—. 2001. Transportation Energy Consumption Survey. Available at:
http://www.eia.gov/emeu/rtecs/
—. 2003. Model Documentation Report: System for the Analysis of Global Energy Markets
(SAGE). August 2003. (DOE/EIA-M072).
http://www.eia.gOv/F TPROOT/modeldoc/m072(2003)2.pdf
—. 2004. State Energy Consumption, Price, and Expenditure Estimates (SEDS). Available at:
http://www.eia.gov/beta/state/seds/
—. 2006. Manufacturing Energy Consumption Survey. Available at:
http ://www. eia.gov/emeu/mecs/
—. 2007. EIA-Technology Forecast Updates - Residential and Commercial Building
Technologies - Reference Case Second Edition (Revised). September 2007. Prepared by
Navigant Consulting. Available at:
http://wpui.wisc.edu/news/EIA%20Posts/EIA%20Reference%20Case%2009-
2007%20Second%20Edition%20Final.pdf
—. 2007a. Commercial Buildings Energy Consumption Survey. Available at:
http://www.eia.gov/consumption/commercial/
—. 2009. The National Energy Modeling System: An Overview 2009. (DOE/EIA-0581). October
2009. Available at: http://www.eia.gov/oiaf/aeo/overview/pdf/0581 (2009).pdf
—. 2009a. Annual Energy Review 2008. (DOE/EIA-0384). June 2009. Available at:
66
-------
http://www.eia.gov/totalenergy/data/annual/archive/038408.pdf
—. 2009c. Reidential Energy Consumption Survey. Available at:
http://www.eia.gov/consumption/residential/data/2009/
—. 2011. Documentation of the Oil and Gas Supply Module (DOE/EIA-M063). July 2011.
Available at: http://www.eia. gov/FTPROOT/modeldoc/m063(2011).pdf
—. 2012a. Natural Gas Transmission and Distribution Module of the National Energy Modeling
System (DOE/EIA-M062). February 2012. Available at:
http://www.eia.gOv/F TPROOT/modeldoc/m062(2011).pdf
—. 2012b. U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Proved Reserves, 2010.
August 2012. Available at:
http://www.eia.gov/naturalgas/crudeoilreserves/pdf/uscrudeoil.pdf
—. 2012c. Annual Energy Outlook 2012 with Projections to 2035 (DOE/EIA-0383). June 2012.
Available at: http://www.eia.gov/forecasts/archive/aeo 12/pdf/0383(2012).pdf
—. 2013. Coal Market Module of the National Energy Modeling System. January 2013.
Available at: http://www.eia.gov/forecasts/nemsdoc/coal/pdf/m060(2012).pdf
EPA (United States Environmental Protection Agency). 2004. Air Quality Data Analysis
Technical Support Document for the Proposed Inter state Air Quality Rule. January 2004.
Available at: http://www.epa.gov/cair/pdfs/tsd0006.pdf
EPA (United States Environmental Protection Agency). —. 2004a. Final Regulatory Analysis:
Control of Emissions from Nonroad Diesel Engines. 2004. (EPA420-R-04-007).
—. 2007. Energy Trends in Selected Manufacturing Sectors: Opportunities and Challenges for
Environmentally Preferable Energy Outcomes, US EPA, Office of Policy, Economics
and Innovation, pp.204.
—. 2010. Moves (Motor Vehicle Emission Simulator). 2010. Available at:
http://www.epa.gov/otaq/models/moves/index.htmtfghg
—. 2010b. Renewable Fuel Standard Program (RFS2) Regulatory Impact Analysis. (EPA-420-
R-10-006).
—. 2011. IPM Analysis of the Proposed Mercury and Air Toxics Standard. 2011. Available at:
http://www.epa.gov/airmarkt/progsregs/epa-ipm/proposed-toxics.html
Foster Associates. "Capital Costs for Gas Pipeline and Storage Project Expansions." Report for
EIA. May 1999.
CPBIS (Georgia Tech Center for Paper Business and Industry Studies). 2007. Mills Online
Database. Available at: http://www.cpbis.gatech.edu/data/mills-online-new
67
-------
Humbird, D., Davis, R., Tao, L., Kinchin, C., Hsu, D., Aden, A., et al. (2011). Process design
and economics for biochemical conversion of lignocellulosic biomass to ethanol (NREL/TP-
5100-47764). Golden, CO: National Renewable Energy Laboratory:
http://www.nrel. gov/docs/fy 11 osti/47764 .pdf.
IPCC (Intergovernmental Panel on Climate Change). 2005. Carbon Dioxide Capture and
Storage. Metz, B., Davidson, O., de Coninck, H., Loos, M., and Meyer, L. (Eds.). NY:
Cambridge University Press.
Kaplan, P. O.; DeCarolis, J.; Thorneloe, S., Is it better to burn or bury waste for clean electricity
generation? Environmental Science & Technology 2009, 43 (6), 1711-1717.
Kocoloski, M., Griffin, W.M. and H.S. Matthews (2010). Impacts of facility size and location
decisions on ethanol production cost. Energy Policy, 39(1), 47-56.
Larson, E.; Consonni, S,; Katofsky, R.; Kreutz, T. Preliminary Economics of Black Liquor
Gasification/Gas Turbine Cogeneration at Pulp and Paper Mills. Journal of Engineering for
Gas Turbines and Power. Volume 122, April 2000.
Larson, E.; Consonni, S,; Katofsky, R.; lisa, K., Fredrick, J. A Cost-Benefit Assessment of
Gasification-based Biorefining in the Draft Pulp and Paper Industry. Volume 1, Main
Report, December 2006. DOE Contract Number: DE-FC26_04NT42260.
McAloon, A., F. Taylor, W. Yee, K. Ibsen and R. Wooley (2000). Determining the Cost of
Producing Ethanol from Corn Starch and Lignocellulosic Feedstocks. (NREL/TP-580-
28893). Golden, CO: National Renewable Energy Laboratory:
http://www.nrel.gov/docs/fv01osti/28893.pdf
Mintz, Marianne, Amgad Elgowainy and Jerry Gillette. "H2A Delivery Scenario Analysis
Model Version 2.0 (HDSAM 2.0) User's Manual." September 2008.
MIT (Massachusetts Institute of Technology). 2007. The Future of Coal: Options for a Carbon-
Constrained World. Cambridge, MA: Massachusetts Institute of Technology.
NCDC (National Climatic Data Center). 2010. State, Regional and National Monthly Heating
Degree Days, Historical Climatography Series No. 5.1 (HCS No 5.1). Available at:
http://www.ncdc.noaa.gov/oa/documentlibrary/hcs/hcs.htmltf51updates
North Carolina State University (NCSU). 2010. Database of State Incentives for Renewables
and Efficiency. Available at: http://www.dsireusa.org/rpsdata/index.cfm
NRC (National Research Council). 2004. The Hydrogen Economy: Opportunities, Costs,
Barriers, andR&D Needs. Washington, D.C.
ORNL (Oak Ridge National Laboratory). 2004. Off-highway Transportation Related Fuel Use.
Washington, D.C.
68
-------
—. 2004b. Transportation Routing Analysis Geographic Information System (TRAGIS).
Available at:
http://www.ornl.gov/sci/gist/TRAGIS 2005.pdf
Perrm, R. K., Fretes, N. F., & Sesmero, J. P. (2009). Efficiency in Midwest US corn ethanol
plants: A plant survey. Energy Policy, 37(4), 1309-1316.
Simmons, Phil, Scott M. Kaufman and Nickolas J. Themelis. "State of Garbage in America
Recycling Data Analysis." BioCycle, October 2006. Vol. 47, No. 10, page 21.
USDA (U.S. Department of Agriculture). 2012. Agricultural Long-term Projections. Released
February 2012. Available at: http://www.usda.gov/oce/commoditv/ag baseline.htm
Wilson, E.J., T.L. Johnson and D.W. Keith. 2003. "Regulating the ultimate sink: Managing the
risks of geologic CC>2 sequestration." Environmental Science and Technology 37:3476-
3483.
Wright, M.M.,, J.A. Satrio, R.C. Brown, D.E. Daugaard, and D.D. Hsu (2010). Techno-
Economic Analysis ofBiomass Fast Pyrolysis to Transportation Fuels (NREL/TP-6A20-
46586). Golden, CO: National Renewable Energy Laboratory:
http://www.nrel.gov/docs/fyllosti/46586.pdf
Zhang, Y., M.A. Dube, D.D. McLean, M. Kates (2003a). "Biodiesel production from waste
cooking oil: Part 1. Process design and technological assessment" Bioresource Technology
(83)1.
Zhang, Y., M.A. Dube, D.D. McLean, M. Kates (2003b). "Biodiesel production from waste
cooking oil: Part 2. Economic assessment and sensitivity analysis" Bioresource Technology
(83)1.
69
-------
Appendices
A: MARKAL Parameter Descriptions
AF(Z)(Y): Availability of relevant conversion technologies in the specified season (Z) and time-
of-day division (Y) allowing for variation in availability according to time division. This
parameter is used primarily for renewable technologies, such as hydroelectric where water levels
may vary by season. AF(Z)(Y) is specified as a decimal fraction of the time available during
each of the six time divisions.
AF: Total annual availability of a process or conversion technology in each period. This
parameter is presented as a decimal fraction of the total number of hours (8760 hours) per year
and, as a result, is dimensionless. If availability by season and time-of-day (AF(Z)(Y)), or a fixed
capacity utilization (CF(Z)(Y)) is specified, then AF is not required. The default value is 1.
AF: Total annual availability of a process technology in each period. This parameter is presented
as a fraction of the total number of hours (8760 hours) per year, and as a result is dimensionless.
The default is 1.
BAS(E)LOAD: Baseload capacity of the electricity generation system as a fraction of the total
night production of electricity (measured in terms of the generation capacity) in each specified
period. In MARKAL baseload power plants are forced to operate at the same levels day and
night within a season, that is, if used they tend not to be turned off and on. This parameter is
expressed as a decimal fraction, and is generally close to 1.
BOUND(BD): A limit (lower, upper, or fixed) on the installed capacity of a technology in a
specified period. The value of this parameter is usually set as the average of the annual installed
capacity of a technology in a specified period. BOUND(BD) can be used to anchor the model to
historical data or where future capacity of a technology may be limited in a period.
BOUND(BD)O: Limit on the activity of a given technology during a time period. An upper,
lower, or fixed bound may be applied to a technology's total output (i.e., the sum of all outputs).
BOUND(BD)Or: If a resource is limited during a period as a result of technical or economic
characteristics, this parameter can provide a lower (LO), upper (UP), or fixed (FX) limit. An
example of a technical reason would be the limit on the amount of oil that can be produced
during any given year from a petroleum reservoir. An economic limit would indicate that, at the
price reflected in the COST parameter, only the stated amount could be produced; more, if
available, would be at a higher COST. BOUND(BD)Or is measured as the production actually
available for injection into the system, e.g., natural gas used in production or maintenance of a
lease would be excluded from the values applied for this parameter.
CAPUNIT: Conversion factor between units of activity and units of production capacity. For
process technologies, this parameter is defined as PJ (annual activity) per PJ per annum (unit of
capacity). For process technologies, this parameter defaults to 1 (PJ/PJa). For demand
technologies, this parameter will vary depending on the basis (e.g., light duty vehicles are
70
-------
presented in vehicle miles traveled, while commercial HVAC is presented in PJ). For conversion
technologies, this parameter is defined as PJ (annual activity) per GW (unit of capacity). For
conversion technologies, this parameter defaults to 31.536 PJ/GW.
CF: Average annual utilization of the installed capacity of a demand technology in a specified
period. Because MARKAL works with average annual device output, but investment in devices
is usually based upon rated maximum output, CF serves to "pump up" the cost of a new
investment to the capacity needed to service the highest demand. For example, an air
conditioner is purchased with a rating that will cool an entire house adequately on the hottest
days, but normally runs at a much lower utilization level. So a CF = 0.35 would indicate that
average use is only 35% of the maximum rated cooling capacity and would result in 1/0.35 units
overcapacity installed. This parameter is expressed as a decimal fraction and defaults to 1 (i.e.,
100% utilization during the year) when created.
COST: Annual cost of energy from a source in a specified period. Sources can include domestic
extraction/mining (MIN), imports (IMP), exports (EXP), renewables (RNW), or stockpiles
(STK). Cost excludes any costs associated with delivery of the carrier from the source to the
user. Further, costs are generally valued at the point that an energy carrier enters the energy
system. For example, coal mined domestically would be valued at the mine-mouth, and
domestically produced natural gas would be valued at the wellhead. Because exports represent a
cost reduction to the system, the level of exports usually requires that an upper limit be applied.
CUM: Total availability of an energy carrier from a resource supply curve step over the entire
forecast horizon. An example would be the total proven reserves from a particular group of
reservoirs. Whereas BOUND(BD)Or limits the annual production level (e.g., pumping rate), the
CUM indicates the total that can be extracted over time.
DELIV(ENT): Annual delivery and handling costs of an energy carrier from source to a
specified technology in a specific period.
DEMAND: Annual demand for an energy service. This parameter is expressed in the units of the
demand service. In the EPAUS9r model, demands may be expressed in various units, including
energy units (PJ) as well as such measures as vehicle miles traveled (VMT).
DISCOUNT: Long term annual discount rate for the economy as a whole. This parameter is
expressed as a decimal fraction and currently, in the U.S. model, this parameter is valued at 5%.
DISCRATE: Technology-specific discount rate. This differential discount rate, often referred to
as a "hurdle" rate, can be used to represent the impediments that a technology may face when
competing in the market strictly on economic terms. These impediments might be a function of
preference (e.g., SUVs vs. compact cars), behavior (e.g., builders put in electric hot water heaters
owing to lower initial cost), or high required rates of return and/or lack of understanding (e.g.,
the extra upfront $ for compact fluorescent light bulbs or high efficiency refrigerators are not
perceived to be justified by the eventual down-the-road energy cost savings). These impediments
to the market can be represented to MARKAL as higher discount rates for such technologies.
The value is a percent.
71
-------
(E)RESERV: Amount of installed electricity generating capacity above the average load of the
season and time-of-day division of peak demand. This parameter is calculated by dividing the
reserve capacity (i.e., sum of scheduled outages, forced outages, and load required to meet the
peak requirement) by the average load of the season/time-of-day peak load, and is expressed as a
decimal fraction. This number is usually higher than the traditional utility reserve margin as it is
the level above the average peak period load, not the peak itself.
EFF: Technical efficiency of a demand technology in a specified period. Efficiency is measured
as the number of units of end-use demand satisfied per unit of input energy carrier consumed.
EFF is a decimal fraction, and usually, with existing technologies, it does not vary over the life.
In the case of some demand technologies, such as commercial HVAC, this measure is PJ in per
PJ out. For other technologies such as light duty vehicles, this measure is PJ in/ number of
vehicle miles traveled. The majority of demand technologies will have efficiencies of less than
one. However, such technologies as heat pumps may have efficiencies greater than one. For
some technologies, such as lighting, where outputs are not easily measured in PJ, the efficiency
of some standard technology has been set to 1, and efficiencies for other technologies and
demand levels have been calibrated accordingly.
ELF: Fraction of the electricity consumed by technologies in an end-use demand category that is
to be included in the electricity peaking constraint. This parameter should be specified if a
demand category has one or more demand technologies using electricity. A value of less than 1
would be provided only to indicate that not all the demand will coincide with the peak moment
(e.g., as a result of off-peak electric pricing to industry). Default is 1 indicating that all the
electricity consumed should be charged in the peaking constraint. The value is a dimensionless
fraction.
ENV_ACT: Emission coefficient per unit of activity. This parameter specifies the quantity of an
environmental emission or other variable that is associated with the actual operation of a
conversion or process technology or demand device. Most emissions associated with the burning
of a fuel (e.g., at a power plant or vehicle) are represented using this parameter. The parameter
specifies the amount of pollutant emitted per unit of technology activity. The units are a function
of the activity unit of the technology (usually PJ) and the unit of the environmental variable (in
the U.S. model, million tons for CC>2 and thousand tons for the other pollutants.)
ENV_COST: The cost ("tax") added to the objective function associated with each unit
emissions.
ENV_MAXEM: Annual maximum (net) quantity of an emission or other environmental
variable that can be released (or used) in a given period. This parameter is used to model such
environmental regulations as limitations on emissions of SC>2 or NOX under CAIR. The units are
those of the environmental variable.
ENV_SEP: Emission coefficient per unit of resource activity. This approach links emissions or
other environmental variables to a unit of resource activity (e.g., tons of CFLi released in the
production of a P J of coal). Resource activities include coal mining, natural gas and oil
72
-------
production, and similar extractive activities. The units are a function of the activity unit of the
technology (PJ) and the unit of the environmental variable
FIXOM: Annual fixed O&M costs associated with the installed capacity of a demand
technology. Costs associated with this parameter are assessed without regard for the annual
capacity utilization rate of a technology. The same FIXOM is charged whether the process
actually operates or not. Examples of fixed O&M include fixed labor costs, rental of a building,
property taxes, and similar costs.
FR(Z)(Y): For each demand sector that can be serviced by demand devices consuming
electricity or heat, FR(Z)(Y) indicates the fraction of annual consumption occurring in each
season and time-of-day subdivision that best describes the end-use demand load through the
typical year. This parameter is required where the load is non-uniform, e.g., space heating and
space cooling. If the demand is not required in a time slice, then the associated FR(Z)(Y) should
be omitted or specified as 0. The sum over all demands for each season/time-of-day time slice
determines the amount of electricity and heat that must be generated by the conversion plants in
that time slice, as well as the peaking and baseload requirements. If this parameter is not
specified, MARKAL defaults to the seasonal/time-of-day energy distribution specified across the
entire system by the global parameter QHR(Z)(Y). The FR(Z)(Y) should sum to 1.
GROWTH: Maximum rate of growth of the capacity of a technology from period to period.
This parameter is expressed as the sum of 1 and the annual growth rate. For example, a growth
rate of 15% over a period would be expressed as 1.15 in the designated period. If a technology
has a 5% annual growth rate, then the period growth rate for a five-year period = 1.05 5 = 1.276.
GROWTH_TID: Incremental quantity of a technology's capacity that is permitted over and
above the GROWTH constraint. This parameter provides flexibility in allowing investment in
new capacity. In addition, if technology capacity could be zero in the first time period that
GROWTH is applied, a GROWTH_TID with the anticipated initial maximum period penetration
level must be provided in order to "seed" the growth constraint. The GROWTH_TID should be
set to the initial level of new capacity expected to be possible when the technology first becomes
available.
IBOND(BD): Limit on the investment in capacity for a technology during a specified period.
Investment in MARKAL is assumed to occur at the beginning of a period. This parameter will
therefore impact investment for an entire period. This constraint may have an upper, lower, or
fixed bound. IBOND(BD) is specified in the units for capacity.
INP(ENT)c: Amount of an energy carrier(s) that is input to a conversion technology in a
specified period. Each energy carrier requires a separate INP(ENT)c entry and can be interpreted
as the number of PJs of an input energy carrier per unit of electricity produced. Another way of
thinking of it is that INP(ENT)c is expressed as the inverse of the efficiency (1/eff) of the power
plant. For renewables and other non-fossil fuels (e.g., nuclear), the standard practice is to use the
average efficiency from fossil technologies to compute the required "fossil fuel equivalent." The
model uses this value when reporting consumption of primary energy (as fossil equivalent).
73
-------
INP(ENT)p: Amount of an energy carrier(s) that is input to a process technology during a
specified period. Each energy carrier requires a separate INP(ENT)p entry, which can be
interpreted as the number of PJs of that input carrier required per unit of activity (output) from
the process. In the EPAUS9r, processes are usually normalized to the sum of the output carriers;
i.e., the outputs should sum to one. However, in a number of limited cases, inputs are normalized
to a subset of inputs (e.g., some of the technologies depicted in the industrial sector). The ratio of
the sum of output energy carriers to the sum of input energy carriers represents the technical
efficiency of the process technology activity. Because INP(ENT)p relates outputs to inputs, its
units are dependent upon the units of the outputs and inputs.
INVCOST: Total cost of investment for one incremental unit of new capacity in a specified
period. The unit of capacity becomes available for production at the beginning of the specified
period, and the investment cost is assumed to be charged at the beginning of that period.
LIFE: Technical life or period of potential operation of a technology. Note that the technical and
economic lifetimes are assumed to be the same in MARKAL. This parameter is expressed in
number of years.
LIMIT: Sets a maximum total output per unit of input for a process having flexible (that is,
model-chosen) outputs, as opposed to the normal fixed proportions. When LIMIT is specified,
the OUT(ENC)p then represents the maximum share of each energy carrier that can be produced
per PJ of input. Thus the OUT(ENC)p in this case may sum to > 1, though the overall level in the
solution will adhere to the LIMIT in accordance to the input levels.
MA(ENT): Fraction of each energy carrier that is input to a demand technology in a given
period. For multiple energy carrier inputs, this parameter is expressed as a fraction of the total
inputs to a demand technology. In most cases, when a single energy carrier feeds a demand
device, MA(ENT) equals 1.
OUT(DM): Fraction of the output per activity of a demand technology that contributes to
servicing an end-use demand in a specified period. If more than one end use is satisfied, the sum
of this parameter must equal 1. This parameter can be used for such demand technologies as heat
pumps that satisfy both heating and cooling demand or heating systems that also produce hot
water.
OUT(ENC)p: Amount of an energy carrier produced by a process technology per unit activity.
Each energy carrier output from a technology requires a separate OUT(ENC)p entry. As with
input energy carriers, output carriers are described as so many PJ output per PJ input.
OUT(ENT)r: Specifies the energy carrier entering the energy system due to a resource activity
(e.g., natural gas from a well or coal from a strip-mine.). The value of this parameter should be
set to 1, the default.
74
-------
PEAK(CON): Fraction of the total capacity of a conversion technology in a specified period that
can be counted on to be available to meet peak demand and reserve margin requirements, based
upon its availability, reliability, and other considerations. Base-loaded technologies will have
values close to 1, while other technologies, such as renewables, may have values less than 0.5.
QHR(Z)(Y): Division of the year into season and time-of-day fractions that describe the
duration of the seasons and time-of-day associated with a typical year. These values are used as
the default when constructing the load curve for electricity and heat.
RESID: Capacity of a technology installed prior to the start of the modeling horizon, or capacity
in place today. The investment cost for previously installed capacity is assumed to be sunk, and
the model is thus encouraged to use existing capacity before new capacity is added. This
parameter is defined in the same units as other capacity measures, and in the U.S. model demand
device capacities are expressed in PJa, billions of vehicle miles traveled per annum, etc.
START: First period of availability of a technology within the modeling horizon. For
technologies already in existence at the beginning of the modeling horizon, START should be set
to the first period, which is the default.
TE(ENT): Average transmission and distribution efficiency of a specified energy carrier in a
given period. This parameter is applicable to all energy carriers and, for the majority of energy
carriers, is set to 1 (i.e., 100% efficiency). The default value is 1. Electricity generally has a
value of less than 1, and in the EPAUS9r has a value of 0.9350.
TRNEFF(Z)(Y): Specifies the average transmission efficiency of low temperature heat from a
coupled-production (CPD) technology during each season/time-of-day. The value is a
dimensionless fraction.
VAROM: Annual variable O&M costs of a technology. These costs are assessed as each unit of
activity occurs and are to be normalized in the same manner as the process itself (usually to
output). VAROM does not include fuel costs or the costs associated with fuel handling and
delivery.
75
-------
B: Sector Workbook Description - Oil Resource Supply
Workbook Name: EPAUS9r_12_Oil_vl.0.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
domestic and imported oil resources and the imported refined products in the EPAUS9r
MARKAL database.
Data Sources
The supply characteristics were taken from the AEO reference case (EIA, 2012c) and the NEMS
Oil and Gas Supply Module (EIA, 2011) output data. Emissions factor data were derived from
the GREET model, version 1.8 (Argonne, 2007).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Emission factors are expressed in ktonnes per PJ with the exception of CC>2, which is expressed
in Mtonnes per PJ.
Workbook Description
The following section gives a description of each of the 25 worksheets found in the oil resource
supply workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all energy carriers used for oil
supply. The following naming conventions are used (where R represents region number, P
represents the PADD number, and FFF represents the 3-letter distinction for fuel type):
OILDR Domestic Crude Oil
OILIPP Imported Crude Oil, all types
OILtfHP Imported Oil, high sulfur, heavy gravity
OILHLP Imported Oil, high sulfur, low gravity
OILHVP Imported Oil, high sulfur, very heavy gravity
OILLLP Imported Oil, low sulfur, low gravity
OILMHP Imported Oil, medium sulfur, heavy gravity
FFFPP Imported Refined Products
Technologies*
Lists the technology names, units, and set memberships for all supply curve steps, import and
export technologies, and transportation technologies for domestic and imported oil and imported
refined products. The following naming conventions are used (where X represents the step
76
-------
number, YY represents the sulfur type and specific gravity, XX represents the PADD and step
number, and FFF represents the fuel type):
MINOILDX Domestic Crude Oil Supply Step
IMPOILYYXX Imported Crude Oil Supply Step
IMPZZZPXX Imported Refined Product Supply
TechData RO*
Contains the costs (COST), supply limits (BOUND(BD)Or), and cumulative supply amounts
(CUM) for each of the supply steps for domestic oil and imported refined products. The data are
drawn from the worksheets described below. In addition, each supply step has a maximum
annual growth rate in activity from 2.5% to 5% (GROWTHr) and a limit rate at which supply
activity can be reduced by 5% annually (DECAYr). Once oil is "purchased" from the supply
curve it passes through respective collectors (SCMINOILD) where emissions associated with
extraction are applied. These emission factors are pulled from the Emissions worksheet.
TechData_ExIm*
Contains the trading links needed to move oil between regions. Export (names begin with EXP)
and import (names begin with IMP) trade technologies are used to export oil out of one region
and import it into another region. The parameter BI_TRD(ENT) is equal to 1 when a region can
receive a particular source.
TechData DelOIL*
Contains the bound on capacity (BOUND) and costs (VAROM) for the crude oil and imported
refined product transportation technologies including transport between regions and transport to
end-use sectors.
TechData_Limits*
Contains the investment costs and flow bounds for pipeline delivery of crude oil and imported
refined products.
Constraints*
There are no user-defined constraints in the Oil workbook.
ConstrData*
There are no user-defined constraints in the Oil workbook.
Domestic Crude Production Trends
Contains data and charts that graph the historical annual production of crude oil in each of the
PADDs in the U.S. Data are used to estimate GROWTH and DECAY values for domestic crude
supply. These data can be found on the EIA's website at the following link:
http://www.eia.gov/dnav/pet/pet crdcrpdnadcmbbla.htm
Domestic Crude Oil CUM
Contains the raw data and calculations for determining the cumulative (CUM) amount of crude
oil reserves in each of the nine-regions. The data are pulled from the EIA proved reserves data
77
-------
(EIA, 2012b). Raw data are converted from million barrels to PJ and multiplied by a factor used
to estimate how the reserves will change over time.
Domestic Crude Oil Prices
Contains the raw data used to calculate crude oil production prices. Data are taken from two
AEO reference case tables: "Lower 48 Crude Oil Production and Wellhead Prices by Supply
Region," and "Oil and Gas Supply." Data are regionalized and converted to MARKAL units. A
growth factor is calculated and used to extend the prices out beyond 2035.
Domestic Crude Oil Prod
Contains the raw data used to calculate crude oil production by state for the years 2002 through
2011. These data can be found at http://www.eia.gov/dnav/pet/pet crdcrpdnadcmbbla.htm.
Average production levels for 2005 and 2010 are calculated by region. Those average values are
then converted to MARKAL units and set as the production bound (BOUND(BD)Or) for Step 3
in the regional supply curves. Steps 1, 2, 4, and 5 are then calculated from Step 3 using supply
curve elasticities taken from AEO.
Imported Prod Trend
Contains data and charts that graph the historical annual production of imported crude oil and of
refined products in each of the PADDs in the U.S. Data are used to estimate GROWTH and
DECAY values for domestic crude supply. These data can be found on the EIA's website at the
following link: http://www.eia.gov/dnav/pet/pet move wklv dc NUS-ZOO mbblpd 4.htm
Imp Refined
Contains calculations used to convert the imported refined supply curve data aggregated in the
worksheet Agg prdcrv to MARKAL units of million dollars per pJ. In addition, standard
naming conventions are applied to each step of the supply curve.
Agg prdcrv
Aggregates the imported refined products supply curve data found in the worksheet,
AEO 2012 prdcrv.
AEO prdcrv
Contains the raw data for the AEO imported refined products supply quantities and prices
provided by NEMS.
Imp Crude
Contains the calculations used to convert crude supply curve data from the worksheet
Agg crdcrv to MARKAL units of million dollars per pJ. In addition, standard naming
conventions are applied to each step of the supply curve.
Agg crdcrv
Aggregates the crude oil supply curve data found in the worksheet, AEO crdcrv.
78
-------
AEO crdcrv
Contains raw data for imported crude oil supply quantities and prices provided by NEMS.
Imported Price Trends
Contains the raw data for refined petroleum product prices found in the AEO reference case
table: "Petroleum Product Prices.'
DelivOIL
Contains the raw data and calculations for the delivery prices of crude oil to the refineries and
the delivery prices of imported refined oil products to the end-use sectors. Raw data tables come
from the NEMS Petroleum Market Module and include Table ID: "Refinery Gate Prices" and
Table 8: "End Use Prices."
OILTrade
Contains raw data from the AEO's Oil and Refining Markets division showing the total capacity
of refined products moved from one region to another through pipelines in the U.S. These data
are used to determine pipeline capacities.
Emissions
Contains emission factors associated with natural gas and oil extraction, processing and
distribution. These factors do not differentiate between conventional and unconventional
resources.
Types
Contains a series of tables matching fuel types and regions with MARKAL designations for
those same fuel types and regions.
Conversion Factors
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) and
the Annual Energy Review Table Al: "Approximate Heat Content of Petroleum Products."
79
-------
C: Sector Workbook Description - Natural Gas Resource Supply
Workbook Name: EPAUS9r_12_NatGas_vl.0.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
domestic and imported natural gas resources in the EPAUS9r MARKAL database.
Data Sources
The supply characteristics were taken from the AEO reference case (EIA, 2012a) and the NEMS
Oil and Gas Supply Module (EIA, 2011) output data. Emissions factor data were derived from
the GREET model, version 1.8 (Argonne, 2007).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Emission factors are expressed in ktonnes per PJ with the exception of CC>2, which is expressed
in Mtonnes per PJ.
Workbook Description
The following section gives a description of each of the 19 worksheets in the natural gas resource
supply workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all energy carriers used for natural
gas supply. The following naming conventions are used:
NGA Natural Gas
NGAD Domestic Natural Gas
NGAC Imported Canadian Natural Gas
NGAL Imported Liquified Natural Gass
NGAM Natural Gas Net Exports to Mexico
NGL Natural Gas Liquids
Technologies*
Lists the technology names, units, and set memberships for all supply curve steps, import and
export technologies, and transportation technologies for domestic and imported natural gas
supplies. The following naming conventions are used (where X represents the step number):
MINNGADX Domestic Natural Gas Supply Step
EVINPNALX Imported NatGas Liquids Supply Step
EVIPNGACX Imported Canadian NatGas Supply Step
80
-------
TechData RO*
Contains the costs (COST), supply limits (BOUND(BD)Or), and cumulative supply amounts
(CUM) for each of the supply steps for natural gas. The data are drawn from the worksheets
described below. In addition, each supply step has a maximum annual growth rate in activity
from 2.5% to 30% (GROWTHr) and a limit rate at which supply activity can be reduced 5%
annually (DECAYr). Domestic natural gas supply passes through respective collectors
(SCMINNGAD) where emissions associated with extraction are applied. These emission factors
are pulled from the Emissions worksheet.
TechData Exlm*
Contains the trading links needed to move natural gas between regions. Export (names being
with EXP) and import (names begin with IMP) trade technologies are used to export gas out of
one region and import gas into another region. The parameter BI_TRD(ENT) is equal to 1 when
a region can receive a particular source.
TechData DelNGA*
Contains the bound on capacity (BOUND) and costs (VAROM parameter) for the natural gas
transportation technologies including transport between regions and transport to end-use sectors.
TechData_Limits*
Contains the investment costs and flow bounds for pipeline delivery of natural gas.
Constraints*
There are no user-defined constraints in the NatGas workbook.
ConstrData*
There are no user-defined constraints in the NatGas workbook.
Types
Contains a series of tables matching fuel types and regions with MARKAL designations for
those same fuel types and regions.
Domestic NG CUM
Contains the raw data and calculations for determining the cumulative (CUM) amount of natural
gas reserves in each of the nine-regions. The data are pulled from EIA proved reserves data
(EIA, 2012b). Raw data are converted from billion cubic feet to petajoules and multiplied by a
factor used to estimate how the reserves will change over time.
Domestic NG Production
Contains the raw data used to calculate natural gas production by state for the years 2002 through
2010. These data can be found at http://www.eia.gov/dnav/ng/ng_prod sum dcu NUS m.htm.
Average production levels for 2005 and 2010 are calculated by region. Those average values are
converted to MARKAL units and set as the production bound for Step 3 in the regional supply
curves. Steps 1, 2, 4, 5 and 6 are then calculated from Step 3 using supply curve elasticities
taken from AEO.
81
-------
Domestic NG Prices
Contains the raw data for natural gas production prices. Data are taken from two AEO reference
case tables: "Lower 48 Natural Gas Production and Wellhead Prices by Supply Region," and
"Oil and Gas Supply." Data are regionalized and converted to MARKAL units. A growth factor
is calculated and used to extend the prices out beyond 2035.
NGATrade
Contains the raw data for regional natural gas imports. Data taken from AEO Table: "Primary
Natural Gas Flows Entering NGTDM Region from Neighboring Regions" provides the raw data
for regional pipeline natural gas imports from Canada. Table 76: "Natural Gas Imports and
Exports" provides the raw data for regional liquified natural gas imports.
AEO NG Imports
Contains the raw data and calculations for the imported natural gas supply curves, both pipeline
gas from Canada and LNG. Raw data comes from the AEO reference case table: "Natural Gas
Imports and Exports." Average import levels for 2005 and 2010 are calculated and converted to
MARKAL units and then set as the production bound (BOUND(BD)Or) for Step 3 in the
imported supply curves. Steps 1, 2, 4, 5 and 6 are then calculated from Step 3 using supply
curve elasticities taken from AEO.
DelivNGA
Contains the raw data and calculations for regional delivery costs to transport natural gas from
production to the end-use sectors. End-use sector prices come from the AEO Table: "Natural
Gas Delivered Prices by End-Use Sector and Census Division." MARKAL delivery costs
(VAROM) are calculated by subtracting the wellhead gas prices, found on worksheet NGADom,
from these end-use sector delivered prices.
NGACap
Contains the raw data and calculations for the upper bound on capacity for existing natural gas
pipelines. Raw data are taken from two AEO tables: "Natural Gas Pipeline Capacity by
NGTDM Region" and "Primary Natural Gas Capacity Entering NGTDM Region from
Neighboring Regions."
NGExpansion
Contains the raw data and calculations for the cost of natural gas pipeline expansion in each
region. The raw data are taken from a report done for EIA on costs for pipeline expansion
projects (Foster Associates).
Emissions
Contains emission factors associated with natural gas and oil extraction, processing and
distribution. These factors do not differentiate between conventional and unconventional
resources.
82
-------
Conv
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) and
the Annual Energy Review Table Al: "Approximate Heat Content of Petroleum Products."
83
-------
D: Sector Workbook Description - Coal
Workbook Name: EPAUS9r_12_Coal_vl.0.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
domestic coal resources in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics were taken from the AEO reference case and NEMS Coal Market
Module (EIA, 2013) output data. Emission factors were derived from the GREET model,
version 1.8 (Argonne, 2007).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Emission factors for CC>2 are in units of MTonnes/PJ. Units for all other pollutants and for CC>2
attributed specifically to this sector, are in units of kTonnes/PJ.
Workbook Description
The following section gives a description of each of the 22 worksheets in the coal resource
supply workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all coal energy carriers. Coal types
follow the naming convention given below:
First letter = C (Coal)
Next 2 letters = Coal Region
Final 3 letters = Coal Type
Coal region and coal type descriptions can be found on the worksheet
AEO CMM BASE YR SUPPLY CURVE
Technologies*
Lists the technology names, units, and set memberships for coal supply curve steps, import
technologies, export technologies, and coal transportation technologies.
Each supply curve step follows the naming convention given below:
First 3 letters = MIN
Next 2 letters = Coal Region
Next 3 letters = Coal Type
Final letter = Step on the curve (from A to K)
84
-------
Import and Export technologies follow the naming convention given below:
First 3 letters = IMP (import) or EXP (export)
4th letter = C (coal)
Next 2 letters = Coal Region
Next 3 letters = Coal Type
Final letter = Region
Transportation technologies follow the naming convention given below:
First letter = X (denotes a transportation technology)
2nd letter = C (coal)
Next 2 letters = Coal Region
Next 3 letters = Coal Type
Final 3 letters = End-use demand distinction (i.e. RES for residential)
TechData RO*
Contains the costs (COST), supply limits (BOUND(BD)Or), and cumulative supply amounts
(CUM) for each of the eleven steps in each of the 41 supply curves. The data are drawn from the
worksheets described below. In addition, each supply curve step is allowed a maximum annual
growth rate in activity of 5% (GROWTHr) and a limit rate at which activity can be reduced 3%
annually (DECAYr).
TechData Exlm*
Contains technology links that transport coal from the dummy supply region (RO) to one of the
nine census regions. The parameter BI_TRD(ENT) is equal to 1 when a region can receive a
particular coal source.
TechData_Trans*
Contains the costs (VAROM) for coal transportation technologies. Data comes from the Trans-
rates worksheet.
Emissions
Contains emissions associated with coal mining and cleaning. Coalbed methane emissions are
included for surface and underground mines in various coal supply regions. These factors are not
currently used within MARKAL. More information about these factors is available upon request.
Data-curves conversions
Converts the supply curve step quantities and prices from the worksheet Agg 5 YR into
MARKAL units of PJ and 2005 U.S. dollars per PJ.
Agg SYR
Sorts the average data from the 5 YR AVG worksheet by supply curve and adds naming
conventions.
85
-------
5YRAVG
Sorts the supply curve data from the AEO Supply Curves Raw worksheet by coal region and
coal type and calculates 5-year averages for the Step Quantity and Step Price.
AEO Supply Curves Raw
Contains the raw data for the coal supply curves from the AEO reference case.
Types
Expands the CMM_BASE_YR_SUPPLY_CURVE worksheet, adding naming conventions,
cumulative productions, weighted sulfur and mercury totals, and reserves.
AEO CMM BASE YR SUPPLY CURVE
Contains the raw coal supply data from the AEO CMM BASE YR SUPPLY CURVE output file.
This file gives the base year production and prices for each of the 40 coal supply curves.
Data-reserves
Contains raw data from the EIA coal reserves data (EIA, 1999) which gives estimated
recoverable reserves by heat and sulfur content.
Coke Imports
Contains the calculations used to develop a supply curve for imported coke. The step quantities
and step prices for the years 2000 and 2005 are taken from historical data of coke imports and
exports found in a EIA report on metallurgical coal and coke supplies (Bonskowsi, 2002) and the
Annual Energy Review (EIA, 2009a). For years out beyond the historical data, the step
quantities are calculated by applying a quantity growth factor calculated from the AEO net
imports data found in Table 15: "Coal Supply, Disposition, and Prices." The step prices of the
out years are calculated by applying an average coal import price to export price factor to the
export price in the out years.
Trade-rules
Contains a manually generated table of supply rules which holds a "Y" if there is trading out of
or in to a region of a particular coal type and a "N" if there is no trading between that coal type
and a particular region.
Rate esc
Contains the raw data from the AEO CMM TRAN RATE ESC COMP output file, which gives
the projections for the change in transportation rates out to the year 2055 for the eastern and
western portion of the U.S. An average is calculated for the east and the west which is applied to
the projections for the VAROM in the TechData_Trans worksheet.
Trans-rates
Takes the data from the AEO CMM ORIGIN DEST RATES raw data file and calculates the cost
in 2005 million dollars per PJ. In addition, transportation technology names are assigned to
represent each node in the coal transportation system (i.e. "XCCABLSRES" represents the
transportation of Central Appalachian low sulfur, surface bituminous coal to the residential
sector).
86
-------
AEO CMM ORIGIN_DEST RATES
Contains the raw data from the AEO CMM ORIGIN DEST RATES output file. This file gives
the costs of different coal types from one region to a particular sector in another region.
Data Production
Contains raw data from the AEO CMM ANN COAL PRODUCTION output file that gives coal
production by year (2011 through 2046) for each coal supply curve. A total production across
the time frame is calculated and used in other worksheets to calculate the cumulative (CUM)
available coal for each supply curve.
Defs
Contains AEO tables that define index numbers used in producing the coal supply curves.
Conversion Factors
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012).
87
-------
E: Sector Workbook Description - Unconventional Fuels
Workbook Name: EPAUS9r_12_UNCNV_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the coal to
liquids and coal to natural gas process technologies in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics were taken from the data gathered by the Southern Research
Institute (SRI) under contract from the EPA. Detailed references are provided on the References
worksheet.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Emission factors are expressed in ktonnes per PJ with the exception of CC>2, which is expressed
in Mtonnes per PJ.
Workbook Description
The following section gives a description of each of the 19 worksheets found in the
unconventional fuels workbook. The worksheets are listed in the order they appear, from left to
right, in the workbook. The worksheet names noted with an asterisk contain the data that are
automatically uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all process energy carriers and
emissions. Energy carrier names include:
DSU_CTL Ultra Low Sulfur Diesel from FT Diesel Process
PFS_CTL Petrochemical Feedstocks from FT Diesel Process
STM_CTL Steam from FT Diesel Process
SNG_CTG Natural Gas from Coal the Gas Process
PFS_CTG Petrochemical Feedstocks from Coal to Gas Process
Technologies*
Lists the technology names, units, and set memberships for the coal to liquids and coal to natural
gas process technologies and collector technologies for the process outputs. Each process
technology follows the naming convention given below:
First letter = P (process)
Next 3 letters = COA (coal)
Next 3 letters = Output (DSU for diesel, SNG for synthetic natural gas)
Final 2 letters = Vintage year (if necessary)
-------
Collector technologies follow the naming convention given below:
First 2 letters = SC (collector)
Next 3 letters = Output (PFS for petrochemical feedstocks, DSU for diesel, SNG
for natural gas)
Next 2-3 letters= Technology Type (FT for Fisher Tropsch, FTN for FT with NG)
Final 2 letters = Vintage year (R for residual, 30 for new)
Constraints*
Lists the names, units, and set memberships for the coal to liquids and coal to natural gas user
constraints.
TechData*
Contains the data for coal to liquids and coal to synthetic natural gas processes. The data are
drawn from worksheets 4A2, 4A3, and 4B3 and converted to MARKAL ready units.
ConstrData*
Contains user defined constraints limiting the amount of coal to liquids and coal to natural gas
that can be produced in a MARKAL model run. Coal to liquids production is based on AEO
results. Coal to synthetic natural gas production is held constant at 2005 levels until 2040, when
it is able to as much as double in production.
Coal to Liquids
Summarizes the collected coal to liquids data and chooses the best representation of the process
for use in MARKAL.
Coal to SNG
Summarizes the collected coal to synthetic natural gas data and chooses the best representation
of the process for use in MARKAL.
4A1
Contains raw data for the production, efficiency, inputs, outputs, and emissions of the Gilberton
Coal-to-Clean Fuels and Power Project.
4A2
Contains raw data and calculations for the cost, efficiency, inputs, outputs, and emissions for the
production of Fischer-Tropsch fuels and other conventional byproducts using Wyoming coal.
4A3
Contains raw calculations for the cost, efficiency, inputs, outputs, and emissions of the Gilberton
Coal-to-Clean Fuels and Power Project.
4B1
Contains coal to synthetic gas raw production data for the Orlando Gasification Project at
Stanton Energy Center.
89
-------
4B2
Contains raw production data for the Dakota Gasification Company coal to synthetic gas process.
4B3
Contains a spreadsheet developed by SRI showing calculations for the costs, efficiencies,
outputs, emissions, and capacities of coal to synthetic plants.
4C1
Contains a spreadsheet developed by SRI showing raw data and calculations for the costs,
efficiencies, and capacities of transport reactor integrated gasification plants, first and nth of its
kind.
AEO2010 Tables and Calibration
Contains raw data from the AEO Table 2: "Energy Consumption by Sector and Source and Table
11: Liquid Fuels Supply and Disposition." Data from these tables are used to determine the
global constraint for coal to liquids in the model.
Conversions
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) chart
used to convert AEO prices to the year 2005.
Tech Authors
Lists the contact information for the researchers from both EPA and SRI who worked on this
workbook.
References
Provides a detailed list of data references, including the full citation, source comments, and
general data assumptions.
Definitions
Provides a list of definitions for various terms used throughout the workbook.
90
-------
F: Sector Workbook Description - Refineries
Workbook Name: EPAUS9r_12_REF_vl. 1 .xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the refinery
process technologies in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics were taken from the AEO reference case. Additional data used to
calculate costs were taken from research done for EPA by Research Triangle Institute (RTI).
Emission factors are derived from the GREET model, version 1.8 (Argonne, 2007). In addition,
speciation factors for black carbon and organic carbon are derived from three studies: Battye et
al., Bond et al., 2004, and Bond et al., 2007.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Emission factors are in kTonnes per PJ, with the exception of CC>2, which is expressed in
MTonnes per PJ.
Workbook Description
The following section gives a description of each of the 18 worksheets found in the refinery
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all refinery input and output fuels
(energy carriers). The refinery fuel outputs are named using a three letter representation of the
fuel name as given below followed by an "E," "N," or "L" to distinguish which refinery type the
fuel was produced by.
LPG Liquid Petroleum Gas
GSC Gasoline - Conventional
GSR Gasoline - Reformulated
JTF Jet Fuel
DSH Distillate Heating Oil No. 2
RFL Low Sulfur Residual Fuel Oil
RFH High Sulfur Residual Fuel Oil
KER Kerosene
PFS Petrochemical Feedstocks
ASP Asphalt
DSL Low Sulfur Highway Diesel (500 ppm)
DSU Ultra-low Sulfur Highway Diesel (15 ppm)
91
-------
PTC Petroleum Coke
DLG Dummy Liquid Fuel
Constraints*
Lists the constraint names for all refinery constraints.
Technologies*
Lists the technology names, units, and set memberships for all refinery and refinery fuel output
collector technologies. The refinery names start with a "P" to indicate a process technology,
followed by "REF" for refinery, and then "E" (existing), "N" (new), or "L" (limit) to distinguish
the type.
The fuel output collector technologies start with "SC" for collector, followed by the three letter
representation of the fuel name as given in the Table above, and then an "E," "N," or "L" to
distinguish which refinery type the fuel was produced by.
ConstrData*
Contains the data for fuel shares for asphalt, petroleum feedstocks, petroleum coke, and LPG
fuels. The constraints force the refineries to produce a minimum amount of these fuels. The
values can range from fifty to seventy-five percent of the historical levels of output that have
been reported by the EIA.
TechData*
Contains the inputs, outputs, and costs for refineries by region. The data are drawn from the
worksheets described below. There are three refinery types in each region: existing refineries,
new refineries (with associated capital costs), and new limit refineries. The new limit refineries
can take up to twenty percent of the output of a refinery and produce higher percentages of
gasoline and diesel fuels.
TechData-Emis*
Contains emission factors associated with refining.
RESID
Contains calculations for the regional residual capacity (RESID) of existing refineries in 2005.
Capacity
Contains calculations for the residual refinery capacity by PADD. Total capacity in million
barrels per day is taken from the AEO reference case table "Domestic Refinery Distillation Base
Capacity, Expansion, and Utilization." From these totals, the capacity used is calculated using
the average utilization over 2010-2012 found on the Utilization worksheet. Process loss and
refinery outputs not used in MARKAL are subtracted from the capacity utilized to come up with
a final residual capacity amount. Data for the process loss and refinery outputs is found in the
Yields worksheet. Finally, based on these residual capacity amounts, the percent of crude oil per
output is calculated as input to the refinery in MARKAL.
92
-------
PADD Products
Takes the yield data by fuel and by PADD from the Yields worksheet. Factors are applied to
give MARKAL a range of fuel output to work from. Gasoline and diesel have higher output
values than the other refinery products.
Yields
Contains raw data from EIA for the refinery and blender percent yields by fuel type and by
PADD. Average values for 2005 and 2010 are calculated. Only the fuels that are tracked in the
database are outputs.
Utilization
Contains raw data from EIA for the refinery inputs and utilization by PADD. Data are used to
calculate the average yearly percent refinery utilization.
DSH Percent
Contains raw data from EIA for the weekly refinery and blender net production by PADD.
These data are used to calculate the percentage of refinery yield of distillate fuel oil that is
greater than 500 ppm sulfur (DSH in MARKAL).
Inputs ELC NG
Contains raw data from the AEO 2010 Table 34: "Refinery Industry Energy Consumption,"
including refinery energy fuel inputs. This information is used to calculate the input
(INP(ENT)p) of electricity and natural gas per unit of refinery output.
RTI Costs
New refinery cost data from the "Task_5_Refmery_investment_costs," prepared by RTI
International.
Conv
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012).
chart used to convert AEO prices to the year 2005.
EmisGREET
Includes refinery related emissions that were extracted from GREET and converted to MARKAL
units.
EmisBC
Includes estimates of the fraction of PM2.5 that can be classified as black carbon. The fractions
can differ by source category,
EmisOC
Includes estimates of the fraction of PM2.5 that can be classified as organic carbon. The
fractions can differ by source category.
93
-------
G: Sector Workbook Description - Biomass
Workbook Name: EPAUS9r_12_BIOMASS_vl.0.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize biomass
resources in the EPAUS9r MARKAL database.
Data Sources
For the feedstock supply curves are derived from the Billion Ton Update (DOE, 201 la). Prices
and quantities for a number of supply steps were derived from this report and data were
downloaded from the website: https://bioenergykdf.net/content/billiontonupdate.
Units
All costs are expressed in millions of 2005 dollars. Biomass feedstock is generally expressed as
million short tons (Mt). For some end-use sectors, such as the electric sector and industrial
sector, the feedstock supplies are converted to PJ based on the energy content of the feedstock.
Workbook Description
The following section gives a description of each of the 32 worksheets in the biomass workbook.
The worksheets are listed in the order they appear, from left to right, in the workbook. The
worksheet names noted with an asterisk contain the data that are automatically uploaded to
ANSWER when importing data from Excel.
Conversions
Contains basic unit conversion data.
Commodities*
The naming for the biomass commodities follows their supply chain from production in the field
to the conversion technologies that convert the biomass into liquid fuels, heat or power. Biomass
feedstocks as produced at the farm, forest, etc. begin with a B. As the biomass moves down the
supply chain emissions are accounted for (noted by EA) and any additional collection costs or
energy use is added (noted by C). Delivered biomass after transport is simply indicated by its
three letter name (STV for corn stover or AGR for agricultural residues), and when they are
converted from Mt to PJ, there is a _PJ appended. Biomass that is delivered to specific sectors is
indicated by a prefix (INDBIO or ELC). Emissions commodities and some fossil energy
commodities are also included. Biomass commodities use the solid synthetics set membership,
with only the initial commodities (e.g., BCRN, BFSR) designated as renewables. These solid
synthetics should not be changed to renewable commodities, as this can lead to errors in the
modeling (i.e., upper bounds from initial supply curves will not necessarily be met).
Technologies*
The naming for the biomass technologies follows the same structure as the commodities:
biomass production, emissions accounting, collection, transportation, and designation to specific
94
-------
sectors. The critical technologies are the "biomass production" technologies, which begin with
RNW. These technologies are essentially the supply curves for each of the biomass categories.
For some biomass, corn and soybean oil, there may simply be one price point and upper bound.
However, for all other cellulosic feedstocks, there are between 4 and 20 incremental price steps
that form the supply curves. These supply curves form the bulk of the biomass workbook, and
are the key component to this sector.
TechData_AgFeedstock*
Includes the agricultural-based biomass feedstocks that have supply curves derived from the
Iowa State University's FAPRI-CARD model (using a modified version of the FAPRI-CARD
baseline, see Elobeid 2011 for background on the model structure) or from USD A Agricultural
Long-term Projections for corn (USDA, 2012. Table 18) and soybean and soybean products
(USDA, 2012. Table 23) These feedstocks include corn grain, soybean oil, and corn stover. The
corn grain and soybean oil include a single price (COST) and quantity (BOUND(UP)Or),
according to the CARD model run that is included or the USDA projection. Corn stover is also
projected in the CARD model, and can be run. However, as a default, these projections are not
used, and the Billion Ton Update supply curves (on another worksheet) are used instead.
Conversion Process for Billion Ton Update Data
A general overview of how the Billion Ton Update data (DOE, 201 la) are utilized in MARKAL
follows.
Although there are a number of different cellulosic feedstocks, the process for converting the raw
data from the Billion Ton Update to the TechData worksheet follow the same basic steps.
1. The quantities reported when the raw data are downloaded from the Bioenergy KDF website
(https://bioenergykdfnet/content/billiontonupdate) are cumulative amount of biomass (tons)
by year (2012, 2015, 2020, 2025 and 2030), price ($10/ton, $20/ton... $200/ton) and location
(county or state).
2. Where multiple biomass feedstocks were aggregated into one category, those feedstocks
quantities were summed for each year, price, and state/county.
3. State level data were then aggregated to the nine-regions.
4. Cumulative amounts were converted into incremental quantities for each time step.
5. Quantities were then converted from tons to million tons (note that these are short tons).
In step 4, for some of the feedstocks based on agricultural lands, higher price steps could result
in a lower supply (or negative price step). This result is somewhat counterintuitive, but is due to
the fact that the Billion Ton Update also represents land use change - shift from one crop to
another crop depending on the price. Therefore, at a higher price step for biomass feedstocks,
there may be land use changing from one feedstock category to another, e.g., conversion of land
from wheat or corn to energy crops. In the cases where this change occurs, the incremental step
was set to zero, resulting in a trivial difference in total feedstock supply, typically on the order of
0.1%-0.3%.
TechData Stover(Bton) *
Stover supply curves reflect the quantity of stover that can be collected profitably, and takes into
consideration factors such as crop yields, tillage, and upper limits on removal due to factors such
95
-------
as erosion. Stover supplies have five price steps from $40-$60 per ton of collected stover and are
available for all regions except Region 2. These prices reflect farm gate prices.
TechData AgRes(Bton)*
Agricultural residues reflect the quantity of "straw and stubble" collected from agricultural lands,
including: wheat straw, barley straw, oats straw and sorghum stubble. Quantities of straw and
stubble for specific crops are small relative to corn stover, and are therefore aggregated to a
single supply curve. Similar to stover, these agricultural residues have five price steps from $40-
$60 per ton of collected residues and are available for all regions except Region 2.
TechData_ForestRes(Bton)*
Forest residues include a variety of forest biomass resources, including residues from logging
and thinning and other removal residues. For these supply curves, all lands, including Federal
lands are included. Removing the Federal lands from the analysis reduces the total potential
forest biomass available by about 1-9 million tons, with the larger differences at the higher price
steps. In the future, we may include differentiated supplies with and without Federal lands.
Note that thinning on federal lands do not currently qualify as renewable feedstocks under the
Renewable Standards Program. However, they are currently aggregated together.
Forest residues have 20 supply steps - from $10 to $200 per ton — substantially more than the
other biomass categories. Prices reflect forest "roadside prices" that a buyer would pay, and
therefore do not include transportation or preprocessing costs. However, our default for the base
EPAUS9r database is to include only up to $70 per ton. We limit the feedstock supplies to this
price step because according to the description in the Billion Ton Update report, estimates for
conventional pulpwood to energy at prices above $80 per ton fall outside the model parameters.
Given the higher level of uncertainty for these higher price steps, we have excluded them from
the TechData sheet. However, the values are available from the BtonU-base worksheet and can
be used in scenario analysis, with the caveat that extrapolating to these higher price steps has a
great deal of uncertainty.
TechData MillResUU(Bton)*
The Billion Ton Update also included Mill Residues, but is careful to distinguish used from
unused mill residues. For the unused mill residues, four price steps range from $10-$40 per ton
and are available for all regions except Region 2. These prices reflect the price at which it is
assumed that they residues can be purchased at the mill and are comparable to the disposal cost.
Although there are four price steps, all of the residues can actually be purchased at $20 or less.
Therefore only the first two price steps are actually used.
TechData MillResUsed(Bton)*
Again, the unused mill residues were the focus of the Billion Ton Update. However, we are also
interested in the approximately 32 million dry tons of used mill residues that are currently used,
mostly for energy. These supply curves are reflected as a single price point but are available
only to the mills themselves in the model. The extrapolations of the used mill residues are also
included on the worksheet BtonU-base.
96
-------
TechData UrbanWood(Bton)*
Urban wood waste (UWW) includes two categories of wood waste: C&D waste and wood from
MSW. UWW has ten price steps, ranging from $10-$ 100 per collected ton, and is available for
all regions. Only the first five steps are used, as all UWW is available at less than $50/ton.
TechData ECG(Bton)*
Three types of perennial grasses (switchgrass, Giant Miscanthus, and sugarcane) are included in
the supply curves for Energy Crops - Grasses (ECG). Perennial grasses have five price steps
from $40-$60 per ton of collected residues and are available for all regions except Region 2. The
prices and quantities are for the "baseline" assumptions from the Billion Ton Update. Note that
there are also high-yield scenarios (for 2%, 3% and 4% yield growth). These additional options
may be incorporated in future workbooks, but are currently not available.
TechData ECW(Bton)*
Three types of woody energy crops (poplar, willow, eucalyptus, southern pines) are included in
the supply curves for Energy Crops - Woody (ECW). Woody crops have five price steps from
$40-$60 per ton of collected residues and are available for all regions except Region 8. The
prices and quantities are for the "baseline" assumptions from the Billion Ton Update. Note that
there are also high-yield scenarios (for 2%, 3% and 4% yield growth). These additional options
may be incorporated in future workbooks, but are currently not available.
TechData ECA(Bton)*
The last of the energy crops is energy sorghum, or a high-yield sorghum, considered an energy
crop in the Billion Ton Update. Energy sorghum is an annual crop but was included in the
Billion Ton Update due to its ability to grow across a wide range of sites. The supply curves are
included as a TechData sheet, but it has not been linked to any potential end uses. However,
energy sorghum is available as a feedstock for consideration under alternative scenarios. The
production potential of energy sorghum is relatively small (on the order of 5-20 Mt per year)
compared to the larger feedstock categories.
TechData ProdCollect*
The technologies in this workbook represent the production and collection of biomass
feedstocks. In general, these are simply "pass through" technologies, but provide a placeholder
if additional analyses regarding changes in production and collection costs and/or energy use are
desired. However, because all of the biomass feedstocks supply curves already take into account
the cost and energy associated with the production and collection of biomass (and therefore costs
are reflective of "farm gate" or "roadside" prices), additional costs and energy use here could be
double counting. The one exception is the inclusion of a factor accounting for some feedstock
degradation and loss for stover and other agricultural residues (with a slightly higher factor for
INP(ENT)p than OUT(ENC)p.
TechData_Transport*
Contains both the transportation of feedstock, as well as sector specific "collector technologies"
that take multiple biomass feedstocks, and send them to individual end use sectors. Feedstock
transportation includes the diesel use for the truck transport of biomass (INP(ENT)p which is PJ
97
-------
of diesel for every Mt or PJ of biomass transported). There is also a VAROM that represents the
non-fuel costs associated with the transport of cellulosic feedstocks in particular.
For electricity production, feedstocks are converted from Mt to PJ based on their energy content.
However, there are two different collectors: biomass to integrated gasification and combined
cycle (IGCC) and combustion. The reason for the separation is that the emissions for the CC>2
and CO2 from biomass feedstock going to combustion is accounted for on the input fuel
(biomass), whereas for IGCC, the emissions are accounted for using a separate emissions
accounting technology (SEELCBIGCC).
TechData_Emissions*
Includes tracking of electric sector biomass-related CC>2 and SC>2 emissions associated with
combustion in the electric sector. There is also a placeholder for looking at CO2 update from
biomass but is not included in the base CC>2 emissions accounting for biomass. Note that the
CO2 emissions are tracked as a separate CO2 biomass source for the electric sector (e.g., CO2BE)
as but are not included in the total CC>2 accounting and are thus assumed to be carbon neutral.
BtonU-base
Contains all of the feedstock supply curves from the Billion Ton Update . As described earlier,
the data from the Billion Ton Update has been aggregated from state-level to the nine-regions
and converted from cumulative to incremental quantities for each price step and from tons to
million tons. The maximum quantity available for all price steps and all regions are summed
after the last row of each set of supply curves. These quantities have been cross-checked against
the tables in the report itself to ensure that all aggregation was correct.
OldStover(CARD+Removal)
Includes the earlier corn stover supply forms used in the 2008 version of the EPAUS9r database.
Because there are some stover related parameters that are included in this sheet, such as the
average losses during feedstock storage and transportation, this sheet has been retained.
Stover(CARD), Corn(CARD) and Soybean(CARD)
These worksheets take the state and regional level data from the CARD_DATA_Sheet and
convert it into the cost and bounds for corn grain, corn stover and soybean oil in the
TechData_AgFeedstock worksheet. Data that are linked directly to the CAKD_DATA_Sheet
are highlighted in yellow, and data that are then linked to the TechData sheet are highlighted in
blue. For the quantities (Mt) available in each EPAUS9r region, we start with total corn
production (million bushels) based on the CARD regions. CARD does not provide total corn
production at the state level, so corn production is disaggregated to states based on CARD
planted acreage, and then re-aggregated from the state to the nine-region level. There are also
three options for how much corn may be utilized for ethanol production. The default is 50% of
total corn production. Other options are to link to the CARD projections for corn used for
ethanol (between 30-38% of total corn production) or allow the model to use as much of the corn
crop as it chooses based on the price (although even 50% must be taken with majors caveats).
Prices are also given by CARD using their regional definitions. Therefore, prices are again
disaggregated from the CARD regions to states, then aggregated up again to a price using a
weighted (by total production in each state) average. The process is similar for soybean oil,
98
-------
although it also takes the additional step at incorporating the share of total soybean production to
crush for oil, and the share of that soybean oil to FAME biodiesel.
Corn(USDA), Soybean(USDA)
To have another set of corn grain and soybean oil prices and quantities to run alternative
scenarios, we have also included agricultural baseline projects from the USDA
(http://www.usda.gov/oce/commodity/ag_baseline.htm). These supply curves are simple prices
and quantities, and again, have defaults of 50%, which can be adjusted by users to account for
the share of crops that can be used toward fuel production. Soybean oil curves are in the process
of being included.
CARD DATA Sheet
This sheet includes the data from the FAPRI-CARD agricultural market model. These are the
raw data that include all agricultural commodity market projections out to 2025. Only a subset
of These data are used and is converted into MARKAL parameters in the earlier CARD
worksheets.
Collect-Trans
Fuel use (diesel) for transportation of biomass feedstocks is calculated using default values from
the GREET model (Argonne, 2007). Fuel use is expressed as PJ for every Mt of biomass
transported, and takes into account one-way and roundtrip distances for transport (miles),
payload (tons/vehicle) by vehicle type (medium or heavy duty truck), and vehicle efficiency
(gal/mi). These assumptions (trip distance) can be modified from these default values. These
are static values and do not take into account vehicle efficiency improvements that would be
reflected in the heavy duty sector of the MARKAL model. Transportation costs are also
modeled as a combination of fixed and variables costs for trucking based on reported data
(Kocoloski et al.)
Feedstock Emissions
Calculates the carbon and sulfur emissions associated with combustion in the electric sector
based on the carbon and sulfur content (by weight, derived from references with ultimate
analysis of the chemical composition of various feedstocks) for each category of biomass
feedstock. Original journal articles referenced are included in the table along with the full
reference listed at the bottom of the worksheet. The carbon content is also used to derive an
optional carbon dioxide uptake factor as well for the growing of the biomass, which can be used
on the TechData Emissions worksheet.
99
-------
H: Sector Workbook Description - Biofuels
Workbook Name: EPAUS9r_12_BIOFUEL_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize biofuel
production in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics for the majority of the biofuel production technologies are based
on techno-economic assessments from the National Renewable Energy Laboratory (NREL).
Data regarding fuel transportation and blending come from a number of sources, including DOE
reports and peer-reviewed journal articles. The renewable fuel volumes from the EPA
Renewable Fuel Standard Program are also included as reference and for potential use as
constraints, as are EIA data from the Annual Energy Review for historic (2005, 2010 volumes).
Additional data sources are specified in the worksheets themselves next to the raw data that were
utilized.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ or
MTons (million short tons) in the case of biomass resource inputs. Most emission factors are in
units of kTonnes per PJ of biofuel that is transported within a region or from one region to
another, with the exception of CC>2 emissions, which are in units of MTonnes per PJ.
Workbook Description
The following section gives a description of each of the 30 worksheets in the biofuels workbook.
The worksheets are listed in the order they appear, from left to right, in the workbook. The
worksheet names noted with an asterisk contain the data that are automatically uploaded to
ANSWER when importing data from Excel.
Conversions
Provides basic unit and price conversions.
Commodities*
Declares the energy carriers that are included in the biofuels workbook, including all biofuel
energy carriers and associated fossil fuel inputs and biomass inputs.
Technologies*
Declares the technologies for biofuel production processes, application of subsidies and credits,
inter-regional trading and transportation of biofuels, blending, application of credits for non-
energy co-products, and international imports/exports of ethanol and biodiesel.
100
-------
Constraints*
Declares the constraints that can be applied to model upper, lower or fixed bounds for specific
biofuel categories, such as corn-based ethanol or total advanced biofuels. There are global,
cross-region constraints that apply upper or lower limits to the aggregated national volumes of
bio-based fuels that are produced or consumed.
TechData_BiochemEth*
Ethanol produced via biochemical processes (as opposed to thermochemical) is included in this
sheet. This includes four variations of corn-based ethanol production: existing wet mills,
existing dry mills, new dry mills, and new dry mills with combined heat and power. The first
two represent the existing ethanol production capacity as of 2005 via the RESID. All of the
corn-based ethanol production technologies utilize a number of fossil inputs (including
electricity, natural gas, gasoline, and coal for the wet mills), and output corn-based ethanol.
These technologies also produce a number of co-products ranging from high-fructose corn syrup
to dried distiller's grain. Those co-products were tracked separated as material flows in earlier
versions of the model. However, for simplicity tracking the energy carrier flows, these non-
energy co-products have been combined and included via a discount (DELIV(ENT)) on the corn-
ethanol energy carriers (the raw data links back to the "co-products" worksheet). There is also a
subsidy for corn-based ethanol applied for the first two time periods (2005, 2010), and the
subsidy is zero for the rest of the time horizon (to 2055), with the assumption that this subsidy
will not be reintroduced.
This sheet also includes cellulosic-based ethanol production via the biochemical platform. This
technology is set up to use up to five different feedstocks, although for the purposes of the base
model run, only stover and other agricultural residues are used. Additional feedstocks for
biochemical ethanol can be added for scenario runs. Biochemical cellulosic ethanol
(CELETHB) then has the subsidy applied (with the parameter DELIV(ENT)) and the subsidy is
assumed to extend throughout the modeling horizon.
Note that some of the technologies have improvements in product yield (declining INP(ENT)
over time) in terms of Mt biomass required for per 1 PJ ethanol produced. Some technologies
also have energy efficiency improvements. For example, the energy efficiency of existing corn-
based ethanol is assumed to improve to begin to approximate the efficiency of new corn-based
dry mills without CHP. The model can also chose to apply CHP. However, the share of CHP is
constrained to less than 30% of the total corn ethanol production (see ConstrData). The share
of CHP can be changed for scenarios analyzing high or lower potential for adoption of CHP.
CornEthProdDry
The majority of current U.S. ethanol production is dry mill corn-grain ethanol. Includes the raw
data from a number of peer-reviewed publications and government reports that provide
information regarding efficiency and costs that use process engineering economic studies, survey
based approaches, or spreadsheet models. The primary source for much of the data are a survey
article (Perrin, 2009). Data on CHP come from the CornEthProdWet worksheet, discussed
below.
101
-------
CornEthProdWet
Contains data related to ethanol production from wet mills. Much of the data are relevant only to
existing wet mill capacity, given that new ethanol capacity in the U.S. has been new dry mills.
Therefore, the residual capacity includes a number of wet mill facilities. Includes data from the
USDA Ethanol Cost-of-Production Survey as referenced in the worksheet. As noted above,
information regarding the CHP option for dry mills also links back to Worksheet.
CellEthProdBio
Includes the detailed raw data that were used for characterizing the biochemical platform for
cellulosic ethanol production. Three related reports (McAloon, 2000; Aden, 2002; Humbird,
2011) were examined and relevant numbers are included for comparison. The Humbird 2011
report provided the final numbers for the technology included in the sheet
TechData_BiochemEth
EthRESID
The state-level existing dry and wet mill ethanol capacity for 2005 was derived from several
tables in the Renewable Fuel Standard Program Regulatory Impact Analysis (EPA, 201 Ob) and
aggregated up to the nine-regions.
EthRESID-old
Contains data used in earlier versions of the EPAUS9r database. The worksheet is retained
because of the state-level capacity data for 2005 by company, city and state that were derived
from publicly available data maintained by the Renewable Fuel Association. These production
data were also used to support GIS analysis that went into the distances by mode for ethanol
transportation from the ethanol plants to blending/end users, for the calculation of ethanol
transportation costs. These transportation routes/distances were derived using the TRAGIS
model (ORNL, 2004b), which was developed and maintained by Oak Ridge National
Laboratory.
TechData_Biodiesel*
Biodiesel within worksheet is defined as FAME derived from soybean and waste oil. Includes
the oil pre-processing steps (soybean crushing), the actual biodiesel production process, and
credits for the co-products (glycerin), although that price credit drops substantially due to the
excess of glycerin on the market and assumes no upgrading to refined glycerin.
Biodiesel
The raw data for the characterization for the biodiesel production process come from two articles
(Zhang, et al 2003 a, 2003b). This characterization of the biodiesel production process also
includes an assessment of waste oil based on population and a related urban waste resources
assessment cited in the worksheet.
TechData_Thermochem*
This sheet contains the data for the thermochemical production of ethanol and a gasoline-
blendstock from several different biomass feedstock sources, primarily, woody biomass. For
each of these technologies, there is an investment cost, fixed and variable O&M, two inputs (a
biomass feedstock and the denaturant), and two outputs (ethanol and mixed higher alcohols that
102
-------
are blended into gasoline). There is also a hurdle rate to reflect the high level of investment
uncertainty regarding the technology itself, but also the uncertainty regarding the availability and
price of the biomass feedstock (note that there is no hurdle rate associated with the biomass
feedstocks production and logistics). This sheet also includes a number of collectors (including
XCELGSL and XCELDSL) that apply a subsidy for cellulosic biofuels, before those biofuels are
sent as a "drop-in" fuel to the gasoline and diesel supply. The raw data for this sheet come from
CellETHTC
Thermochem(TBD)
This sheet is a TechData sheet intended for use in scenarios and is not included in the baseline.
This sheet provides a number of additional thermochemical pathways for the production of
renewable gasoline and diesel and additional co-products such as LPG. These pathways are
considered more uncertain technologies, but are of potential interest and use in exploring
additional scenarios regarding biofuel production pathways. If the user decides to do model runs
with these technologies as an option, the sheet name would need to be changed to
TechData_Thermochem(2), and all technologies and cells above the parameter would need to
have the * removed. The raw data for this sheet comes from CellGSLTC and CellGSLPYR.
CellEthTC
The primary thermochemical biofuel technology was modeled as woody biomass to ethanol
(Dutta et al., 2011). Provides details regarding the raw data and detailed information regarding
the fuel and product yields, capacity, capital and operating costs. The worksheet also includes
references and comparisons to earlier reports with similar techno-economics analysis (Phillips,
2007) and also shows the unit conversions to MARKAL units and conversion to 2005 dollars.
CellGSLTC
Worksheet and the following worksheet highlight two additional thermochemical biofuel options
that can be included in scenario analysis. This sheet summarizes a wood to gasoline via the
Methanol to Gasoline (MTG) intermediate process (Phillips et al., 2011). The Provides details
regarding the raw data and detailed information regarding the fuel and product yields, capacity,
capital and operating costs and also shows the unit conversions to MARKAL units and
conversion to 2005 dollars.
CellGSLPYR
Includes the characterization of a fast pyrolysis to gasoline pathway (Wright et al., 2010).
Again, pyrolysis is an additional thermochemical biofuel option that can be included in scenario
analyses. This technology is translated into two potential options for linking to the TechData
sheet, on-site hydrogen production and hydrogen purchase (included as a cost, not an actual H2
input).
TechData_FuelTransport*
This sheet establishes the structure for the inter-regional trading, transport and blending of fuels.
The underlying structure is such that there is a full origin-destination matrix that connects all
regions to all regions for the transportation of ethanol. However, when the ethanol is moved by
individual transport modes (truck, barge, rail) there are limitations on the capacity to move
between regions. In the basic set up, each region is able to export to the other regions via
103
-------
EXPETH# for the exporting regions IMPETHR## for the importing regions, where the #
indicates the regions. Within the importing regions, there are additional technologies that
designate whether that ethanol was imported via barge, truck or rail. For each combination of
mode, there are also energy inputs, variable O&M costs, and upper bounds on the capacity for
transport. Currently the transportation bounds are set to an upper limit for rail and barge, with
much greater flexibility to expand transportation by trucking. However, with corn-based and
corn-stover based ethanol being the key feedstocks for much of the ethanol production, trucking
is limited to go from Regions 3 and 4 to the other Regions. This can be modified for scenario
analyses.
Worksheet also captures the blending of denatured ethanol and gasoline to an E85 blend,
including the costs associated with conversion of dispensers and storage tanks to handle E85
blends, as well as a blending of biodiesel into a B20 blend. Technologies for the international
exports and imports of both ethanol and biodiesel are also captured here, and are derived from
the FAPRI-CARD agricultural model. Because there are no supply curves (only a single price
and quantity point are available), these volumes are fixed in the ConstrData sheet. These prices
are the cost to import (with fixed amounts of imports), as well as the credit to export (again, with
fixed amounts of exports).
TechData_Emissions*
This TechData sheet includes emissions of CO2, NOX and SO2 associated with the transportation
of biofuels, specifically ethanol.
ConstrData*
The majority of the constraints are for specifying lower, upper or fixed bounds on the production
of particular fuels, such as corn-based ethanol or biodiesel, or categories of fuels, that either go
through a collector (such as CELETH) or are specified by a technology filter in the database,
such as advanced biofuels (BIOF_ADV2) or all exported biodiesel (BDL_IMP). These
constraints are global, cross-region constraints, meaning the sum of the activity in all 9 regions
has to meet the "GLOBAL" constraint which is specified as the cross-regional right hand side of
the equation. Some constraints are included as "NON" or non-binding constraints, but can be
used in scenario analyses. There are two share constraints: the share of new corn-based ethanol
production that can come from CFtP facilities, and the share of E85 in the total gasoline/ElO
pool. E85 share constraint is applied to the blending technology (XBLNDE85) that represents
gasoline station retrofits to dispense E85, and restricts the total dispensing of E85 to 1/3 of the
total gasoline/ElO pool. Both constrains can be changed.
RFS2 Volumes
Includes the original volume standards for the EPA's Renewable Fuel Standard program (RFS2),
as well as historical renewable fuel production data from the EIA's Annual Energy Reviews that
are used to set the 2005 RESID and 2010 production volumes.
104
-------
Co-Products
Includes non-energy co-products from both wet and dry mill production and biodiesel production
to capture the economic value of producing these products. These non-energy co-products are
calculated based on data from the FAPRI-CARD agricultural market model, and translate the
price per unit of co-product ($/ton or $/lb) and co-product yield (Mt co-product per PJ fuel
produced) into a cost credit (M$/PJ fuel produced). For each fuel product, dry mill ethanol, wet
mill ethanol or biodiesel, the multiple feedstock credits are then summed and added as a negative
delivery cost (DELIV(ENT)) in the corresponding TechData sheet.
Int'l Exp+Imp
Summarizes the raw data from the CARD model for the imports and exports of ethanol and
biodiesel. Because these are point estimates for prices and quantities, and not supply curves,
these are included as fixed constraints on the ConstrData worksheet. Their inclusion is to
account for potential imports and exports of biofuels, and is primarily intended for use when run
iteratively with the CARD model.
EthTrans2
Contains data from a number of different sources, including an in-house geospatial analysis. At
the core are the origin-destination matrices that provide the travel distances by mode (rail, barge,
truck) from each region to each region. These distances were obtained using the TRAGIS model
(ORNL, 2004b), which provided the approximate routing distance by rail, barge and truck for
each region to region combination. Barge and rail capacity was considered to be constrained to
existing limits, using the capacity limits from the PMM documentation. However, capacity is
allowed to grow for all trucking transport from Regions 3 and 4 to the other Regions, given that
with corn based ethanol and biofuels from other Midwest feedstock, such as corn stover, the
majority of interregional transport will come from these regions. Other regions can produce and
use biofuels but there are more constraints on their ability to export to other regions. Those
assumptions can be changed for scenarios analysis by changing the upper bound for specific
region-region pairs. Transportation costs were updated from: Kocoloski et al., (2010). Based on
the transportation distances for truck from Kockoloski et al, a $/ton-mi is then translated into a
VAROM for transportation. These distances and vehicles efficiencies and capacities are also
used to calculate the fuel use (diesel) for the transportation of ethanol - interregionally and
locally.
EthTrans
Contains data from the EIA. Specifically, from the PMM Documentation: Table 14. 2012 New
Ethanol Shipments and Freight Costs by Census Divisions. That this is an older version, and is
primarily for reference. All of the cost numbers have been updated in the EthTrans2 worksheet,
but some of the capacities - for rail and barge, specifically - still link back to these data.
Emissions-Trans
Includes emissions factors for the transportation sector for calculation of emissions associated
with the transporting ethanol.
105
-------
Blend
Provides the data for different ethanol blend levels, including denatured ethanol, E10, and E85.
The infrastructure costs for upgrading retail stations to provide E85 in dispensers are also
included.
Credits
Federal credits for ethanol blending include the Volumetric Ethanol Excise Tax Credit (VEETC),
also known as the 'blenders credit' and cellulosic ethanol income tax credit for cellulosic alcohol
and other cellulosic biofuels as specified in the Food, Conservation and Energy Act (FCEA) of
2008 (H.R. 2419). The corn ethanol and biodiesel credits expired December 31, 2011. The
expiration of credits is reflected starting in model year 2015. Cellulosic ethanol credits are
assumed to remain in place to 2055.
CARD CellETH Yields
Includes projections of changes in yields (gallons of ethanol per ton of biomass) for cellulosic
ethanol production from the FAPRI-CARD model.
SoybeanOO
Provides basic state level data on soybean production and was used to derive 2005 RESID
biodiesel capacity (linked to sheet Biodiesel) for each of the regions.
106
-------
I: Sector Workbook Description - Municipal Solid Waste
Workbook Name: EPAUS9r_12_MSW_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
conversion of MSW to electricity in the EPAUS9r MARKAL database.
Data Sources
The MSW resource supply curve data were taken from The State of Garbage (Simmons et al.).
The emissions data are derived from the EPA's MSW-DST (Kaplan et al.).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Workbook Description
The following section gives a description of each of the six worksheets in the MSW workbook.
The worksheets are listed in the order they appear, from left to right, in the workbook. The
worksheet names noted with an asterisk contain the data that are automatically uploaded to
ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers) associated
with MSW, including supply, transportation, and emissions.
Technologies*
Lists the names, units, and set memberships for the resource supply curves, collection
technologies, transportation to end-use sector technologies, and emissions tracking technologies
for MSW.
TechData MSW*
Contains the supply curve costs and bounds for MSW by region. In addition, the worksheet has
emissions data for emission tracking technologies. The methodology to calculate the emissions
is presented in the paper Is It Better To Burn or Bury Waste for Clean Electricity Generation?
(Kaplan et al., 2009)
MSW
Contains the raw data and calculation for the cost and supply bounds for the MSW supply
curves.
Pop
Regional population data were taken from the U.S. Census Bureau state population projections
released in 2005 and based on the Census 2000. Data can be found at
www.census.gov/population/www/projections/projectionsagesex.htmal. Table 6: Total
107
-------
population for regions, divisions, and states: 2000 to 2030. Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are also taken
from EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (IPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. The data have been adjusted to
include Alaska and Hawaii.
Conversion Factors
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) chart
used to convert AEO prices to the year 2005 and biomass energy conversions.
108
-------
J: Sector Workbook Description - Electric Sector
Workbook Name: EPAUS9r_12_ELC_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the electric
sector in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics for electric power production technologies come from a number
of different sources, with the primary one being the AEO reference case (EIA, 2012c).
Additional sources are specified in the worksheets themselves next to the raw data that were
utilized.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ or
GW. Most emission factors are in units of kTonnes per PJ of fuel input with the exception of
CC>2, which is expressed in MTonnes per PJ.
Workbook Description
The following section gives a description of each of the 47 worksheets in the electric sector
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers) and
emissions in the electric sector.
CommData*
Contains the reserve margin (15%) and the transmission efficiency (93.5%) for electricity.
CommData-CCS*
Contains the upper bound for sequestered CO2 from CCS technologies. Values are based on
electric sector expert judgement.
CCS Capacity Raw Data
Contains CCS storage capacity potential raw data for use in determining the upper bounds for
CO2 sequestration.
Technologies*
Lists the technology names, units, and set memberships for all technologies in the electric sector.
109
-------
TechData_DMY*
Contains a dummy electric supply technology with a high investment and variable O&M cost.
This electricity supply will be used only by the model when there are no other sources for
electricity, allowing the modeler to complete an otherwise infeasible model run and then go back
and look for the cause of the use of a dummy fuel. Also includes a number of dummy
technologies for use with the retirement constraints for residual coal steam technologies.
TechData NewCap*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for new electric production technologies (not including CCS, wind, solar, and landfill
gas).
New Cap Raw Data
Contains raw data for the costs and efficiencies of new electric generation technologies. The
data comes from AEO Table: "Cost and Performance Characteristics of New Central Station
Electricity Generating Technologies."
Geothermal Raw Data
Contains raw data for the costs and efficiencies of new geothermal technologies.
TechData CCS*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for carbon capture and sequestration electric production technologies.
CCS Calcs
Contains the raw data and calculation for new CCS electric production technologies. The data
are drawn from a number of sources including AEO and the MIT Future of Coal (MIT, 2007).
TechData_Uranium*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for mined uranium conversion and nuclear power plant technologies.
Nuclear Flowcharts
Describes the flow of nuclear power production from mined uranium through conversion of the
uranium and the associated waste products. Parameter values are placed in each of the
corresponding boxes in the flowchart.
Nuclear Raw Data
Contains raw data for nuclear power plant production. Data comes from a number of sources
referenced in the worksheet.
TechData Solar&Wind*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for wind and solar energy technologies.
110
-------
Solar Raw Data
Contains raw data for wind power technologies taken from the AEO 2006 solar input files.
Wind Raw Data
Contains raw data for wind power technologies taken from the AEO 2006 wind input files.
TechData_LFG&MSW*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for landfill gas to energy technologies and municipal solid waste to energy
technologies.
LFG Raw Data
Contains raw data for costs, efficiencies, and residual capacities for existing and new LFG
technologies.
MSW Raw Data
Contains raw data for costs, efficiencies, and residual capacities for existing and new MSW
technologies.
TechData_CHP*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for CHP technologies.
CHP raw data
Contains the raw data for combined heat and power technologies.
TechData NonCoalResid*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for existing non-coal electric generation technologies (not including wind, solar,
nuclear, CHP, and municipal solid waste).
Resid Raw Data
Contains old residual capacity mined directly from NEMS.
RESID Calcs
Contains the residual capacity calculations by region and by fuel type. Data for the calculations
come from the ExistingUnits2007 worksheet.
ExistingUnits2007
Contains the raw data, including region, fuel type and plant capacity, for all existing generating
units in the United States in 2007. Data comes from Form EIA-860 data (Energy Information
Administration's "Annual Electric Generator Report").
ExistingUnits2007 NG STM and CC
Contains the raw data from EIA-860 for natural gas steam and combined cycle plants only.
Ill
-------
ExistingUnits2007 Nuclear
Contains the raw data from EIA-860 for nuclear plants only.
File Description
Describes the column headings and various codes contained in the ExistingUnits2007
worksheet.
TechData CoalResid*
Contains the costs, efficiencies, residual capacities, availability factors, and peak demand
parameters for existing coal electric generation and selected air pollution control technologies
(S02, NOX, PM).
Resid Coal Boilers
Contains the raw data used to populate information in the TechData_CoalResid worksheet.
Data in worksheet originate from the National Electric Energy Data System (NEEDS) v.4.10
database, and were expanded/refined by E.H. Pechan & Associates under EPA Contract No. EP-
D-07-097.
Bin Table - Resid Coal
Contains calculations used to populate residual capacities and input requirements of the existing
coal generation technologies in the TechData_CoalResid worksheet. Development of the
regional qualifiers, required in the Technologies worksheet, is also performed in worksheet.
Bin Table - SOx
Contains calculations used to populate residual capacities of the 862 control technologies in the
TechData_CoalResid worksheet. Data for the calculations comes from the Resid Coal Boilers
worksheet.
Bin Table - NOX
Contains calculations used to populate residual capacities of the NOX control technologies in the
TechData_CoalResid worksheet. Data for the calculations comes from the Resid Coal Boilers
worksheet.
Bin Table - PM
Contains calculations used to populate residual capacities of the PM control technologies in the
TechData_CoalResid worksheet. Data for the calculations comes from the Resid Coal Boilers
worksheet.
PM Emissions
Contains the calculations used to populate environmental accounting parameter for PMio
emissions in the TechData_CoalResid worksheet. Data for the calculations comes from the
Resid Coal Boilers worksheet.
112
-------
PM Control Data
Contains the calculations used to derive investment and O&M costs for the PM control
technologies in the TechData_CoalResid worksheet. Data for the calculations comes from the
Integrated Environmental Control Model (IECM) Version 6.2.4, E.H. Pechan & Associates, and
the Resid Coal Boilers worksheet.
Retrofit Resid Raw Data
Contains the calculations used to derive investment and O&M costs, as well as environmental
accounting parameter values for the SO2 and NOX control technologies in the
TechData_CoalResid worksheet. Data in Worksheet were generated with the EPA CUECost
spreadsheet model.
Constraints*
Lists the share constraint names for all user defined constraints.
ConstrData*
Contains the constraint data for regional fuel shares, technology shares, and RPS standards.
Fuel share constraints are taken from the Raw Regional Projected Shares worksheet. Lower
values for wind technology in each region and upper bounds on each class of wind technology
are taken from the Wind Raw Data worksheet. Nuclear, solar, MSW, and gas turbine
constraints are based on electric sector expert judgement.
ConstrData-CoalRetire*
Contains the constraint data used to keep retired coal capacity from coming back on-line. This
keeps coal plants in the model from starting and stopping throughout the time horizon.
ConstrData-RPS*
Contains the constraint data for regional tiered RPS constraints. The data is pulled from the
Regional RPS Calcs worksheet.
Regional RPS Calcs
Aggregates the data from the State RPS Data worksheet into regions and calculates a regional
RPS for each tier.
State RPS Data
Contains state RPS raw data from the DSIRE website (NCSU 2010). The data includes accepted
technologies and RPS goals.
Raw Regional Projected Shares
Contains the fuel use percentages by region used on the Constraints worksheet. Percentages are
calculated from the data on the AEO Regional Fuel worksheet.
AEO Regional Fuel
Contains raw data from the AEO Supplemental tables for Regional Data, Tables 1-8, "Energy
Consumption by Sector by Source". The data for the fuel use in the electric sector is pulled from
113
-------
these tables. These data are used to determine fuel splits for the first three time periods after
2005 in the model.
TechData EmisCoal*
Contains pre-control emission factors for existing and new pulverized coal-fired electric
generation units (EGUs), as well as factors for IGCC units. All factors are in kTonnes per PJ of
coal input, with the exception of CO2, which is in MTonnes per PJ of coal input. PM2.5
emissions are assumed to be 85% of PM10 emissions. All other factors on this workbook
originate from the GREET model (Argonne, 2007)) or were derived using PM2.5 speciation
estimates from the RTI-BC and RTI-OC worksheets.
TechData_EmisOther*
Includes emission factors for non-coal sources within the electric sector. Emission factors are
derived from the GREET model or were calculated using PM2.5 speciation estimates from the
RTI-BC and RTI-OC worksheets. Fuel oil and diesel controls on NOX and PM are assumed to
reduce 75% of uncontrolled emissions. The fractions of PMio that are PM2.5, black carbon (BC)
and organic carbon (OC) are assumed to be equivalent to those for natural gas turbines.
GREET2012-UtilityFactors
Summarizes available emission factors for utilites.
NewEmissionFactors
Displays the new MARKAL emission factors as original data (value, unit, and data source) for
each pollutant-technology combination..
GREET-EF TS
This table was extracted from the GREET model, version l.S.c.O. Cells were added to convert
factors from grams/MMBT to kTonnes/PJ.
RTI-BC
Contains factors that represent the fraction of PIVb.s that is BC for various fuels and source
categories. The factors were derived from the literature. More detailed information about the
source of these factors is available upon request.
RTI-OC
Contains factors that represent the fraction of PM2.5 that is OC for various fuels and source
categories. The factors were derived from the literature. More detailed information about the
source of these factors is available upon request.
GREET_Summary
This table was extracted from the GREET model, version l.S.c.O. The data shown here are
equivalent to those in the GREET-EF_TS tab, but have been reformatted to facilitate use in other
tabs. Also, SO2 factors have been added.
114
-------
GREET FuelSpecs
Includes data from the GREET model, version l.S.c.O. These values are used to calculate CC>2
and 862 emission factors for various fuels.
115
-------
LFG_MSW-Emis
Includes emission factors for waste-to-energy and landfill gas combustion.
Water Data
Contains the water consumption and water withdrawal factors for electric generation
technologies.
Conversions
Contains common conversion factors including the Implicit Price Deflator (DOC, 2010) chart
used to convert AEO prices to the year 2005 and conversion technology heating values.
116
-------
K: Sector Workbook Description - Residential Sector
Workbook Name: EPAUS9r_12_RES_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
residential sector in the EPAUS9r MARKAL database.
Data Sources
The technology and end-use demand characteristics were taken from the AEO reference case.
Additional data used to calculate demands were taken from RECS (EIA, 2009c). Lighting data
were taken from a report provided to the EIA from Navigant Consulting (EIA, 2007). Regional
HDD and CDD values were taken from NOAA's National Climate Data Center (NCDC)
Historical Climatological Series - HCS 5.1. Regional population data were taken from the U.S.
Census Bureau state population projections released in 2005 and based on the 2000 Census.
Emission factors were developed from several sources, including the Climate Registry (TCR),
EPA's WebFIRE emission factor database, and the Easter Regional Technical Advisory
Committee (ERTAC).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of PJ with the exceptions of refrigeration and freezing, which
are given in terms of million units, and lighting, which is given in terms of billion lumens per
year. Emission factors are represented in units of ktonnes per PJ of fuel or energy input.
Workbook Description
The following section gives a description of each of the 27 worksheets in the residential
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers) and
demands in the commercial sector. Fuels coming into the residential sector have passed through
a collector technology that tracks the emissions from the fuels. The "EA" used in the fuel name
stands for "Emissions Accounting." Fuels used include electricity (RESELC), diesel
(RESDSLEA), kerosene (RESKEREA), LGP (RESLPGEA), natural gas (RESNGAEA),
biomass for wood heating (RESBIOEA), and solar (SOL).
For the naming convention for the demand technologies, the first letter in the technology name
reflects the technology category. In this case 'C' is used for 'Commercial'. The remaining
letters represent the type of end-use demand. The demands in the database are:
117
-------
RSC = Space Cooling equipment
RSH = Space Heating equipment
RWH = Water Heating equipment
RLT = Lighting equipment
RRF = Refrigeration equipment
RFZ = Freezing equipment
ROE = Miscellaneous Electric equipment
ROG = Miscellaneous Natural Gas equipment
Constraints*
Lists the constraint names for all user constraints in the residential sector.
Technologies*
Lists the technology names, units, and set memberships for all end-use and collector
technologies in the commercial sector. The naming convention starts with the end-use demand,
followed by the fuel type, the technology type, the efficiency level, and the model year. For
example, "RSHEHPV110" refers to the 2010 model year of the lowest efficiency electric heat
pump for residential space heating. The efficiency level goes from VI for base model efficiency
to V4 or V5 for the highest efficiency.
ConstrData*
Contains the data for fuel shares and technology shares by end-use demands. Values for 2010
were calculated in the RECS RESID worksheet for all end-uses except lighting. Electricity and
natural gas fuel shares are relaxed 3% per time period out to 2055. Diesel, LPG, and kerosene
fuel shares are relaxed over 5% per time period out to 2055 (for a total of 50%). Lighting
technology shares were calculated in the RLT Splits worksheet. Incandescent lighting shares
are reduced 50% by 2020 in keeping with the Energy Independence and Security Act (EISA) of
2007. Compact fluorescent and LED lighting technologies increase in shares by a minimum of
20% by 2055. Purchased electric and geothermal heat pumps provide both heating and cooling
for residential homes, therefore a constraint is used to force any purchased technologies to be
used for both end-uses.
CommData_Demand*
Contains the demand values by end-use and by region. The data are taken from the
RegionalDmds worksheet which is explained below. The fraction of capacity entering
electricity peak equations and the fractional demand by season and time of day (FR(Z)(Y)) are
based on expert judgement.
TechData*
Contains the parameter values for all end-use technologies (with the exception of solar
photovoltaics). The data for the technology investment cost, efficiency, and capacity factor are
drawn from the AEO technology data aggregated in RTEKTY Conv. The data for the residual
capacity are calculated in the RESID CALC worksheet. The values for the implied discount
rate (or hurdle rate) for each technology are determined based on financial and non-financial
factors that affect the choice of various residential technologies. The discount rates range from
0.18 to 0.45 with the lower rates being attached to standard technologies and the higher rates
118
-------
being attached to higher efficiency technologies, new technologies such as LED lighting and
technologies that use diesel fuel.
TechData_Emis*
Contains technology definitions for the collectors that are used to assign emissions to various
fuels used within the residential sector. The emission factors are fuel specific and do not
currently change as a function of time. Several species include an "R" added to the end, which is
used in sectoral emissions accounting. The emissions factors that populate Worksheet are
obtained from the following worksheets: EIACO2Coef, EmisBC, EmisOC, and EmisRes.
TechData SESC*
Contains the parameter values for a collector technology carrying electricity to residential
technologies.
TechData _ZZ*
Contains the parameter values for dummy technologies for each of the end-use demands. These
technologies have a high variable O&M cost and will be used in a model run only when the
model cannot meet its demand with available technologies. The use of any of these technologies
indicates an infeasibility in the model.
Sol_PV*
Contains the parameters for residential solar photovoltaics. The investment costs, fixed O&M
costs, and seasonally adjusted technology availability factors are taken from the electric sector
workbook and are based on values from the AEO2006 Solar Input data. Fixed bounds are
implemented forcing the use of solar photovoltaics based on the AEO projections. Nationally
projected use, found in the AEO12 Renew worksheet, is evenly distributed across the nine-
regions.
AEO12 Renew
Contains raw data from the AEO main reference case table: "Renewable Energy Consumption
by Sector and Source."
RESID CALC
Contains calculations for the regional residual capacity (RESID) of existing technologies in
2005. RESID is calculated by multiplying the market share (taken from RECS RESID) for each
end-use technology times the total demand for that end-use and dividing by the capacity factor
(CF) for that technology.
RECS RESID
Regional market shares by technology are calculated from the EIA 2005 RECS detailed tables
for space heating, air conditioning, and water heating. Fuel and aggregated technology shares
are also calculated for use as model constraints.
RLT Splits
Data from the Navigant Consulting (EIA, 2007) are used to determine the 2005 market shares for
lighting technologies.
119
-------
RTEKTY Conv
Contains an aggregated listing of the end-use technologies from the AEO RTEKTY file and
from Res Lighting with their data characterizations. The technologies are given MARKAL-
specific names and descriptions using standard MARKAL naming conventions and the
technology efficiency and capital cost are converted to MARKAL units. Technology capacity
factors are determined from expert elicitation.
Res Lighting
This residential lighting raw data come from an EIA report prepared by Navigant Consulting
(EIA, 2007).
RTEKTY12 agg
Worksheet takes the technology data from the RTEKTY12 worksheet for Region 2. Decisions
are then made about which technologies to represent in MARKAL. Where certain technologies
do not exist in Region 2, Region 5 data are used.
RTEKTY12
Contains raw data from the AEO Residential Technology Equipment Type Description File
(RTEKTY). Each record gives specifications for a specific model year of a specific end-use
technology.
Regional Dmds
Demand values calculated in ResDemand are re-organized into regions.
ResDemand
End-use demands are calculated for each region by taking the national demands calculated in
USDemand by regional data calculated in ResCalc as follows:
Regional Cooling Demand = cooling coefficient * square footage of AC space * CDD
Regional Heating Demand = heating coefficient * square footage of heated space * HDD
Regional Refrigeration Demand = number of households * refrigerators per household
Regional Freezer Demand = number of households * freezers per household
Regional Lighting Demand = national lighting demand * regional percent of households
Regional Misc Electric Demand = national misc electric * regional percent of households
Regional Misc NG Demand = national misc NG * regional percent of households
Regional Water Heating Demand = national water heating * regional percent of households
ResCalc
Contains a number of preliminary calculations used to determine the end-use demands found in
workbooks USDemand and ResDemand.
The grey and green shaded areas hold the data and calculations for regional household heating
degree day (HDD) and cooling degree day (CDD) coefficients. Regional HDD and CDD values
for the 10 year average from 1998-2008 were take from the NCDC Historical Climatological
Series - HCS 5.1. Years recorded and used for our calculations end in June 2009. These data
can be found at www.ncdc.noaa.gov/oa/documentlibrary/hcs/hcs.htmltf51updates. National
120
-------
CDD and HDD adjustment factors were calculated by time period by doing a weighted average
based on regional population data. These adjustment factors, along with the national demands,
average household square footage, and CDDs and HDDs were used to calculate the national
housing cooling and heating coefficients by time period.
The blue shaded area in the workbook holds the calculations for the percentage of households
with cooling technologies, refrigerators, and freezers. Air conditioning usage was taken from
DOE's 2005 Residential Energy Consumption Survey (RECS). Regional refrigerator and
freezer distributions were calculated based on data from the 2005 Annual Energy Review.
USDemand
National energy demands for space cooling, space heating, and water heating (given in PJ) are
calculated using the data for energy consumption by end-use demand and fuel and the average
stock equipment efficiency from the AEO reference case. National energy demands for
refrigerators and freezers (given in million units) are calculated from the AEO equipment stock
data found in the table: "Residential Sector Equipment Stock and Efficiency." National energy
demands for miscellaneous electric and natural gas demands are taken from the AEO delivered
energy data found in the table: "Residential Sector Key Indicators and Consumption." National
lighting demands (given in billion lumens per year) are calculated by multiplying the lighting
energy use data found in the AEO by a billion lumens per PJ conversion factor, and then
applying that conversion factor by regional square footage per household.
AEOHW
Water heater data from AEO are used to calculate a factor for the average number of PJ of
demand met by a water heater unit. This information is used to convert AEO costs per water
heater unit into cost per PJ of demand met on the RTEKTY Conv worksheet.
Pop
Regional population data were taken from the U.S. Census Bureau state population projections
released in 2005 and based on the 2000 Census. Data can be found at
www.census, gov/population/www/proj ections/proj ectionsagesex.html. Table 6: "Total
population for regions, divisions, and states: 2000 to 2030." Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are also taken
from EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (IPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. These data have been adjusted to
include Alaska and Hawaii.
Conversion Factors
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) chart
used to convert AEO prices to the year 2005.
EmisCO2Coef
Contains CO2 coefficients for various fuels, as reported by the DOE (DOE. 2008).
121
-------
EmisBC
Includes estimates of the fraction of PM2.5 that can be classified as BC. The fractions, which can
differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and Bond
et al., 2007.
EmisOC
Includes estimates of the fraction of PM2.5 that can be classified as OC. The fractions, which
can differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and
Bond et al., 2007.
EmisRes
Includes commercial emission factors. These factors were derived from a number of sources,
including the Climate Registry and the EPA WebFIRE emission factor database. More
information about the derivation of the emission factors is available upon request.
122
-------
L: Sector Workbook Description - Commercial Sector
Workbook Name: EPAUS9r_12_COM_vl.0.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
commercial sector in the EPAUS9r MARKAL database.
Data Sources
The technology and end-use demand characteristics were taken from the AEO reference case.
Additional data used to calculate demands were taken from the CBECS database (EIA, 2007a).
Regional HDD and CDD values were taken from the NCDC Historical Climatological Series -
HCS 5.1. Regional population data were taken from the U.S. Census Bureau state population
projections released in 2005 and based on the 2000 Census. Emission factors were developed
from several sources, including the Climate Registry (TCR), the EPA WebFIRE emission factor
database, and the Easter Regional Technical Advisory Committee (ERTAC).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of PJ with the exceptions of ventilation, given in terms of
thousand cubic feet per minute, and lighting, given in terms of billion lumens per year. Emission
factors are represented in units of ktonnes per PJ of fuel or energy input.
Workbook Description
The following section gives a description of each of the 25 worksheets in the commercial
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers) and
demands in the commercial sector. Fuels coming into the commercial sector have passed
through a collector technology that tracks the emissions from the fuels. The "EA" used in the
fuel name stands for "Emissions Accounting." Fuels used include electricity (COMELC), diesel
(COMDSLEA), kerosene (COMKEREA), LGP (COMLPGEA), natural gas (COMNGAEA),
residual fuel oil (COMRFLEA), and solar (COMSOL).
For the naming convention for the demand technologies, the first letter in the technology name
reflects the technology category. In this case 'C' is used for 'Commercial'. The remaining
letters represent the type of end-use demand. The demands in the database are:
CSC = Space Cooling equipment
CCK = Cooking equipment
CSH = Space Heating equipment
123
-------
CWH= Water Heating equipment
CLT = Lighting equipment
CRF = Refrigeration equipment
CVT = Ventilation equipment
COF = Office Equipment
CMD = Miscellaneous Diesel equipment
CME = Miscellaneous Electric equipment
CMN = Miscellaneous Natural Gas equipment
CML = Miscellaneous LPG equipment
CMR = Miscellaneous Residual Fuel Oil equipment
Constraints*
Lists the constraint names for all user constraints in the commercial sector.
Technologies*
Lists the technology names, units, and set memberships for all end-use and collector
technologies in the commercial sector. The naming convention starts with the end-use
demand, followed by the fuel type, the technology type, the efficiency level, and the model year.
For example, "CSH.NG.FURNACE.ST.10" refers to the 2010 model year of a standard
efficiency natural gas furnace for commercial space heating. The efficiency level is either: BS
for base model efficiency, ST for standard efficiency, or HE for high efficiency. Some of the
technology abbreviations used are given below:
Heating and Cooling
AHP = Air Source Heat Pump
GHP = Ground Source Heat Pump
ELO = Other Electric
CSC = Commercial Scroll Chiller
CSW = Commercial Screw Chiller
Ventillation
CAV = Constant Air Volume
Refrigeration
CEN = Supermarket Central Refrigeration
WIR = Walk-in Refrigerator
WIF = Walk-in Freezer
RIR = Reach-in Refrigerator
Lighting
INC = Incandescent
CFL = Compact Fluorescent
HAL = Halogen
LFL = Linear Fluorescent
LED = Light-emitting Diode
CRC = Commercial Reciprocating Chiller
CCC = Commercial Centrifugal Chiller
RAC = Rooftop Air Conditioning
WAC = Window Air Conditioning
CAC = Central Air Conditioning
VAV = Variable Air Volume
RIF = Reach-in Freezer
ICM = Ice Machine
BVM = Beverage Machine
RVM = Vending Machine
MH = Metal Halide
MV = Mercury Vapor
MAG = Magnetic
HPS = High Pressure Sodium
124
-------
ConstrData*
Contains the data for fuel shares and technology shares by end-use demands. Values for 2010
were calculated from the AEO market shares found in the Mkt Shares raw worksheet. Shares
are relaxed 3% per time period out to 2055.
CommData_Demand*
Contains the demand values by end-use and by region. The data are taken from DMD ByRegion
worksheet which is explained below. The fraction of capacity entering electricity peak equations
and the fractional demand by season and time of day (FR(Z)(Y)) are based on expert judgement.
TechData_COM*
Contains the parameter values for all end-use technologies (with the exception of solar
photovoltaics). The data for the start year, technology lifetime, investment cost, efficiency, and
fixed operating and maintenance cost are drawn from the AEO technology data aggregated in the
Aggregated Data worksheet. The data for the residual capacity are calculated in the RESID
worksheet. The values for the implied discount rate (or hurdle rate) for each technology are
determined based on financial and non-financial factors that affect the choice of various
commercial technologies. The discount rates range from 0.18 to 1.25 with the lower rates being
attached to standard technologies and the higher rates being attached to higher efficiency
technologies, new technologies such as ground source heat pumps, and technologies that use
diesel fuel.
TechData_Emis*
Contains technology definitions for the collectors that are used to assign emissions to various
fuels used within the commercial sector. The emission factors are fuel specific and do not
currently change as a function of time. Several species include a "C" added to the end, which is
used in sectoral emissions accounting. The emissions factors that populate Worksheet are
obtained from the following worksheets: EIACO2Coef, EmisBC, EmisOC, and EmisCom.
TechData SESC*
Contains the parameter values for two collector technologies: one carrying electricity to
commercial technologies and one carrying solar to the photovoltaics.
TechData _ZZ*
Contains the parameter values for dummy technologies for each of the end-use demands. These
technologies have a high variable O&M cost and will only be used in a model run when the
model cannot meet its demand with available technologies. The use of any of these technologies
indicates an infeasibility in the model.
Sol_PV*
Contains the parameters for commercial solar photovoltaics. The investment costs, fixed O&M
costs, and seasonally adjusted technology availability factors are taken from the electric sector
workbook and are based on values from the AEO2006 Solar Input data. Fixed bounds are
implemented forcing the use of solar photovoltaics based on the AEO 2010 projections.
Nationally projected use, found in the AEO12 Renew worksheet, is evenly distributed across the
nine-regions. After 2035, solar photovoltaics are projected to grow 30% per 5-year time period.
125
-------
AEO10 Renew
Contains raw data from the AEO main reference case table: "Renewable Energy Consumption
by Sector and Source."
RESID
Contains calculations for the residual capacity (RESID) of existing technologies in 2005. RESID
is calculated by multiplying the market share for each end-use technology times the total demand
for that end-use and dividing by the capacity factor (CF) for that technology.
Mkt Shares raw
Regional market shares by technology are taken from the AEO KTECH file and organized for
use in the Mkt Shares worksheet. Market shares for refrigeration, ventilation, and lighting are
given for the 11 building types instead of the nine-regions. Therefore, a national market share is
calculated.
Agg CLT
Lighting data are aggregated into 29 technologies.
CLT raw data
Worksheet has the lighting specific data from the AEO KTECH file.
Aggregated Data
Contains an aggregated listing of the end-use technologies from the AEO KTECH file.
Technologies grouped and shaded in grey are averaged into a single technology for the
MARKAL database. The technologies going into the EPA database are then given MARKAL
specific names and descriptions using standard MARKAL naming conventions.
AEO10 Com Tech
Contains raw data from the AEO Commercial Technology Characterization Database (KTECH
file). Each record gives specifications for a specific model year of a specific end-use technology
in a specific Census Division.
DMD ByRegion
Demand values calculated in the ComDemand worksheet are re-organized into regions.
ComDemand
End-use demands are calculated for each region using data from the CMCalc worksheet. For
space heating and space cooling, demands are calculated by multiplying the demand coefficient
by the regional commercial floorspace and the national HDD value. For all other end-uses, the
demands are calculated by multiplying the demand intensity by the regional commercial
floorspace. The national AEO demand is calculated using the delivered energy by end-use and
the stock efficiency from the AEO. These values are compared and graphed to the summed
regional demands as a reference.
126
-------
CMCalc
Contains a number of preliminary calculations used to determine the end-use demands found in
the worksheet ComDemand.
The grey shaded area of the worksheet holds the data and calculations for regional commercial
square footage. National square footage per capita is determined by dividing AEO total
commercial floorspace for each 5-year time period found in the table: "Commercial Sector Key
Indicators and Consumption" by the national population projections. From those Figures, a
national annual rate of change is calculated. The regional square footage per capita is calculated
for 2005 by dividing the square footage data found in the CBECS Table A2: "Census Region,
Number of Buildings and Floorspace for All Buildings" by the population for each region.
Subsequent years are then calculated by applying the annual rate of change to the previous year's
value. These regional square footage per capita are then multiplied by the regional population
projections to give the regional square footage, which are used in the end-use demand
calculations.
The green shaded area of the workbook holds the calculations for the percentage of buildings
with cooling technologies. The 2003 regional percentages were taken from the CBECS tables
C7, C8, C9: "Consumption and Gross Energy Intensity by Census Division for Sum of Major
Fuels for Non-Mall Buildings." Subsequent years were then calculated using an assumed annual
growth rate of 0.42%. This annual growth rate was calculated based on historical CBECS data.
The blue shaded area of the workbook holds the calculations for HDD and CDD and demand
intensity. Regional HDD and CDD values were take from the NCDC Historical Climatological
Series - HCS 5.1. Years recorded and used for our calculations end in June 2009. These data
can be found at www.ncdc.noaa.gov/oa/documentlibrary/hcs/hcs.htmltf51updates. Building
heating and cooling coefficients were calculated in the first two time periods by taking AEO
national space heating and space cooling demands and dividing them by the national floorspace
and the national HDD or CDD. Subsequent years were left equal to the year 2010. The demand
intensity for the other end-use demand categories were calculated simply by taking the AEO
national demands and dividing them by the national floorspace. For water heating and cooking,
the demand intensity was calculated by taking the major fuel energy intensity by end-use demand
from the CBECS Table E2A: "Major Fuel Consumption (Btu) Intensities by End Use for All
Buildings."
Pop
Regional population data were taken from the U.S. Census Bureau state population projections
released in 2005 and based on the 2000 Census. Data can be found at
www.census.gov/population/www/projections/projectionsagesex.htmal, Table 6: "Total
population for regions, divisions, and states: 2000 to 2030." Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are also taken
from EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (TPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. The data have been adjusted to
include Alaska and Hawaii.
127
-------
Conversion Factors
Contains a number of conversion factors including the Implicit Price Deflator (DOC, 2012) chart
used to convert AEO prices to the year 2005.
EmisCO2Coef
Contains CC>2 coefficients for various fuels, as reported by the DOE (DOE. 2008).
EmisBC
Includes estimates of the fraction of PM2.5 that can be classified as BC. The fractions, which can
differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and Bond
et al., 2007.
EmisOC
Includes estimates of the fraction of PM2.5 that can be classified as OC. The fractions, which
can differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and
Bond et al., 2007.
EmisCom
Includes commercial emission factors. These factors were derived from a number of sources,
including the Climate Registry and the EPA WebFIRE emission factor database. More
information about the derivation of the emission factors is available upon request.
128
-------
M: Sector Workbook Description - Industrial Sector
Workbook Name: EPAUS9r_10_IND_vl.3.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
industrial sector in the EPAUS9r MARKAL database. The also provides a brief description of
each worksheet in the Industrial workbook. The industrial sector workbook characterizes and
accounts for the energy consumption of many industries in the economy. In this documentation,
following terminology will be used.
• An industry sector is a group of industries that produce similar end-products or uses similar
raw materials. Examples include primary metals, chemicals, cement, food, etc.
• ^process (plant design/configuration) is defined as the set of technologies that would make
up the complete system (i.e. sets of equipment) to produce end-product/s in an industry.
• The energy used by the process will be categorized under energy service categories such as
boilers/steam/cogen, process heat (direct/indirect heat, electrical heating etc.), machine drive,
facility (or building), electrochemical, feedstock and other heat.
• A technology could be a piece of equipment (e.g. boiler) or set of equipments (e.g. a boiler
with control equipments) in a process.
Data Sources
The largest source of end-use energy consumption data for the industrial sector comes from the
MECS database (EIA, 2006). EIA State Energy Consumption, Price, and Expenditure Estimates
(SEDS) data (EIA, 2004) were used along with the MECS data to estimate the MARKAL
regional distribution of industrial energy use. Because these data set includes non-manufacturing
energy consumption, these data were used to make non-manufacturing demand assumptions in
the model. Additional data were taken from the AEO reference case. Some of the technology
definitions and parameter determinations for the industrial sector were based on EIA System for
Analysis of Global Energy markets (SAGE) model (EIA, 2003). Emissions data for energy
carriers used in the industrial sector is taken from the GREET model version 1.8 (Argonne,
2007).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of PJ. Emission factors for CC>2 are in units of MTonnes/PJ.
Units for all other pollutants and for CC>2 attributed specifically to this sector are in units of
kTonnes/PJ.
Workbook Descriptions
The following section gives a description of each of the 34 worksheets in the industrial
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
129
-------
ReadMe
Includes tips for adding new process and demand technologies, CHP characterization and
troubleshooting pointers.
Demand
Calculates the future energy demands based on AEO value of shipments pulled from the
AEOData worksheet. These energy demands are used in the CommData_Dmd worksheets.
AEOData
Presents raw AEO data for value of shipment from each industry sector.
Commodities*
Declares the commodity names, units, and set memberships for all fuels (energy carriers),
emissions, and demands in the industrial sector.
The energy carriers for industrial sector follow the naming convention below:
First 3 letters: IND for industrial
Next letters: indicates the fuel type used in all industrial sectors (e.g. COA for coal,
DSL for diesel, etc.)
Next 3-5 letters: indicate the particular fuel carrier (i.e. CNG for compressed natural gas,
E85 for 85% ethanol, DSL for diesel, JTF for jet fuel, etc.)
The demands for industrial sector follow the naming convention below:
First letter: I for Industrial
Second letter: (When not IND) indicates the industry sector (i.e. C for chemicals, P for
pulp & paper, M for metals. F for food, N for non-metallic manufacturing,
T for transportation equipment, O for all other industries, and X for non-
manufacturing)
Next 3-7 letters: indicate the energy service demand category (i.e., STM for steam, PRH
for process heat, MDR for machine drive, FAC for facility, FEED for
feedstock, CHEAT for other heat)
Technologies*
Declares the technology names, units, and set memberships for all end-use demand and collector
technologies for each industry in the industrial sector. For each industry sector, a variety of
process technologies using a variety of fuel types is represented to fulfill energy service demands
such as steam, process heat, machine drive, etc.
The process technologies for the industrial sector follow the naming convention below:
First letter: I for Industrial
Second letter: Indicates the industry sector (i.e. C for chemicals, P for pulp & paper, M
for metals. F for food, N for non-metallic manufacturing, T for
transportation equipment, O for all other industries, and X for non-
manufacturing)
130
-------
Next 2-3 letters: indicate the energy service demand technologies (i.e., BOR for boiler,
PRH for process heat, MDR for machine drive, ECM for electrochemical,
FAC for facility, FST for feedstock, OH for other heat)
Next 3-5 letters: indicate the type of fuel used in that category (i.e., COA for coal, DST for
distilates, RFL for residual fuel oil, NGA for natural gas, ELC for
electricity, etc.)
Last 2 letters: 01 for existing technologies; new technologies do not include any number
suffix
Besides process technologies, industries may use CFIP technologies. The naming conventions
for these technologies are:
First letter: I for Industrial
Next 1-2 letters: Indicates the industry sector (i.e. CM for chemicals, PP for pulp & paper,
M for metals. FD for food, NM for non-metallic manufacturing, TN for
transportation equipment, and OT for all other industries)
Next letter Indicates an existing (E) or new (N) technology
Next 3 letters: indicate the CHP technology type (i.e., BST for boiler/steam turbine, CCT
for combined cycle/combustion turbine, FCL for fuel cell, REG for
reciprocating engine, MTT for microturbine)
Next 3 letters: indicate the type of fuel used in that category (i.e., COA for coal, BIO for
biofuels, OIL for oil, NGA for natural gas, WOD for wood)
In addition there is a technology which represents each industry sector as a whole. The naming
convention for this technology is
First letter: I for Industrial
Second letter: Indicates the industry sector (i.e. C for chemicals, P for pulp & paper, M
for metals. F for food, N for non-metallic manufacturing, T for
transportation equipment, O for all other industries, and X for non-
manufacturing)
Next 7 letters: TECHEXT
In addition, two types of dummy technologies are utilized in the database. The first are dummy
backstops for each demand technology. These avoid model infeasibility from inadequate energy
carrier supply to a certain demand; if enough energy is not available to meet the demand the
model will use "ZZDMY". This aids the modeler in determining where problems exist within
the specific energy chains.
The second are dummy fuel collectors. Because the MECS data, the SEDS data, and the fuels
entering the industrial sector all have different names and characteristics, it is sometimes
necessary to combine fuels into one generic energy source.
TechData_X&ZZ*
Contains the parameter values for all dummy backstop technologies.
131
-------
CHP_resids
Contains raw data calculations for the technology characterization of the existing CHP units in
each industry sector. Data are pulled from the CHP-raw data worksheet.
CHP-raw data
Includes the raw data on the individual existing CHP units taken from the CHP database. See
CHPDATA_22108LCD.xls workbook for details. Worksheet also contains the AEO raw data on
CHP usage for all industries.
TechData CHP*
Includes technology parameter values to be input to MARKAL for both existing and new CHP in
each industry sector, i.e., food, paper, chemical, other, primary metals, non-metals, and
transportation equipment.
TechData DMD*
Contains parameter values for all end-use demand technologies in the industrial sector as defined
in the Technologies worksheet. Worksheet also includes the fractional distribution of each
process category (boiler/steam, machine drive, process heat, electrochemical, other heat,
feedstock, and facilities) which is required to meet a unit of demand for a given industry. (This is
the xxTECHEXT technology) The parameter values are pulled from Demand, RESID_TDATA,
National_EndUse, and Steam worksheets.
TechData_Emis*
Contains emission factors associated with industrial fuel combustion. These factors are pulled
from the Emissions, RTI-BC and RTI-OC worksheets.
Emissions
Worksheet compiles relevant emissions from the GREETl.SbEmis worksheet into a format that
can be readily used in the TechData_Emis worksheet.
GREETl.SbEmis
Includes industrial emission factors that were extracted from the GREET model, version 1.8, and
converted to MARKAL units.
RTI-BC
Includes estimates of the fraction of PM2.5 that can be classified as BC. The fractions, which can
differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and Bond
et al., 2007.
RTI-OC
Includes estimates of the fraction of PM2.5 that can be classified as OC. The fractions, which
can differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and
Bond et al., 2007.
132
-------
Steam
Worksheet allocates CHP steam residuals to the regional steam demands. The balance of the
steam demands and other process heat demands are distributed across the steam and other
process heat technologies, respectively, using national fuel ratios to establish residual capicity for
these processes.
RESID TDATA
Worksheet compiles the raw data from the Steam worksheet and does additional calculations to
have residual values for all other existing technologies.
old_CensusDiv
Contains the original MECS data on fuel consumption. The original MECS data are presented
for four-regions, i.e., northeast, southeast, northwest and southwest.
CensusDiv
Worksheet allocates the fuel consumption presented in the old_CensusDiv worksheet from
MECS four-regions into MARKAL's nine census regions by using SEDS data Table S6 of
estimated industrial energy use in 2004 by state. Because SEDS data includes non-manufacturing
demand, a ratio of total energy use between MECS data and SEDS data was calculated to
estimate the manufacturing-only energy use by state. Those state data could then be organized on
a nine-region basis and compared directly with the census four-region data available from
MECS. The fraction of nine-region energy use to four-region energy use was calculated by
dividing the total manufacturing energy use in a particular nine-region by the total manufacturing
energy use in the four-region (i.e., energy use in MARKAL Region 2 (New England) over
energy use in the Northeast census division). For the paper sub-sector, regional distribution was
calculated using the state-based mill data from the CPBIS database (CPBIS, 2007). Mills in both
the nine-region and four-region areas were counted, and the same ratio for energy use was
calculated.
Feedstock
Contains MECS data on feedstock type fuel consumption in industrial sector. Original data are
presented in four census regions. In the spreadsheet the data are allocated into nine census
regions using SEDS data as above.
CommData_Dmd_Food*
Contains parameter data for the total food sector end-use demand. The data are pulled from
CensusDiv and Demand worksheets.
CommData_Dmd_Prmry Metals*
Contains parameter data for the total primary metals sector end-use demand. The data are pulled
from CensusDiv and Demand worksheets.
CommData_Dmd_Chem*
Contains parameter data for the total chemicals sector end-use demand. The data are pulled from
CensusDiv and Demand worksheets.
133
-------
CommData_Dmd_Paper*
Contains parameter data for the total pulp and paper sector end-use demand. The data are pulled
from CensusDiv and Demand worksheets.
CommData_Dmd_Non-Metal Minrls*
Contains parameter data for the total non-metallic manufacturing sector end-use demand. The
data are pulled from CensusDiv and Demand worksheets.
CommData_Dmd_Tranprt Equip*
Contains parameter data for the total transportation equipment sector end-use demand. The data
are pulled from CensusDiv and Demand worksheets.
CommData_Dmd_Oth*
Contains parameter data for all the other industrial sectors end-use demand. The data are pulled
from CensusDiv and Demand worksheets.
CommData_Dmd_NonMan*
Contains parameter data for the total non-metal manufacturing sector end-use demand. The data
are pulled from CensusDiv and Demand worksheets.
Constraints*
Worksheet declares the share constraint names for all user constraints in the industrial sector.
ConstrData*
Contains the data for fuel share splits for each energy service demand in industrial sectors.
Constraints in the industrial sector are added to all process technologies from the years 2010-
2055. There are no constraints in the year 2005 because the RESID parameter acts as a
constraint. Similar to the RESID parameter, constraints are based on the percentage a specific
energy carrier accounts for the total energy use of a technology based on MECS data Table 5.2.
When MECS data are not available, AEO data are used. Constraints apply over all regions unless
the technology does not exist in a particular region. Constraints in the industrial sector are all
lower bounds, meaning the technology has to use equal to the constraint value for a particular
energy carrier, but above that the technology can chose another carrier if it provides a less-cost
solution. Because of the end-use driven nature of the industrial sector, energy carriers and
process technologies are tightly constrained. Some of the technologies are allowed fuel
switching, and for these technologies associated fuel constraints are relaxed somewhat up to
2055.
Worksheet also contains national level constraints for CHP usage in each industry. CHP usage
through time grows at the rates given by AEO which are pulled from the CHP-raw data
worksheet.
UC Shares
Worksheet calculates the fuel splits for each energy service end-use category in each sector. The
values generated in this workbook are linked to the ConstrData worksheet to implement lower
bounds on the fuel shares. Worksheet also contains the raw data for fuel switching which sets the
134
-------
relaxation limits for the constraints. The fuel switching data comes from OPEI report (EPA,
2007).
National_EndUsel
Contains raw data from MECS on the fuels used by each industry at the national end use. The
worksheet aggregates the MECS categories to MARKAL process categories and brings the CHP
steam data and feedstock data from their worksheets.
135
-------
N: Sector Workbook Description - Industrial Biofuels
Workbook Name: EPAUS9r_12_INDBIO_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the pulp and
paper process to convert biomass into process steam and black liquor production in the EPAUS9r
MARKAL database.
Data Sources
The technology characteristics were taken from the work of E.D. Larson et al. (Larson et al.,
2000 and Larson et al., 2006). and the AEO. Detailed references are provided on the
BiomassCHPforlPS worksheet.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Workbook Description
The following section gives a description of each of the five worksheets in the industrial biofuels
workbook. The worksheets are listed in the order they appear, from left to right, in the
workbook. The worksheet names noted with an asterisk contain the data that are automatically
uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all process energy carriers.
Technologies*
Lists the technology names, units, and set memberships for pulp and paper black liquor
production, black liquor gasifier, and biomass boiler.
TechData*
Contains the MARKAL ready data for pulp and paper black liquor production, black liquor
gasifier, and biomass boiler.
BiomassCHPforlPSupdated
Contains the calculations for the residual capacity for the biomass boiler and the black liquor
resource supply. Data are drawn from the EIA, Office of Energy Markets and End Use, Energy
Consumption Division, Form EIA-846, "2002 Manufacturing Energy Consumption Survey."
BiomassCHPforlPS
Charts the raw data for the capital and operating costs of a black liquor cogeneration system.
Data are drawn from the reports Larson et al.
136
-------
O: Sector Workbook Description - Light Duty Transportation
Workbook Name: EPAUS9r_12_TRN_LDV_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
transportation light duty vehicle sector in the EPAUS9r MARKAL database.
Data Sources
The technology characteristics for traditional light-duty vehicle technologies were taken from the
AEO. Additional data used were taken from the Transportation Energy Consumption Survey
(EIA, 2001) and the Transportation Energy Data Book (DOE, 2010). Data for advanced
technologies were taken from inputs provided by National Renewable Energy Laboratory, ERG,
and EPA's Office of Transportation Air Quality. Emissions data were developed using the
MOVES model (EPA, 2010), although black carbon and organic carbon emission speciations
were developed by RTI from a literature review. CO2 emission factors were developed from EIA
estimates.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of billion vehicle miles traveled (bn-vmt). Emissions are
represented in grams per mile, which is equivalent to kTonnes/billion-vmt.
Workbook Description
The following section gives a description of each of the 35 worksheets in the light duty
transportation workbook. The worksheets are listed in the order they appear, from left to right,
in the workbook. The worksheet names noted with an asterisk contain the data that is
automatically uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers), emissions,
and demands in the transportation light-duty vehicle sector. The demand for this sector is given
the name TL for "transportation light-duty." The energy carriers follow the following naming
convention:
First 1-2 letters: TR or TRN for Transportation
Next 3-5 letters: point to the particular fuel carrier (i.e. CNG for compressed natural gas,
ETH for ethanol, GSLR for reformulated gasoline, etc.)
Technologies*
Lists the technology names, units, and set memberships for all end-use and collector
technologies in the commercial sector. The naming convention for the demand technologies
starts with the end-use demand (TL), followed by the car class (E for existing, MC for
minicompact, C for compact, F for full-size. M for minivan, P for pickup, SS for small SUV, and
137
-------
LS for large SUV), the fuel type, and finally the first year of technology availability. For
example, "TLPHEV15" refers to hybrid electric pickups with an initial availability in the year
2015.
Collector technologies start with an SC, followed by the input energy carrier name and then the
output energy carrier name.
Constraints*
Lists the share and investment constraint names for all user constraints in the transportation light-
duty vehicle sector.
CommData*
Contains the demand values for transportation light-duty vehicles by region. The data are taken
from the VMT By Region worksheet which is explained below. The fraction of capacity
entering electricity peak equations and the fractional demand by season and time of day
(FR(Z)(Y)) are based on expert judgement.
TechData All*
Contains the parameter values for all light-duty vehicle technologies drawn from the
ConsolidatedVehData worksheet explained below. The discount rates range from 0.40 to 0.44:
with the lower rate attached to conventional gasoline vehicles and the higher rate being attached
to new technologies.
TechData_RES*
Contains the parameter values for existing light-duty vehicle technologies. Efficiency and O&M
costs come from the ConsolidatedVehData worksheet and the residual capacity values come
from the ResidDataAndCalcs worksheet.
TechData ZZ*
Contains the parameter values for a dummy technology for the end-use demand. This
technology has a high variable O&M cost and will be used only in a model run when the model
cannot meet its demand with available technologies. The use of any of these technologies
indicates an infeasibility in the model.
ConstrData*
Contains the constraint data for fuel shares, car class splits, technology shares, and CAFE
standard by region. Approximated CAFE standards are pulled from the CAFE10 worksheet.
TechData ResidEmis*
Represents emissions from the existing stock of vehicles. These emission factors are obtained
from the EmisLDV worksheet.
TechData_NewEmis *
Represents emissions from new vehicles, vintage 2010 and beyond. These emission factors are
obtained from the EmisLDV worksheet.
138
-------
EmisLDV
Represents light duty emission factors derived from runs of the MOVES model. Emission factors
for existing vehicles include consideration of pre-2005 vintages that leave the fleet, as well as
degredation of vehicle emission controls over time. Emission factors for 2010 and later vintages
represent lifetime average emissions and do not otherwise incorporate degredation. The
calculations that underlie these emission factors are available upon request.
EmisOC
Includes estimates of the fraction of PM2.5 that can be classified as OC. The fractions, which
can differ by source category, originate from three studies: Battye et al., Bond et al., 2004, and
Bond et al., 2007.
ERG_ElectricVeh
Contains vehicle cost and efficiency data for electric vehicle technologies. The data were
provided by Rick Baker of ERG.
ConsolidatedVehData
Worksheet consolidate vehicle efficiency and cost from the Costs, Efficiencies, and
AdditionalVehs worksheets into a single large table.
AdditionalVehs
Contains cost and efficiency data for vehicle technology-classification pairs not in the original
AEO data, including E85 hybrids, PHEV20, E85 PHEV20, and E85 PHEV40. These tables
consolidate efficiency and cost data on advanced vehicle technologies from OTAQ and AEO
sources. Additional technologies are also added, including moderate ICE vehicles, running both
on gasoline and E85, advanced ICE vehicles running on E85, PHEV20s running on gasoline and
E85, and PHEV40s, running on E85. Many of these additional vehicles currently are commented
out in the Answer upload sheets.
Costs
Contains the vehicle technology purchase cost estimates calculated by layering OTAQ vehicle
assumptions for advanced technologies onto AEO assumptions regarding conventional vehicle
technologies. Data from 2040 through 2055 are held constant to the 2035 value.
Efficiencies
Contains the vehicle technology efficiency estimates calculated by layering OTAQ's vehicle
assumptions for advanced technologies on AEO assumptions regarding conventional vehicle
technologies. Data from 2040 through 2055 are held constant to the 2035 value. AEO
adjustment factors are used to reduce efficiencies to account for real-world driving conditions
and vehicle degredation over time.
CAFE10
Includes calculations to derive the national energy-to-LDV constraints to reflect the CAFE
standards. This is a preliminary implementation. No credit is given towards the standard for
alternative fuel vehicles. Similarly, this constraint does not represent different efficiency targets
for cars and trucks.
139
-------
NREL PHEV
Contains raw data estimating the fraction of plug-in hybrid electric vehicle operation under
electric-power only. Data come from the National Renewable Energy Laboratory (Denholm and
Short).
AEO10 T59
Contains the raw data for vehicle efficiency projections by technology and vehicle classification
taken from the AEO table: "New Light-Duty Vehicle Fuel Economy."
AEO10 T60
Contains the raw data for vehicle cost projections by technology and vehicle classification taken
from the AEO table: "New Light-Duty Vehicle Prices."
OTAQ Car
Contains data for light duty car technology cost and efficiency assumptions as used with SGM,
provided by the EPA Office of Transportation and Air Quality. Plug-in hybrid electric vehicle
assumptions for efficiency gain and gasoline utilization are taken from worksheet.
OTAQ Truck
Contains data for light duty truck technology cost and efficiency assumptions as used with SGM,
provided by the EPA Office of Transportation and Air Quality. Plug-in hybrid electric vehicle
assumptions for efficiency gain and gasoline utilization are taken from Worksheet.
AEO10 Salesdata
Contains the raw data and calculations for determining the car class splits in 2005 and 2025.
Total new car and truck sales, in thousands, come from the AEO table: "Light-Duty Vehicle
Sales by Technology Type." Percent of new vehicle shares by car class comes from the AEO
table: "Summary of New Light-Duty Vehicle Class Attributes." Car class splits are calculated
by multiplying the percent of new vehicle shares by car class by the total new sales, and then
getting the fraction of the total by car class.
AEO10 T7
Contains raw data from the AEO table: "Transportation Sector Key Indicators and Delivered
Energy Consumption."
ResidDataAndCalcs
Contains the calculations for regional residual capacity for existing light-duty vehicle types.
Residual capacity in the year 2005 by technology and by region, given in billion vehicle miles
traveled (VMT) per year, is calculated by multiplying the regional total VMT by the fraction of
VMT that is either passenger cars or light trucks. The resulting value is then multiplied by the
fraction of total VMT for each technology class (i.e. MiniCompact, Full, Minivan, etc.). Data for
the calculations come from a number of sources. The 2005 regional total VMT are calculated in
the VMT By Region worksheet. The fraction of VMT that is either passenger cars or light
trucks is found on the TED10 worksheet. The fraction of total VMT for each technology class is
calculated in worksheet from data for millions of vehicles by type and fractional VMT taken
from the Transportation Energy Consumption Survey table: "Light Duty BVMT by Technology,
140
-------
2005." Residual capacity for the time periods after 2005 are calculated by multiplying the 2005
value by the estimated scrappage rates of 30% after 5-years, 59% after 10 years, 84% after 15-
years, and 92% after 20 years.
TED10
Contains data from the Transportation Energy Data Book. Vehicle miles traveled for 2005 are
taken from Table 4.1 "Summary Statistics for Passenger Cars" and Table 4.2 "Summary
Statistics for Two-Axle, Four-Tire Trucks" and used for residual capacity splits for passenger
cars and trucks on the ResidDataAndCalcs worksheet. Table 2.3 "Alternative Fuel and
Oxygenate Consumption" and Table 6.4 "Number of Alternative Refuel Sites by State and Fuel
Type" provide raw data for calculating the residual capacity for alternative fuel vehicles on the
ResidDataAndCalcs worksheet.
AEO08 T50
Contains raw data from the AEO table: "Light-Duty Vehicle Miles Traveled by Technology
Type." Data for 2005 vehicle miles traveled is used on the ResidDataAndCalcs worksheet to
calculate residual capacity of the different technology types within passenger and light truck
classes.
VMT By Region
Contains the raw data and calculations for determining the regional demand for light duty vehicle
miles traveled. Raw data for the regional percentage of national vehicle miles traveled comes
from the Transportation Energy Consumption Survey Table Al: "U.S. Number of Vehicles,
Vehicle-Miles, Motor Fuel Consumption and Expenditures, 2001." These values are multiplied
by the national VMT demand and then divided by the population to get the regional VMT per
capita. These new regional splits are then multiplied by the regional population from the
worksheet Pop to obtain the regional demands.
AEO10 DMD
Contains the raw data and calculations for determining the base light duty vehicle miles traveled
per capita. Raw data for national vehicle miles traveled (VMT) comes from the AEO table:
"Transportation Sector Key Indicators and Delivered Energy Consumption." The base VMT per
person is determined by dividing the national VMT by the population from the Pop worksheet.
Beyond 2035, the VMT per person is held constant.
Pop
Regional population data was taken from the U.S. Census Bureau state population projections
released in 2005 and based on the Census 2000. Data can be found at
www.census.gov/population/www/projections/projectionsagesex.htmal, Table 6: Total
population for regions, divisions, and states: 2000 to 2030. Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are also taken
from EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (TPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. The data have been adjusted to
include Alaska and Hawaii.
141
-------
GDP Deflator
Contains the Implicit Price Deflator (DOC, 2012) chart used to convert AEO prices to the year
2005.
142
-------
P: Sector Workbook Description - Heavy Duty Transportation
Workbook Name: EPAUS9r_12_TRN_HDV_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Document describes the sources of the data and the calculations used to characterize the
transportation heavy duty vehicle sector in the EPAUS9r MARKAL database.
Data Sources
The demands for heavy-duty subsectors were taken from the AEO. Some characteristics of on-
road heavy-duty vehicle technologies were taken from the NEMS Advanced Technology
Options File. Additional data are referenced on the worksheets where data is utilized.
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of billion vehicle miles traveled (bn-vmt) for on-road vehicles,
billion ton miles (bn-t-m) for rail and marine freight, and billion passenger miles (bn-pass) for air
and rail passenger travel.
Workbook Description
The following section gives a description of each of the 21 worksheets in the heavy duty
transportation workbook. The worksheets are listed in the order they appear, from left to right,
in the workbook. The worksheet names noted with an asterisk contain the data that are
automatically uploaded to ANSWER when importing data from Excel.
Commodities*
Declares the commodity names, units, and set memberships for all fuels (energy carriers),
emissions, and demands in the transportation heavy-duty vehicle sector. The demands for
heavy-duty sector are given the following names:
TA Transportation Air
TB Transportation Bus
TC Transportation Commercial Trucks (Class 2b)
TM Transportation Medium Duty Trucks (Class 3-6)
THS Transportation Heavy Duty Trucks - Short Haul (Class 7-8)
THL Transportation Heavy Duty Trucks - Long Haul (Class 7-8)
TRF Transportation Rail - Freight
TRP Transportation Rail - Passenger
TS Transportation Shipping (Marine)
The energy carriers follow the following naming convention:
First 1-3 letters: T or TRN for Transportation
Second letter: (When not R) indicates the subsector using the fuel (i.e. B for bus, C for
143
-------
commercial truck, or H for heavy duty)
Next 3-5 letters: indicate the particular fuel carrier (i.e. CNG for compressed natural gas,
E85 for 85% ethanol, DSL for diesel, JTF for jet fuel, etc.)
Technologies*
Declares the technology names, units, and set memberships for all end-use and collector
technologies in the commercial sector. The naming convention for technologies starts with the
end-use demand (TA, TB, THL, etc.), followed by the fuel type, and finally either (a) "E" for
existing technology, (b) the first year of availability for conventional technologies, or (c) an
indicator of efficiency for improved (EVI) or advanced (ADV) technologies. For example,
"TBCNG15" refers to conventional CNG fueled buses with an initial availability in the year
2015. "TCGSLADV" refers to a gasoline commercial truck with advanced efficiency
technology.
The sheet includes two emission collectors "SETRDSL" and "SETRB20" where criteria
emissions are counted from diesel and B20 fuels used in rail.
TechData RES*
Contains the parameter values for all heavy-duty vehicle technologies. Efficiency and operating
and maintenance costs come from the sub sector worksheets (Air& Marine, Freight, Trucks,
Rail&Bus, Rail&Bus2) and the residual capacity values come from the Resid worksheet.
CommData*
Contains the demand values for transportation heavy-duty vehicles by region. The data are taken
from By Region worksheet.
Constraints*
Declares the share constraint names for all user constraints in the transportation heavy-duty
vehicle sector.
ConstrData*
Contains the data for fuel share splits and the passenger rail share splits (subway, intercity, and
commuter) for all user constraints in the transportation heavy-duty vehicle sector. The 2010
values are pulled from the AEOData, Resid, and Rail&Bus worksheets. Fuel shares are
allowed to shift through time. In most cases, constraints are relaxed 1-3% per period. For some
fuels AEO projects increased usage in the future (for example, CNG usage in buses). For these
fuels, the share is constrained to match AEO in 2035.
TechData EMIS*
Contains the emission factors for all technologies listed in the TechData_RES worksheet.
Values are pulled from the Emissions worksheet described below.
Air&Marine
Contains the technology characteristics data for ships and for passenger and cargo airplanes
imported in final MARKAL units from ERG files. The data for the post-2010 technologies for
these heavy duty classes were gathered from various sources. Raw data, calculations, and
144
-------
references are listed in the ERG workbooks: "MARKAL MV ERG Final" and "MARKAL
Aviation ERG Final". Technology characteristics from worksheets are passed to the MARKAL
TechData_Res worksheet which uploads into ANSWER.
Freight
Contains the technology characteristics data for rail freight imported in final MARKAL units
from the ERG file "MARKAL Rail ERG Final". The data for post-2010 technologies come from
various sources. Raw data, references, and calculations are shown in the ERG workbook. The
Freight worksheet also contains three tables of raw state-level data for commodity shipments
used to determine the regional shares of rail, truck, and marine freight. The references for these
tables are located in the Freight worksheet. Technology characteristics from worksheet are
passed to the MARKAL TechData_Res worksheet which uploads into ANSWER. Regional
shares from worksheet are passed to the Transcalc worksheet.
Trucks
Contains the technology characteristics data for all classes of on-road trucks from ERG files.
Most of the data for post-2010 truck technologies were derived by ERG using the NEMS
Advanced Technology Options file. Other sources used by ERG are referenced in the ERG
workbook: "Onroad FID Trucks and Buses ERG Final". Technology characteristics from
worksheet are passed to the MARKAL TechData_Res worksheet which uploads into ANSWER.
Rail&Bus
Contains the raw data and calculations to determine national demands and regional shares for bus
and passenger rail services. AEO provides data on the fuel energy used by buses and passenger
rail but does not provide end-use services (See AEO table in the AEO Data worksheet.)
Worksheet converts the fuel used to billion VMT for buses and billion passenger miles for rail.
National demands and regional shares from worksheet are passed to the Transcalc worksheet.
Rail&Bus2
Contains the technology characteristics data for buses and the three types of passenger rail
(subway, intercity, and commuter). Most of the data for the post-2010 bus technologies were
derived by ERG using the NEMS Advanced Technology Options file. Data and references from
other sources are listed in the ERG workbook: "Onroad FID Trucks and Buses ERG Final".
Passenger rail technologies are limited to those provided in the original 1997 MARKAL
workbook. However, characteristics have been updated. Efficiencies have been updated from
data sources on the Rail&Bus worksheet, and cost data updated from the "MARKAL Rail ERG
Final" workbook. Technology characteristics from Worksheet are passed to the MARKAL
TechData_Res worksheet which uploads into ANSWER.
Res id
Contains the calculations for regional residual capacity for existing heavy-duty vehicle types.
Residual capacity in the year 2005 by fuel and by region [given in billion vehicle miles traveled
(VMT) per year, billion ton-miles (T-M) per year, or billion passenger miles (PASS) per year] is
calculated by multiplying the regional total demand (from the TransDemand worksheet) by the
fraction of demand that is delivered by a specific fuel. The raw data for the fuel shares on this
sheet come from various sources including AEO, the Transportation Energy Data book, and the
145
-------
Census Bureau 2002 Vehicle Inventory and Use survey. Data from worksheet are passed to the
MARKAL TechData_Res worksheet which uploads into ANSWER.
By Region
Organizes the data for all end-use demands by region. Demand data calculations are done on the
TransDemand worksheet described below. Data from the ByRegion worksheet are passed to
the MARKAL CommData worksheet which uploads into ANSWER.
TransDemand
Contains the calculations for the regional end-use demands. Regional demand is equal to the
national demand times the regional fraction. These two values are pulled from the Transcalc
worksheet.
TransCalc
Contains intermediate calculations for the national end-use demand and collects the regional
fractions for all end use demands from the various technology worksheets. Initial values of
national end-use demands are taken from the AEO Data worksheet. These values are used in
the TransCalc worksheet to determine demand per capita so that overall national demand can
reflect changes in population under this scenario condition. Population data are drawn from the
Pop worksheet.
Pop
Regional population data were taken from the U.S. Census Bureau state population projections
released in 2005 and based on the Census 2000. Data can be found at
www.census.gov/population/www/projections/projectionsagesex.htmal. Table 6: Total
population for regions, divisions, and states: 2000 to 2030. Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are taken
from EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (IPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. The data have been adjusted to
include Alaska and Hawaii.
AEO Data
Contains the raw data taken from the AEO tables for use in calculating national demands and
fuel shares. Data extracted from the tables are: "Freight Transportation Energy Use" - medium
and heavy duty VMT demand, medium and heavy duty fuel shares, freight rail ton-miles
demand, domestic marine shipping ton-miles demand; "Air Travel Energy Use" - billion
passenger miles demand; "Transportation Sector Energy Use by Fuel Type within a Mode" -
fuel shares for commercial trucks, fuel shares for marine shipping, fuel shares for air
transportation, fuel shares for buses, fuel shares for passenger rail; and "Transportation Sector
Key Indicators and Delivered Energy Consumption" - commercial truck VMT demand.
Raw Emissions
Contains the raw emissions factors collected and developed by Pechan. Original data and
references from Pechan are shown in the following workbooks: "4 On-Road Emission Factors
146
-------
Global Summary," "CMV Emission Factors," "Rail Emission Factors," and "Aircraft Emission
Factors." As noted above, Pechan used data from ERG for aircraft emissions.
Emissions
Worksheet maps the technology classes from the Raw Emissions worksheet to the technology
classes in the heavy duty sectors of this workbook. Data from this Emissions worksheet are
passed to the MARKAL TechData_EMIS worksheet which uploads into ANSWER.
Conv Factors
Contains various conversion factors used throughout this workbook.
147
-------
Q: Sector Workbook Description - Off-Highway Transportation
Document describes the sources of the data and the calculations used to characterize the
transportation off-highway sector in the EPAUS9r MARKAL database.
Workbook Name: EPAUS9r_12_TRH_OH_vl .O.xlsx
Description Revision: 1.0
Revision Date: 12/31/12
Data Sources
The technology and end-use demand characteristics were taken from the AEO reference case.
Emission factors were derived from the EPA's Clearinghouse for Inventories and Emissions
Factors (CHIEF) National Emissions Inventory (NEI) Air Pollutant Trends Data and EPA diesel
emissions analysis (EPA, 2004a).
Units
All costs are expressed in millions of 2005 dollars. All energy quantities are expressed in PJ.
Demand units are given in terms of PJ. Emission factors are in kTonnes/PJ, except for system-
wide CC>2, which is reported in MTonnes/PJ.
Workbook Description
The following section gives a description of each of the 12 worksheets in the off-highway
transportation workbook. The worksheets are listed in the order they appear, from left to right,
in the workbook. The worksheet names noted with an asterisk contain the data that are
automatically uploaded to ANSWER when importing data from Excel.
Commodities*
Lists the commodity names, units, and set memberships for all fuels (energy carriers), emissions,
and demands in the transportation off-highway sector.
There are two demand technologies:
TOHDSL = Off-highway diesel use
TOHGSL = Off-highway gasoline use
Technologies*
Lists the technology names, units, and set memberships for all end-use and collector
technologies in the transportation off-highway sector.
Similar to the demands, there are two end-use demand technologies"
TODSL = Off-highway diesel technology
TOGSL = Off-highway gasoline technology
CommData_Demand*
Contains the demand values by end-use and by region.
148
-------
TechData_RES*
Contains the parameter values for diesel and gasoline off-highway technologies.
By Region
Regional demand is re-sorted by region.
Reg Demand
Contains regional demand data calculated from the national demands and the regional shares for
off-highway petroleum consumption.
Natl Demand
Contains national demands for off-highway petroleum consumption recalculated using a per
capita measure to allow for automatic changes to the demands under different population
assumptions in the Pop worksheet.
Off-Highway
Contains calculations for the national demands and regional shares for off-highway petroleum
consumption. The demands for 2005 are calculated using data from the ORNL data (ORNL,
2004). Future year demands are calculated using a growth rate developed from AEO projections
of agriculture, construction, and recreational transportation gasoline and diesel fuel use.
Regional demand shares are calculated from state level data from the U.S. Department of
Transportation's Highway Statistics 2005, Table MF-24: Private and Commercial Nonhighway
Use of Gasoline-2005."
Pop
Regional population data was taken from the U.S. Census Bureau state population projections
released in 2005 and based on the Census 2000. Data can be found at
www.census.gov/population/www/projections/projectionsagesex.htmal, Table 6: Total
population for regions, divisions, and states: 2000 to 2030. Subsequent years out to 2055 were
calculated using a linear extrapolation. Additional regional population projections are also taken
from the EPA's Integrated Climate and Land-Use Scenarios (ICLUS) projections based on the
Intergovernmental Panel on Climate Change's (TPCC) Al Emissions Scenario. These data are
found in the ICLUS workbook ICLUS'populationpopulation. The data have been adjusted to
include Alaska and Hawaii.
NOX, SO2, VOC, PM10, CO, PM25, and CO2
These worksheets contain the calculations for off-highway gasoline and diesel technolgies. Data
are taken from the NEI.
regulation
Contains the raw emissions data for land-based nonroad diesel engines. The data are drawn from
the EPA regulatory analysis (EPA, 2004a).
149
-------
reference
Contains raw data for off-highway gasoline and diesel use and national criteria air pollutant
emissions taken from the Transportation Energy Data Book Edition 24.
150
------- |