Section G:
Summary of the Models used for the Analysis
Description of the Integrated Planning Model (IPM)
Analytical Framework of IPM
• EPA uses the Integrated Planning Model (IPM) to analyze the
projected impact of environmental policies on the electric power
sector in the 48 contiguous states and the District of Columbia.
Developed by ICF Resources Incorporated and used to support public
and private sector clients, IPM is a multi-regional, dynamic,
deterministic linear programming model of the U.S. electric power
sector.
• The model provides forecasts of least-cost capacity
expansion, electricity dispatch, and emission control strategies
for meeting energy demand and environmental, transmission,
dispatch, and reliability constraints. IPM can be used to evaluate
the cost and emissions impacts of proposed policies to limit
emissions of sulfur dioxide (SO2), nitrogen oxides (NOx), carbon
dioxide (CO2), and mercury (Hg) from the electric power
sector.
• IPM was a key analytical tool in developing the President's
Clear Skies proposal.
IPM Is Well Suited to Model Multi-Emission Control Programs
• Among the factors that make IPM particularly well suited to
model multi-emissions control programs are (1) its ability to
capture complex interactions among the electric power, fuel, and
environmental markets, (2) its detail-rich representation of
emission control options encompassing a broad array of retrofit
technologies along with emission reductions through fuel switching,
changes in capacity mix, and electricity dispatch strategies, and
(3) its capability to model a variety of environmental market
mechanisms, such as emissions caps, allowances, trading, and
banking.
• IPM is particularly well suited for modeling Clear Skies
because the program relies on the operation of an allowance market,
the availability of a broad range of emissions reduction options,
and empowerment of economic actors to achieve emission limits.
Extensive documentation of the IPM is available at
http://www.epa.gov/airmarkets/epa-ipm/index.html.
Description of Air Quality Modeling
• The results for fine particle concentrations, visibility,
sulfur deposition, and nitrogen deposition are based on the
Regional Modeling System for Aerosols and Deposition (REMSAD).
• REMSAD is an Eulerian air quality model developed to simulate
regional-scale distributions, sources, formation, transport, and
removal processes for fine particles and other airborne pollutants.
This analysis used REMSAD version 6.4 with meteorological inputs
previously developed for 1996 using the Mesoscale Meteorological
Model (MM-5).
• The results for ozone concentrations are based on the
Comprehensive Air Quality Model with Extensions (CAMx).
• CAMx is an Eulerian air quality model developed to simulate
local and regional-scale distributions, sources, formation
transport, and removal processes for ozone and other photochemical
pollutants. This analysis used CAMx version 3.1 with meteorological
inputs previously developed using the Regional Atmospheric Modeling
System (RAMS) for episodes in June, July, and August 1995.
• The Integrated Planning Model (IPM) was used to derive all
future projections of electricity generation source emissions.
• Emissions inputs for non-electric generating facilities for
REMSAD and CAMx were derived from the 1996 National Emissions
Inventory (NEI). In addition, inventories prepared for the Heavy
Duty Diesel Engine rulemaking were the basis for future year
emissions projections.
• For the most part, the modeling results are analyzed in terms
of the change in future year air quality relative to predictions
under baseline conditions. In this way, effects of any
uncertainties in emissions forecasts and air quality modeling are
minimized.
• Results for projected annual PM2.5 and 8-hour ozone
nonattainment were determined by "rolling back" current air quality
levels. This was based on the change in air quality between the
1996 Base Year and each future year scenario. Since no ozone
modeling was performed for the Western U.S., future ozone
nonattainment in the West was determined through an emissions
scaling analysis that used forecast changes in NOx emissions in the
West coupled with the response of ozone to emissions changes, as
modeled in the East.
• Maps which display the impacts on PM2.5 concentrations and
deposition are reported as a percent reduction. A positive percent
reduction (e.g. 30%) is a decrease in concentration or deposition
compared to current conditions (an improvement); a negative percent
reduction (e.g. -30%) is an increase in concentration or deposition
compared to current conditions.
• Visibility results are reported as a change in deciviews.
"Perfect" visibility is represented by a deciview of zero, so a
decrease in deciview is an increase or improvement in visibility.
An increase in deciview is a decrease in visibility.
Description of Benefits Modeling
• The Criteria Air Pollutant Modeling System (CAPMS) is used to
quantify human health benefits due to the changes in a population's
exposure to fine particulate matter and ozone.
• Using the air quality modeling results, the change in
pollutant concentration based on modeling for each CAPMS grid cell
is determined. This is the level at which the population living in
that grid cell is assumed to be exposed.
• Concentration-response functions from epidemiological studies
are applied to each grid cell to predict the changes in incidences
of health outcomes (e.g. asthma attacks) that would occur with the
projected changes in air quality.
• The grid cells are aggregated to estimate the health impact
of the change in air quality across the study region.
• The estimated economic value of an avoided health outcome
(e.g. $41 per asthma attack day) is multiplied by total change in
events to determine the health benefits of air quality improvements
for the entire region.
• For visibility, benefits were calculated based on changes in
fine particle concentrations, presented as deciviews, which are
provided by the REMSAD air quality model.
• Individuals place a value on visibility improvements in
recreational areas, such as National Parks and wilderness
areas
• The economic value that people place on improved visibility
on a day that they visit a Class I area is applied to the predicted
deciview changes and projected number of park visitors affected to
attain recreational visibility monetary benefits.
Description of Freshwater Modeling
• The Model of Acidification of Groundwater in Catchments
(MAGIC) is used to examine changes in surface freshwater chemistry
as indicated by changes in acid neutralizing capacity (ANC) in the
waterbody.
• ANC represents the ability of a lake or stream to neutralize,
or buffer, acid. The condition of a lake or stream improves as the
the ANC increases, moving from chronically acidic � episodically
acidic � not acidic.
• Episodically acidic lakes (ANC of 0-50 µeq/l) have a greater
capacity to neutralize acid deposition than chronically acidic
ones. However, these lakes remain susceptible to becoming
chronically acidic if acid deposition increases.
• Watershed characteristics (e.g., soils, bedrock type,
geologic history) affect the rate of water chemistry response to
acid deposition.
• "Direct response" lakes or streams manifest changes more
quickly, whereas "delayed response" lakes or streams manifest
changes over a longer period of time.
• MAGIC results show the distribution of lakes and streams (by
percentage) over the three ANC classes
• Three regions were modeled (the Adirondacks, the Northeast
(including the Adirondacks), and the Southeast).
• Results are reported for current conditions (2000) and in
2030 under the Base Case and the Clear Skies Act.
Results are based on a model called the "Model of Acidification
of Groundwater in Catchments" (MAGIC) used by the National Acid
Precipitation Assessment Program (NAPAP) to estimate the long-term
effects of acidic deposition (sulfur) on lakes and streams. The
model simulates the size of the pool of exchangeable base cations
in the soil. As the fluxes to and from the pool change over time
due to changes in atmospheric deposition, the chemical equilibria
between soil and soil solution shift to give changes in surface
water chemistry. Changes in surface water chemistry are
characterized by changes in Acid Neutralizing Capacity (ANC) - the
ability of a waterbody to neutralize strong acids added from
atmospheric deposition.
Description of the Technology Retrofit and Updating Model
Uses of the Technology Retrofit and Updating Model
• At this time, IPM does not model price elasticity of demand
and the effect of multiple allowance allocation mechanisms. To
study the effect of these variables on electricity prices and
markets, ICF developed a macro-driven spreadsheet program termed
the "technology retrofit and updating model."
• The model is used to discern trends in marginal costs and
retrofits, the approximate magnitudes of those trends, and the
reasons for those trends.
Modeling Approach
• The technology retrofit and updating model consists of a set
of approximately six hundred "sample" generating units with varying
characteristics. The mix of generation types and sizes was chosen
to mirror, in general terms, the nationwide mix of capacities. Each
unit is assumed to choose emission control retrofit options, fuels,
and generation levels so as to maximize its own net profit in
response to fuel prices, emission allowance prices, and prices of
electricity for various demand segments. Prices of fuels can be
adjusted in the model in response to demand; prices of electricity
by demand segment is set in the model so as to meet demand; and
allowance prices can be adjusted to cause the industry to meet
given caps.
• To simulate the effects of demand elasticity, the quantity of
electricity demanded in each segment can be set as a function of
electricity prices using an elasticity value that is entered as an
input to the model. Finally, to simulate the effects of allowance
updating, the value of reallocated allowances can be calculated and
subtracted from each unit's cost of generation - thereby inducing
each unit to change its profit-maximizing level of generation in
response to a given set of fuel, allowance, and electricity prices.
Readjusting the allowance prices to meet the same emission caps
then generates results showing the costs of meeting given caps with
and without updating.
• An important limitation of this model is that it does not
simulate changes over time in the demand for electricity, prices,
technology, or other factors considered within IPM. Instead, it is
run as though every year is the same as every other year and is
therefore static in its outlook. In addition, it does not recognize
the distinctions among electricity demand regions and the
transmission constraints that can keep them separate. Thus, only
one price of electricity is determined for each demand segment for
the entire set of sample plants.
Description of the Retail Electricity Price Model
Primary Attributes of the Model
• The Model provides a forecast of average retail electricity
prices from 2005 to 2020 for 13 regions and the contiguous U.S.,
and considers areas of the country that (1) will have competitive
pricing of power generation and, (2) are likely to price retail
power based on a cost-of-service basis.
• Combines IPM and EIA information with data from the National
Regulatory Research Institute and Center for Advanced Energy
Markets regarding the restructuring of the power industry.
• "Main Case" is EPA's forecast of "likely deregulation"
considering areas of the country that should price generation
services for retail customers competitively and those that most
likely to use cost-of-service pricing principles.
• The Model readily analyzes alternative multi-pollutant and
base case scenarios modeled with IPM, alternative assumptions on
deregulation and future savings/costs, and the implications of
different allowance allocation approaches. The strongest
application of the Calculator occurs from examining the relative
price differences between two or more scenarios.
The Limitations of the Model Include
• The Model combines IPM and EIA cost elements that use similar
-- but not identical -- assumptions on capital recovery and
aggregate cost data in a similar -- but not identical -- regional
manner that needs adjustment.
• The Model assumes public and private companies seek the same
return and have the same tax treatment, which overstates prices in
areas where there are large amounts of public power.
• The Model focuses on major costs. It assumes for
cost-of-service areas (where most of power sales are likely to
occur) that allowance allocations will not alter pricing of
electricity.
• Uses EIA's limited (but best available) data in some areas
(e.g. rate base with stranded assets).
• The Model cannot address the uncertainty of deregulation
created by California's experience -- where competition may
increase or decrease in the future. With the phasing in of
competition and limited experience, the full benefits and costs of
deregulation still remain unknown.
