&EPA
Roadmap for Incorporating Energy
Efficiency/Renewable Energy
Policies and Programs into State and Tribal
Implementation Plans
Appendix I: Methods for Quantifying Energy
Efficiency and Renewable Energy Emission
Reductions
-------
EPA-456/D-12-001J
July 2012
Roadmap for Incorporating Energy Efficiency/Renewable Energy
Policies and Programs into State and Tribal Implementation Plans
Appendix I: Methods for Quantifying Energy Efficiency and Renewable Energy Emission
Reductions
By:
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Outreach and Information Division
Research Triangle Park, North Carolina
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Outreach and Information Division
Research Triangle Park, North Carolina
-------
ACKNOWLEDGMENTS
We would like to acknowledge substantial contributions from members of an inter-office EPA
team that included the Office of Atmospheric Programs, the Office of Policy Analysis and
Review, the Office of General Counsel and Regions 1 and 6. This document also reflects
comments received from a number of stakeholders, including state and local air quality
agencies.
1-2
-------
Contents
FIGURES 1-4
TABLES 1-5
SECTION 1.1: PURPOSE 1-6
SECTION 1.2: GETTING STARTED: IMPORTANT FACTORS 1-6
Availability of Staff and Budgetary Resources 1-6
Energy Data Availability and Relationship between Air and Energy Agencies 1-6
Potential Magnitude of Emission Reductions 1-7
Multi-Jurisdictional Collaboration 1-7
SECTION 1.3: STEPS FOR QUANTIFYING EMISSIONS OF ENERGY EFFICIENCY AND RENEWABLE ENERGY POLICIES
AND PROGRAMS 1-8
Step 1: Develop a Future Projected Baseline Emissions Inventory 1-8
Step 2a: Estimate the Energy Savings of Energy Efficiency and Combined Heat and Power Policies and Programs I-
10
Energy Efficiency 1-10
Combined Heat and Power 1-11
Sources of Information for Energy Efficiency and Combined Heat and Power Policies and Programs 1-12
Step 2b: Estimate the Energy Impacts of Demand Response Policies and Programs 1-13
Step 2c: Estimate the Renewable Energy Generated from Policies 1-14
Important Considerations 1-15
Step 3: Understand How EE/RE Policies and Programs Impact Emissions in a Nonattainment Area 1-15
Determining the Geographic Boundary of Emissions Analysis 1-16
Steps to Analyze Energy Efficiency/Renewable Energy Policy and Program Impacts 1-16
Step 4: Choose an Approach to Quantify Avoided or Displaced Electric Generating Unit Emissions 1-17
Four Emission Quantification Approaches 1-18
Suggested Quantification Approaches for Each State and Tribal Implementation Plan Pathway 1-31
Important Considerations for Emission Quantification Approaches 1-31
Managing Uncertainty 1-32
References 1-34
1-3
-------
FIGURES
Figure 1: Steps for Quantifying Emissions of Energy Efficiency and Renewable Energy Policies and Programs 1-8
Figure 2: Resources for Emissions Inventory Development 1-9
Figure 3: Resource for Emissions Projections 1-10
Figure 4: Emissions Quantification Approaches 1-18
Figure 5: eGRID2010Subregion Representational Map 1-20
Figure 6: Emissions Quantification Using an eGRID Approach 1-21
Figure 7: Sample Curve for Relating Displacement to Capacity Factors 1-22
Figure 8: Steps for Historical Hourly Emissions Rate Approach 1-26
1-4
-------
TABLES
Table 1: How to Choose an Emissions Quantification Approach 1-19
Table 2: Allocating Displaced Energy Using the Capacity Factor Approach 1-24
Table 3: Suggested Emissions Quantification Approaches for Each Pathway 1-31
1-5
-------
SECTION I.I: PURPOSE
The purpose of this appendix is to help jurisdictions determine the best emissions
quantification approach when accounting for energy efficiency/renewable energy (EE/RE)
policies and programs in state implementation plans/tribal implementation plans (SIPs/TIPs).
This appendix starts with an overview of the important factors to consider prior to quantifying
emission of EE/RE policies and programs. Then, the emission quantification process is
described with links to resources, tools and key information to help state, tribal and local
agencies select an appropriate emissions quantification approach. After reviewing this
appendix, jurisdictions should understand the methods and options for quantifying EE/RE
policies and programs in their SIP/TIP.
SECTION 1.2: GETTING STARTED: IMPORTANT FACTORS
Important factors to consider before quantifying emissions of EE/RE policies and programs
include:
Availability of staff, budgetary resources, and energy data;
Energy data availability and air agency's relationship with state, tribal and local energy
experts;
Potential magnitude of emission reductions for the electric generating unit (ECU) sector
associated with EE/RE policies and programs; and
Whether to engage in multi-jurisdictional collaboration.
Understanding these factors upfront will aid the decision making process for how to account for
emission reductions in a SIP/TIP.
Availability of Staff and Budgetary Resources
Understanding the range and capability of the emission quantification approaches will help a
jurisdiction judge the level of resources needed to address analytical questions on the
emissions impact of EE/RE policies and programs. It is possible to estimate the emission
impacts of EE/RE policies and programs with low staff resources. Basic approaches are easy to
employ and are often useful in justifying more effort if there is a significant emission reduction
potential. Methods that are more sophisticated could mean increased staff time, air-energy
trainings and monetary resources devoted to paying for fees associated with proprietary energy
models. Reviewing the existing tools and resources that complement each emission
quantification approach described in this appendix is a good first step to determining the level
of effort needed to quantify the emission impacts of EE/RE policies and programs
Energy Data Availability and Relationship between Air and Energy Agencies
A significant piece of information necessary for quantifying the emissions impact of EE/RE
policies and programs is data on energy impacts from state energy offices, utility commissions,
or other data sources. The strength of relationship between a state's air agency and its energy
agencies may influence whether the appropriate type of data is available for a SIP/TIP emissions
analysis. For example, some approaches will require hourly energy impacts information, while
1-6
-------
others only need annual or season energy impact information. It is possible that, even in cases
where relationships between air and energy agencies are strong, the types of data needed for
specific emissions analysis may not be available at the level of detail compatible with the
desired emissions quantification approach. In these cases, air agencies may need to rearrange
the data to use it with a specific approach or choose the emissions quantification approach that
is compatible with the available energy impacts information. Reviewing this appendix will help
determine what energy impacts information will be needed depending upon the state, tribal or
local air agency's policy objectives and analytical questions.
Potential Magnitude of Emission Reductions
One of the most important objectives in understanding the potential magnitude of emission
reductions of EE/RE policies and programs is comparing the potential emission reductions for a
particular pollutant resulting from an EE/RE policy or program to existing emissions for that
pollutant:
1) From the ECU sector in the state,
2) From all sources in that state, and
3) In a nonattainment area for which the SIP/TIP is intended.
This comparative analysis will help illuminate how large an impact the EE/RE policies and
program may potentially have in an area relative to other contributors. Generally, state, tribal
and local air agencies inventory emissions on an annual basis to review the emission reduction
potential of different control strategies. However, EE/RE policies affect EGUs that operate
marginally at shorter time scales. Therefore, a typical annual emissions profile of the ECU
sector will not be completely indicative of the emission reduction potential of EE/RE policies
and programs within a jurisdiction. Some emissions quantification approaches provided in this
appendix, such as the capacity factor and hourly emissions approaches, can be used at the
same time as the emissions inventory development process to help determine the emission
reduction potential of EE/RE policies and programs over shorter time periods without
expending large resources.
Multi-Juris dictional Collaboration
Since power plants are interconnected through the larger electric power system, to fully
understand the emissions impact of an EE/RE policy, it is often important to consider
collaborating with neighboring states to develop joint ECU analyses. Understanding the goals
and intentions of neighboring state, tribal and local air agencies could be an important factor in
deciding the most appropriate emissions quantification approach. For example, a highly
resource intensive approach for one state may make sense if multiple states are contributing
resources. Including collective EE/RE policy goals may also increase the emissions reduction
potential within the region. In many cases, a collective quantification effort would enhance
efficiencies in costs and would better capture cross state benefits in the air quality planning
process.
1-7
-------
SECTION 1.3: STEPS FOR QUANTIFYING EMISSIONS OF ENERGY
EFFICIENCY AND RENEWABLE ENERGY POLICIES AND PROGRAMS
There are four general steps state, tribal and local agencies should consider when quantifying
emission reductions of new EE/RE policies and programs (see Figure 1). (If all of the EE/RE
policies and programs sought for inclusion in the SIP/TIP process are included in baseline
emission projections, then steps 2,3 and 4 are not needed.)
Figure 1: Steps for Quantifying Emissions of Energy Efficiency and Renewable Energy Policies and
Programs
Steo 2a:
Estimate energy
savings of energy
efficiency policies and
programs
Step 1:
Develop a future
projected baseline
emissions inventory
Steo 2b:
Estimate energy
impacts of demand
response programs
Estimate generation
from renewable
energy policies and
programs
Understand how
EE/RE policies and
programs impact
emissions in
nonattainment areas
Quantify displaced
ECU emissions
resulting from EE/RE
policies and programs
Step 1: Develop a Future Projected Baseline Emissions Inventory
As a part of the SIP/TIP process, a jurisdiction must prepare an emissions inventory that
documents the emissions in a selected base year for the EGUs within the area of analysis (e.g.,
state-level or multi-state level emission inventories). The information necessary to develop an
1-8
-------
emission inventory for the ECU sector includes electricity generation data, and emission factors
to convert estimates of energy use into emissions or measured reports from continuous
emissions monitoring (CEMS) equipment. For both SIPs/TIPs and developing a future projected
baseline emission inventory, a bottom-up emissions inventory is the most appropriate and
involves collecting activity data, emissions profiles and pollution control device information for
each ECU in the state, tribal area, locality, or region. Developing a future projected baseline
emissions inventory provides a comprehensive assessment of emissions and details regarding
spatial and temporal attributes that are required for air quality modeling. (See Figure 2 for
resources on emissions inventory development.)
Emissions relevant to SIPs/TIPs include all six criteria air pollutants, including precursors to
ozone (nitrogen oxides (NOX) and volatile organic compounds (VOC)), as well as carbon
monoxide (CO), sulfur dioxide (S02), lead (Pb), and coarse and fine particulate matter (PM).
Figure 2: Resources for Emissions Inventory Development
Emissions Collection and Monitoring Plan
System (ECMPS)
The EPA system for reporting of
emissions data, monitoring plans, and
certification data.
It replaces the Emission Tracking System
that previously served as a repository of
S02, MCL, and cabon dioxide (C02)
emissions data from the utility industry.
The EPA collects data annually in five-
minute, hourly, daily, monthly and
annual intervals from continuous
emissions monitors at all large U.S.
power plants.
Available at:
http://camddataandmaps.epa.gov/gdm/
Emissions & Generation Resource
Integrated Database (eGRID)
This free, publicly available software
from EPA has data on annual S02, NO..,
and C02, and mercury (Hg) emissions for
most power plants in the United States.
It also provides annual average non-
base load emission rates, which may
better characterize the emissions of
load-following resources.
Available at: http://www.epa.gov/egrid
A baseline emissions projection forecasts what emissions will occur in the future, in the
absence of additional policies. This baseline projection is a reference case that includes "on the
books" EE/RE policies or programs, against which the impacts of new policies and programs are
measured. When state, tribal or local agencies take into account certain EE/RE policies and
programs in developing a baseline emissions projection, then such activities are considered
already included in the SIP/TIP, as part of the SIP/TIP baseline pathway, and additional emission
reductions cannot be granted through another SIP/TIP pathway. Only those policies and
programs not included in the future baseline may be included in the subsequent steps. In other
words, if a state, tribal or local agency incorporates the EE/RE policies and programs in the SIP
baseline emissions projections, then any emission reductions from those EE/RE policies and
1-9
-------
programs are already counted and cannot be "double counted" in other parts of the SIP/TIP.
(Refer to Figure 3 for resources for emissions projections and Appendix E for more information
on the SIP/TIP baseline emission projections pathway.)
State, tribal and local agencies can project future emissions based on historic trends and
expectations about the impact of numerous factors, including projections of population growth
and migration, economic growth, fuel availability and prices, and technological progress. Some
of these factors will rely on
data about electricity
imports and exports, power
plant construction or
Figure 3: Resource for Emissions Projections
retirement, transmission
constraints and
environmental regulations.
EPA Emissions Inventory Improvement Program Technical Report
Series, Volume X: Emissions Projections.
This is guidance that provides information and procedures to
state and local agencies for projecting future air pollution
emissions for the point, area, and mobile source sectors.
It describes data sources and tools states can use for emissions
projections.
Available at:
http://www.epa.gov/ttn/chief/eiip/techreport/volumelO/x01.pdf
The underlying
assumptions for the
baseline emissions
projection include an
electricity demand forecast.
The forecast documents the historical, current, and projected pattern of energy demand in the
state or region, as well as how future generation projections will meet forecasted demand.
Step 2a: Estimate the Energy Savings of Energy Efficiency and Combined Heat
and Power Policies and Programs
Energy Efficiency
The purpose of this step is to determine the energy savings from the specific EE policy/program,
which can then be used to estimate any associated emission reductions from the power sector.1
Energy savings refer to the expected reduction in energy demand that are the result of a
specific EE policy and/or program. For SIP/TIP purposes, energy savings can be a reduction in
the base year energy demand, and/or a reduction in the projection of future energy demand.
The impacts of EE policies or programs are typically measured in kilowatt-hours (kwh) or
gigawatt-hours (GWhs) and are generally reported on an annual basis, but some EE programs
may be available in smaller time intervals, such as a seasonal or hourly scale.
A general approach for determining the amount of energy saved for EE policies and programs,
although each EE policy and/or program will have individual factors to be considered. The
general approach is:
Estimated savings from EE policies and programs can be calculated in many different ways. State energy
regulators typically have methods for estimating energy savings in your jurisdiction. Methods may vary in
precision and rigor. Energy regulators can provide the estimates of the energy savings in your jurisdiction from the
EE policies and programs they oversee.
1-10
-------
Step A: Determine the baseline energy usage subject to the EE policy and/or program.2
Step B: Determine the projected energy use after implementation of the EE policy
and/or program.
Step C: Subtract the result of Step A from the result from Step B. The result yields the
projected energy savings due to the EE policy and/or program.
For later verification purposes, state, tribal and local agencies may need to collect and compare
the evaluated (e.g., ex-post)3 estimate of energy savings to initial forecast impacts. This added
step can result in more accurate estimates of energy savings and associated emission
reductions. The appropriate level of energy savings verification depends primarily on the SIP
pathway selected. Other factors to consider are the magnitude of projected energy impact and
the nature and type of efficiency program under consideration (for more information, consult
the appendices on each SIP pathway).
Combined Heat and Power
Combined heat and power (CHP) is an efficient method of providing energy services (electricity
and thermal energy, including cooling) to the end user. Instead of purchasing electricity from
the grid and simultaneously burning fuel in an on-site boiler or furnace to produce needed
thermal energy, an industrial or commercial user can use a CHP plant on site to provide both
services in one energy efficient step. CHP's inherent efficiency and the elimination of demand
for electricity from the power system with associated transmission and distribution losses can
provide significant energy savings and lower emissions that can be accounted for in a SIP/TIP.
Calculating the net energy savings from CHP requires estimating the expected reduction in the
amount of electricity generated by an existing utility system as a result of the specific CHP
policy and/or program and subtracting from that the incremental fuel used by CHP systems
over and above the existing boilers and/or furnaces that were replaced. The resultant savings
can reduce current energy demand, future demand, or both. The purpose of this step is to
determine the energy saving impacts of a specific CHP policy/program.
For determining the amount of energy saved for CHP policies and programs, although each
policy and/or program will have individual factors to be taken into account, the general
approach is as follows:
Step A: Determine the baseline forecast of energy use for the activity subject to the CHP
policy and/or program.
Step B: Determine the projected energy use after implementation of the CHP policy
and/or program. The projected energy use must reflect:
o The reduction in energy use at the utility power plant due to the reduction in
demand caused by the generation of power on-site.
2EPA(2010c), Chapter 2.
3 Evaluations of ratepayer funded energy-efficiency programs - sometimes referred to as evaluation,
measurement, and verification (EM&V) - provide retrospective estimates of energy savings. They are typically
conducted on an annual basis by state Public Utility Commissions (PUCs). It is important to note that energy-
efficiency forecast data - not ex-post EM&V data - are used to calculate avoided-emissions for SIP purposes.
1-11
-------
o The incremental increase in fuel use at the industrial or commercial facilities that
the CHP system uses over and above the boilers and/or furnaces that CHP
replaced.
Step C: Subtract the result of Step A from the result from Step B. The result yields the
projected energy savings due to the CHP policy and/or program.
For later verification purposes, state, tribal and local agencies may need to collect and compare
the evaluated (e.g., ex-post) estimate of energy and fuel savings to initial forecast impacts. This
added step can result in more accurate estimates of energy and fuel savings as well as
associated emission reductions. The appropriate level of verification depends primarily on the
SIP pathway selected. Other factors to consider are the magnitude of projected energy impact
and the nature and type of CHP program under consideration. (See the SIP Pathway
appendices for more information.)
Operational and performance data for CHP technologies are also important factors in this
analysis. This information is available from the EPA Combined Heat and Power Partnership,4
National Renewable Energy Laboratory (NREL), the Department of Energy's (DOE) eight
Regional Clean Energy Application Centers5 and universities and other energy associations that
promote or conduct research on the applications of CHP.
Sources of Information for Energy Efficiency and Combined Heat and Power Policies
and Programs
Energy experts within a jurisdiction can help state, tribal and local air agencies understand
current and future policy opportunities, as well as obtain estimates of the kilowatt-hour (KWh)
impacts (projected and historical)
from the EE/CHP policy/program of
interest. The EPA recommends
starting with the public utility
commission staff and energy offices.
These agencies are typically
responsible for the evaluation,
measurement and verification
Sources of Information on Energy
Impacts of EE/CHP Policies
State Public Utility Commissions
State or Local Energy Offices
Energy Information Administration
Electric Grid Operators
Regional Transmission Organizations
Independent Service Organizations
Utilities
(EM&V) of EE programs, and the
tracking of RE generation. If
jurisdictions need further
information, the Energy Information
Administration (EIA) and electric grid operators can also be good sources. For example, a large
utility that controls the dispatch of resources within the area of analysis could be an electric
grid operator and could provide electricity demand and supply forecasts. If a state, tribal or
local jurisdiction is located within a regional transmission organization (RTO) or an independent
system operator (ISO), these entities can also be helpful resources. These organizations can
help develop the energy impacts information or serve as a resource for assistance in developing
For more information, go to: http://www.epa.gov/chp/.
1 For more information, go to: http://wwwl.eere.energy.gov/industrv/distributedenergy/racs.html.
1-12
-------
the energy savings or generation estimates for particular EE/CHP policies or programs. Refer to
Appendix B, for more information on the roles and responsibilities of these organizations.
When gathering information from energy organizations in a state or region, air agencies will
want to ask specific questions about the policy or program design characteristics. .6 The
questions below are intended to help air agencies ask questions up front to avoid confusion and
ensure the energy savings information is provided in the proper SIP/TIP context.
Program period: What year does the policy/program producing the energy savings start
and end?
Anticipated compliance or penetration rate: How many utilities will achieve the target
or standard called for? How many consumers will invest in new EE/CHP equipment
based on the initiative? How will this rate change over the time period?
Compliance assurances: Are there compliance penalties, monetary incentives or other
assurances included in the policy or program design?
Annual degradation factor: How quickly will the performance of the EE measure
installed become less efficient?
Transmission and distribution loss (T&D): Is there an increase or decrease in T&D losses
that would require adjustment of the energy savings estimate?
Post Performance: What is the success rate of existing or past EE/RE policies and
programs?
Evaluation, Measurement and Verification (EM&V): Which programs have EM&V
procedures in place, and how will energy savings be measured and verified over time?
Step 2b: Estimate the Energy Impacts of Demand Response Policies and
Programs
"Demand response" is a broad term encompassing a range of programs designed to reduce
electricity demand during peak periods of electricity use. Demand response initiatives range
from programs that provide customer incentives for voluntary or mandatory load curtailment
based on contractual arrangements, to dynamic pricing structures that charge higher rates for
energy consumed during peak periods.
Demand response policies and programs can effectively reduce emissions during "high electric
demand days."7 For example, hot summer days are conducive to ground-level ozone
formation, and air conditioning loads on such days are often major contributors to electricity
demand spikes. At the same time, some EGUs called "peaking units" only operate during
periods of peak demand when the electric grid requires maximum generating capacity, and
could be high-emitting sources. Peaking units may lack NOXcontrols because they have low
6EPA(2010c), p. 42.
7 On the hot, hazy days of summer, higher demand for electricity can result in a dramatic increase in ozone-
forming air pollution. These are called "high electric demand days" or HEDD.
1-13
-------
emissions on a seasonal basis, even if hourly emissions are high during periods when they are in
use.
Since demand response programs target peak demand hours, air agencies will want to collect
hourly energy savings data, during peak hours, to translate the energy savings from these
programs into emission reductions. Demand response programs may not always provide
emission reduction benefits. For instance, some demand response programs relieve demand
from the electric grid by replacing that generation with back-up generators that emit emissions.
(e.g., diesel generators) To properly account for the emission impact of demand response
programs in a SIP/TIP, emission calculations should account for the net emissions (e.g.,
emission reductions from the electricity grid and any emission increases from back-up
generation).
To avoid a net emissions increase that can result from the use of emissions-intensive sources of
backup power generation, state, tribal and local agencies should combine demand response
efforts that focus on passive approaches, such as cycling down equipment to curtail electricity
demand, with efforts to promote clean backup power that can be an effective strategy for
achieving this objective. Some program administrators have addressed this issue by including
requirements for the types of load reductions that are eligible for demand response incentives.
Step 2c: Estimate the Renewable Energy Generated from Policies
This step determines how much energy would be displaced by the RE policy and/or program,
including less polluting sources of new energy, such as cogeneration and fuel cells. In general,
for renewable sources, state, tribal and local agencies will need data showing the total amount
of energy provided to the grid by the RE source. Keep in mind, for later verification purposes,
data on the amount of renewable energy generated may need to be collected and compared to
original estimates.
Operational and performance data for renewable technologies are also important factors in this
analysis. This information is available from National Renewable Energy Lab (NREL), as well as
universities and other RE associations that promote or conduct research on the applications of
RE. In addition, generation-related data and RE potential information can be obtained from
many sources, including:
State energy offices
Utility Integrated Resource Planning filings
Public utility commissions
ISOs
North American Electric Reliability Corporation (NERC)
EPA's Emissions & Generation Resource Integrated Database (eGRID) 8
DOE's EIA
DOE's NREL
' For more information, go to: www.epa.gov/egrid.
1-14
-------
Important Considerations
The jurisdiction's analysis of RE policy and program impacts needs to account for how various
RE resources generate energy at different times of the day and year. For example, wind power
facilities may be unable to provide energy during high ozone episodes due to stagnation
conditions, as compared to solar power facilities. Land-based wind energy facilities have to be
sited in locations where winds speeds are sufficiently strong to produce power on as many days
as possible, which can be in locations outside of the urban corridor where air quality tends to
be the poorest. Nevertheless, these remote RE facilities may affect emissions levels at fossil-
fired EGUs that are positioned upwind of jurisdictions with poor air quality. Any air-quality
analysis of RE policy impacts should account for these emissions impacts. Off-shore wind
generation - which may be closer to coastal load centers and areas of poor air quality - should
likewise be analyzed to determine how generation and emissions from surrounding EGUs are
affected, including the magnitude of such impacts.
Step 3: Understand How EE/RE Policies and Programs Impact Emissions in a
Nonattainment Area
Air quality planners need to determine the displaced emissions attributable to each applicable
ECU in order to understand how those emission reductions will improve the air quality in the
nonattainment area.9 Emission reductions cannot be assigned to the nonattainment area
based on where the EE/RE policy is implemented. Rather, a jurisdiction should assign emission
reductions based on where the displacement of electrical generation will likely occur. All
emission quantification approaches described below, except for the eGRID non-baseload
emissions rate approach, can allocate emission reductions to an individual ECU or power plant.
To determine if an EE/RE policy affects a nonattainment area, air agencies should (1) estimate
which EGUs and EE/RE policy or program will likely affect generation and (2) use the emission
rates at those EGUs.
For example, if the nonattainment area imports a significant amount of electricity from
locations outside and downwind of the area, reduced demand from EE could result in less
electricity being imported, rather than reduced production (and consequently reduced
emissions) within the nonattainment area, or in areas affecting its air quality. Conversely, if the
energy savings reduce emissions at upwind sources, then the measure may produce some air
quality benefits to the area. (For more details, see the section below on determining the
geographical area where emissions occur.)
After state, tribal and local agencies quantify the emission reductions of EE/RE
policies/program, as described in step four of this appendix, the state should use the
appropriate air quality model to evaluate the extent to which reductions will improve air quality
9 The current policy with respect to taking credit for emission reductions outside nonattainment areas for purposes
of Reasonable Further Progress in ozone SIPs is as follows: RFP credit can be taken for VOC and NOX emission
reductions within 100 kilometers (km) and 200 km, respectively, outside the nonattainment area under certain
circumstances. This policy is currently under reconsideration. See "Reasonable Further Progress Requirements for
the 1997 8-Hour Ozone National Ambient Air Quality," 75 Federal Register 80420-80425, 80421,
http://www.gpo.gov/fdsys/pkg/FR-2010-12-22/pdf/2010-32139.pdf,.
1-15
-------
in the nonattainment area from the selected EE/RE policy and/or program. The EPA has
separate modeling guidance documents to help state, tribal and local agencies choose the best
air quality model for SIPs/TIPs.10
Determining the Geographic Boundary of Emissions Analysis
Determining the appropriate geographical boundaries for an emissions quantification analysis
requires an understanding of the electric power grid relative to the nonattainment area and
EE/RE policy of interest. (Refer to Appendix B for more information on the electric grid.) A
state, tribal or local agency will need to choose the most appropriate boundary for their
analysis based on key factors:
Whether adjacent states are analyzing EE/RE policies within the same NERC region11
The location of the area in which the EE and/or RE policy is being implemented
Whether there are any transmission constraints where the EE/RE policy is being
implemented
How energy is transferred within the state and across state boundaries
Determining the location of the emission reductions that occur at fossil fuel fired generation
units is challenging because electricity from numerous generators is fed into an electrical grid
from which many different consumers at various locations will draw power. There typically is
no direct connection between a specific facility generating electricity and the end user of that
electricity. Understanding how the electric grid operates within a jurisdiction is the first
important step in making an approximation as to which units would be affected by a certain
EE/RE policy and program.
In areas of the country where several states in close proximity to one another implement EE
and/or RE policies and programs, it may be advantageous for these states to work together in
conjunction with their electric grid operation, ISO/RTO and EPA regional office to identify the
overall impact of the EE/RE policy and programs on the electrical grid in the future. Ideally,
such a process will yield a technically valid solution that attributes the emission reductions from
decreased reliance on fossil fuel fired EGUs in an equitable manner between the states, and
also ensures that double counting of emission reductions does not occur.
The EPA understands that conducting this type of analysis may be beyond the means of the
jurisdictions that implement these EE/RE policies and programs. Accordingly, EPA encourages
any state, tribal and local agencies that need assistance with this to contact the relevant EPA
regional office for assistance.
Steps to Analyze Energy Efficiency/Renewable Energy Policy and Program Impacts
Determining the location of the fossil fuel fired units that can operate less as EE or RE becomes
available can be a complex task, particularly when the efficiency or renewable resources are
located outside the nonattainment area that seeks to use the emissions reduction from that
10 For more information, go to: http://www.epa.gov/ttn/scram/aqmindex.htm.
11 For more information, go to: http://www.eia.gov/cneaf/electricity/page/prim2/figure7.html.
1-16
-------
reduced generation for SIP/TIP purposes. The following three steps can be used to analyze
EE/RE policies and programs:
Step 1: Review historical analysis to determine the location of the fossil fuel fired EGUs
that could reduce their output as EE or RE resources were made available. This
information should already exist at the balancing authority or reliability organization
such as an, ISO or RTO that oversees the electrical grid for the area or the utility that is
complying with the EE or RE policy.
Step 2: Understand how the grid has responded in the past as efficiency or renewable
resources have come on-line in order to develop planning assumptions for how the grid
will respond in the future.
Step 3: Obtain and review the results from existing dispatch modeling conducted by the
grid operator of the balancing authority, ISO or RTO. The grid operators have the most
pressing need to accurately determine the impact that EE or RE resources will have on
the future operation of the electrical network.
The next section will describe four different emission quantification approaches that state,
tribal and local agencies can consider.
Step 4: Choose an Approach to Quantify Avoided or Displaced Electric
Generating Unit Emissions
State, tribal and local agencies will need to choose the most appropriate emission
quantification approach based on the agency's policy objectives. Each approach answers
different analytical questions, has varying levels of rigor, assumptions, resource requirements,
data needs, and varied temporal and spatial scales of emission outputs. They all are intended
to quantify the avoided or displaced emissions from fossil fuel generation as a result of EE/RE
policy/program implementation.12
The level of sophistication of each emission quantification approach is inversely related to the
degree of uncertainty inherent in each approach. Basic approaches apply simple assumptions
and tend to have a higher degree of imprecision compared to more sophisticated approaches.
Figure 4 shows how the level of sophistication of those approaches increases as you go from
straightforward emissions calculations to complex modeling.
Basic approaches, such as eGRID emission rates, are generally more appropriate for preliminary
screening analyses or to be used for analyzing the incorporating of emission reductions in
specific SIP/TIP pathways, such as the weight of evidence (WOE) or emerging/voluntary
measures SIP pathway (Refer to Appendix H for the WOE pathway and Appendix G for the
emerging/voluntary measures pathway). More sophisticated approaches, such as energy
modeling can simulate how the system of EGUs will change generation and emissions given
specific assumed conditions and capture multiple assumptions for multiple factors such as:
12 Lifecycle emissions are not addressed in this appendix and do not apply to quantifying emissions for SIP/TIP
purposes.
1-17
-------
electricity imports, exports, fuel prices, unit availability, and technology performance. (Refer to
Appendix B for more information on how the electric grid operates) Therefore, more
sophisticated approaches are most appropriate for quantifying emissions for the baseline
emissions projection and control strategy pathways. (Refer to Appendix E for information on
the baseline emissions projection pathway and Appendix F for information on the control
strategy pathway.)
Figure 4: Emissions Quantification Approaches
Assumptions are Simpler
eGRID
Subregion
on-Base Load
Emission Rates
Approach
Historical
Hourly
Emission Rates
Approach
Energy
Modeling
Approach
Capacity Factor
Emission Rates
Approach
Four Emission Quantification Approaches
This section explains four emission quantification approaches, including how each approach
works, when to use it, the advantages and limitations, available tools to support each, and an
13
emission quantification example . The approaches are:
Basic approach: eGRID sub region "non-base load" emission rates
Basic approach: capacity factor emission rates
Midrange approach: historical hourly emission rates
Sophisticated approach: energy models
Agencies will have different policy objectives and associated analytical questions associated
with incorporating EE/RE policies and programs in their SIPs/TIPs. Some of the policy objectives
and analytical questions state, tribal and local agencies may want to address can be satisfied by
one or more of the quantification approaches. Table 1 provides analytical questions commonly
asked during the SIP/TIP process and the corresponding emissions quantification approach.
13 Emission quantification examples are currently under development. For more information on specific SIP/TIP
examples, go to Appendix K.
1-18
-------
Table 1: How to Choose an Emissions Quantification Approach
Analytical Questions
Emission
Quantification
Approaches
Energy
Data Needs
Emissions
Outputs
Pathways
What is the relative
magnitude of SO2 or NOx
emission reductions of an
EE/RE policy or program
in my jurisdiction?
Which EGUs in my state
are on the margin and
how much will emissions
be displaced on a
seasonal or annual basis?
How can 1 quantify hourly
emission reductions?
How much are emissions
reduced during peak
electricity demand?
How will EGU emissions
change in future years?
How can 1 compare
baseline emissions
forecasts and emissions
of new EE/RE policies?
How can 1 simulate
emission changes of
EE/RE policies when also
subject to cap and trade
program(s)?
eGRID region
non-baseload
emission rates
approach
Capacity factor
emission rates
approach
Historical hourly
emission rates
approach
Energy models
approach
Annual or season
energy impacts
(megawatt hours
(MWh))
Annual or seasonal
energy impacts
(MWh)
Hourly energy
impacts, peak
and/or base load
effects (MW or
MWh)
Hourly, seasonal or
annual energy
impacts depending
upon model
capabilities (MW or
MWh)
eGRID regional
average non-
base load
emissions
EGU emissions
of marginal
units on an
ozone season
or annual basis
Quantify EGU
hourly
emission rates
and hourly
emission
reductions
Average
emissions
based on
dispatch
order, hourly
emissions or
seasonal
emissions
WOE
WOE,
Emerging/
Voluntary
Measures, or
Control
Strategy
Pathways
Baseline, WOE,
or Control
Strategy
Pathways
WOE,
Emerging/
Voluntary
Measures, or
Control
Strategy
Pathways
Basic Approach: eGRID Sub region "Non-Base Load" Emission Rates
This approach entails a simple calculation. An agency would multiply the amount of generation
or electricity consumption displaced by the EE/RE policy/program by the "non-base load"
emission rate indicated for a specific pollutant in an eGRID subregion. The non-base load
emission rate for an eGRID subregion represents an average emission rate for the EGUs that are
likely to be displaced by an EE/RE policy and program.
What is eGRID?
The EPA's eGRID is a data source that provides information on the environmental
characteristics of almost all electric power generated in the U.S. The eGRID includes
1-19
-------
operational data such as total annual emissions, emission rates, generation, resource mix,
capacity factors, and heat input.
14
Figure 5: eGRIDZOlO Subregion Representational Map
The eGRID includes emissions of GHGs , NOX, and S02, and Hg. The eGRID emissions data are
associated with the generation of electricity, not with the consumption of electricity; therefore,
the values do not account
for transmission and
distribution losses,
imports and
exports across
eGRID subregions
(or any other
geographic area),
transmission
constraints within
any geographic
area, or life cycle
emissions at EGUs
(e.g., emissions
from the
extraction,
processing, and
transportation of
fuels).
Source:
http://www.epa.gov/cleanenergy/documents/egridzips/eGRID2010Vl 1 vearO? Summa
rvTables.pdf>
The eGRID
subregions are
identified and
defined by EPA, using the NERC regions and power control areas as a guide. An eGRID
subregion is often, but not always, equivalent to an Integrated Planning Model (IPM)15 sub
region. The 26 eGRID subregions in eGRID2010 are subsets of the NERC regions as shown in
Figure 5.
When to use the eGRID approach
The EPA recommends that state, tribal and local agencies use this approach as a method for
estimating potential emission reductions for the WOE SIP pathway. Alternatively, the approach
can be used as a preliminary, screening analysis before carrying out a more sophisticated
analysis that could be used under another SIP/TIP pathway submission or to justify a more
sophisticated analysis under the WOE pathway.
14 For more information, go to: http://www.epa.gov/cleanenergy/energy-resources/egrid/index.html.
15 The IPM is a multi-regional, dynamic, deterministic linear programming model of the U.S. electric power sector.
It provides forecasts of least cost capacity expansion, electricity dispatch, and emission control strategies while
meeting energy demand and environmental, transmission, dispatch, and reliability constraints.
1-20
-------
Jurisdictions can estimate the emissions reduction potential for S02 emissions, ozone season or
annual NOX emissions in their region by applying eGRID subregion emission factors to the EE/RE
policy energy impacts. This method requires minimal resources, simple assumptions, and
provides a rough estimate of average emissions at a regional scale. This approach is most
appropriate when EGUs within the eGRID subregion contribute to air quality in the
nonattainment area, either by being in or upwind to the nonattainment area.)
How does this approach work?
State, tribal and local agencies can use the eGRID subregion non-base load output emission
rates to estimate emission reductions of EE/RE policies and programs that reduce consumption
of grid-supplied electricity. In particular, eGRID will help planners figure out the non-baseload
output emission rates. This type of rate is useful to understand because it is associated with
emissions from the EGUs most likely to be displaced when EE/RE policies and programs are
implemented. These emissions data are derived from plant level data and are aggregated up to
the eGRID subregion level.
16
This basic eGRID method entails a simple calculation where a jurisdiction would multiply the
amount of generation or
electricity sales displaced by the
EE/RE policy/program by the
"non-base load" emission rate
indicated fora specific pollutant
in an eGRID subregion.17 The
non-base load emission rate for
an eGRID subregion represents
an average emission rate for the
EGUs that are likely to be
displaced by an EE/RE policy and
program. This method is
recommended to help state,
tribal and local agencies estimate
Figure 6: Emissions Quantification using an eGRID Approach
Non-base load emission rate (Ib/MWh)
(1/1-grid loss factor)
Reduced consumption or supply in energy of EE policy and
program (MWh)
(2,000lbs/l short ton conversion for criteria pollutants)
Tons of emissions reduced from EE/RE policy and program
the relative magnitude of
emission impacts from a
potential EE/RE policy or program by using the equation in Figure 6.
What are the advantages and limitations of the eGRID approach?
Advantages:
Easy "back of the envelope" calculation
Non-baseload output emission rates gives a basic understanding of how much ECU
emissions could likely be avoided or displaced
16 EPA (2010a).
17 Grid loss factors should be included in this calculation. EPA (2010c).
1-21
-------
eGRID data is derived from reported data submitted to EPA and the Energy Information
Administration
Limitations:
Future looking ECU capacity is not represented
o Some EGUs in base year may have already or will shut down in future years.
The eGRID approach does not show which EGUs could have its emissions displaced
Information is from 2007 and is generally on a three year time lag
The eGRID only accounts for generation within a specific area and does not include
information about imports/exports of electricity (except for state level net imports)
o Approach assumes EE/RE policies will affect all non-baseload plants
proportionally to each plant's non-base load generation.
What tools and resources are available for this approach?
The eGRID contains information in several formats, including spreadsheets, and summary
tables in Portable Document Format (PDF) format.
18
Figure 7: Sample Curve for Relating Displacement to
Capacity Factors
Basic Approach: Capacity Factor Emission Rates
This approach is based on the assumption that a power plant's capacity factor is an indicator for
the amount of generation subject to
displacement. This approach makes
general assumptions based on the
capacity factor of a power plant to
show, which power plants may
reduce generation with additional
EE/RE policies or programs. Capacity
factor information can be found n
the eGRID or a state, tribal or local
agency can use their own capacity
factors from other sources. This
approach does not account for the
complex nature of the electrical grid
such as electricity imports, exports,
and transmission constraints.
What is a capacity factor?
A power plant's capacity factor is a
ratio in a given period of time: "the
actual electricity produced by a
generating unit" to "the electricity
~ 8fl%
ra n?
* M 60% -
go.
^ =i
O o 9f)% -
£ o jara>
*
Q
\
\
\
\
\.
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Unit Capacity Factor
Source: http://www.svnapse-
energv.com/Downloads/SvnapseReport.2005-07.PQA-EPA.Displaced-
Emissions-Renewables-and-Efficiencv-EPA.04-55.pdf
that could have been produced at continuous full-power operation.'
19
' For more information on eGRID, go to: http://www.epa.gov/cleanenergy/energy-resources/egrid/index.html.
1 For more information, go to: http://www.eia.doe.gov/electricity/page/prim2/charts.html.
1-22
-------
The capacity factor of a power plant can be used as a proxy for how likely the power plant is to
be displaced by an EE/RE policy or program. Figure 7 shows an example of a displacement
curve, or a "rule of thumb," that relates the likelihood that a power plant or unit's output would
be displaced to its capacity factor. Baseload plants, such as nuclear units or large coal-fired
plants, have capacity factors close to 1, so they are represented on the lower right hand side of
Figure 7. The capacity factor approach assumes that these plants will displace zero percent of
their generation from the addition of EE/RE policies and programs. As a power plant's capacity
factor gets closer to zero, the likelihood that a plant will displace generation increases. For
example, the capacity factor approach assumes that peak load plants with low capacity factors,
such as combustion turbines are represented on the upper left hand side of Figure 7, and
therefore, are much more likely to be displaced.
When to use the capacity factor approach
State, tribal and local agencies could use a capacity factor approach to incorporate EE/RE
policies and programs in their SIP/TIP as a control measure, as long as future generation
characteristics, exports, and imports are included in the analysis. The approach could also be
used in an analysis of the benefits of EE/RE policies and programs in support of a WOE
demonstration or voluntary/emerging measure pathway. This approach is most appropriate to
answer the questions such as:
What is the relative dispatch order of the power plants within a state or region?
Which power plants are on the margin and what are the emission rates of those plants?
Estimating the emission impacts of energy efficiency policies or policies or solar policies or
programs that generally affect non-baseload power plants more so than baseload power plants
would be the appropriate type of EE/RE policy or program to apply for this approach (e.g., using
this approach for specific RE technologies, such as on-shore wind, that displace baseload EGUs
would not produce satisfactory results).
How does this approach work?
This approach is based on the assumption that a power plant's capacity factor is an indicator for
the amount of generation and emissions subject to displacement by an EE/RE policy/program.
This approach makes general assumptions about power plant generation and displacement
based on an power plant's historical annual or seasonal generation. By using a general rule of
thumb shown in Figure 7, the impacts of EE/RE policies and programs are allocated to each
power plant's or unit's annual or seasonal emission rate.
For example, the approach assumes that power plants with low capacity factors (operating at
equal to or less than 20 percent of capacity) are most likely to be displaced by the EE/RE
policy/program and power plants with high capacity factors (operating at equal to or greater
than 80 percent of maximum capacity) would not be displaced by the EE/RE policy/program.
The power plants with capacity factors between 20-80 percent are displaceable using a linear
relationship shown in Figure 7.
1-23
-------
Seasonal Variation Considerations
While the annual capacity factors described above are helpful in determining which power
plants are likely to displace generation by EE/RE programs/policies, EPA recommends using
seasonable capacity factors when possible for summer-time ozone because annual capacity
factors ignore seasonal weather variations. For example, many combustion turbines only
operate during summer daytime hours in a typical year. Using an annual capacity factor would
incorrectly allocate displaced emissions to these units during seasons when they are not
operating. If you are assessing the impact of EE/RE programs on ground-level ozone, the EPA's
eGRID has ozone season capacity factors available for this type of analysis.
Table 2 illustrates how an example could work in practice for an efficiency program projected to
save 1,000 MWhs pear year. There are seven generating units in this hypothetical power
system, labeled A through G in the first column. The second column shows the percentage of
each unit's production that could be displaced by the efficiency program, based on the rule of
thumb from Figure 7. The third column shows each unit's actual generation in the historical
year being used. The fourth column shows the amount of energy that could be displaced at
each unit, which is the second column multiplied by the first column. The fifth column shows
the percentage of the saved energy that is allocated to each unit. This is done by dividing the
displaceable energy for each unit by the total available displaced energy (e.g., Unit A's displaced
energy is 50,000 MWhs, which is 6.5 percent of the total 768,100 MWhs displaceable energy)
and the sixth column shows the MWhs displaced at each generating unit (column five
multiplied by 1,000 MWhs). The final step would be to multiply the MWhs displaced in Column
6 with the appropriate emission rates for each unit.
Table 2: Allocating Displaced Energy Using the Capacity Factor Approach
Percentage
Displaceable
(2)
Historical
Generation (MWh)
(3)
MWhs
Displaceable
Percentage of Energy
Saved Allocated to
MWhs Displaced
(6)
A
B
C
D
E
F
G
Totals
100%
82%
79%
48%
22%
0%
0%
50,000
65,000
120,000
500,000
1,500,000
1,800,000
2,000,000
6.035,000
50,000
53,000
94,800
240,000
330,000
0
0
768,100
6.5%
6.9%
12%
31%
43%
0%
0%
100%
65
69
123
312
430
0
0
1,000
Source: http://www.svnapse-energv.com/Downloads/SynapseReport.2005-07.PQA-EPA.Displaced-Ernissions-
Renewables-and-Efficiencv-EPA.04-55.pdf
What are the advantages and limitations of this approach?
Advantages:
Emissions can be assigned to each power plant or generating unit
Relatively easy calculation if infrastructure is set up
1-24
-------
Simple way to get a relative sense of the marginal unit in area of analysis
Limitations:
This is a simplified approach assuming power plants are operating the same throughout
the year or ozone season
Emissions estimates are approximate, based on annual or seasonal capacity factors
o Does not account for maintenance and outages
Imported and exported power is not considered
Assumes power plant generation characteristics are the same in the base year and
future year.
Assumes a general rule of thumb where all energy savings or generation affect all
peaking units first, which is not always true with some EE programs or RE technologies
(e.g., street lighting programs, wind power)
What tools and resources are available for this approach?
The EPA is providing a Capacity Factor Emissions Calculator (CFEC) for state, tribal and local
agencies to use for this approach. The CFEC uses eGRID annual and ozone season capacity
factors for the power plants represented in this database. Jurisdictions can follow the
instructions in this appendix and the CFEC when using this emission quantification approach.20
Midrange Approach: Historical Hourly Emission Rates
This approach requires technical manipulation of data about historical generation, including
load and emission rates to determine ECU dispatch order and marginal emissions rates. By
applying this approach, state, tribal and local agencies will understand where EGUs lie within
the dispatch order for each hour, day, and month of a historical year. Depending upon where
geographical boundaries are set, this approach may not account for electricity imports, exports,
and transmission constraints. And understanding these things will be useful in quantifying
emissions in a SIP/TIP because the electricity grid is a large interconnected system that crosses
nonattainment area and state boundaries. For example, if an ECU in a nonattainment area is
exporting it's generation outside of the nonattainment area, then EE/RE policies implemented
within the nonattainment area will have little effect on the EGU's emissions.
The EPA has information state, tribal and local agencies can use for this approach. The EPA
collects generation, emissions and heat input data in hourly intervals from CEMS for all large
EGUs subject to EPA's trading programs.21 State, tribal and local agencies can use reported
hourly generation and emissions data for each ECU to derive hourly emission rates for one hour
or group of hours. For example, 2 p.m. to 6 p.m. in the Electric Reliability Council of Texas
(ERCOT) NERC region22 is a typical group of hours that represents times of peak demand.
20 Capacity factors are located in the eGRID excel file called: eGRID2010Vl_l_year07_PLANT.xls. EPA (2010b).
21 For more information, go to:
http://camddataandmaps.epa.gov/gdm/index.cfm?fuseaction=emissions.prepackaged select
22 The Electric Reliability Council of Texas manages the flow of electricity in Texas. For more information, go to:
http://www.ercot.com/about/.
1-25
-------
Taking into account the emission rates from the EGUs that typically operate during these hours
would represent the potential emission reduction during peak hours within ERGOT.
When to use historical hourly emission rate approach
By applying this approach state, tribal and local agencies will understand (for every hour or
segment of hours of an historical year) which EGUs are base load (operating all hours of the
day), which EGUs are load following (EGUs that ramp up or down depending upon demand),
and which EGUs are peaking units (EGUs that only operate at high demand periods). This
approach is most appropriate to answer the questions such as:
How much emissions are reduced in blocks of hours, or during periods of peak electricity
demand?
How much emissions are reduced for demand response policies or programs?
Agencies can use an historical hourly emissions rate approach to incorporate EE/RE policies and
programs in their SIP/TIP as a control measure, as long as future generation characteristics,
exports, and imports are included in the analysis. The approach could also be used in an
analysis of the benefits of EE/RE policies and programs in support of a WOE demonstration or
baseline analysis.
How does this approach work?
The state, tribal or local agency first needs to evaluate how the EE/RE policy reduces load or
displaces generation in the area of analysis. Most importantly, the agency needs to identify if
the EE/RE policy impacts peak hours and/or base load energy use. From this, air quality
planners will be able to determine which EGUs within the dispatch order will be affected by the
policies.
It is possible for multiple EE/RE policies/programs to affect both the base load and peak hours
of a day. In that case, the agency should add the programs "bottom up" to obtain an aggregate
level of energy savings and generation on an hourly basis and then apply their impacts to the
Figure 8: Steps for Historical Hourly Emissions Rate Approach
Collect ECU
hourly
information
from EPA,
DOE or
other
Analyze
ECU
dispatch
order and
hourly
emissions
SteoS:
Identify
"hourly
segments
of ECU
emission
Apply
hourly
generation
or savings
profile of
EE/RE
Quantify
hourly
emission
changes of
new EE/RE
program
23
predicted displaced EGUs. Refer to Figure 8 for steps associated with this approach.
Synapse (2005).
1-26
-------
What are the advantages and limitations to the approach?
Advantages:
Reported data is easy to find on EPA's website on an hourly, daily, and quarterly basis
ECU emission rates from any group of hours can be derived from the hourly data
Emission impacts of different RE technologies can be compared against other RE
technologies (e.g., wind vs. solar) and EE policies and programs
Limitations:
Setting up hourly emissions database can be resource intensive if infrastructure is not in
place
Future ECU capacity changes are not represented.
Only EGUs subject to EPA's national reporting requirements are represented in EPA's
hourly ECU database24
Energy import and export exchanges and transmission constraints are not captured
What tools and resources are available?
The EPA has available hourly emissions and generation data.25 EGUs greater than 25
megawatts (MW) must report hourly emissions data to EPA on quarterly basis to comply with
EPA's National Acid Rain Program.26
The Mid-Atlantic Regional Air Management Association (MARAMA) completed an hourly
emissions analysis of the states in the Northeast and Mid-Atlantic region.27 The purpose of the
analyses was to assist states in the Northeast and mid-Atlantic regions with addressing policy
relevant questions concerning emissions during periods of peak electricity demand.
The Washington Council of Governments used a time-matched marginal emissions approach
that matches certain EE or RE technologies or measures with historical hourly emissions
information from the EPA hourly database. The emissions tool is applicable for the
Virginia/Maryland/ Washington, DC area.28
24 For more information, go to: http://www.epa.gov/airmarkets/.
25 For more information, go to: http://www.epa.gov/airmarkets/.
26 For more information on the hourly emissions and generation information, go to:
http://camddataandmaps.epa.gov/gdm/index.cfm ?fuseaction=iss.progressresults
27 For more information, go to:
http://www.marama.org/RegionalEmissionslnventorv/2007hourlypoint/FinalDoc mar2011 Analysis of Hrly CA
MD Emissions Data.pdf.
28 For more information, go to: http://www.mwcog.org/environment/air/EERE/default.asp.
1-27
-------
Sophisticated Approach: Energy Models
This approach employs dynamic simulation models that forecast which EGUs will be displaced
in the future based on inputs and assumptions in the model. Energy models can produce
emissions output information at various levels and account for the complex interactions of the
grid such as, transmission constraints, import/export dynamics, fuel prices, air pollution control
equipment and wide range of energy policies, and environmental regulations.
What is an energy model?
Energy models require extensive underlying data and complex formulation that represents the
engineering and economic decisions made by the energy system. They capture the complex
interactions within the electricity market and simulate what might happen given a set of
assumptions. These models are equipped to:
Simulate energy transfers among different regions
Optimize system dispatch from EGUs
Incorporate transmission constraints, forced outages, environmental regulations, plant
retirements, new generation, and limitations on specific power plants (e.g., ramp rates,
start-up constraints minimum down time)
Dispatch models
Dispatch models specifically replicate least-cost system dispatch, with the lowest cost resources
dispatched first and the highest cost last. Dispatch models determine which generating units
are displaced and when they are displaced based on economic and operating constraints.
Capacity expansion models
Capacity expansion models (also called system-planning models) can examine the potential
long-term impacts on the electric sector or upon the entire energy system of an EE/RE policy
and program. These models predict how the electric system will evolve over time; including
what capacity will be added through the construction of new generating units and what units
will be retired, in response to changes in new regulations, demand, and prices.
When to use an energy model approach
If a jurisdiction wants to incorporate EE/RE policies and programs in their SIP/TIP in the baseline
emissions projection pathway or as a control strategy, using an energy model, dispatch model,
capacity expansion model, or comparable model, is recommended. Either energy model or an
alternative emission projection tool can handle multiple iterations of different EE/RE
policy/program scenarios, therefore, state, tribal and local agencies can use a single emission
quantification method to compare emissions of a baseline emissions projections and varying
policy cases with new EE/RE policies.
Dispatch models are generally good to use for analysis of one to five years into the future
especially when the future ECU fleet is not changing substantially. An hourly dispatch model
can be used to determine hourly emission rates, which can then be aggregated by time period
1-28
-------
and applied to a range of EE/RE policies and programs according to their production
characteristics.
Capacity expansion models forecast future generation and retirements, as well as the dynamic
fluctuation within the electric grid and are generally useful to use for analysis five to 30 years
into the future. This approach is most appropriate to use when an EE/RE policy or program is at
a large enough scale to completely change electric system operations (e.g., EE/RE policy
transforms the electric sector market or future ECU operational characteristics)
How does this approach work?
Dispatch models
Energy modeling experts or consultants normally operate dispatch models. These models
determine which EGUs operate on the "margin" in the electrical power system (typically the
most expensive unit needed to meet demand is the "marginal ECU" in a given time period).
States can use hourly dispatch or energy models to determine hourly marginal emission rates
(pounds/kWh), which can then be aggregated by time period and applied to a portfolio of
programs used to achieve the EE/RE policy requirement.29
There are important considerations when using dynamic simulation models such as dispatch
models. Since this method can be less transparent than other methods, jurisdictions should
inform the EPA regional office of important input assumptions for any dispatch or energy model
used to measure displaced emissions.
The following information should accompany a state, tribal or local agency's SIP/TIP submittal
under this pathway for any quantification of emission reductions using a dispatch or similar
type of model:
Type and amount of energy savings/generation information used
o Specify if peak (MW), annual (MWh), seasonal, and/or hourly load information was
applied for EE/RE policy
Fuel prices assumed for all fuels and technologies
Emission rates for each applicable ECU
What are the advantages and limitations of dispatch models?
Advantages:
Electricity transfers are well represented
Can provide very detailed estimations about specific plant and plant type effects
Highly detailed, geographically specific hourly emissions data at the ECU level
Can simulate emission changes for jurisdictions subject to cap and trade programs
Limitations:
The model is only as good as the assumptions
29 EPA (2010c), pp. 69-70.
1-29
-------
All dispatch models are proprietary and require more resources than other approaches
Expertise in energy modeling is normally recommended
Future changes in electric system are not factored into model
Capacity expansion models
Energy modeling experts or consultants normally operate capacity expansion models. These
models are typically used for longer-term studies (e.g., five to 20 years), where the impacts are
dominated by long-term investment and retirement decisions. They are also typically used to
evaluate large geographic areas. This method involves allowing the model to predict what will
likely happen to the electric energy resource mix based on costs of new technology, growth,
existing fleet of generating assets, environmental regulations (current and planned), and
considering dispatch both with and without the new clean energy resource.30
The following documentation should accompany a state, tribal or local agency's SIP/TIP
submittal under this pathway for any quantification of emission reductions using a capacity
expansion model or similar type of model:
Fuel price forecasts, ECU retirements, and EE/RE regulatory requirements (e.g.,
renewable portfolio standards)
Plant type and emission rates of assumed new generation for all applicable future years
Information on whether model outputs were validated or calibrated against actual data
or another projection model
What are the advantages and limitations of capacity expansion models?
Advantages:
This is the most sophisticated way to capture how the electrical grid will react to EE/RE
policies
Future ECU generation and retirements are represented
Provides very detailed estimations about specific plant and plant type effects
Highly detailed geographically data at ECU or emission unit level
Can simulate emission changes for jurisdictions subject to cap and trade programs.
Limitations:
The outputs of energy models are only as good as the assumptions
Energy models are proprietary, require significant resources to run than other
approaches, and can be data intensive
Input assumptions for capacity expansion models can be difficult to discern due to the
proprietary nature of the models
Expertise in energy modeling is normally recommended
Hourly emissions information is not always available
30 EPA (2010c), p. 71-72.
1-30
-------
What tools and resources are available?
There are many different energy models available. Dispatch models include: PROSYM,
PROMOD, and Ventyx Market Analytics. Capacity expansion or planning models include:
NEMS, IPM, and ENERGY 2020.
Suggested Quantification Approaches for Each State and Tribal Implementation Plan
Pathway
The EPA is providing suggested emission quantification approaches state, tribal and local
agencies can use as guidelines when accounting for emission impacts of EE/RE policies and
programs within a certain SIP/TIP pathway (see Table 3). These approaches are suggestions
only, so a jurisdiction can choose to use an alternative approach, not listed here, that has
comparable rigor and emission results. Before getting too deep into any EE/RE emissions
analysis, contact the air program in an EPA regional office31 to discuss options for the emission
quantification approach that is appropriate for the EE/RE policies and programs at hand.
Table 3: Suggested Emissions Quantification Approaches for Each Pathway
Pathway Suggested Quantification Methods32
Baseline Emissions Projection Pathway
Control Strategy Pathway
Emerging/
Voluntary Measures Pathway
WOE Pathway
Energy models approach
Historical hourly emission rate approach
Alternative emissions projection tools or analysis
Energy models approach
Historical hourly emission rate approach
Capacity factor approach
Capacity factor approach
Energy models approach
Historical hourly emission rate approach
Capacity factor approach
eGRID subregion non-base load emission rates
Important Considerations for Emission Quantification Approaches
All emission quantification approaches should address the following key information:
Defining geographic boundaries at the appropriate regional level (see section below)
Identifying ECU dispatch order
Accounting for electricity imports/exports
Transparency of data sources and analytical steps
Recognition of any transmission constraints
Environmental regulations affecting EGUs
If a state, tribal or local agency is not using an energy modeling approach, the following
information should be specifically addressed in any SIP/TIP submission:
31 For more information, go to: http://www.epa.gov/aboutepa/where.html.
32 The quantification methods are only suggestions. Please note that the capacity factor and historical hourly
emissions approach are under a third-party peer review process as of June 2012.
1-31
-------
Future Generation
If the projections for EE/RE policies and programs extend out more than five years, then, for
more sophisticated analyses, a state, tribal or local agency should develop assumptions for how
future generation will change over time. The jurisdiction should examine each nonattainment
area33 and assign emission rates to new units expected to come online or exclude planned
retired plants in the jurisdiction's future emission rates. There are multiple organizations that
project how EGUs will meet future demand and react to new environmental regulations. The
EPA recommends obtaining projections of future ECU growth from EPA, EIA, electric grid
operators, RTOs, ISOs, or NERC.
It is also important to consider which new resources may be entering an area and whether
there are plans for transmission upgrades. Energy efficiency resources can help an area avoid
the need for new or upgraded transmission lines. Depending upon the region, upgrades could
encourage further development of RE, or may permit greater access by older, high-emitting
sources that may be more likely to run if the new transmission is built.
Energy Exports and Imports
The EGUs located in the area of analysis may import or export significant amounts of energy.
The first step in addressing electricity transfers is to determine whether there has been
significant movement in recent years between the area of analysis and other areas. The
following data sources are available for electricity import/export information:
Data on total generation and export/import percentages will indicate whether the area
is a net importer or exporter, as well as the magnitude of transfers relative to total
generation34
Most system operators (balancing authorities, RTOs /ISOs) release information annually
about generation, loads, and interchange on their system
Long-term power purchase agreements that underlie exports and import transfer
information
If EGUs collectively in an area of analysis are net exporters or importers, then the next step is to
determine if the transfer level follows a daily load pattern, a seasonal load pattern or is a
consistent source of energy to an outside importer throughout the year. Once typical energy
transfers are characterized, the dispatch order in the area of analysis should be adjusted to
account for these transfers within the relevant time frames.35
Managing Uncertainty
All SIP/TIP emission reduction measures have some level of uncertainty, whether it comes from
the uncertainty associated with the emissions factors for certain sources, the level of
33 For more information on areas not meeting the National Ambient Air Quality Standards, go to:
http://www.epa.gov/oar/oaqps/greenbook/.
34 For more information, go to: http://www.epa.gov/cleanenergv/energy-resources/egrid/index.html.
35 Synapse (2005).
1-32
-------
compliance with existing SIP/TIP measures, or air quality modeling for an attainment
demonstration. By using conservative assumptions, appropriate discount factors or verification
techniques, emission reductions from EE/RE policies and programs can be applied for SIP/TIP
purposes. The EPA recognizes that there will likely always be some level of uncertainty
regarding the exact quantity and location of emission reductions resulting from EE or RE
measures. However, in many cases EPA also believe that you can apply existing tools with
sufficient rigor to be able to quantify estimated emission reductions with acceptable certainty
to allow the reductions to be credited.
1-33
-------
References
EPA (2010a). eGRID Technical Support Document. Available online at
EPA (2010b). eGRID Table in: eGRID2010Vl_0_STIE_USGC. Available online at
EPA (2010c). Assessing the Multiple Benefits of Clean Energy. September 2011. Available online at
Synapse Energy Economics, Inc. (2005). Methods for Estimating Emissions Avoided by Renewable Energy
and Energy Efficiency. July 8, 2005. Available online at
1-34
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-456/D-12-001J
Environmental Protection Outreach and Information Division July 2012
Agency Research Triangle Park, NC
------- |