United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued November 16, 1999 December 21, 1999
No. 98-1525
Lignite Energy Council, et al.,
Petitioners
v.
U.S. Environmental Protection Agency,
Respondent
Natural Gas Supply Association, et al.,
Intervenors
Consolidated with
98-1529, 98-1533, 98-1541, 98-1543
On Petitions for Review of an Order of the
Environmental Protection Agency
F. Willam Brownell and William F. Pedersen argued the
cause for petitioners. With them on the briefs were Craig S.
Harrison, Jeffrey A. Knight, Harold P. Quinn, Jr., Gene E.
Godley, Scott H. Segal, Brian R. Bjella, and Charles S.
Miller, Jr.
Heidi Heitkamp, Attorney General, State of North Dakota,
and Carmen Miller, Assistant Attorney General, were on the
brief for amicus curiae the State of North Dakota.
Wendy L. Blake, Attorney, U.S. Department of Justice,
argued the cause for respondent. With her on the brief was
Lois J. Schiffer, Assistant Attorney General.
Armond M. Cohen was on the brief for amicus curiae
Conservation Law Foundation, et al.
John H. Sharp, Michael R. Barr and Michael A. Conley
were on the brief for intervenors.
Before: Edwards, Chief Judge, Silberman and Henderson,
Circuit Judges.
Opinion for the Court filed Per Curiam.
Per Curiam: Petitioners challenge EPA's new source per-
formance standards for nitrogen oxides emissions from utility
and industrial boilers. We conclude that EPA did not exceed
its discretion under section 111 of the Clean Air Act in
promulgating these standards, and therefore deny the peti-
tions.
* * * *
Fossil-fuel fired steam generating units ("boilers") emit
nitrogen oxides (NOx), air pollutants that can cause deleteri-
ous health effects and contribute to the formation of acid rain.
Section 111 of the Clean Air Act requires EPA to establish
performance standards for the emission of NOx from newly
constructed boilers; these "new source performance stan-
dards" are to be set at a level that
reflects the degree of emission limitation achievable
through the application of the best system of emission
reduction which (taking into account the cost of achieving
such reduction and any nonair quality health and envi-
ronmental impact and energy requirements) the Admin-
istrator determines has been adequately demonstrated.
42 U.S.C. s 7411(a)(1). In its 1990 Clean Air Act Amend-
ments Congress specifically directed EPA to exercise its
section 111 authority and establish new NOx standards that
incorporate "improvements in methods for the reduction of
emissions of oxides of nitrogen." 42 U.S.C. s 7651f(c)(1).
In response to these statutory mandates, EPA promulgated
a rule lowering its NOx new source performance standards to
.15 lb/MMBtu (pounds of NOx emitted per million BTU
burned) for utility boilers1 and .20 lb/MMBtu for industrial
boilers. See 63 Fed. Reg. 49,442, 49,443 (1998) (to be codified
at 40 C.F.R. pt. 60). These standards reflect the level of NOx
emissions achievable by what EPA considers to be the "best
demonstrated system" of emissions reduction: the use of
selective catalytic reduction (SCR) in combination with com-
bustion control technologies.2 Petitioners' central claim is
that EPA selected SCR as the basis for its NOx standards
without properly balancing the factors that section 111 re-
quires it to "take into account." Because section 111 does not
__________
1 To be precise, the emission standard for utility boilers is an
output-based standard of 1.6 pounds of NOx emitted per megawatt-
hour of electricity generated. However, as this output-based stan-
dard was intended by EPA to correlate with a .15 lb/MMBtu input-
based standard, we refer to its input-based equivalent for simplici-
ty's sake throughout this opinion. We reject petitioners' argument
that EPA's decision to shift to an output-based standard for utility
boilers unfairly "penalizes" the use of low-energy coals, like lignite;
it would seem just as easy to argue that an input-based standard
"penalizes" high-energy fuels.
2 SCR is a "flue gas treatment technology"; it reduces NOx
after combustion by injecting ammonia into the flue gas in the
presence of a catalyst, breaking down NOx and producing nitrogen
and water. In setting past standards, EPA had focused solely on
combustion control technologies, which instead reduce NOx by
suppressing its formation during the combustion process. See 62
Fed. Reg. 36,948, 36,949-50 (1997).
set forth the weight that be should assigned to each of these
factors, we have granted the agency a great degree of discre-
tion in balancing them, see, e.g., New York v. Reilly, 969 F.2d
1147, 1150 (D.C. Cir. 1992); EPA's choice will be sustained
unless the environmental or economic costs of using the
technology are exorbitant. See National Asphalt Pavement
Ass'n v. Train, 539 F.2d 775, 786 (D.C. Cir. 1976).
Petitioners argue that SCR is not the "best demonstrated
system" under section 111 because the incremental cost of
reducing NOx emissions is considerably higher with SCR than
with combustion controls. Recent improvements in combus-
tion controls will enable many boilers to attain emissions
levels close to EPA's SCR-based standards; accordingly,
petitioners assert that EPA should have based its standards
on these less expensive technologies. However, in light of
EPA's unchallenged findings showing that the new standards
will only modestly increase the cost of producing electricity in
newly constructed boilers, see 62 Fed. Reg. 36,948, 36,958
(1997) (proposed NOx revisions), we do not think that EPA
exceeded its considerable discretion under section 111.
Moreover, petitioners' argument stressing the comparable
environmental merits of advanced combustion controls is to a
certain extent self-defeating, since the new source perfor-
mance standards set by EPA are not technology-forcing, and
continuing advances in combustion control technologies will
reduce the amount of NOx reduction that must be captured by
the more expensive SCR technology.
It was also within EPA's discretion to issue uniform stan-
dards for all utility boilers, rather than adhering to its past
practice of setting a range of standards based on boiler and
fuel type. See, e.g., 44 Fed. Reg. 33,580 (1979) (establishing
varying NOx emissions standards for utility boilers). Peti-
tioners recognize that EPA is not required by law to subcate-
gorize--section 111 merely states that "the Administrator
may distinguish among classes, types, and sizes within cate-
gories of new sources," 42 U.S.C. s 7411(b)(2) (emphasis
added)--but argue that it was arbitrary and capricious for
EPA to decline to do so. EPA explains that its change to
uniform standards is justified by SCR's performance charac-
teristics: Unlike the technologies on which past new source
performance standards were based, flue gas treatment tech-
nologies like SCR limit NOx emissions after combustion, and
the effectiveness of SCR is thus far less dependent upon
boiler design or fuel type. Petitioners respond that there are
reasons to expect SCR to perform less adequately on boilers
burning high-sulfur coals, but EPA collected continuous emis-
sions monitoring data on two high-sulfur coal-fired utility
boilers that showed that the .15 lb/MMBtu standard was
achievable, and supplemented this study with similar evidence
from foreign utility boilers. EPA also considered petitioners'
concerns about the impact of alkaline metals on the perfor-
mance of the catalyst used in the SCR process, and concluded
that such "catalyst poisoning" is not a significant problem in
coal-fired boilers. See 63 Fed. Reg. at 49,445. Mindful of the
high degree of deference we must show to EPA's scientific
judgment, see, e.g., Appalachian Power Co. v. EPA, 135 F.3d
791, 801-02 (D.C. Cir. 1998), we accept these determinations
and sustain EPA's uniform standard for utility boilers.
Petitioners offer a broader challenge to EPA's .20 lb/
MMBtu standard for industrial boilers, claiming that SCR is
not "adequately demonstrated" for any coal-fired industrial
boilers. EPA was unable to collect emissions data for the
application of SCR to these boilers, but this absence of data is
not surprising for a new technology like SCR, nor does it in
and of itself defeat EPA's standard. Because it applies only
to new sources, we have recognized that section 111 "looks
toward what may fairly be projected for the regulated future,
rather than the state of the art at present." Portland
Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 391 (D.C. Cir.
1973). Of course, where data are unavailable, EPA may not
base its determination that a technology is adequately demon-
strated or that a standard is achievable on mere speculation
or conjecture, see, e.g., National Asphalt Pavement Ass'n,
539 F.2d at 787, but EPA may compensate for a shortage of
data through the use of other qualitative methods, including
the reasonable extrapolation of a technology's performance in
other industries. See, e.g., Weyerhaeuser Co. v. Costle, 590
F.2d 1011, 1054 n.70 (D.C. Cir. 1978).
EPA has done precisely that here, concluding from its
study of utility boilers that SCR is "adequately demonstrat-
ed" and the .20 lb/MMBtu standard is "achievable" for coal-
fired industrial boilers as well. Utility and industrial boilers
are similar in design and both categories of boilers can attain
similar levels of NOx emissions reduction through combustion
controls, which means that SCR will be required to capture
comparable quantities of NOx for both boiler types. While
petitioners argue that SCR is less likely to be effective on
industrial boilers because they have widely fluctuating load
cycles, EPA has shown that SCR can be successfully applied
to coal-fired utility boilers under a "wide range of operating
conditions" including those analogous to the load cycles of
industrial boilers. 63 Fed. Reg. at 49,444. We think that it
was reasonable for EPA to extrapolate from its studies of
utility boilers in setting an SCR-based new source perfor-
mance standard for coal-fired industrial boilers.3
We also sustain EPA's application of the .20 lb/MMBtu
standard to combination boilers, which simultaneously com-
bust a mixture of fuels. The preexisting NOx emissions
standards established a range of values for combustion boil-
ers that varied by fuel type: while combination boilers burn-
ing natural gas with non-coal solid fuels (e.g., wood) were
subject to a .30 lb/MMBtu standard, the performance stan-
dards for combination boilers combusting coal with oil or
natural gas were determined based upon the proportion of
the boiler's total heat input provided by each fuel. See 51
Fed. Reg. 42,768, 42,790 (1986). It is difficult to understand
petitioners' objection to the application of the industrial boiler
standard to boilers burning natural gas and wood. A reduc-
tion of that standard from .30 to .20 lb/MMBtu is perfectly
__________
3 For similar reasons, we do not think that EPA's lack of data
on domestic SCR applications to boilers burning lignite renders its
standards unlawful. In assessing a new technology like SCR, EPA
is not required to provide evidence of its application to boilers
burning every type of coal from every geographical location. It is
acceptable for EPA to extrapolate from the successful applications
of SCR to domestic high-sulfur coal-fired boilers and to foreign
boilers burning lignite.
reasonable in light of the significant advances in NOx emis-
sions technology since 1986; indeed, EPA studies show that
wood-fired boilers can reach emissions levels far lower than
.20 lb/MMBtu through the application of flue gas treatment
technologies. And our conclusion that the .20 lb/MMBtu
standard is achievable for boilers burning only coal necessari-
ly defeats petitioners' objection that the industrial boiler
standard is unreasonable as applied to combination boilers
burning coal simultaneously with other fuels with lower NOx
emissions characteristics.
Petitioners' final objection is to EPA's valuation of steam
energy produced by "cogeneration facilities." EPA's adop-
tion of an output-based standard for utility boilers raised the
question of how to calculate the energy produced by these
units, which generate thermal steam energy in addition to
electrical energy. Steam energy produced by cogeneration
facilities is exported for several different industrial uses;
however, because of inefficiencies in transporting and con-
verting steam, only a fraction of steam energy produced by
cogeneration facilities is actually used in the industrial pro-
cess. EPA resolved this problem by assigning a 50% credit
for steam energy when determining a cogeneration unit's
output. See 63 Fed. Reg. at 49,447. Petitioners describe this
credit as an arbitrary and capricious "discounting" of steam
energy's value, but it just as easily could be called a subsidy:
The maximum efficiency for the conversion of steam to elec-
trical energy is only 38%, and EPA's final rule justifies the
50% credit on the ground that it will encourage cogeneration.
Id. In light of the difficulties that would attend calculating
the useful energy of steam heat produced by cogeneration
facilities on a unit-by-unit basis, we conclude that EPA's
resolution of this issue was acceptable.
The petitions for review are denied.
So ordered.