BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
NOTICE
TO:
Dorothy Gunn
Clerk
Illinois Pollution Control Board
James R. Thompson Center
100 West Randolph St., Suite 11-500
Chicago, IL 60601-3218
SEE ATTACHED SERVICE LIST
PLEASE TAKE NOTICE that I have today filed with the Office of the Clerk of the
Illinois Pollution Control Board the TESTIMONY OF DAVID C. FOERTER, EZRA D.
HAUSMAN, Ph.D., ROBERT J. KALEEL, SID NELSON JR., and JEFFREY W. SPRAGUE,
and AMENDED TESTIMONY OF CHRISTOPHER ROMAINE, a copy of which is herewith
served upon you.
ILLINOIS ENVIRONMENTAL
PROTECTION AGENCY
By: ______________________
Gina Roccaforte
Assistant Counsel
Division of Legal Counsel
DATED: April 28, 2006
1021 North Grand Avenue East
P. O. Box 19276
Springfield, IL 62794-9276
THIS FILING IS SUBMITTED
217/782-5544
ON RECYCLED PAPER
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
TESTIMONY OF DAVID C. FOERTER
My name is David C. Foerter and I am the Executive Director of the Institute of Clean
Air Companies (ICAC). I am testifying on behalf of the Illinois Environmental
Protection Agency. ICAC is the national trade association of companies that supply air
pollution emission control and monitoring technologies for electric power plants and
other stationary sources. ICAC represents eighty-five of the leading manufacturers
providing emissions control solutions for affected industries as well as employment
opportunities across the U.S. ICAC recommends that any program to control mercury
emissions be designed with consideration of the following important goals: ensuring
substantial environmental improvements, providing flexibility to all affected parties,
minimizing costs, maintaining fuel diversity, and continuing reliable electric generation.
To that end, ICAC would like to provide the following comments.
Commercially Available Technology
Despite the lack of a strong national mercury requirement for coal-fired utilities, a
number of mercury control technology options are commercially available while other
options are still in development and testing phases and their deployment can benefit from
regulatory certainty. A rapid development of mercury control technologies over the last
several years produced a number of technologies available for implementation of
mercury control programs for coal-fired power plants. A large number of full-scale
demonstrations and even more laboratory tests were conducted and provide a foundation
of information on the effectiveness of controls for various coal types and existing
emissions control configurations. Much of the improvements in technology are
documented in the Illinois technical background document. In general, there are a suite of
options available to cost-effectively control mercury emissions from power plants of
different configurations and coal types.
Based on recent demonstration results, significant amounts of mercury can be removed
through the use of existing controls. Existing control installations such as fabric filters,
electrostatic precipitators, SO
2
scrubbers, selective catalytic reduction (SCR), and others
are currently achieving high levels of mercury reductions even though these processes
were not originally intended, designed, nor optimized for mercury capture. With the
implementation of mercury regulatory requirements beyond incidental co-benefit levels
of control, a number of options for optimizing existing controls will be implemented to
provide cost effective reductions.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
2
Mercury specific control technologies such as sorbent injection systems are commercially
available and have been demonstrated at full-scale on various coal-fired boilers, coal
types, and control configurations. Multipollutant control approaches as well as other
mercury specific technologies provide additional low cost, innovative approaches to
mercury control. Advances in control technologies have overturned the assumptions that
sub-bituminous coals are the most difficult and expensive to control. For example, a
better understanding of sub-bituminous coals has led to successes in dramatically
reducing the cost of sorbents while increasing the control effectiveness.
Over the last year, ICAC members reported booking new contracts for equipment for
sixteen power plant boilers, with engineering companies reporting new contracts on
several facilities to develop design specifications for procurement of mercury specific
control technology. These contracts are for controlling mercury on new and existing
sources, burning bituminous and subbituminous coal, with different particulate capture
equipment such as fabric filters and electrostatic precipitators (ESP). The contracts for
commercial systems are attributed to federal and state regulations, including new source
permit requirements and consent decrees, which specify very high levels of mercury
capture. ICAC strongly believes that the strength of these bookings effectively ends any
debate on the commercial availability of mercury specific control technologies.
Providing Flexibility
The Institute advocates setting requirements that effective use available control
technologies, and then using flexible approaches to promote innovation and early
compliance with those requirements. Flexible approaches should establish an
environmental goal that allows affected sources to choose among control options and
seek a least cost approach to achieving that goal. Examples of flexible approaches
include market-based approaches, capital recovery programs, plant-wide averaging,
annual emissions averaging, early reduction incentives, safety valves, or other
approaches. Incentives combined with concrete goals can encourage further technology
innovation and offer opportunities to focus controls on the most cost effective sources
and coal types.
Regulatory programs that allow a larger pool of affected sources to work towards one
goal will produce higher emission reductions at lower costs. An averaging program that
permits company wide or plant level averaging, or alternatively, a cap-and-trade program,
provide the most cost-effective means to achieve substantial mercury emission reductions
from the power generation industry. These types of programs compel utilities to target
reductions from the units where controls are most cost-effective, and focus in nearly all
cases on the larger units with the highest emissions. These flexibilities within a program
encourage economically efficient decisions. Typically, a unit with the highest mercury
emissions will be among the first to be controlled since the cost per pound of mercury
controlled will be the lowest at these units.
Regulations that mandate reductions through the use of a dual limit of an emission rate
and a maximum control efficiency would also provide some flexibility to utilities. It also
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
3
compliments the fundamental capabilities of the control technologies which become
increasingly more cost effective and yield higher percentage removals the higher the
mercury concentration is going into the plant. The way that this would function for
utilities, for units that have a high inlet mercury concentration, it would be more cost
effective to achieve a higher maximum control efficiency while units that have lower
inlet mercury concentrations may find it more cost effective to meet lower emission rates.
Given a longer compliance period, such as a yearly average, utilities will have the
opportunity to meet stringent limits as it will provide more ability for utilities to take into
account changes in mercury content in coal or other operational changes at the boiler
compared to shorter averaging periods.
Encouraging early adoption of control technology will create benefits to both the power
generator and the environment. This sort of mechanism has been implemented in other
market-based programs such as the US EPA’s Acid Rain and NO
x
Budget Programs.
Under the Acid Rain Program, allowance credits were banked by those plants that
installed controls in advance of the compliance date and achieved emission reductions
greater than their historical baseline emissions. Under the NO
x
Budget Program, a
limited pool of allowance credits were distributed to plants that installed controls in
advance of the compliance date and reduced their emissions below a specific NO
x
emission rate. This method not only encourages plants to install controls early but also
gives incentives to maximize the performance of their emissions control strategy.
Banking of credits at individual plants (not for sale to other plants but to offset future
emissions) will lead to greater mercury reductions earlier and will also significantly
reduce costs for the plant.
Maintaining Electric Reliability
Electricity is essential to our modern economy. Advancements in technology have
increased U.S. productivity and driven growth, and many technologies have increased
electricity demand. Currently, coal generation provides more than fifty percent of the
nation’s electricity supply with the remainder being provided by nuclear, natural gas, oil,
hydro, and other renewables. Certain fuels in the electricity generation mix are better
suited than others for particular applications. That’s why a variety of fuels – as well as
increasingly more cost-effective and efficient ways to use and conserve energy – is
needed.
Low-cost, reliable electricity results in part from our ability to utilize a variety of readily
available energy resources – coal, nuclear energy, natural gas and hydropower, and other
renewable energy resources. Fuel diversity is key to affordable and reliable electricity. A
diverse fuel mix also helps to protect consumers from contingencies such as fuel
unavailability, price fluctuations and changes in regulatory practices.
Creation of Jobs
The installation of air pollution control equipment on power plants also creates crucial
job opportunities for a variety of professions. Many of the jobs in the air pollution
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
4
control industry are high quality, high-tech jobs, such as engineering and computer aided
design positions. In addition to the high tech jobs, the following types of labor are
required for the installation of the technologies on coal-fired power plants including:
general construction workers for site preparation and storage facility installation; skilled
metal workers for specialized hardware assembly; other skilled workers such as
electricians, pipe fitters, millwrights, painters, and truck drivers; and other unskilled labor
to assist with hauling of materials and cleanup. There are more than 150,000 air
pollution control professionals working in the U.S. today that are continually advancing
the capabilities of the industry to met environmental requirements.
Conclusion
ICAC acknowledges even in light of rapid success in the development of effective
mercury capture technology and the initiation of a commercial market, plant specific
engineering and technical challenges may exist as they have for any emission control
program. However, the technologies that exist today can be deployed along with
regulatory flexibilities to produce a cost effective and sound control program.
The Institute supports the development of regulations that provide cost-effective
reductions in mercury emissions through the development of emissions targets that
maximize mercury removal while minimizing risk to all stakeholders. In order to protect
electric reliability and reduce compliance costs, the implementation of market-based
programs and other flexibility mechanisms are recommended. The development of
regulations that go beyond the U.S. EPA requirements will further the development of
control technologies, spur development of even more cost effective technologies, and will
create jobs for the design and construction of emission control equipment.
.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES(MERCURY) )
TESTIMONY OF EZRA D. HAUSMAN, Ph.D.
Qualifications
My name is Ezra D. Hausman, Ph.D. I am a Senior Associate with Synapse Energy
Economics, Inc. (“Synapse”), a research and consulting firm specializing in energy and
environmental issues. Synapse’s areas of expertise include electricity market analysis;
generation, transmission and distribution system reliability; market power analysis;
electricity market price forecasting; valuation of stranded costs and benefits; and
integration of energy efficiency and renewable energy in wholesale electricity and
capacity markets.
I have been employed by Synapse since July of 2005. Prior to this I was employed as a
Senior Associate with Tabors Caramanis & Associates (TCA) since 1997, performing a
wide range of electricity market and economic analyses and price forecast modeling
studies, including asset valuation studies, market transition cost/benefit studies, market
power analyses, and litigation support studies. I have extensive personal experience with
market simulation software including GE-MAPS, and I have strong familiarity with a
number of other market simulation environments and approaches to electricity market,
and economic analysis.
I hold a B.A. from Wesleyan University, a M.S. in civil engineering from Tufts
University, an S.M. in applied physics from Harvard University and a Ph.D. in
atmospheric chemistry from Harvard University.
Purpose and Summary of Testimony
I was asked to testify today by the Illinois Environmental Protection Agency (Illinois
EPA) in order to offer my expert analysis of how the proposed Mercury emissions rule in
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 2
Illinois will impact the Illinois electricity market and the Illinois economy. Much of my
analysis is based upon information contained in the Technical Support Document (TSD)
provided by the Illinois EPA in support of the rule. I have also relied upon data provided
by ICF Corporation relating to their use of the IPM model to analyze the impacts of this
rule. While I do rely upon the same underlying data used by ICF, in many cases, my
analysis differs from the conclusions reached by ICF using this model. I will explain
these differences, and why I feel my analysis to be more realistic, as appropriate
throughout my testimony.
I begin with my analysis of the TSD's conclusions regarding the proposed rule's expected
impact on wholesale and retail electricity prices, and on the competitiveness of Illinois
generating units. I will address the question of whether existing coal-fired generating
plants would be likely to “retire” as a result of the proposed rule, and whether this would
cause reliability concerns in the state of Illinois. Finally, I will offer some analysis of the
economic impact of the proposed rule on the economy and employment in the State of
Illinois, as well as health-related impacts. My conclusions may be summarized as
follows:
•
The cost of producing electricity at Illinois coal plants is likely to increase by
about 0.0375 cents/kwH;
•
The impact on retail prices is likely to be much smaller than the impact on
production cost because coal units in Illinois only set the price of electricity for a
fraction of the hours of any year. I calculate that the
total
price impact of the rule
for Illinois ratepayers will be between zero and $11 million per year, in 2006
dollars;
•
I calculate that the
total
price impact on consumers in the broader region (Illinois
and the surrounding states) will be up to $60 million annually, which is roughly
twice the total annual cost of compliance for Illinois generators;
•
In terms of reliability impacts due to retirements of plants that would be rendered
uneconomic by the rule, I conclude that a very small number of plants are likely
to retire, if any, and that the impact on system reliability is negligible;
•
In terms of economic impacts, I find that any direct job losses due to the proposed
rule are likely to be more than offset by economic benefits, including
construction, installation and operational employment increases, and new jobs in
the tourism and recreational fishing industries;
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 3
•
The health and avoided premature death benefits of reducing mercury emissions
under the rule will be hundreds of millions of dollars per year, well in excess of
the cost of implementing the rule.
Impact of Proposed Rule on wholesale and retail electricity prices
I first address the TSD's analysis and conclusions regarding the proposed rule's expected
impact on wholesale and retail electricity prices and on the competitiveness of Illinois
generating units, relating especially to the material on that point contained in Chapter 9 of
the TSD. This section of my testimony supports the following conclusions:
1. The analysis of electricity markets summarized in Chapter 9 of the TSD
overstates the effects of the proposed rule on electricity market prices and costs to
Illinois electricity consumers.
2. The retail electricity cost impact for Illinois is likely to fall somewhere between
zero and $11 million per year.
3. The effect of the proposed rule on the competitiveness of coal-fired generating
units in Illinois is likely to be quite modest, smaller than the effects of other
factors.
Electric Power System Modeling
The operation and evolution of electric power systems are complex processes subject to a
wide variety of technical, economic, and regulatory factors. Computer models are used to
understand these processes, and to estimate the impacts of changes to the system upon the
characteristics and costs of the system; one such change would be a proposed
environmental regulation such as the proposed mercury rule. When analyzing such a
change, it is useful to consider the short-term and the long-term separately, since the roles
of various factors and the uncertainty associated with market simulations differ for the
two situations.
In the short-term, the set of capacity resources is largely fixed and impact analyses can
focus upon the
operation
of the system. System operations are complex but relatively
well understood and subject to rules and procedures that are implemented by grid
operators such as PJM and MISO
1
. Regional wholesale power markets are dispatched
1
The PJM Interconnection (“PJM”) and the Midwest Independent Transmission System Operator
(“MISO”) are centrally-dispatched Regional Transmission Organizations (RTOs) which control large
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 4
based upon bids submitted by generators in order to minimize the total costs, subject to
various physical constraints such as transmission limitations and generator ramp rates.
Dispatch models such as MAPS and MARKETSYM simulate the dispatch using very
detailed inputs on the available resources, their costs and heat rates, their locations
relative to transmission constraints, and chronological electricity demand by customers,
typically on an hourly basis. The inputs that matter to operations analysis are mainly
“variable costs” which include fuel and some O&M costs including the variable costs of
pollution controls. The outcome of these models depends directly upon the input data, so
any uncertainty in forecasting future conditions, of which there is a great deal, results in
implicit uncertainty in the forecast. Nonetheless, the algorithms are generally accepted to
be good for representing the phenomena that they attempt to simulate – the deterministic
operation of the electric power system given a certain set of input assumptions.
In the long-term, say five years or more, the set of capacity resources can be changed by
capital investment decisions. In this case impact analysis must address the
capacity mix
of the system as it evolves over a period of years, with power plant additions and
retirements as well as capital investment in the generating plants, including investments
in air emissions controls. Capital investment and plant retirement decisions are quite
complex and notoriously difficult to represent in a computer model. In the simplest sense,
they depend upon reasonably well understood fundamentals such as discounted cash flow
analysis. For example, a unit retirement decision would, at its simplest, be a
straightforward matter of projecting forward-going costs (e.g., for fuel, O&M, and
required investments for continued operation) and expected revenue (e.g., for selling
capacity, energy, and any ancillary services into the market), and applying a discount rate
to compute the present value of the net revenues. However, these decisions also depend
upon a number of highly complex factors and considerations such as:
•
Selection of the discount rate to use in the present value calculation in any
particular situation;
regions of the eastern United States electricity market. Illinois is split between these two, with the northern
area (including Chicago) controlled by PJM, while the southern part of the state is controlled by MISO.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 5
•
Consideration of strategic factors such as market power;
•
Consideration of uncertainty, risk, and option value, which are generally not
represented in market simulation models;
•
Regulatory constraints and risks.
Attempts have been made to incorporate some of these into computer models, but this is
quite challenging, and the accuracy of such models will be relatively poor. The results
depend upon the input data assumptions as well as algorithms to represent complex
corporate investment behavior, which at times can turn on little more than a decision-
maker’s hunch about the future. The assumptions and algorithms may or may not be
realistic in any particular situation.
In this case, the Integrated Planning Model (IPM) was used to estimate the electric
market effects that are reported in Chapter 9 of the TSD. The IPM model has previously
been applied by the U.S. EPA for environmental policy impact analysis, including
analysis of the impact of the Clean Air Mercury Rule on a nationwide basis. The model,
developed by ICF Consulting, attempts to represent both system operations and capital
investment decisions over a multi-year planning horizon; in tackling both complex
problems at once, of necessity it does both in a highly simplified manner.
IPM is a linear programming model that develops a single scenario for system capacity
additions and retirements by finding the set of decisions that minimizes the present value
total costs to operate the entire electric power system over a specified period, subject to
various constraints. For example, demand must be met in each region for each time
period, and capacity requirements must be satisfied. Limitations on air emissions must be
observed, transmission limits on key interfaces must be respected, and so on. IPM is a
deterministic model that works with perfect foresight, by which I mean that it makes its
internal choices about operations and investment as if decision-makers knew (or
believed) that the modeler’s input assumptions about future load conditions and
technology costs were guaranteed to be perfectly accurate. It calculates some costs
endogenously, such as fuel costs and emissions allowance prices, based on input
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 6
assumptions about emission caps and well-head and minemouth prices, plus its internally
calculated demand for those items. However, both the deterministic inputs and the
algorithms used to calculate prices involve some amount of uncertainty that is not
publicly estimated by IPM, but which must be assumed to exist by anyone using these
results.
In order to accommodate the large geographic scope and the ambitious incorporation of
capital investment decisions into the model, IPM used aggregated and simplified data in
its unit dispatch function. For example, IPM represents system load conditions using a
very limited number of “segments” (i.e., a year of customer loads is represented by six
load segments in each of two seasons.) Generating units are not dispatched
chronologically as they are in a real market; rather, the generators are dispatched to meet
each of these load segments as part of the single, all-encompassing optimization problem,
and the resulting unit operation is extrapolated and interpreted to represent annual
operations. IPM simulates generating unit forced outages as capacity deratings. That is,
rather than simulate actual random outages of generating units during dispatch, the model
reduces the capacity of each generating unit to approximate the effect of forced outages.
IPM predicts plant additions and retirements such that total present value system cost in
the model, over the entire planning period, is minimized, but selects additions only from
among a list of potential resource additions with specified cost and operational
characteristics. All of these simplifications should be kept in mind when interpreting the
results of one or more IPM model runs.
The nature and extent of the simplifying assumptions suggest that the dispatch
representation in the model is quite coarse-grained. For capital investment and retirement
decisions, the model has some problematic differences with the way that such decisions
are actually made; for example, IPM allows fractions of units to be built or retired in
order to reach an “optimal” result, and does not take uncertainty about the future into
account. The model results show a considerable degree of lumpiness, as is inevitable in a
model that represents hourly dispatch in such a highly aggregated fashion. This may not
be a problem in some cases where the policy being analyzed in much larger in scope or
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 7
impact, such as a national CAMR analysis, for which one could argue that the errors tend
to cancel each other out. However, when trying to discern impacts on the scale of a single
state, this lumpiness can obscure the information one seeks to obtain from the model. In
sum, my judgment is that IPM is ill-suited for analysis of a rule in a limited geographic
area (e.g., Illinois), affecting a small number of generating units (e.g., the existing coal-
fired units in Illinois), with a relatively small compliance cost (e.g., estimated at $33
million per year in Table 8.7 of the TSD). The model results in this case bear out this
judgment.
My concern about the coarse resolution of the model is particularly acute in this case
because the result of interest is the
difference
between two model runs, one with and one
without the Illinois rule. If the inaccuracy in an individual model run is, say, plus or
minus 5 percent for the output variable of interest (e.g., the market price in a particular
location) that might be perfectly acceptable for some purposes. But for understanding the
difference between two such model runs where the policy is a relatively small effect (e.g.,
less than 1 percent) then it is impossible to get a meaningful result by comparing two
individual runs each with 5% uncertainty. The “noise” simply overwhelms the “signal”.
In a national scenario, IPM is simulating a system of more than 10,000 generating units
in 48 states, representing a total electricity industry with a total capacity of about 950,000
MW and annual plant expenditures of about $90
billion
. In contrast, the Illinois rule
which it is attempting to analyze, will effect 25 coal-fired generating units and will have a
compliance cost of about $33
million
on an annualized basis. This level of precision is
simply far too much to ask from such a coarse-grained and large scale modeling exercise.
Impacts of the Proposed Rule on Costs and Electricity Market Price
The application of the Illinois mercury control rule will reduce the mercury emissions
from Illinois coal plants, but will also add to the costs of those plants and to the variable
cost of their generated electricity. A detailed analysis of the mercury control costs on a
unit by unit basis is contained in Chapter 8 of the TSD. The analysis supporting these
findings was carried out by Dr. James Staudt of Andover Technology Partners. The
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 8
starting point for our cost impact analysis is summarized in Table 8.7 of that report and
presented below, in a slightly different format.
Mercury Control Costs for Illinois Coal Plants
In thousands of 2006 $
CAMR 2010
Illinois Rule
Difference
Capital Investment
$35,515
$75,593
$40,078
Annual costs:
Sorbent Cost
$18,665
$41,729
$23,064
Toxecon O&M
$0
$425
$425
Ash Disposal
$9,900
$13,403
$3,503
Annualized Capital Cost
*
$4,972
$10,583
$5,611
Total Annual Cost
$33,537
$66,140
$32,603
*Assumes 14% capital recovery factor
The yearly additional control costs associated with the Illinois rule are $33 million, of
which most of the cost is for sorbent. This is the cost borne by the generating unit owners
to retrofit and operate their units with mercury emissions controls; it does not translate
directly into electricity prices and costs to consumers.
The historic generation from the Illinois coal plants, from TSD Chapter 8, is 86,997
GWh. That converts into an average cost increase for the Illinois coal plants of
$0.375/MWh. For comparison, current retail prices in Illinois are about $70.00/MWh and
are likely to increase if price caps are removed as proposed.
In order to determine how this increase in coal plant costs will affect electricity market
prices, it is necessary to estimate the amount of time that the coal units bearing these
extra costs are “on the margin” and therefore influencing the market price for electricity
in the regional dispatch. That, in turn, depends upon regional operation of the electricity
grid and the dynamics of new entry to the electricity market. I believe that a reasonable
range for the annual electricity wholesale market cost to Illinois customers is between
zero and $11 million.
I calculate the upper end of this range as follows.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 9
1. I estimate that coal generation is “on the margin” in the regional dispatch 85% of
the time.
2
2. The Illinois electricity market is a tightly interconnected part of a much broader
wholesale electricity market. For current purposes I will conservatively assume
that this market includes Illinois, Indiana, Wisconsin, Iowa, Missouri, and
Michigan, of which Illinois contains about 20% of the regional coal generation.
3
3. I multiply the $0.375/MWh cost increase from the rule for Illinois coal by 0.85
and 0.20 to reflect the contribution of Illinois coal units to the regional electricity
market price, yielding a wholesale electricity market price impact of
$0.064/MWh.
4. At annual electricity sales of about 166,000 GWh,
4
the wholesale cost impact to
Illinois electricity customers amounts to $11 million. I assume that this increase is
passed directly through to the retail cost impact.
I believe this is a conservatively high estimate of the price impact of the rule on Illinois
electricity consumers, for the following reasons.
First, my calculations include the variable costs (sorbent and ash disposal) and fixed costs
(annualized investment) of compliance with the rule. A dispatch model simulation would
only apply the increased variable costs to calculate the energy price effects, since the
fixed costs would generally not be included in the generators’ energy supply offers.
Second, the compliance scenario introduced in the TSD is a very simple one, and there
may be ways that the market could respond to the rule that would achieve compliance at a
lower cost. These might include increasing electricity imports (or decreasing electricity
exports), retiring inefficient generators, and installing other emission control technologies
(e.g., FGD and SCR, where Hg reductions would be a co-benefit). To the extent that
2
Based upon the PJM Market Monitoring Unit’s “2005 State of the Market Report,” March 8, 2006 (which
indicates on page 86 that coal was on the margin 62% of the time in PJM in 2005); MISO’s “March
Monthly Report: April 20, 2006” (which indicates on page 77 that coal was on the margin about 86% of the
time in MISO in that month); and inspection of the capacity supply curve compared with load levels.
3
A broader region would probably be more appropriate for wholesale market price calculations, given the
strong transmission interconnections in the MAIN and ECAR reliability regions. Ideally, multi-area
electricity market simulation model analysis would be used to determine the impact of an electricity market
price increase upon the regional market, and the extent to which the price increases occur in different
portions of the Eastern Interconnection. However, given the lack of transmission constraints inhibiting the
import
of electricity to Illinois, we believe that the six-state region is a reasonably conservative proxy.
4
According to EIA data, retail electricity sales in Illinois were 139,254 GWh in 2004 and 144,554 GWh in
2005. Extrapolating this growth rate (3.8% annual) to 2009 yields 166,000 GWh.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 10
some of these approaches were found to be lower cost and to contribute to a mixed
compliance approach, the overall total cost would tend to be reduced. There is no
indication that Dr. Staudt attempted to find the
least cost
compliance scenario, nor was it
his task to do so.
Third, and most importantly, is the impact of new market entry in response to anticipated
electricity market price increases. In regional electricity markets, the long-term price of
electricity is generally expected to be equal to the levelized annual cost of building and
operating a new power plant. While there will be excursions above and below this
“equilibrium price” set by market entry, there are strong forces working to bring prices
into line. If market prices fall below the cost of entry, then developers will defer and
cancel generating facility construction projects. If market prices exceed the cost of entry
for a prolonged period, then developers will initiate and accelerate capacity construction
projects, in order to earn the high profits available under such conditions.
This dynamic of market entry disciplining price increases that would otherwise occur is
one reason that I put the low end of potential market price effect at zero. The other reason
is that, with excess generating capacity in the region and no relevant and binding limits
on power imports (or decreased exports), it may be that existing generators simply cannot
increase market prices in order to pass along compliance costs to customers.
The modest impact of the proposed rule on electricity prices can also be seen in the
“supply curves” provided here as Exhibits EDH-1 and EDH-2. These were derived from
the IPM model files provided by ICF. The graph in Exhibit EDH-1 shows the cost of
electricity from Illinois generators only, with and without the proposed rule. The line for
the case with the rule shows a slight cost increase relative to the case without the rule, in
the middle range of the supply curve—this represents the increased production cost for
specific coal units under the proposed rule relative to CAMR. At some load levels, the
cost of electricity is actually
lower
with the proposed rule in place, according to the IPM
model results. This would occur if certain plants opted to invest in emission control
technology as a result of the rule and thereby eliminated the need to purchase allowances
for NO
x
and SO
2
emissions. The associated capital investment is not a variable cost of
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 11
production, so it is not reflected in the figure. The graph in Exhibit EDH-2 is an
analogous supply curve but for the multi-state region, including Illinois, Indiana,
Wisconsin, Iowa, Missouri, and Michigan. Here the effect of the Illinois rule is
considerably more subtle, reflecting the fact that the marginal cost of electricity at just
about any regional load level would be largely unaffected by the rule. If the curve in
Exhibit EDH-2 represents the highly competitive generation mix that serves a large
interconnected market including Illinois, then the only time prices can be affected by this
rule is when the load falls in the area where these two curves diverge. Even in those
cases, the price impact of the rule can be no more than the vertical distance between the
lines in that region.
There are large differences in production cost between coal units and the lower-
operating-cost nuclear units, and also between coal units and the high-operating-cost oil
and gas units. There are, in fact, some substantial differences in production cost among
the coal-fired units, which inhabit the range between about $16/MWh and $23/MWh in
production cost as shown on the vertical axis. These differences among the coal units
have to do with variations in age, size, efficiency, fuel supply and other factors. The cost
implications of compliance with the Illinois rule, 37.5 cents per MWh, are quite small in
the overall context of variation among generating unit costs of production, and thus the
effect on the supply curve for energy is, as shown in the Exhibits, quite small.
Impacts of the Proposed Rule on Generators
The cost impacts are of importance from the perspective of Illinois electricity customers,
and they will be of use in estimating the direct impact on the Illinois economy later in this
testimony. For generators, the range of impacts differs, but the likely impacts are also
quite modest on a net basis.
Consider first the scenario in which market prices are not increased as a result of the rule.
In that case, Illinois generators would bear the full compliance cost impact of $33 million
per year. While not a trival sum, in the context of the overall electricity markets it is
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 12
almost negligible. To put it in context, the total cost of fuel to electric power plants
located in Illinois amounts to about $2 billion per year.
5
It is also interesting to consider a scenario in which electricity market prices do increase
as a result of the proposed rule. I have proposed a high case in which regional market
prices increase by $0.064/MWh. In this case, the total annual cost increase to customers
in the multi-state region amounts to about $60 million annually. This is roughly twice the
estimated annual compliance cost of $33 million, indicating that generators as a group
will be better off financially with the rule than without it. Of course, in this scenario there
are winners and losers within the group of generating companies, but on average the
generation owners would be more than made whole.
IPM Results
The IPM model was discussed earlier in my testimony, where I highlighted how the
coarse resolution of the model limit its utility for simulating a market effect as subtle as
the one under consideration here. As may be seen in Chapter 9 of the TSD, the electricity
price and cost results obtained using the IPM model are dramatically higher than those I
have calculated. Specifically, ICF reported incremental price increases associated with
the Illinois rule, relative to the CAIR/CAMR case, of $0.57/MWh, $1.67/MWh, and
$1.15/MWh for the years 2009, 2015, and 2018, respectively.
6
In terms of costs to Illinois electricity consumers, the same IPM runs put the totals for
2009, 2015, and 2018 at $99 million, $311 million, and $221 million.
7
For costs to
5
Source: IPM model reports.
6
The prices reported in Exhibit A.3 on page 4 of “Analysis of the Proposed Illinois Mercury Rule,
Appendix A: Summary Results Tables,” March 10, 2006, by ICF, are $0.50/MWh, $1.46/MWh, and
$1.00/MWh for the three years. But these are reported in 1999 dollars. The “Price Indexes for the Gross
Domestic Product” reported by the US Department of Commerce, Bureau of Economic Analysis, indicate
an inflator of 14.6 percent from 1999 to 2005. I applied the 14.6 percent inflation factor to convert the
prices from 1999 dollars to 2005 dollars.
7
The costs reported in Exhibit A.5 on page 6 of “Analysis of the Proposed Illinois Mercury Rule,
Appendix A: Summary Results Tables,” March 10, 2006, by ICF, are separately by customer class. I
totaled the customer classes, and got $86 million, $271 million, and $193 million. But because these are in
1999 dollars I applied the 14.6 percent inflation factor to convert the costs from 1999 dollars to 2005
dollars.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 13
“national” electricity consumers, the IPM runs put the totals for 2009, 2015,and 2018 at
$332 million, $529 million, and $786 million.
8
The IPM results imply that $33 million in annual production cost increases would
translate into hundreds of millions of dollars of annual costs to consumers. Electricity
markets may have their flaws, but the idea that they could be so inefficient and so
punitive to customers, and that we reside on a precipice where a small increase in the cost
of coal generation will tip us into the abyss, defies credulity. The implication of these
results suggests a tremendous windfall to generators flowing from the Illinois rule. My
judgment is that that this windfall will not occur, but rather that the impacted coal
generators in Illinois may or may not recover the costs of compliance, that the net impact
on these entities will be small.
There are various aspects of the way the Illinois rule is modeled in the IPM model that
make the IPM impact results conservatively high, including the representation of the
emission caps (maximum total emissions instead of maximum emission rates), decreased
flexibility in compliance (relative to the Proposed Rule’s actual provisions), and the
accelerated compliance date (at the beginning of 2009 rather than at mid-year 2009). I
review some of these in greater detail in the next section of my testimony. But while
these aspects of the IPM modeling approach will tend to exaggerate the impact of the
rule, it remains difficult to see how they could account for ICF’s results in terms of
market prices and customer costs. I can only conclude that the results are an artifact of
the model structure, which is designed more for wide-ranging analysis of national policy
than for highlighting the smaller-scale impacts of a regional rule.
One effect predicted by the IPM model with which I do concur is that energy exports
from Illinois may be decreased as a result of this rule. Units with production costs just
below the marginal cost of electricity in the absence of sorbent costs, for example, may
be rendered uneconomic to run during certain hours given this small additional expense.
8
As with the Illinois costs to consumers, I totaled the customer classes (yielding $290 million, $462
million, and $686 million for the 3 years, respectively) and applied 14.6 inflation factor to convert to 2005
dollars.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 14
However, in this case the revenue they would have earned during these hours would have
just barely covered their running costs, so the impact of this output reduction on their
overall economic performance will be minimal. Further, if other states adopt similar rules
in the future, this effect will be abated.
Finally, I find that IPM is unrealistic in its treatment of power plant retirement decisions.
This issue is discussed in the next section.
Electric System Reliability Impacts of the Proposed Rule
The IPM model runs performed in support of the Illinois rulemaking predict the
retirement of a number of older, coal-fired generating units as a result of the proposed
rule. My judgment is that this prediction is overstated, and that in any case that this level
of potential retirements raises no reliability concerns. My reasons are as follows:
•
The total MW capacity of retirements predicted by the IPM model runs as a result
of this rule is quite modest, representing less than 1% of in-state capacity and a
much smaller share of regional capacity. Even this is likely to be an overestimate,
given differences between the model implementation and the realities of the
marketplace;
•
Illinois is strongly interconnected and shares capacity requirements with a large
region that will be unaffected by this rule
9
, making the total MW capacity of
possible retirements comparatively even less significant;
•
The reliability regions and market operators encompassing Illinois have rules in
place to prevent retirements if the specified units are required for reliability
reasons;
•
If there are any subregions within Illinois that have capacity shortages, these
subregions are likely to have higher electricity prices, meaning that units located
in these areas will receive extra revenues and are less likely to retire;
•
The units identified for retirement by the IPM model are small and relatively
inefficient, and may well be nearing the end of their operating lives in any case; it
is likely that they would be retired, upgraded or replaced with more efficient and
cleaner technology with or without the proposed rule within the next several
years;
9
A number of other states in the region are considering similar state-specific mercury rules , but the current
analysis is focused upon the effect of Illinois’ proposed.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 15
•
The cost of new entry is only lightly impacted by the rule, so the rule will hasten,
if anything, the development of new generating capacity in Illinois and the
surrounding region.
Retirements likely to be minimal
The IPM forecast predicts 345 MW of coal plant retirements in Illinois by 2009 in the
case including national CAIR/CAMR rule, and an additional 252 MW of coal plant
retirements in Illinois by 2009 under the proposed, more stringent Illinois EPA mercury
rule. For perspective, the model represents a total of 16,000 MW coal-fired capacity in
Illinois and 43,000 MW of total in-state generating capacity. Thus the retirements
represent 1.6% of in-state coal generation capacity and 0.6% of total in-state generating
capacity.
I believe the IPM model over-predicts retirement in this case, and that the actual number
is likely to be far smaller. There are a few reasons for this. One reason is that it is much
easier to build and retire generating units in a model than it is in the real world. For
example, the IPM model can and does predict “partial” retirements, which is to say that it
finds an optimal number of units to retire which may include part of some unit even if
that is physically impossible. My understanding is that the IPM model does not have the
ability to “mothball” a unit (maintain it in a standby mode) instead of retiring it, which
would otherwise allow it to be returned to service much more easily in the future should
conditions render that profitable. Mothballing of generating units is quite common in real
electricity markets. This is because the option of returning a unit to service in the future,
should market conditions become favorable, is valued in a way that is difficult to capture
in electricity market models with “perfect foresight”.
Another reason that the model overstates likely retirement is that the gas prices calculated
by the model are very low compared to today’s gas and gas futures prices. Gas prices are
calculated endogenously in the IPM model, presumably based on a formula that has been
fit to historical gas price trends. Unfortunately, the current gas market prices are well
above the historical norm for reasons that reflect the unprecedented growth in demand,
increasingly costly domestic gas production, and the globalization of the gas market. ICF
provided the gas prices to Synapse upon request, and they come out at about $4.25 per
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 16
MMBtu in 1999 dollars. This is perhaps half the cost of gas in today’s market or less. As
a result, the IPM model would seriously understate electricity prices and revenues coal
units would receive when gas is on the margin. If more realistic gas prices were
considered in the model, the economics of coal units would look quite a bit more
attractive.
Finally, the implementation of the Illinois rule in the IPM model is unrealistically
stringent. The rule as proposed allows some flexibility in meeting the requirement,
including some averaging of emissions among plants under certain conditions. In the
model implementation a hard cap is in place for each plant. Clearly, generation owners
who are able to reallocate their emissions among plants will find more economical ways
of controlling emissions than they would were they required to rigidly reduce emissions
equally on all plants. Specifically, it would often make sense to leave uncontrolled those
units that run infrequently, taking advantage of the ability to average overall emissions
instead of investing in emissions controls that would be underutilized. While it is hard to
draw solid conclusions on capacity factor due to the low resolution of the IPM model, the
units that are slated for retirement are those with the lowest reported non-zero capacity
factors of all coal units in Illinois. These units would be particularly subject to this
particular distortion.
Thus I conclude that the 252 MW of coal retirements predicted by the IPM model are
unlikely to occur, but even if they did the implications for reliability would be negligible.
Despite this judgment on my part, it is important to consider the implications in case this
level of retirement did occur. I do not believe that this would present a problem in terms
of reliability, because both the local and regional systems have considerable reserve
capacity. The most recent projection of reserve capacity in the MAIN region,
10
for
example, indicates that for the coming summer MAIN has a planning reserve of 17.6%
without including “uncommitted resources”. When uncommitted resources are
10
The MAIN region has been superseded as of January 1, 2006, so that Illinois is now divided between the
Reliability First and SERC reliability regions. However, the most recent market reports concerning capacity
margins were issued under the previous configuration, under which all of Illinois was in the MAIN region.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 17
considered, the planning reserve increases to 21.4%. This compares favorably to the
recommended long-term planning reserve margin of 16 to 19% based on the most recent
NERC Long-term Reliability Assessment. The same report lists projections of planning
reserve margin for the summer of 2010 at 14.8%, without including uncommitted
resources. With the uncommitted resources included the planning reserve, the 2010
margin rises to 20.5%. Since MAIN is a summer-peaking region, the winter reserve
margins are much higher.
The surrounding regions also report planning reserves ranging from 14.5 to 14.9%, of the
same magnitude as MAIN’s 14.8% projection, and as these areas have aggregated into
larger regions the reserve requirements have decreased (see discussion below). Thus, I
conclude that even if 252 MW of Illinois coal generation were to retire, this would not
present a reliability problem from a regional capacity perspective.
Sharing of reserves
Any impact of retirements on the ability of Illinois entities to meet their capacity
requirement is further diminished by at least three regional initiatives, each of which will
increase the effectiveness of resources from the large regional area surrounding Illinois to
support system adequacy. First, the former ECAR, MAIN and MAAC reliability regions
have been reformed into a single large region reliability organization known as
Reliability First, meaning that Illinois (which was in MAIN) now has a much broader
capacity pool from which to draw. The MRO region
11
, to the north and west of Illinois, is
considering consolidation into this larger region. This kind of consolidation generally
results in greater levels of reserve sharing and thus boosts reliability throughout the
system. Essentially, the diversity of use of resources – i.e., varying times at which
systems experience their peak loads – allows for a more efficient sharing of resources
across the broader area. This is evidenced by the way in which capacity reserve
obligations in PJM have been steadily lowered (on a percentage basis) as the PJM RTO
region expanded to include additional utility areas such as American Electric Power,
11
Formerly known as MAPP
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 18
Commonwealth Edison, and Dominion Power--PJM reserve obligations have been
reduced from 19% in 1998 to 15.0 in 2005.
Second, the Midwest ISO region is currently planning to coordinate an explicit operating
reserve market. This will allow for more efficient use of capacity to support adequacy
needs.
Third, the Midwest ISO and the PJM RTO continue to discuss way in which to
coordinate their respective market operations, and they have signed agreements with the
SPP RTO to further coordinate operations. Such coordination can increase the ability of
capacity resources in one region to serve needs in adjacent regions, especially given the
diversity in peak load use across such large regions.
Because of all of these trends towards greater cooperation in reserves sharing, and
because Illinois is almost invariably is an exporter of power so there are no import
constraints, we do not believe that reliability will be threatened by lack of access to
adequate reserves with or without the proposed mercury rule.
Rules governing retirements
If, despite all of these factors, a generating unit which is needed for reliability reasons
were to be nominated for retirement, either PJM or the Midwest ISO (MISO) can take
steps to keep the unit operating, depending on the location of the unit. Indeed, RTOs have
done so from time to time. Generally, this involves entering into an agreement with the
unit’s owner to ensure that the costs of continuing to operate a unit will be recovered,
even if the RTO must supplement market or regulated payments for operation with
additional compensation.
Thus I conclude that if units are rendered uneconomic by the proposed Illinois mercury
rule, but are needed for reliability reasons, either the MISO or PJM market operator has
the necessary authority and the procedures in place to either compel or adequately
compensate the generation owner to keep the unit on line until an alternative solution can
be found.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 19
Retirements unlikely to occur in load pockets due to price signals
The IPM model runs which form the basis for the analysis in the TSD do not represent
transmission constraints within, for example, the MANO area which contains Illinois.
Thus it may be that there are certain subregions which, for reasons of local transmission
or distribution constraints, are particularly vulnerable to reliability problems should
generating units in these areas retire. If this were the case, perhaps retirements of small,
aging coal plants in these areas would raise some reliability concerns.
However, I do not believe that these concerns would be justified for two reasons. The
first is outlined above, which is that the RTOs have the tools and structure in place to
prevent retirements of units that are needed for reliability reasons. Secondly, both RTOs
operate under a locational electricity pricing system known as LMP, which is designed to
produce higher electricity prices in regions that are more expensive to serve. If this is not
enough, PJM is moving towards implementation of a locational capacity compensation
scheme in PJM, under which generation owners will be paid for their capacity (in
addition to their energy) in a way that is designed to compensate generators in capacity-
short regions sufficiently to deter retirements and encourage new entry. Thus, I once
again conclude that if any specific units are needed for reliability reasons—in this case to
ensure local reliability—it would be compensated at a greater rate than would be
predicted by the IPM model, and would be unlikely to retire.
Predicted retirements are not unusual
The specific generating units which are predicted by the IPM model to retire as a result of
the Illinois mercury rule are Hutsonville Units 5 and 6 (partial) and Meredosia Units 1
through 4. As noted in the TSD, these units are all at least 50 years old and may well be
nearing the end of their operating life. Based on data from the IPM model, these units are
about 10% less efficient than average for coal-fired power plants in Illinois, and
considerably less efficient than newer units. It would not be surprising, especially under
conditions of surplus such as those seen in the Illinois region today, to find that such
plants are no longer economically justified with or without the proposed rule, especially
if there is not some specific reliability-based need for these particular assets. In some
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 20
cases, units such as these would be replaced with more efficient new plants such as
natural gas combined cycle units, or with gas-fired peaking units, at the same location.
12
Such units would offer greater operational flexibility and much lower emissions. Thus,
while it is possible that the owners of certain inefficient coal units would find it
preferable to take the units out of service rather than to bring them into compliance with
new emissions rules, my judgment is that this would not be out of line with the normal
evolution of generating assets, may in fact make way for the construction of newer, more
efficient units, and in any case would not pose a reliability problem for the state or the
region.
New entry unaffected by proposed rule
The final point I would like to make with regard to the reliability impact of the proposed
rule is that new generating units are unlikely to be significantly affected by this rule,
because they are already required to meet stringent emissions criteria. To the contrary,
this rule may give a slight economic boost to new entry if it does, indeed, cause a small
number of retirements to be accelerated, or raise local electricity prices by a small
amount. Along with this comes greater efficiency, greater operational flexibility, greater
unit reliability, and lower emissions. If anything, this would provide a net benefit in terms
of electric system reliability in Illinois and the surrounding region.
Conclusions
Based on my analysis of the IPM model data and results, and based on my understanding
of market conditions in the MISO and PJM regions, I conclude that the proposed rule will
have little if any effect on electric system reliability in the region. I conclude that the
number of generating unit retirements caused or accelerated as a result of the rule is very
small if any, that there is more than adequate capacity in the region to accommodate any
retirements that may occur, and that there are safeguards in place to make sure that
retirements will not occur if they would raise reliability concerns.
12
For example, all units over 50 MW retired between 1999 and 2004 identified EIA Form 860 filing
database were replaced with new units mostly gas combine cycle units.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 21
Impacts on the Economy and Employment in Illinois
I have not performed a specific economic modeling study on the impacts of the proposed
mercury rule on the economy of the State of Illinois. I do not believe that such a study
would be particularly informative as to the impacts attributable to changes in generation
costs or prices. As I explained above, the direct impact of the rule in terms of electricity
prices and costs to consumers will be quite modest. However, based on a range of
existing studies and on closely related modeling analysis performed by Synapse staff in
similar cases, I am able to estimate certain of the effects of this rule on the Illinois
economy. Many of these direct and indirect economic costs and benefits to the state of
Illinois are also modest in scale relative to their uncertainty. As will be explained below,
the situation with regard to economic value of the public health benefits of the Proposed
Rule is quite another story.
Retail Rate Impact
My estimate of the incremental total retail cost impact from the proposed rule is between
zero to $11 million per year in 2006 dollars, over and above the cost impact of the
CAIR/CAMR requirements. For perspective, this range represents approximately 0% to
0.1% of the Illinois retail electric bill of $9,359 million per year.
13
In an earlier study performed for the Citizens Utility Board of Illinois, Synapse estimated
the effect on employment in Illinois associated with changes in the total cost of
electricity. On the basis of that relationship, this Proposed Rule’s estimated incremental
retail cost impact would result in the loss of between zero and about 69 jobs in the state
of Illinois. I would caution that even the upper end of this range is quite small relative to
the precision of macroeconomic models, and relative to what I assume to be the day-to-
day variation in employment in the state. Thus I would conclude that the magnitude of
this effect may be statistically indistinguishable from zero impact on employment.
13
Total retail electric costs in 2003 according to the US EIA:
http://www.eia.doe.gov/cneaf/electricity/esr/esr_tabs.html
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 22
Impact of Potential Generating Unit Retirements
The IPM model results reported in the TSD predict that the Proposed Rule may lead to
retirement of certain coal plants now operating in Illinois, over and above the potential
impact of the CAIR/CAMR requirements, totaling 243 MW or about 0.6% of in-state
generating capacity. This includes one "partial retirement" which is wholly an artifact of
the model structure. The employment at these units is estimated to total 160 jobs.
14
In
addition, these plants purchase goods and services from the economy to operate. If there
were a net loss of generating capacity in Illinois, the loss of jobs would exceed the direct
decrease in generating plant employment by some factor.
As discussed above, however, my judgment is that the unit retirements are probably
overestimated due to limitations of the model, and in any case the prediction that these
small, aging, and inefficient coal plants will retire during the next decade is not out of
line with normal evolution of the generation stock. Furthermore, were these retirements
to occur, it is likely that the retired capacity would be replaced by newer, more efficient
plants, quite possibly even in the same location, providing employment benefits
associated with plant construction for up to several years. Thus I judge the direct impact
in terms of employment at Illinois generating units to be small, if any.
Other employment impacts in Illinois
In addition to the direct impact flowing from employment at generating units, there are a
number of beneficial impacts on employment in Illinois that may be expected from the
Proposed Rule. These include:
1. Employment gains due to installation and operation of mercury controls.
Compliance with the Proposed Rule will require certain capital investments in the
early years of the Proposed Rule and certain ongoing operating and maintenance
costs, the largest of which is likely to be purchase of sorbent. The TSD provides an
estimate of the capital cost of incremental Hg controls. Table 8.7 of the TSD
estimates this cost at $35.5 million over the CAMR cost of $75.6 million in 2006
14
TSD at 201
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 23
dollars. The Proposed Rule provides considerable flexibility in implementation timing
and details, and as a result it is likely that the employment benefit of the capital
investment portion of compliance costs, as well as any multiplier effect through the
state economy, will be spread over several years.
There are also likely to be annual O&M costs for mercury controls; estimates of these
are provided in Table 8.7. These costs include:
•
Sorbent cost
$23.064
million per year
•
Toxecon O&M cost $0.435
million per year
•
Ash disposal cost
$3.503
million per year
I cannot predict what fraction of this annual expenditure will be for Illinois goods and
services, but there will clearly be some local benefit.
2. Impact of Increased Demand for Illinois Coal
The IPM model predicts that there will be an increase in generation from Illinois coal
under the Proposed Rule compared to the CAIR/CAMR scenario. The increase varies
with the scenario year, ranging from 40 TBtu to 67 TBtu, assuming that all
bituminous coal burned in Illinois coal-burning EGUs comes from Illinois.
15
On the
other hand, the analysis in Chapter 8 of the TSD suggests that all existing Illinois
coal-burning EGUs can meet the Proposed Rule's requirements without changing the
type of coal they burn. For purposes of this analysis, I assume a range of incremental
use of Illinois coal of zero to 50 TBtu.
The US average as received heat content of bituminous coal is 24 MMBtu/ton.
16
A
range of incremental consumption of Illinois coal from zero to 50 TBtu then
corresponds to an increase in consumption of Illinois bituminous coal of zero to 2.08
million tons. The 2004 production of bituminous coal in Illinois was 31.5 million
tons, which was 79% of total coal production in Illinois that year.
17
This is a
15
TSD Table 9.10; TSD at 201.
16
US EIA online glossary, at http://www.eia.doe.gov/glossary/glossary_b.htm
.
17
US EIA Annual Coal Report 2004.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 24
percentage increase of zero to 6.6%. Average direct employment in coal mining in
Illinois for 2004 was 3573 jobs.
18
If we assume that 79%, or 2823 jobs, were in
bituminous coal mines, a 6.6% increase in production of Illinois bituminous coal
production would result in an approximate increase in employment of 186 direct jobs,
along with a substantial number of additional jobs in secondary employment.
3. Impact of Enhanced Sport Fishing and Other Wildlife Activities
According to the TSD, "Any improvement, or prevention of loss, to Illinois’ fish and
wildlife activities through implementation of Illinois’ mercury rule could have a
positive impact to this important industry."
19
I agree. The relevant wildlife-associated
recreation activities include at least sport fishing said to be worth "more than $1.6
billion to the State economy when considering the salaries from jobs created, as well
as sales and motor fuel taxes, and State and federal income taxes."
20
However, this
includes all such activities in the state, including those on the Great Lakes. Preparing
a specific estimate of the likely impact of the Proposed Rule on these expenditures is
beyond the scope of this study. However, for illustrative purposes, consider the
scenario where the Proposed Rule results in an incremental 1% increase in sport
fishing. This would produce an additional stimulus to the state economy of about $16
million per year, and to an incremental increase in employment of about 130 jobs.
4. Impact of Health Benefits
The most significant economic impact to be anticipated from the proposed rule flows
from the long-term health and avoided premature mortality benefits of reducing the
environmental loading of mercury. Chapter 3 of the TSD reviews several studies of
the direct economic cost of mercury exposure, whether due to power plant emissions
or other sources. There is wide variation in these estimates, and they were not
necessarily specific to Illinois or the Proposed Rule. However, they may be taken as
indicative of the hidden costs of mercury emissions associated with human health,
18
EIA
Annual Coal Report
, 2004.
19
TSD at 189.
20
TSD at 189.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 25
lost IQ points and premature mortality impacts that will be partially abated by the
proposed rule.
To estimate the economic value of this benefit, I rely on the results of the very recent
NESCAUM/Harvard study
21
cited in the TSD. This study concludes that the
annual
economic benefit of avoided health and mortality impacts is about $182 to $194.5
million per ton of mercury removed.
22
The TSD (p. 29) indicates that the incremental
mercury removal due to the proposed rule is approximately 2400 pounds per year
relative to the CAMR limit for the first ten years of implementation; the benefits of
each ton of avoided mercury emissions will continue to accrue year after year. This
leads to a health and avoided mortality benefit from the Proposed Rule which may be
valued well into the billions of dollars over 10 years. Estimating the considerable
secondary effects of this reduction in adverse health impacts, avoided intellectual
impairment and improved productivity is beyond the scope of this study.
In summary, I find that the direct economic losses associated with implementation of the
proposed rule to be between zero and 69 jobs due to the retail rate impact of between zero
and $11 million, and a maximum of 160 jobs associated with unit retirements should they
occur as projected by the IPM model. On the other hand, I find that there are likely to be
employment benefits associated with installation, operation and maintenance of mercury
control technologies, construction and operation of replacement generation technology,
and possibly with the increased use of Illinois bituminous coal. I find that employment
benefits associated with increased fishing and tourism due to decreased mercury loading
in the Illinois environment and waterways will be on the order of 130 jobs for each 1%
increase in recreational fishing. My judgment is that the increase in employment
associated with these benefits is likely to more than offset the potential decreases.
Additional employment benefits associated with installation and maintenance of mercury
controls, and short- and long-term employment associated with construction and
21
NESCAUM, 2005.
Economic Valuation of Human Health Benefits of Controlling Mercury Emissions
from U.S. Coal-Fired Power Plants.
http://www.nescaum.org/documents/rpt050315mercuryhealth.pdf/
22
Ibid. p. 193.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
Page 26
operation of new power plants, would be likely to put employment gains associated with
this rule well in excess of employment losses.
Finally, I conclude that there will be significant long-term economic benefits associated
with reduced health care costs, improved productivity and avoided premature mortality
due to the proposed rule. These combined benefits would be worth several billion dollars
over the first ten years, alone exceeding by over an order of magnitude the
implementation cost of the plan during that period. Thus, I expect the Proposed Rule to
have a significantly positive and long lasting effect on the economy of the State of
Illinois.
Overall Conclusions
My overall conclusions may be summarized as follows:
•
The proposed rule will have at most a modest impact on the cost of electricity to
consumers, reflecting the relatively modest cost of compliance for Illinois
generating companies;
•
The proposed rule will not raise system reliability concerns because its overall
effect on plant retirements will be small;
•
While there may be some negative direct economic and employment-related
impacts of the rule flowing from the increased price of electricity and the loss of
jobs if power plants do close down, these effects are small and will be
overwhelmed by the positive economic impacts of the rule. These positive
impacts include (but are not limited to) direct employment benefits, benefits in
tourism and recreational fishing, and the substantial economic value of
improvements in human health and avoided premature mortality.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
0
10
20
30
40
50
60
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
Cumulative Capacity (MW)
Variable Production Cost ($/MWh)
Average Illinois Output
Supply Curve with Proposed Hg Rule
CAIR/CAMR Case Supply Curve
Exhibit EDH-1.
Variable cost supply curve for Illinois generating units under the proposed mercury rule, compared with the supply
curve under CAIR/CAMR only case. Based on IPM model data and results.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
0
10
20
30
40
50
60
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
Cumulative Capacity (MW)
Variable Production Cost ($/MWh)
Average Regional Output
Regional Supply Curve with Proposed Rule
CAIR/CAMR Case Regional Supply Curve
Exhibit EDH-2.
Comparison of variable cost supply curves for combined generating stock in six-state region, including Illinois,
Indiana, Wisconsin, Iowa, Missouri, and Michigan. Based on IPM model data and results.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
TESTIMONY OF ROBERT J. KALEEL
My name is Robert Kaleel. I am the Manager of the Air Quality Planning Section in the
Division of Air Pollution Control, Bureau of Air at the Illinois Environmental Protection
Agency (“Agency”). I have been employed by the Agency for twenty-five years in the
areas of air quality modeling, planning, and regulatory development. I have also worked
for a private consulting company in the fields of atmospheric modeling, air permitting,
and air pollution meteorology. I have a Bachelor of Science degree in meteorology from
Northern Illinois University.
In my current position at the Agency, I am responsible for overseeing the development of
technical analyses in support various regulatory proposals, including the proposed
mercury regulations that are the subject of this rulemaking. I am available to provide
testimony in this matter.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
TESTIMONY OF SID NELSON JR.
QUALIFICATIONS
My name is Sid Nelson Jr. I am here today for Sorbent Technologies Corporation, a provider of
power plant mercury emission control, where I serve as President.
I have a Bachelor of Science degree with highest distinction in engineering from The
Pennsylvania State University and a law degree from the University of Arizona. I also have a
Masters degree in Public Policy analysis from Harvard University, where I was a Kennedy
Fellow of Science, Technology and Public Policy.
I have spent the vast bulk of my career in the coal and energy field, including early positions
with the former U.S. Steel Corporation in their coal and resource development group and at
Booz-Allen & Hamilton in their coal-industry management-consulting practice.
I have been with Sorbent Technologies Corporation for over fifteen years, the last five years as
its President. I have been specifically involved in
mercury
emission control activities at Sorbent
Technologies for over ten years.
I appear today to testify on some of our full-scale results, on my understanding of the results of
others, and on the commercial availability of these technologies.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
2
RESULTS OF FULL-SCALE MERCURY CONTROL DEMONSTRATIONS TO DATE.
Is it technologically possible to reduce mercury emissions at each Illinois power plant by 90%
today?
Of course it is. At a minimum selective catalytic reduction (SCR) systems and wet or dry sulfur
dioxide scrubbers could be installed at these plants and these have been shown to be widely
capable of 90%+ mercury removal. While SCR and scrubbers would also have the large benefits
of reducing sulfur dioxide and NOx, they would be expensive for mercury alone, although the
ability to sell SO2 and NOx allowances could mitigate their costs.
Are there inexpensive retrofit technologies available to get 90%+ at Illinois plants? For the vast
bulk of Illinois plants, the answer to this is also yes, even today.
I am going to concentrate my remarks on sorbent injection technologies, in particular on
activated carbon injection technologies for mercury reductions. Variations of this technology
have now been tested at full scale at probably thirty different plants - and counting - so there is a
large knowledge base of its capabilities. With this technology the resulting costs and results vary
somewhat with the particular plant situation: first with the coal that is being burned and, second,
with the type of existing pollution control equipment at the plant.
Subbituminous Coals with Cold-Side ESPs.
The dominant configuration for coal-fired power plants in Illinois is a subbituminous coal with a
cold-side electrostatic precipitator (ESP). Fortunately, the technology exists to retrofit this
particular configuration very inexpensively. My company has been involved in two full-scale
thirty-day continuous runs on this combustion with a product that we produce, called B-PAC (for
brominated powdered activated carbon). Very high mercury removal rates were demonstrated at
very low sorbent consumption rates and costs.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
3
In a Department of Energy sponsored demonstration two years ago at the St. Clair Power Plant
of Detroit Edison, my company injected our B-PACs sorbents continuously for thirty days at a
low injection rate of 3 lbs per million cubic feet of gas into a flue gas. The St. Clair boilers are
primarily fired by subbituminous coal, with about 15% of bituminous coal blended in. We
averaged 94% mercury removal over the period with the plant operating as it usually does,
ramping up and down, changing the coal fired as necessary.
Similarly last fall, in another DOE program with the Energy and Environment Research Center
of the University of North Dakota, the URS Corporation injected our B-PAC sorbent into a
100% subbituminous-coal-fired flue gas stream with a cold-side ESP at Great River Energy’s
Stanton Station continuously for thirty-days. Here they weren’t after 90% mercury removal, just
80%, so they injected at a lower rate of only 2 lbs of sorbent per million cubic feet of gas. Even
at the lower rate they averaged over 80% mercury removal. In earlier short-term parametric tests
at this plant, when they injected at 3 lb/MMacf, they achieved over 90% reductions, as a St.
Clair.
A competitor of ours, called ADA-ES, injected a Norit carbon, similarly halogenated,
brominated like ours. Its called Darco Hg LH. They injected that sorbent at 3.3 lbs per million
cubic feet of gas for nearly thirty-days at the Meramec Station of Ameren, one of the utilities
involved here in Illinois and achieved over 90% average removal over that period as well. These
brominated carbons have been demonstrated at other plants with western coals and dry
scrubbers, as well, and have uniformly achieved 90% mercury removal or better.
With sorbent injection, if you wanted 95% removal you would simply inject more sorbent. The
quantity of sorbent, particularly a brominated sorbent, in a subbituminous plant that you inject is
directly proportional to the mercury removal that you will achieve.
This technology is commercially available today. We supply the sorbent. Norit Americas
supplies the sorbent. Calgon Carbon can supply similar sorbents. In the future, there will likely
be others as well.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
4
Hot-Side ESPs with Subbituminous Coal
There are three boilers in Illinois that burn subbituminous coal with Hot-Side ESPs, which
operate at elevated temperatures. One of them is under a consent decree to install a fabric filter
on the cold-side and this will mean that it can operate very cheaply with traditional sorbents.
The two others, Will County #3 and a boiler in Waukegan, however, can inject hot-side sorbents.
My company has a variation of our B-PAC sorbents called H-PAC, that on bituminous coals at
full-scale we have demonstrated high mercury removals even at the elevated temperatures of hot-
side ESPs. This was demonstrated in short-term tests at Duke Power’s Cliffside Station and then
in a thirty-day test at Duke Power’s Buck Plant. Because these plants burn bituminous coal, a
higher injection rate of carbon was needed to reach equivalent mercury removal rates.
Our H-PAC sorbent for the hot-side ESPs was able to achieve high mercury removals, but we
weren’t injecting at rates capable of 90% at these plants. We weren’t trying to achieve 90% at
these plants, but at an injection rate of 10 lbs of H-PAC per million cubic feet of gas in the
testing at Buck we achieved about 70% mercury removal. Now there are things that we could
have done at Buck, some relatively low-cost capital equipment changes that would have allowed
us to achieve higher mercury removal. However, because it was only a temporary test we did
not do those things.
We believe that because it is so much easier to achieve high mercury reductions with
subbituminous coals with these sorbents, that at Waukegan and Will County we will be able to
achieve 90% at significantly lower injection rates than needed at Buck. This would save the
utility a little bit of money. We will in fact be testing at Will County here in Illinois in a DOE
program for a thirty-day continuous test on the hot-side there early next year to confirm this. Of
course, even if we are unsuccessful, there are other technology options that can be applied.
Turning Will County into a cold-side ESP arrangement and injecting the standard sorbent is just
one example and can be done relatively inexpensively.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
5
“Concrete-Friendly” Sorbents
A number of Illinois plants are able to sell their fly ash as a substitute for cement in concrete.
This is a subset of plants, but it is socially beneficial in that there is less cement that has to be
produced and do not have to dispose of the fly ash. Unfortunately, with our particular
technology, activated carbon injection, the slightest bit of plain activated carbon that gets into
that fly ash generally makes the fly ash un-useable for this re-use application.
Previous testimony may have previously described a few technologies to get around this
problem, the Toxicon and Toxicon II technologies. There is also the possibility of inorganic
sorbents, non-carbon based sorbents, which a number of manufacturers are testing. Amended
Silicates was recently involved in a thirty-day test at an Ohio plant with their inorganic sorbent.
Englehard Corporation has one as well. There are also sacrificial chemicals that are under
development to be added to the slurry or new air entrainment admixture chemicals that are
unaffected by PAC that are under development.
My company has a new product called C-PAC which is a version of the B-PAC which does not
adversely affect the air entrainment admixtures that cause the problem with use of fly ash
containing carbon in concrete. We are going to be demonstrating this C-PAC product in just a
few months at full-scale in a DOE program and the Crawford Plant of Midwest Generation in the
Chicago area. This will involve a thirty-day continuous test at injection rates capable or
targeting 90% mercury removal or better and then extensive testing of the concretes made from
the fly ash. We are confident, based on our prior testing, that we will be able to achieve this, but
we will know for sure at full-scale at the end of the summer. That product will be available to
plants in Illinois that will allow them to continue to sell their fly ash.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
6
Plants That Burn Southern Illinois Bituminous Coal
Activated carbon injection is also applicable to the plants that continue to burn Southern Illinois
bituminous coal. I am not aware of any large scale, long-term testing at plants that burn these
coals, so I am going to have to extrapolate from other bituminous coals that are burned in the
East.
My company has done full-scale tests of our B-PAC with various eastern bituminous coals and
had good results, albeit with somewhat higher injection rates than with subbituminous coals. For
example, we recently completed a thirty-day run with a bituminous at Progress Energy’s Boiler
#1 at their Lee Plant. These we averaged 85% mercury removal, 80% due to the sorbent, over
the month at an injection rate of 8 lbs. of sorbent per million-cubic-feet of gas.
The only Illinois boilers that I am aware of where 90% mercury removal with ACI may be an
issue are those with high sulfur trioxide in their flue gas. These could be boilers burning Illinois
coal and having selective catalytic reduction, or the co-injection of trona with PAC, may be the
lowest-cost route for these few boilers. We have also done short-term tests at Duke Power’s
Allen Plant and at Public Service of New Hampshire’s Merrimack Station. It is not anticipated
in North Carolina that they are going to require 90% mercury removal, so we have targeted about
80% mercury removal in our North Carolina tests. To achieve 90%, we would simply inject the
sorbent at a slightly higher rate.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
7
COMMERCIAL AVAILABILITY OF ACTIVATED CARBON INJECTION
Is activated carbon injection (ACI) technology commercially-available today? Of course it is.
Our silo system subcontractor, for example, has installed 40 such systems for mercury reductions
at waste incinerators and they have operated reliably for many years.
Moreover, these systems and sorbents are commercially available today for power plants as well,
and from multiple suppliers, so they will certainly be available for installation before 2009.
Two things are required for ACI: first, the silo/feeder systems to inject the carbons, and second,
the carbons themselves.
Silo/Feeder/Injection Systems
The capital equipment – the silo/feeder/injection systems – are simple and tiny compared to
typical utility air pollution control equipment. They store the carbon, which is delivered by bulk
truck, and meter it through a pipe to the ductwork, where it is blown into the flowing flue gas
through simple open-ended lances.
Such low-tech equipment is at least fifty years old and can probably be supplied by hundreds of
U.S. companies if necessary. Working with our experienced silo/feeding system subcontractor,
Sorbent Technologies has begun bidding on the installation of such systems for a number of
utilities. A competitor of ours, ADA-ES, has already accepted orders for eight of these systems
at five power plants and is currently in the process of supplying them.
On a non-expedited basis, it takes about six months from order to actual system operation at the
plant site. Little to no source specialized labor is required. The silo/feeder systems are
constructed off-site and simply installed on a concrete platform at the utility site. They can be
installed while the plant is still operating.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
8
Activated Carbon Sorbents
The activated carbons are commercially available in bulk as well. Three large U.S. activated
carbon manufacturers are Norit, Calgon, MeadWestvaco, who each supply tens of thousands of
tons of sorbents annually to many different industries. Smaller carbon producers and importers
of foreign-produced carbons also supply these materials. My company's plant in Ohio
brominates carbons that we purchase from others and its capacity can be easily expanded to meet
any future increases in demand.
Characteristics of Activated Carbon Injection Technologies
In addition to their low costs, there are numerous advantages to activated carbon injection for
mercury control at power plants.
- Installation times are quick
- No specialized trade labor is required
- As most of the costs are in the sorbents actually consumed, costs are proportional to the
level of actual plant operations, so smaller plants or peaking plants are not disadvantaged.
- The same capital equipment can be used as technology advances and better sorbents are
developed.
- Because so little capital equipment is needed, there are no sunk costs to be wasted if a scrubber
is installed later. Mercury reductions can be started today.
SUMMARY
The conclusion of my testimony is that the technologies for mercury removal for the plants in
Illinois are commercially available today and will be even more available and lower in cost in
2009. Sorbent Technologies Corporation and other companies have developed a number of
power plant mercury emission reduction technologies which have been demonstrated to achieve
90% mercury removal at full-scale at numerous power plants around the country. We can
commercially supply, today, the sorbents and systems necessary for easily-retrofitable power
plant mercury reductions in Illinois.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
TESTIMONY OF JEFFREY W. SPRAGUE
My name is Jeffrey W. Sprague. I am an Environmental Protection Specialist and
Assistant Modeling Unit Manager for the Air Quality Planning Section, Division of Air
Pollution Control, Illinois Environmental Protection Agency. I have been employed with the
Agency since 1988. I graduated from Western Washington University with a Bachelor of
Science degree in Geology and minors in chemistry and environmental studies. I have pursued
graduate studies at Western Washington University in geology and the University of Illinois at
Champaign-Urbana in geology and soil science. My duties at IEPA have included emissions
processing for ozone and PM2.5 SIP development, emission inventory preparation, dispersion
modeling in support of Agency permit activity and PM-10 SIP development, enhancement of
Agency air toxics modeling, ecological risk assessment reviews, and receptor modeling
capabilities, and addressing agriculture-related air quality issues. I have served on the
Agency’s Mercury Workgroup, and I currently represent the Agency on the Great Lakes
Commission’s Great Lakes Air Deposition Program Management Team.
The discussion of mercury impacts on human health in Section 3.0 of the Technical
Support Document for Reducing Mercury Emissions from Coal-Fired Electric Generating
Units (AQPSTR 06-02) is a synthesis of statements from Michigan’s Mercury Electric Utility
Workgroup Final Report on Mercury Emissions from Coal-Fired Power Plants (June 20, 2005)
and those from a report prepared for the Illinois EPA by Deborah C. Rice, Ph.D. entitled
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
2
Review of the Nervous System and Cardiovascular Effects of Methylmercury Exposure
(March, 2006). It is generally acknowledged that mercury in various forms can induce toxic
responses in the human body, however, methylmercury exposure resulting from fish ingestion
poses the greatest exposure risk to human beings. Historical acute exposure incidents as well as
evidence from low level exposures have provided information on the symptoms and
neurological effects of methylmercury poisoning. Such effects include sensory impairment,
mental deficits, muscle weekness, tremor, hypersensitive reflexes and other neuropathological
manifestations. Longitudinal prospective epidemiological studies for populations in New
Zealand, the Seychelles Islands, and the Faroe Islands have yielded results that markedly
contrast, but which are not discordant with respect to mercury effects on IQ. An integrative
analysis of these three studies showed a decrement of 0.13 IQ point for each 1 ppm increase in
maternal hair mercury. Cross-sectional studies assessing development in children have shown
test outcomes significantly associated with mercury hair concentrations or blood mercury
levels. Similarly, cognitive function and motor function tests on adults have shown
associations with total urinary mercury, mercury in blood, and/or hair mercury content. The
impact on the human body of methylmercury exposure includes potential cardiovascular and
coronary disease. In a recent study of 2500 men in Finland, the highest measured hair mercury
concentrations were associated with increased incidences of myocardial infarction. The results
of this study also indicate that high levels of methylmercury in the body may negate the
beneficial effects of fish oils in protecting against coronary disease. The NHANES survey and
other studies intended to provide information on mercury body burdens in the U.S. population
provide evidence of a strong association between fish consumption and increased mercury
levels. For some populations, a substantial percentage of individuals have methylmercury body
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
3
burdens greater than that associated with the reference dose (0.1 micrograms/kg/day). The
reference dose represents an estimation of a daily exposure to the human population (including
sensitive subgroups) that is likely to be without appreciable risk of deleterious effects during a
lifetime. The Centers for Disease Control has estimated that approximately 6% of women of
childbearing age have blood mercury levels at or exceeding the reference dose.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
BEFORE THE ILLINOIS POLLUTION CONTROL BOARD
IN THE MATTER OF:
)
)
R06-25
PROPOSED NEW 35 ILL. ADM. CODE 225
)
(Rulemaking – Air)
CONTROL OF EMISSIONS FROM
)
LARGE COMBUSTION SOURCES (MERCURY) )
AMENDED TESTIMONY OF CHRISTOPHER ROMAINE
Qualifications
My name is Christopher Romaine. I am here today for the Illinois Environmental Protection
Agency (Agency), where I am the Manager of the Construction Unit in the Permit Section in the
Bureau of Air.
I have a Bachelor of Science degree in engineering from Brown University and have completed
coursework towards a Masters Degree in Environmental Engineering from Southern Illinois
University. I am a Registered Professional Engineer in the State of Illinois.
I joined the Agency in June 1976, at a junior level in the Permit Section in the Division of Air
Pollution Control. I am currently the Manager of the Construction Unit in the Permit Section. I
previously served as the Manager of the New Source Review Unit, Manager of the Utility Unit,
and Manager of the Joint Utility/Construction Unit, all in the Permit Section.
In particular, in 1999, I became Manager of the Utility Unit in the Permit Section, after about a
year and a half serving as the Acting Manager. As the Manager of the Utility Unit, I supervised
the permit engineers who reviewed air pollution control permit applications for electric power
plants, including applications for proposed new power plants, construction permit applications for
projects at existing power plants, and applications for operating permits for power plants.
My involvement in the permitting of coal-fired power plants has continued to the present day.
When the Joint Utility-Construction Unit was formed in 2001, consolidating the separate Utility
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
2
and Construction Units, I continued to supervise the engineers who worked on power plant
applications. It is only recently, with the issuance of the Clean Air Act Permit Program (CAAPP)
Permits to Illinois’ coal-fired power plants, that formal responsibility for future work on CAAPP
applications for power plants has begun to be transferred over to the CAAPP Unit in the Permit
Section. However, the Construction Unit continues to process construction permit applications
involving power plants.
In addition to my duties related to permitting, in my tenure with the Agency, I have assisted in the
development of a number of regulatory programs for stationary sources. These programs include
Nonattainment New Source Review (NA NSR) for proposed construction projects in
nonattainment areas, Reasonable Available Control Technology (RACT) for volatile organic
material emissions for certain categories of emissions units, the Clean Air Act Permit Program
(CAAPP), and the Emissions Reduction Market System (ERMS).
The purpose of my testimony is to provide additional explanation for various provisions of the
proposed rules.
Definitions
Many of the definitions in proposed 35 IAC 225.130 are transferred over from the Clean Air
Mercury Rule (CAMR). The proposed rule also includes definitions for five terms that are not
found in CAMR: Averaging Demonstration, Gross Electrical Output, Input Mercury, Output-
Based Emission Standard, and Rolling 12-Month Basis. Definitions of these terms were
developed by the Agency to facilitate the understanding and implementation of the proposed rules.
With the exception of the definition of “Rolling 12-Month Basis,” these definitions are self-
explanatory.
The definition of “Rolling 12-Month Basis” is a particularly important definition in the proposed
rules, as it defines the compliance time period associated with the proposed emission standards.
The compliance time period for these emission standards is 12 successive months. A separate
compliance time period applies for each month, as a new month is added and the oldest month is
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
3
dropped from the 12 month long period. In other words, compliance is determined on the basis of
12-months of data, rolled on a monthly basis or a “rolling 12-month basis.”
There are two exceptions or refinements to this general approach. The first addresses months in
which a unit does not operate when compliance is being determined for an individual unit. The
proposed rules would not include a month in which such a unit does not operate in the compliance
determination. Instead, the compliance determination in subsequent months would “skip” a month
in which the unit did not operate and would be based on data from 12 months in which the unit
actually operated. This is consistent with the approach taken by USEPA to compliance
determinations for mercury emissions under the New Source Performance Standards for Electric
Utility Steam Generating Units, which are also made on a 12-month rolling basis. (Refer to 40
CFR 60.50Da(h)(2)(iii).) The second exception addresses a month in which a unit does not
operate when compliance is being determined for a group of units on an aggregate basis. When
compliance is being determined for a group of units on an aggregate basis, the proposed rule
would only skip a month in which all
units covered by the compliance demonstration do not
operate. Compliance determinations would include months in which any of the units covered by a
demonstration is operated, as is necessary so that a compliance determination is made for each
month in which any unit in the group is operated.
Incidentally, one consequence of applying the emission standards in the proposed rules on a
rolling 12-month basis is that compliance with the numerical emission standards will not be able
to be determined for the first 11 months after July 1, 2009, when the standards become “effective.”
The earliest date that the first formal determinations of compliance with these standards can occur
is July 1, 2010. This is the earliest date on which a total 12 months of data will be available for an
existing unit or group of units, for the period from July 1, 2009 through June 30, 2010, from which
compliance with the emission standards can be numerically determined.
Compliance Requirements
The compliance requirements set forth in proposed 35 IAC 225.210 summarize the obligations
that would apply to the operation of a unit under the proposed rules. This Section reflects the
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
4
approach to such matters in CAMR, with the specific provisions paralleling the relevant
provisions in CAMR.
Permitting Provisions
The permitting requirements set forth in proposed 35 IAC 225.220 address the implications of the
proposed rules for permitting. Like the compliance requirements in proposed 35 IAC 225.210,
this Section is based on provision found in CAMR. However, unlike CAMR, proposed 35 IAC
225.220 would not require sources to obtain separate mercury budget permits for units subject to
the proposed rules. The requirements of the proposed rules would be addressed in other permits
that are already required for the units, most commonly CAAPP Permits, along with the other
applicable emission standards and requirements that apply to the units. As related to control of
mercury emissions, these permits and the applications for such permits would have to address the
informational and procedural requirements that USEPA has determined are appropriate for these
applications and permits under CAMR. For example, the applications for such permits must
include the Identification Number assigned to the source by the federal Office of Regulatory
Information Systems, the identification of the units at the source, the intended approach to
required monitoring under the rules, and the intended approach to compliance with the emissions
standards under the rules.
Emission Standards for Units at Existing Sources
Proposed 35 IAC 225.230 sets forth the mercury emission standards for units at existing sources.
At a basic level, these standards require that mercury emissions from an existing unit either not
exceed 0.0080 lb mercury/GWh gross electrical output or be controlled by a minimum of 90-
percent reduction from the input mercury contained in the coal that is burned in the unit. The
proposed rules have an emission standard expressed in terms of electrical output, as well as an
emission standard expressed in terms of the reduction in input mercury, to specifically address
units that are burning washed Illinois basin coal, as has been discussed in detail in the testimony of
other Agency witnesses. These standards take effect beginning July 1, 2009. However, as
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
5
already explained, because the standards apply on a rolling 12-month basis, the earliest date that a
formal determination of compliance could be made with these standards would be July 1, 2010.
Proposed 35 IAC 225.230 becomes more involved because it addresses how these basic standards
would actually be applied to a unit or units at source under difference circumstances. It also
includes specific equations setting forth how relevant data is to be handled for the purpose of
determining compliance with the applicable emission standards.
The simplest circumstance is that compliance with the emission standards is demonstrated for an
individual unit and
the applicable emission standard does not change during the particular 12-
month rolling period. In such case, as addressed by proposed 35 IAC 225.230(a), compliance can
be directly determined in the terms of the applicable standard. If the unit is complying with the
output based standard, the total emissions of mercury during each 12-month rolling compliance
time period would be divided by the total gross electrical output during the same period. The
result would be the actual mercury emission of the unit, in lbs per GWh, which would then be
compared to the applicable standard, 0.0080 lb/GWh. If the unit is complying with the emission
reduction standard, the total emissions of mercury during the 12-month rolling compliance time
period would be compared to the total amount of input mercury during the same period. The
result would be expressed in terms of the reduction efficiency for input mercury, in percent, which
would then be compared to the applicable standard, 90 percent.
The next circumstance, as addressed by proposed 35 IAC 225.230(b), is that compliance with the
emission standards is demonstrated for an individual unit and the applicable emission standard
changes during the particular 12-month rolling period. This circumstance must be considered
because a source could switch the coal supply to a unit during a 12-month period, going to Illinois
basin coal from western coal or vice versa. This would also likely also trigger a shift in the
emission standard with which the unit elects to comply. In this circumstance, it is not possible to
directly determine compliance in terms of an applicable emission standard, since both standards
applied during the 12-month period. To address this circumstance, the compliance of the unit
would be evaluated by comparing the actual emissions of mercury for the 12-month rolling period
to the allowable emission of mercury during the same period. If the actual emissions for the 12-
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
6
month period are equal to or less than the allowable emissions for the 12-month period, a unit
would be in compliance. This restructuring of the compliance demonstration does not effect the
determination of actual emissions. However, the allowable emissions must be calculated. For a
month in which the source elected to comply with the output based standard, the allowable
emissions would be calculated as the product of the gross electrical output for the month and the
output based standard, i.e., 0.008 lb/GWh. For a month in which the source elected to comply
with the reduction standard, the allowable emissions would be calculated as the product of the
input mercury and the allowed emissions after the required reduction, i.e., 10 percent or 0.10. The
allowable emissions for the 12-month rolling period are the sum of the allowable emissions for
each of the 12 months in the rolling period.
The next circumstance, as addressed by proposed 35 IAC 225.230(c), is that compliance with the
emission standards must be demonstrated for two or more units at a plant as a group
because
mercury emission data is only available for the units as a group. This circumstance occurs
because certain units share common stacks. For these units, mercury emission monitoring will be
most reliably and economically conducted with a single monitoring system for the emissions of
mercury from the common stack. This is the approach that is currently being used for units with
shared stacks for monitoring for emissions of sulfur dioxide (SO
2
) and nitrogen oxide (NOx)
under the Acid Rain and NOx Trading Programs. For units that are served by a single emissions
monitoring system in a common stack, it will be necessary to approach the physically distinct
units as if they were a single unit for the purpose of demonstrating compliance. Given the need to
combine data for gross electrical output or input mercury from two or more units, it will again be
simpler if this compliance demonstration is made by comparing the actual and allowable
emissions of mercury during each 12-month running compliance time period.
The last circumstance, as addressed by proposed 35 IAC 225.230(d), is that a source elects to
demonstrate compliance on a source-wide basis. In this circumstance, the compliance
demonstration must be carried out by comparing the total actual and allowable emissions of
mercury of all the units at the source for each 12-month rolling compliance time period. The
proposed rules further specify that if a source has elected to show compliance on a source-wide
basis and fails to show compliance, all units shall be considered to be out of compliance for the
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
7
final month of the 12-month rolling period. This approach was taken because a single party, the
owner or operator, will be responsible for all units covered by the compliance demonstration. It is
not necessary to specifically assign culpabity for noncompliance to a particular unit at the source,
and indeed noncompliance can be broadly attributed to the collective performance of the units.
Averaging Demonstrations for Existing Units
Proposed 35 IAC 225.232 provides the additional flexibility that is present in the proposed rules
for Phase 1, through December 31, 2013. (In fact, when one considers the consequences of the
12-month rolling compliance time period, this additional flexibility is only fully available through
January 31, 2013.) This additional flexibility is present during Phase 1 as companies with
existing sources that meet a minimum criterion for control of mercury emissions can also show
compliance with the proposed standards using system-wide compliance demonstrations, which are
referred to as “averaging demonstrations” in the proposed rules. These compliance
demonstrations would include units at more than one source. Compliance would be determined by
comparing the total actual and allowable emissions of mercury of all the units covered by the
averaging demonstration for each 12-month rolling compliance time period.
While averaging demonstrations can be used through December 31, 2013, companies will have to
be taking any actions that are needed to show compliance on a source-wide basis, without reliance
on an averaging demonstration, well before December 31, 2013. This is because the first 12-
month rolling period after averaging demonstrations cease to be available will actually cover the
12-month period from February 1, 2012 through January 31, 2013.
The minimum requirement for a source to be included in an averaging demonstration is that the
source by itself must also comply with one of the following emission standards on a source-wide
basis for the period covered by the demonstration: (1) An emission standard of 0.020 lb
mercury/GWh gross electrical output; or (2) A minimum 75-percent reduction of input mercury.
This requirement assures that technology for control of mercury emissions is utilized on each
source, and most likely each unit, that is covered by a multi-source compliance demonstration.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
8
Companies with more than one source can only participate in averaging demonstrations that
include other sources that they own or operate. Companies or organizations with only a single
source subject to the proposed rules (i.e., City, Water, Light & Power, City of Springfield;
Electric Energy, Inc.; Kincaid Generating Station; and Southern Illinois Power
Cooperative/Marion Generating Station) can only participate in demonstrations with other
companies or organizations that only have a single source. In addition, for a company or
organization with only a single source, participation in averaging demonstrations must be
authorized through federally enforceable permit conditions for each source participating in the
demonstration. This is intended to assure that the role and the responsibilities of the different
entities involved in the averaging demonstration are well defined.
If averaging is used to demonstrate compliance, the effect of a failure to demonstrate compliance
will be that the compliance status of each source will be determined as if the plant were not
covered by an averaging demonstration.
Existing Units Scheduled for Permanent Shutdown
Proposed 35 IAC 225.235 contains provisions that would allow a source to obtain an exemption
from the proposed emission standards for an existing unit that will be permanently shut down soon
after July 1, 2009. This exemption was included in the rules because the cost for installation of
activated carbon injection systems were not believed to be warranted for units that would shortly
be permanently shut down. A unit for which such an exemption had been obtained could not be
included when determining whether any other units at a source or other sources are in compliance
with the proposed emission standards under a source-wide compliance demonstration or averaging
demonstration.
This exemption specifically responds to the circumstances of City Water, Light and Power
(CWLP), the municipally owned and operated power supply for the residents of the City of
Springfield. CWLP has submitted a construction permit application to build a new coal-fired
generating unit, Dallman Unit 4, at its existing power plant adjacent to Lake Springfield. The new
unit is being developed to meet the future power needs of Springfield. The new unit would also
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
9
replace the two Lakeside Units at the plant, which are the oldest units now at the plant. The
Lakeside Units would be permanently shut down after new Dallman 4 is constructed and
operational. As part of the air pollution control permitting for new Dallman Unit 4, CWLP is
relying upon decreases in emissions from the shut down of the Lakeside Units
Depending on the circumstances surrounding the shut down unit, different dates are proposed for
when existing units must be scheduled to be shut down to qualify for exemption from the emission
standards, For existing units for which the owner operator is constructing a new unit or other
generating units to specifically replace the existing units, as is occurring with CWLP’s two
Lakeside Units, the existing units must be scheduled to be shut down by December 31, 2011. For
existing units for which the owner or operator is not constructing a new unit or other generating
units to specifically replace the existing units, the existing units must be scheduled to be shut
down a year earlier, by December 31, 2010 to qualify for the exemption. This distinction was
made because construction of a new generating unit is a challenging and lengthy undertaking. The
source that is replacing a unit should be provided more time to carry out the construction of a
replacement unit or other new generating capacity than the source that is only retiring a unit and
relying on other existing units, which are already operating, to make up for the shut down unit.
This distinction also recognizes the significant efforts and commitment of a source that is actively
engaged in developing a replacement unit or developing other new generating capacity.
The exemption does include a provision for unforeseen events that would delay the scheduled shut
down of an exempted unit. A source must permanently shut down the unit by the applicable
deadline unless the source submits a demonstration to Agency before such date showing that
circumstances beyond its reasonable control (such as protracted delays in construction activity for
the new replacement units, unanticipated outage of another unit, or protracted shakedown of a
replacement unit) have occurred that interfere with the plan for permanent shut down of the
existing unit. In such circumstances, the deadline for shut down of the existing unit may be
extended for up to one year if the unit is not being replaced or up to 18 months if the unit is being
replaced, provided, however, that after December 31, 2012, the existing unit shall only operate as
a back-up unit.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
10
Procedural requirements accompany the exemption to ensure that reliance upon the exemption is
carefully considered by sources and occurs in a timely manner and that that the exemption is not
misused by sources. Before the effective date of the proposed emission standards, i.e., by June 30,
2009, a source must notify the Agency that it is planning to permanently shut down the unit. In
addition, the source must have applied for a construction permit or be actively pursuing a federally
enforceable agreement that requires the unit to be permanently shut down as needed to qualify for
the exemption. The source must also have applied for revisions to the operating permit(s) for the
unit to include provisions that terminate the authorization to operate the unit in a manner
consistent with the exemption. By July 1, 2010, the earliest date that compliance with the
proposed numerical emission standards can be determined, the requirement to permanently shut
down a source must be embodied in a federally enforceable permit or other enforceable
agreement.
Lastly, the exemption specifies that if a unit for which the exemption is relied upon is not shut
down in a timely manner, the unit shall thereafter be considered a new unit for the purposes of the
proposed rules. This is a final measure to address any possible abuse of the exemption.
Emission Standards for New Sources
Proposed 35 IAC 225.237 sets forth the mercury emission standards for units at new sources.
While the standards are numerically identical to those for units at existing source, units at new
sources must comply with applicable emission standards on an individual, unit-by-unit basis.
Unlike the provisions for units at existing sources, compliance cannot be shown on a source-wide
basis or, as allowed during Phase 1 of the proposed rules, with an averaging demonstration. This
is appropriate because the units at new sources will be controlled with Best Available Control
Technology (BACT) under the federal rules for Prevention of Significant Deterioration (PSD) 40
CFR 52.21. As a result, control of mercury emissions through co-benefit will be maximized.
Activated carbon injection systems can also be installed on these units as original equipment, so as
to maximize the capabilities of this technology to control mercury emissions.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
11
The effective date of the emission standards for new units is set to match the effective date of the
mercury emission standard that would also apply to a new units under the federal New Source
Performance Standards (NSPS) for Electric Utility Steam Generating Units, 40 CFR 60.45Da.
This was done for administrative convenience, so that the compliance determinations under the
NSPS and proposed rules for new units start at the same time and continue on the same schedule.
Since the emission standard of the NSPS does not immediately become effective upon initial
startup of a new unit, this approach also allows time for the shakedown of the new unit as related
to the control of mercury emissions and the orderly shakedown and certification of the continuous
emissions monitoring system for mercury. During the period before the emission standards take
effect, a source is under the general obligation to operate a unit in accordance with good air
pollution control practices to minimize emissions.
Emissions Monitoring
As explained in the Technical Support Document, the requirements in the proposed rule for
monitoring of mercury emissions are essentially identical to the monitoring that would be required
under CAMR. Accordingly, the proposed rule refers to the provisions for emissions monitoring
adopted by USEPA at 40 CFR Part 75, Subpart I, Hg Mass Emission Provisions, and 40 CFR Part
75, Appendix K, Quality Assurance and Operating Procedures for Sorbent Trap Monitoring
Systems. The proposed rule also allows use of the excepted “low mass” monitoring methodology,
as adopted by USEPA at 40 CFR 75.81(b), for units that have annual emissions of no more than
29.0 pounds of mercury. Other aspects of the proposed rule related to emissions monitoring are
also consistent with provisions of the CAMR. For example, the owner or operator of an existing
unit must begin monitoring for mercury emissions no later than January 1, 2009, as would be
required under CAMR. Emissions monitors must be certified and generally operated as would be
required under CAMR. Any alternative emission monitoring methods must be approved by
USEPA. The technical feasibility of the required monitoring for mercury emissions was
addressed by USEPA as part of its rulemaking to adopt the CAMR. (For example, refer to 70 FR
28633, May 18, 2005.)
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
12
Other Monitoring Requirements
The proposed rule also includes monitoring requirements that are not directly related to mercury
emissions but are necessary to determine compliance with the proposed emission standards. The
owner or operator of a unit complying with the output-based emission standard would be
“required” to conduct operational monitoring for the electrical output from the unit, as measured at
the generator. Since accurate information on electrical output is already needed by a source for
operational reasons and this data is readily collected by watt meters designed for this specific
purpose, the proposed rule does not specify particular monitoring methodology that must be used
for the collection of this data. The rule also does not require that this data be collected until this
data will be needed to determine compliance.
The owner or operator of a unit complying with the 90 percent reduction standard would be
required to conduct analyses of the coal being burned in the unit to determine its mercury content.
This information would then be used to determine the amount of mercury in the coal going into
the unit so that the mercury removal efficiency achieved by the control devices on the unit could
be calculated, as related to the 90 percent reduction standard. While most sources that will be
subject to this proposed rule already collect and analyze coal samples on a routine basis for
operational reasons, this activity does not extend to analysis for mercury content. The provisions
for coal sampling in the proposed rule are intended to ensure an accurate determination of the
input mercury to the subject units. Since the mercury content of coal varies, even when coming
from a single mine and coal seam, and the amount of coal consumed by an unit can vary from day
to day, daily sampling of the coal supply to units is necessary. The coal supply must be sampled
at a point after long-term storage, where the sample will be representative of the coal being burned
in the unit on the day that the sample is taken. This location for coal sampling was selected after
consultation with industry representatives to provide flexibility in the point at which samples are
collected while ensuring that the resulting data accurately reflects the coal that is actually being
burned in a unit.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
13
Certain ASTM Methods were selected for the required analyses of coal. For mercury, these are
ASTM D6414-01, “Standard Test Method for Total Mercury in Coal and Coal Combustion
Residues by Acid Extraction or Wet Oxidation/Cold Vapor Atomic Absorption,” and ASTM
D3684-01, “Standard Test Method for Total Mercury in Coal by the Oxygen Bomb
Combustion/Atomic Absorption Method.” These methods were chosen by the Agency after
consultation with industry representatives and experts on coal analysis because these methods are
accurate, sources and commercial laboratories are familiar with these methods, and the costs of
these methods are reasonable.
The proposed rule would require that a source begin collecting and analyzing coal samples at least
30 days before data is needed to determine compliance, if it is reasonably possible to do so. This
approach was taken to reasonably ensure that data is being properly collected when it is finally
needed for the purpose of determining compliance. However, if a unit has been out of service,
such that meaningful sampling and analysis of the coal supply could not be conducted in advance
of the time at which data is needed, a source must begin to conduct this monitoring when the unit
is returned to service.
Conclusion
In conclusion, the provisions of the proposed rules have been carefully considered and developed
to carry out the objectives for the proposed rulemaking.
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006
STATE OF ILLINOIS
)
)
SS
COUNTY OF SANGAMON
)
)
CERTIFICATE OF SERVICE
I, the undersigned, an attorney, state that I have served electronically the attached
TESTIMONY OF DAVID C. FOERTER, EZRA D. HAUSMAN, Ph.D., ROBERT J.
KALEEL, SID NELSON JR., and JEFFREY W. SPRAGUE, and AMENDED
TESTIMONY OF CHRISTOPHER ROMAINE, upon the following person:
Dorothy Gunn
Clerk
Illinois Pollution Control Board
James R. Thompson Center
100 West Randolph St., Suite 11-500
Chicago, IL 60601-3218
and mailing it by first-class mail from Springfield, Illinois, with sufficient postage affixed
to the following persons:
SEE ATTACHED SERVICE LIST
ILLINOIS ENVIRONMENTAL
PROTECTION AGENCY,
__________________________
Gina Roccaforte
Assistant Counsel
Division of Legal Counsel
Dated: April 28, 2006
1021 North Grand Avenue East
Springfield, Illinois 62794-9276
(217) 782-5544
ELECTRONIC FILING, RECEIVED, CLERK'S OFFICE, APRIL 28, 2006