AES Puerto Rico L.P. (Board decision, May 27, 1999)

Plain-English summary

AES Puerto Rico sought to build a 454-megawatt coal-fired power plant in Guayama, Puerto Rico, under a PSD permit issued by EPA Region II. Three petitions challenged the permit's sulfur-dioxide controls and modeling, fine-particulate-matter data and limits, and environmental-justice analysis. The Environmental Appeals Board found that the Region reasonably selected the emissions controls and modeling methods, properly used the available particulate-matter data, and addressed environmental-justice concerns through its analysis and permit conditions. The Board denied review, leaving the permit in place.

Decision snapshot

• Cited authorities: Clean Air Act §§ 107, 109, 165, and 169, 42 U.S.C. §§ 7407, 7409, 7475, and 7479; 40 C.F.R. §§ 50.4, 50.12, 52.21, 81.355, and 124.19.
• Outcome: Review of the PSD permit was denied in all three consolidated appeals.
• Key point: A permit challenger must show clear error, an important policy matter, or another reviewable defect to obtain EAB review of the Region's PSD analysis.

Full text (EPA EAB public release)

187-274/Sections16-18 10/15/01 3:32 PM Page 324

324
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

IN RE AES PUERTO RICO L.P.
PSD Appeal Nos. 98–29, 98–30 & 98–31

ORDER DENYING REVIEW

Decided May 27, 1999

Syllabus
This decision addresses three petitions for review filed with the Environmental Appeals Board challenging the Clean Air Act prevention of significant deterioration (“PSD”) permit issued by EPA Region II to AES Puerto Rico L.P. (“AES”) for construction of a 454 megawatt coal-fired power plant in Guayama, Puerto Rico.

Petitioners challenge aspects of the Region’s permit decision relating to sulfur diox- ide (“SO2”), fine particulate matter (“PM10”), and environmental justice.

Held: Review is denied of the petitions for review for the following reasons:

• It was appropriate for the Region to use the permitted SO2 emission rate in the SO2 air quality analysis even though the emission rate was based on a unique combination of three SO2 controls. Each of the three controls has been previously demonstrated but not used in this particular combination. The Region adequately justified its decision to require the particular combination of controls at issue and thus, it was proper to use the permit- ted emission rate in the air quality analysis. (Section II.C.1.)

• Alternative SO2 modeling results presented in Petition No. 98–29 and the critique of AES’s SO2 modeling fail to establish that the SO2 modeling conducted by AES involved clear error of fact or law or an important policy matter that warrants review. (Section II.C.2.)

• The SO2 multi-source modeling analysis contained in Petition No. 98–29 was not preserved for review because the analysis was not submitted to the Region during the pub- lic comment period. (Section II.C.2.)

• The Region acted within its discretion in not requiring a multi-source air quality analysis for SO2 in that the proposed facility’s predicted SO2 impact does not exceed “sig- nificant impact level” thresholds found in Agency guidance. The Region’s decision not to require multi-source modeling in this case is supported both by established policy regard- ing significant impact level thresholds and the quality of the modeling that produced the SO2 impact estimate for the proposed facility. (Section II.C.3.)

• The Region validly applied the regulatory exemption from preconstruction mon- itoring of ambient SO2 levels in this case. An exemption was justified because the pre- dicted SO2 impacts from the proposed facility are lower than the de minimis monitoring levels established in the PSD regulations. In addition, the Region’s decision to require an

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 325

AES PUERTO RICO L.P.
325

evaluation of SO2 air quality in Guayama after permit issuance and its accompanying commitment to promptly address any actual SO2 air quality problems that may be dis- covered appropriately responds to petitioners’ concerns about the SO2 attainment status of Guayama. (Section II.C.4.)

• The Region made use of appropriate PM10 background monitoring data in the PM10 air quality compliance demonstration. Additional data identified by petitioners postdate the permit application and also have been flagged for quality assurance/quality control pur- poses. Thus, the Region’s choice of PM10 monitoring data does not indicate a clear error or an important policy matter that warrants a grant of review. (Section II.D.1.)

• The PM10 emission limit contained in the final permit decision reflects a reasonable approach on the part of the Region to ensure that the permit controls the condensible frac- tion of PM10. The final permit limit was based on the fact that little guidance was available regarding the achievability of a PM10 emission limit (that includes both condensible and non-condensible particulate matter) for this type of source. (Section II.D.2.)

• The environmental justice concerns raised in the petitions for review were ade- quately addressed by the Region during the permit process. The Region prepared an envi- ronmental justice analysis and incorporated additional conditions into the final permit deci- sion as a tangible response to the community’s concerns about air quality and to fulfill the goals of the Executive Order on environmental justice. (Section II.E).

Before Environmental Appeals Judges Ronald L. McCallum, Edward E. Reich, and Kathie A. Stein.
Opinion of the Board by Judge Stein:

EPA Region II issued a federal prevention of significant deterioration (“PSD”) permit, pursuant to Clean Air Act (“CAA”) § 165, 42 U.S.C. § 7475, to AES Puerto Rico L.P. (“AES”) on September 18, 1998, authorizing con- struction of a 454-megawatt coal-fired power plant in Guayama, Puerto Rico. The Environmental Appeals Board (“the Board”) received petitions for review of the Region’s PSD permit decision from: Dr. Jorge E.
González of the University of Puerto Rico—Mayaguez (Petition No.
98–29), Sur Contra la Contaminación (“SURCCo”), a local community organization (Petition No. 98–30), and Pedro J. Saade Llorens, on behalf of five individuals (Petition No. 98–31). The petitioners challenge aspects of the Region’s permit decision relating to sulfur dioxide (“SO2”), partic- ulate matter less than 10 micrometers in diameter (“PM10”), and environ- mental justice.

I. BACKGROUND

The proposed AES facility would be the first coal-fired power plant in Puerto Rico. The facility is designed with two circulating fluidized bed (“CFB”) boilers with a combined maximum heat input rate of 4,922.7 mil- lion British Thermal Units per hour (MMBTU/hr). The plant will produce both electricity to be sold to Puerto Rico’s electric utility and steam to be

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 326

326
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

used by local industries. The proposed facility is subject to PSD review because it is a major new stationary source of pollutants including nitro- gen oxides (“NOx”), SO2, carbon monoxide (“CO”), and fine particulate matter (“PM10”). See 40 C.F.R. § 52.21(b)(1)(i)(a) (defining major stationary source to include “fossil fuel-fired steam electric plants of more than 250 million British thermal units per hour heat input” that will emit more than 100 tons per year of any pollutant subject to PSD regulation).

PSD review is a preconstruction permitting program for major sta- tionary sources located in areas where ambient air quality meets or exceeds national ambient air quality standards (“NAAQS”).1 See In re Kawaihae Cogeneration Project, 7 E.A.D. 107, 111–12 (EAB 1997). These areas are referred to as “attainment” areas. PSD review also applies in areas that cannot be classified as either “attainment” or “non-attainment” areas. Such areas are termed “unclassifiable” areas. Kawaihae, 7 E.A.D. at 112. Guayama, Puerto Rico is considered to be in attainment for SO2, unclassifiable/attainment for CO and nitrogen dioxide (“NO2”), and unclassifiable for PM10. 40 C.F.R. § 81.355. Therefore, a PSD permit is required before a facility such as the proposed AES facility may be con- structed. Two of the most critical elements of the PSD permit process are:
1) the requirement that emissions of certain pollutants be controlled by “best available control technology” (“BACT”)2 and 2) that an air quality analysis be conducted to determine whether a proposed project would cause or contribute to exceedances of NAAQS or PSD increments.3 See In re Knauf Fiber Glass, GmbH, 8 E.A.D. 121, 123–24 (EAB 1999).

The PSD permit issued to AES by the Region contains BACT limits on SO2 emissions by requiring a combination of three control strategies:
1) CFB boilers with limestone injection, 2) low sulfur coal (maximum sul- fur content of 1.0%), and 3) an add-on dry scrubber. See Permit ¶¶ VI.1,

1
NAAQS have been established for six “criteria” pollutants: sulfur dioxide, particulate matter, carbon monoxide, ozone, nitrogen dioxide, and lead. 40 C.F.R. §§ 50.4–50.12.
2
The PSD regulations define BACT as follows:

Best available control technology means an emissions limitation * * * based on the maximum degree of reduction for each pollutant subject to regulation under [the Clean Air Act] which would be emitted from any proposed major stationary source * * * which the Administrator, on a case-by-case basis, taking into account energy, environmental, and economic impacts and other costs, determines is achievable for such source * * *.

40 C.F.R. § 52.21(b)(12).
3
PSD increments are maximum allowable increases in pollutant concentrations per- missible by regulation. See 40 C.F.R. § 52.21(c). See infra Section II.D.1 for a discussion of the PSD increments applicable in this case.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 327

AES PUERTO RICO L.P.
327

VII.2. Emissions of particulate matter and PM10 from the CFB boilers are to be controlled by an electrostatic precipitator (“ESP”). Id. ¶ VII.3. The requirements for control of NOx , CO and other pollutant emissions are not at issue in this appeal.

The petitioners’ primary focus is on the air quality analyses conduct- ed during the permitting process for SO2. All three petitioners believe that Region II should have required a full impact analysis, including multi- source modeling, for SO2. González Petition at 10; SURRCo Petition at 10;
Saade Llorens Petition at 4. Dr. González conducted alternative SO2 air quality modeling and submitted the results with his petition for review.
SURCCo and Mr. Saade Llorens distrust the air quality analysis performed by AES and the Region because it is contingent upon proper operation of the trio of SO2 emission controls specified in the permit. They point out that this combination of controls has never been tested at another facility. SURCCo Petition at 12; Saade Llorens Petition at 2. The petition- ers also question whether Guayama is indeed in attainment for SO2 and request preconstruction ambient air quality monitoring. González Petition at 10; SURCCo Petition at 23.

SURCCo also challenges the adequacy of the air quality analysis for PM10 as well as the limit on PM10 emissions from the CFB boilers. SURCCo Petition at 27, 29. SURCCo claims that local PM10 data indicate that com- pliance with the NAAQS for PM10 will be threatened by the AES facility.
Id. at 27. SURCCo also questions the BACT determination for PM10 from the CFB boilers. The numerical emissions limit was changed from the draft to the final permit, and SURRCo contends that the change consti- tutes an illegitimate increase in the PM10 emission rate. Id. at 29.

The third item at issue in this case is an environmental justice chal- lenge contained in SURCCo’s petition claiming that Region II’s permit decision for this facility does not meet the standards of the Executive Order on environmental justice.4 SURCCo Petition at 30–33. The proposed facility is to be located in an industrial area of Guayama, a city on the south coast of Puerto Rico. According to the petitioners, Guayama is a low-income community and home to several pharmaceutical and petro- chemical plants. Id. at 30–31. SURCCo believes that some of Region II’s decisions regarding the air quality analysis conducted for this facility rep- resent a failure of environmental justice. Id. at 32.

4
See Federal Actions to Address Environmental Justice in Minority and Low-Income Populations, Exec. Order No. 12,898, 59 Fed. Reg. 7629 (Feb. 16, 1994). The Executive Order is discussed infra Section II.E.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 328

328
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

At the request of the Board, Region II filed a response to the three petitions for review. U.S. EPA Region II, Response to Petitions for Review (“Region’s Response”). The Region also provided relevant portions of the administrative record, including its response to comments document. AES Puerto Rico Cogeneration Project Responsiveness Summary (“RS”). In addition, the Board granted a request by AES to file a response to the petitions for review. See AES Puerto Rico L.P., Memorandum of Law in Opposition to Petitions for Review (“AES Response”).

II. DISCUSSION

A. Standard of Review

The regulation governing this proceeding provides that the Board may grant review of a permit decision if some aspect of the decision was based on either a clearly erroneous finding of fact or conclusion of law, or if the decision involves an important matter of policy or exercise of discretion that warrants review. 40 C.F.R. § 124.19(a); see In re Knauf Fiber Glass, GmbH, 8 E.A.D. 121, 126–27 (EAB 1999). The Board is guid- ed by language in the preamble to section 124.19 that states the “power of review should be only sparingly exercised,” and “most permit condi- tions should be finally determined at the Regional level.” 45 Fed. Reg.
33,290, 33,412 (May 19, 1980); accord Knauf, 8 E.A.D at 127; In re Hawaii Elec. Light Co., 8 E.A.D. 66, 71 (EAB 1998) (“HELCO”). The peti- tioners bear the burden of establishing that review is warranted. HELCO, 8 E.A.D. at 71; In re Kawaihae Cogeneration Project, 7 E.A.D. 107, 114 (EAB 1997).

B. Timeliness

Petitions for review of a permit decision must be filed with the Board within thirty (30) days after the decision is issued by the permitting authority. See 40 C.F.R. § 124.19(a). Failure to submit a timely petition is grounds for dismissal. See, e.g., In re Envotech, L.P., 6 E.A.D. 260, 266 (EAB 1996); In re Beckman Prod. Servs., 5 E.A.D. 10, 16 (EAB 1994). In this case, Region II issued its permit decision on September 18, 1998. On September 21, 1998, hurricane “Georges” hit Puerto Rico and caused severe disruptions in power, communications, and other basic services.
Within the original thirty-day appeals period, the Board received requests from SURCCo and Mr. Saade Llorens for extensions of time in which to file petitions for review. In light of the extraordinary circumstances cre- ated by the hurricane and its aftermath, the Board granted a rare exten- sion of the appeals deadline to November 18, 1998. Order Granting

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 329

AES PUERTO RICO L.P.
329

Extension of Time (EAB, Oct. 14, 1998). The extension order specifically stated that “[p]etitions must be received by the Board on or before Wednesday, November 18, 1998.” Id. at 2 (emphasis added).

It is a petitioner’s responsibility to ensure that filing deadlines are met, and the Board will generally dismiss petitions for review that are received after a filing deadline. See, e.g., In re Kawaihae Cogeneration Project, 7 E.A.D. 107, 123 (EAB 1997); Envotech, 6 E.A.D. at 266;
Beckman, 5 E.A.D. at 15–16. The Board’s expectation of timely filings is especially heightened where a previous extension of the deadline has been granted. The Board will relax a filing deadline only where special circumstances exist. In re B&B Wrecking & Excavating, Inc., 4 E.A.D. 16, 17 (EAB 1992). In this case, Dr. González’s petition was received by the Board on November 18, 1998.5 SURCCO’s petition arrived on November 19, 1998, and the petition submitted by Mr. Saade Llorens was received by the Board on November 30, 1998.

The petition filed by SURCCo arrived one day after the deadline established by the Board due to aircraft problems experienced by Federal Express, the overnight package delivery service selected by SURCCo.6 SURCCo promptly informed the Board of the unanticipated delay and provided a copy of a letter of apology from Federal Express explaining the reason for the late delivery. Under these special circumstances, we agree to treat SURCCo’s petition as timely.

The petition filed by Mr. Saade Llorens was also affected by the Federal Express aircraft problems that delayed the SURRCo petition.
However, Mr. Saade Llorens’ petition was further delayed in that the peti- tion was addressed to Region II in New York City rather than to the Board in Washington, D.C. Once the error was discovered by Region II person- nel, the petition was forwarded to the Board. Consequently, the petition was not received by the Board (and therefore was not filed) until November 30, 1998, twelve days after the filing deadline. The reason for the error in delivery is not immediately apparent. The Region’s

5
A letter from the Clerk of the Board acknowledging receipt of Dr. González’s peti- tion for review and requesting a response from Region II erroneously stated that the peti- tion was received on November 19, 1998. The original petition was in fact received on November 18, 1998, and a second copy was received on November 19, 1998. The Board’s receipt of the original petition for review establishes the official filing date, and in this case, the petition from Dr. González was timely filed.
6
SURCCO’s petition and exhibits were tendered to Federal Express on November 17, 1998, and but for Federal Express’s internal problems, should have been delivered to the Board in a timely fashion on November 18, 1998.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 330

330
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

September 18, 1998 permit decision explains the procedures for filing a petition for review with the Board and clearly sets forth the addresses to be used for deliveries to the Board by either regular mail or “Federal Express Mail.” Further, Mr. Saade Llorens was aware that the Board has jurisdiction over this matter as he had previously requested an extension of time to file a petition for review in a letter that was properly sent directly to the Board. It appears that the incorrect address on the petition for review was an unfortunate oversight by the petitioner that nonethe- less caused a substantial delay in filing. Consistent with our rulings in other cases involving late filings, the petition for review filed by Mr.
Saades Llorens is hereby dismissed as untimely.7 See, e.g., Apex Microtechnology, Inc., EPCRA Appeal No. 93–2 (EAB, July 8, 1994) (appeal was dismissed as untimely when filing received by the Board after the filing deadline had been originally sent to a Regional Hearing Clerk in error).

C. Issues Pertaining to Sulfur Dioxide (SO2) Emissions and Air Quality Analysis

The petitions for review are largely focused on the potential impacts of SO2 emissions from the proposed AES facility. The PSD regulations require that an air quality analysis be performed for each regulated pol- lutant that a new source has the potential to emit in significant amounts.
40 C.F.R. § 52.21(m)(1)(i)(a). The significance level for SO2 is 40 tons/year.
40 C.F.R. § 52.21(b)(23). The AES facility has a potential to emit 453 tons/year of SO2. U.S. EPA Region II, Project Description for AES Puerto Rico L.P. at 3. Project Description at 3. Thus, AES prepared an SO2 air quality analysis. Petitioners are dissatisfied with several aspects of the AES analysis and have either offered alternative analyses or have request- ed that additional analyses be conducted prior to final approval of the AES permit. Petitioners’ objections are in part motivated by their concern about ambient SO2 levels in Guayama and limited current data quantify- ing those levels.

An air quality analysis generally proceeds in stages. EPA has issued a guidance document that outlines various elements of the PSD review process, including the air quality analysis. New Source Review Workshop

7
We note that the issues raised in Mr. Saades Llorens’ petition for review are largely covered in the other two petitions.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 331

AES PUERTO RICO L.P.
331

Manual (Draft Oct. 1990) (“NSR Manual”).8 The air quality analysis typi- cally begins with a preliminary analysis that uses modeling to predict air quality impacts based solely on the proposed facility’s emissions. See NSR Manual at C.24. The preliminary analysis does not take into account exist- ing ambient air quality or emissions from other sources. Knauf, 8 E.A.D. at 149. The results are used to determine whether additional analyses are necessary and to define the scope of any additional analyses.

The results of the preliminary analysis are compared to a set of val- ues often referred to as “monitoring de minimis levels” because they are used to determine whether a permit applicant may be exempted from the requirement to obtain preconstruction ambient air monitoring data. In re EcoEléctrica, L.P., 7 E.A.D. 56, 61–62 (EAB 1997). The PSD regulations contain a list of the monitoring de minimis levels and provide that a per- mit applicant may be exempted from preconstruction monitoring require- ments if air quality impacts from the proposed source are less than the de minimis levels. 40 C.F.R. § 52.21(i)(8)(i). The monitoring de minimis level for SO2 is 13 µg/m3 (24-hour average). Id. The maximum modeled SO2 (24-hour average) impact from the proposed AES facility is 4.97 µg/m3 . RS at 102. Thus, AES qualifies for the exemption from preconstruction ambi- ent air monitoring for SO2. AES requested such an exemption from Region II and the Region determined that an exemption was appropriate. See Region’s Response at 42.

According to the approach outlined in the NSR Manual, the results of the preliminary analysis may also be used to determine whether a full impact analysis, which includes multi-source modeling, should be con- ducted. NSR Manual at C.25. The modeled pollutant concentration from a proposed source is compared to a set of significance levels found in the NSR Manual. Id. at C.28. These levels are referred to as “significant ambient impact levels” or “significant impact levels” (“SILs”). If the mod- eled impacts from the proposed facility are less than the SIL for a partic- ular pollutant, the permit applicant is generally not required to conduct a full impact analysis. Table 1 presents AES’s SO2 modeling results and the corresponding SILs for SO2 (at various averaging times):

8
Although the NSR Manual is not entitled to the same weight as a binding regulation, the Board often refers to the NSR Manual for a statement of the Agency’s thinking on cer- tain PSD issues. In re Knauf Fiber Glass, GmbH, 8 E.A.D. 121, 129 n.13 (EAB 1999); In re Hawaii Elec. Light Co., 8 E.A.D. 66, 72 n.7 (EAB 1998).

VOLUME 8
187-274/Sections16-18
10/15/01 3:32 PM Page 332

332
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

TABLE 1
SO2 Air Quality Impacts from AES

        Averaging Time              AES Modeling Results9      Significant Impact

(in µg/m3)
Level10
(in µg/m3)
Annual 0.55 1
24-hour 4.97 5
3-hour 20.0 25

As can be seen from the table, the modeled SO2 impacts from the pro- posed AES facility are all less than corresponding SILs. Thus, the Region did not require, and AES did not conduct, a full impact analysis for the proposed facility.

Petitioners do not dispute that PSD regulations and guidance author- ize the Region’s decisions to exempt AES from preconstruction monitor- ing and to accept the preliminary air quality analysis in lieu of a full impact analysis. However, the petitioners argue that the Region should have used its discretion to require preconstruction monitoring and a full impact analysis in light of a variety of factors that, in petitioners’ opinion, warrant these studies. In addition, the petitioners specifically challenge the validity of the AES modeling. Dr. González has conducted his own modeling and submits the results as support for his argument that AES’s modeling grossly underestimates the potential SO2 impacts from the pro- posed facility. We discuss petitioners’ arguments in support of additional SO2 analyses below.

1. The Combination of SO2 Controls to Be Used on the Circulating Fluidized Bed Boilers Are Untested

In the course of assessing options for control of SO2 from the boilers, the Region determined that a combination of three controls would constitute the best available control technology (“BACT”) for this facility.
The required controls include: 1) circulating fluidized bed boilers with limestone injection, 2) low sulfur (<1% sulfur) coal, and 3) a dry scrub- ber. RS at 3. The resulting emission limit is 0.022 lb/MMBTU. Permit

9
See RS at 102.
10
See NSR Manual at C.28.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 333

AES PUERTO RICO L.P.
333

¶ VIII.4–CFB.a.; RS at 2. This emission rate was used to determine SO2 impacts from the proposed facility for purposes of the preliminary air quality analysis. See supra Table 1. Petitioners do not believe that it is appropriate to presume that AES’s emissions will meet the emission limit specified in the permit in light of the fact that the three required SO2 con- trols have not previously been used in combination. SURCCo Petition at 12. Petitioners believe that the preliminary air quality analysis should have taken into account emission rates in excess of the BACT limit.

The Region and AES defend the BACT determination as a combina- tion of control strategies that will result in stringent limits on SO2 emis- sions. AES states that the trio of controls arose out of its desire to “build one of the cleanest coal-fired power plants in the world.” AES Response at 6. The Region claims that the emission limit of 0.022 lb/MMBTU is technically feasible even if it is lower than the emission limits at previ- ously permitted facilities. Region’s Response at 48–50.

In response to comments about the feasibility of the SO2 control strat- egy, the Region pointed out that CFB boilers with limestone injection have been paired with low sulfur coal in a number of previously permitted proj- ects. This combination of just two controls results in very low SO2 emis- sion rates. RS at 3. At the same time, add-on dry scrubbers “are recognized SO2 pollution control equipment with a proven track record.” Id.
Generally, dry scrubbers have not been used with CFB boilers because the CFB boiler design inherently reduces SO2 emissions and a dry scrubber is a costly addition. However, there is no technological reason that would prevent use of a dry scrubber on CFB boilers, and the Region expects that such a scrubber will successfully provide additional SO2 removal.11 Id.;
Region’s Response at 51. Further, the Region asserts that the reason the combination of three SO2 control strategies required by this permit have not been used before is due to the costs involved rather than any tech- nological feasibility problem. RS at 3; Region’s Response at 51.

The petitioners contend that the only reason AES proposed this com- bination of SO2 controls was to avoid the obligation to prepare a full impact air quality analysis. SURCCo Petition at 18, att. 2. The controls reduce the maximum permitted SO2 emissions from the proposed facility to the point where SO2 impacts attributable to the facility are at levels

11
The Region acknowledges that the typical removal efficiency for dry scrubbers of 70%–90% may not be achieved at the AES facility because the concentration of SO2 com- ing into the scrubbers will be lower than at other facilities due to AES’s use of CFB boil- ers and low sulfur coal. Region’s Response at 51 n.30. However, in order to meet the per- mitted SO2 emission limit, the Region calculates that the dry scrubber at the AES facility would only have to operate at a removal efficiency in the 40%–45% range. Id. at 50.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 334

334
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

below the SILs. Since EPA uses the SILs to guide its decision as to whether a full impact analysis should be performed, one consequence of the selected SO2 controls and emission limit is that AES was able to forego additional air quality analyses in its permit application.

The concern articulated in SURCCo’s petition for review was addressed by the Region in the Responsiveness Summary. The Region explained that the SO2 emission rate in AES’s permit involves three meth- ods of controlling SO2 emissions and is federally enforceable. RS at 26, 40. These stringent controls on SO2 emissions involve a “real decrease” in the potential impacts from the facility. Id. at 26. As a result, air quality will be better than it would have been without the combination of controls.
Id. at 40. The Region argues that “it is within the applicant’s prerogative to accept lower emission limits so that the resultant impacts are below the significant impact level.” Region’s Response at 24.

We do not disagree with the Region’s assessment on this issue. This permit requires a level of SO2 control that appears to be unprecedented for this type of facility. Consequently, the Region’s decision breaks new ground on potentially available control options for similar facilities and may be replicated, yielding beneficial decreases in SO2 emissions at other facilities. In this respect, the BACT determination is to be commended. In addition, the emission limit is an enforceable standard.12 It is legitimate for the air quality modeling to reflect the actual manner in which the facility is required to operate. Because the Region has adequately justi- fied its BACT determination for SO2 from the boilers, it was proper to use the permitted emission limit for SO2 in the air quality analysis. We find no clear error of fact or law or an important matter of policy in the Region’s decision to use the permitted emission rate for SO2 in the air quality analysis. Therefore, we deny review of this issue.

12
When a permit is issued, the permitting authority expects that the permittee will achieve the specified emission limits. Any violation of those limits is subject to a subse- quent enforcement action. See CAA § 113, 42 U.S.C. § 7413. See also In re Kawaihae Cogeneration Project, 7 E.A.D. 107, 128 (EAB 1997) (denying review of a challenge to a so-called “unproven” air pollution control technology and noting the permitting authority’s assurance that failure to comply with the permit emission limit would subject the permit- tee to an enforcement action).

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 335

AES PUERTO RICO L.P.
335

2. Technical Challenges to AES’s SO2 Air Quality Modeling

The González petition raises several technical challenges to the SO2 modeling conducted by AES. Dr. González discusses alleged flaws in the AES modeling protocol and offers alternative modeling results from his own effort to model SO2 impacts from the proposed facility. Dr. González also presents the results of multi-source modeling that he performed. He claims, through the results of his modeling, that the SO2 NAAQS may be threatened in Guayama. González Petition at 7.

The initial response from the Region and AES with regard to the alter- native modeling results contained in Dr. González’s petition is that at least some of his results and analyses were not submitted to the Region during the public comment period and therefore these arguments may not be raised for the first time on appeal. Region’s Response at 25–26; AES Response at 14. The regulations governing the permit process require that “all persons * * * who believe any condition of a draft permit is inappro- priate * * * must raise all reasonably ascertainable issues and submit all rea- sonably available arguments supporting their position by the close of the public comment period.” 40 C.F.R. § 124.13. In addition, the provision that provides for appeal of permit decisions to the Board requires petitioners to demonstrate “that any issues being raised [on appeal] were raised dur- ing the public comment period * * * to the extent required by these regu- lations.” Id. § 124.19(a). In order to preserve an issue for review, these reg- ulatory requirements must be satisfied. In re Maui Elec. Co., 8 E.A.D. 1, 7 (EAB 1998); EcoEléctrica, 7 E.A.D at 63–64 n.9. The reason that issues must first be raised during the public comment period is to ensure that the permit issuer has an opportunity to adjust its permit decision or to pro- vide an explanation of why no adjustment is necessary. The Board has often denied review of issues that have not been properly preserved for review. See, e.g., Maui Elec., 8 E.A.D. at 9–10 (petitioner’s appeal of BACT for SO2 was not preserved for review when comments submitted to permitting authority focused on BACT for NOx not BACT for SO2);
In re Florida Pulp & Paper Ass’n, 6 E.A.D. 49, 56–57 (EAB 1995) (petition- er’s comment on one section of a draft permit was insufficient to preserve for review a challenge to another permit provision). But cf. EcoEléctrica, 7 E.A.D. at 63–64 n.9 (Board declined to deny review of “data-currentness” issue due to a failure to raise the issue during the public comment period because the data-currentness issue was sufficiently related to other issues that were properly preserved for review).

Dr. González filed comments on the draft permit that included alter- native modeling of AES’s SO2 impacts, including a comparison of those impacts to the SO2 SILs. Dr. González’s petition for review renews some

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 336

336
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

of the arguments and analyses regarding an appropriate estimation of SO2 impacts from the proposed AES facility. To the extent that Dr. González’s petition addresses his previous comments and the Region’s response thereto, we find that those issues are preserved for review and properly before us.

The petition also contains results of a multi-source modeling exer- cise. Multi-source modeling is different from the modeling of AES emis- sions alone that Dr. González submitted during the public comment peri- od. Multi-source modeling involves estimation of pollutant concentrations in ambient air, taking into account not only emissions from AES, but also certain other sources and background concentrations. The multi-source modeling information in Dr. González’s petition for review was not sub- mitted to the Region during the public comment period and is sufficient- ly distinct from the modeling of the proposed AES facility alone that it constitutes a new issue or argument not previously presented to the Region. Dr. González has not established that the multi-source modeling analysis was not reasonably ascertainable during the public comment period. As such, the multi-source modeling information is not properly before us in this appeal, and we will not review the details of that analy- sis.13 The following discussion addresses those portions of the González petition that have been properly preserved for review.

Dr. González first notes that the air quality model used by AES to predict pollutant concentrations in complex terrain,14 CTDMPLUS, has not been calibrated for tropical conditions such as exist in Puerto Rico.
González Petition at 2. Dr. González recommends that EPA conduct a cal- ibration process before using CTDMPLUS to model conditions in Puerto Rico. Id. He also points out that an alternative model, referred to as the Puerto Rico Air Quality Model (“PRAQM”), has been calibrated in Puerto Rico. Id. The Region describes CTDMPLUS as “the most refined complex

Petitioners’ general argument about the need for multi-source modeling (i.e., a full 13
impact analysis) was raised during the public comment period and therefore was preserved for review. See infra Section II.C.3. It is only the specific modeling results submitted by Dr.
González that have not been preserved.
14
Air quality models are organized into various categories based on the characteristics of the source, area, and pollutants to be modeled. Two of the most common model cate- gories are those for “simple terrain” and “complex terrain.” For purposes of air quality mod- eling in the PSD program, EPA defines “simple terrain” as “an area where terrain features are lower in elevation than the top of the stack of the source.” 40 C.F.R. pt. 51 app. W (“Guideline on Air Quality Models”) § 4.1. “Complex terrain” is defined as “terrain exceed- ing the height of the stack.” Id. § 5.1. The modeling conducted by AES as part of its per- mit application made use of models for both simple terrain and complex terrain.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 337

AES PUERTO RICO L.P.
337

terrain EPA-approved model.” Region’s Response at 22–23. AES notes that the CTDMPLUS has undergone enhancements since it was first intro- duced such that it can be used for unstable atmospheric conditions. AES Response at 17.

The PSD regulations generally require use of air quality models spec- ified in EPA’s Guideline on Air Quality Models, 40 C.F.R. pt. 51 app. W.
40 C.F.R. § 52.21(l)(1). The regulations also permit substitution of a dif- ferent model in cases where the preferred model is inappropriate. Id.
§ 52.21(l)(2). Any model substitution must be approved by the permitting authority. Id. In this case, CTDMPLUS is the preferred model for predict- ing pollutant concentrations in complex terrain. Guideline on Air Quality Models § 5.1, app. A.9. PRAQM is not mentioned in the Guideline on Air Quality Models and none of the parties to this case sought Region II’s approval to use an alternative model. To the extent that petitioners are challenging the use of CTDMPLUS, review is denied on that issue.
Petitioners have not established either a clear error of law or fact or an important matter of policy that warrants review.

Dr. González also challenges AES’s modeling results based on the preferred models. Dr. González provided alternative modeling results in both his comments to the Region and in his petition for review. See RS at 44–48; González Petition at 3–5. Dr. González submits these results to support the proposition that the SO2 impacts from the proposed facility will exceed the SILs, despite the Region’s conclusions to the contrary.

The Region provided a detailed and thorough response to the alter- native modeling submitted during the public comment period. RS at 48–50. The Region pointed out that Dr. González’s alternative modeling did not use CTDMPLUS for complex terrain, but another EPA-approved model called COMPLEX 1. RS at 49. The difference between COMPLEX 1 and CTDMPLUS is that COMPLEX 1 is a screening model for complex ter- rain and CTDMPLUS is a refined model for complex terrain. The refined model, CTDMPLUS, requires more site-specific terrain and meterological data. Id. at 50. The Region concluded that the modeling results produced by the CTDMPLUS model as provided by AES appropriately predicted SO2 impacts for the proposed facility.15 Id. The predicted SO2 impacts were all below the SILs.

15
We note that AES was required to collect one year of on-site meteorological data for use in the air quality modeling exercise. RS at 27. These data were used in the CTDMPLUS model and as such, AES’s modeling results reflect a relevant and representative record of atmospheric conditions.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 338

338
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

In his petition for review, Dr. González states that he performed a “sensitivity analysis” on the CTDMPLUS protocols used by AES. González Petition at 4. Dr. González challenges the CTDMPLUS results because the model inputs used an SO2 emission rate equal to the rate specified by the permitted emission limit. Dr. González calculated potential SO2 impacts for emission rates higher than the permitted rate. Id. at 5. He notes that “when the SO2 base emission rate * * * is exceeded, significant levels [SILs] are exceeded.” Id. The SO2 emission rate is dependent upon the sulfur content of the coal and the efficiency of the sulfur removal technology.
Dr. González believes that it is unlikely that AES will be able to effectively control these two factors and therefore concludes that AES’s SO2 emis- sions will have a significant impact on the surrounding area.

The scenarios modeled by Dr. González presume that AES will not operate within the established permit limits. As such, this argument is similar to the issue addressed supra Section II.C.1., regarding the achiev- ability of the BACT limit for SO2. While Dr. González may be correct that AES does not have much room for error in controlling the sulfur content of the coal and the operation of the sulfur removal equipment, the SO2 permit limit requires AES to avoid errors that would result in a permit vio- lation. If AES operates the facility in accordance with the permit require- ments, the SO2 impacts are predicted to be less than the SO2 SILs. Because AES is expected and required to operate the facility in such a fashion, it is appropriate to rely upon modeling results that presume compliance.

Dr. González challenges another element of the AES modeling, name- ly the number and placement of receptors.16 González Petition at 3.
Although both the Region and AES argue that the receptor placement issue was not properly preserved for review because it was not raised dur- ing the public comment period, we believe that it is sufficiently related to Dr. González’s critique of AES’s SO2 modeling for the proposed facility.17

“Receptors are locations at which ambient air quality is measured or estimated.” 16

Region’s Response at 26 n.9.
17
In his comments submitted to the Region during the public comment period, Dr.
González included results from a modeling scenario using different receptor heights. Dr.
González pointed out how slight changes in a modeling scenario, such as a change in receptor height, changes the overall estimation of AES’s SO2 impact. Dr. González’s petition for review contains a further analysis of how receptor variables affect the estimated SO2 impact for the AES facility. The challenge articulated in the petition for review, regarding number and placement of receptors, is not identical to the issue raised in Dr. González’s comments, but it is closely related. It is a close call as to whether the issue raised in the petition for review is sufficiently related to the issue raised during the public comment peri- od in order to be considered preserved for review. See EcoEléctrica, 7 E.A.D. at 63–64 n.9.
However, in an exercise of our discretion, we decline to deny review of the receptor issue based on an alleged failure to preserve the challenge.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 339

AES PUERTO RICO L.P.
339

Dr. González performed an air quality analysis using what he terms “a large number of receptors” (i.e., over 400). Id. For this analysis, Dr.
González used the same air quality model and the same stack conditions that AES used. Dr. González presents results which he claims demon- strate that the SO2 SILs will be exceeded by the proposed facility. Id. at 3–4. He believes that AES would have also generated such results had AES considered more receptors in its analysis. Id. at 5.

Both AES and the Region respond to Dr. González’s results. AES points out that while Dr. González used over 400 receptors, it used approximately 1,250 receptors in its analogous modeling exercise. AES Response at 19. The Region explains that although Dr. González employed the same model as used by AES, he did not follow the model protocol in the placement of his receptors. Region’s Response at 27. In particular, Dr. González did not assign receptor locations based upon the actual terrain surrounding the proposed facility. Rather, Dr. González assumed uniform terrain elevations, which according to the Region, do not yield a realistic simulation. Id. As a result, the model yielded estimates of SO2 concentrations higher than those predicted by AES. Id. at 28. In light of these explanations, we find that Dr. González has failed to estab- lish that AES’s SO2 modeling involved a clear error of fact or law or an important policy matter that warrants review.

3. Multi-Source Air Quality Analysis for SO2

Petitioners argue that the Region should have used its discretion to require AES to conduct a multi-source air quality analysis, i.e., a full impact analysis, for SO2. González Petition at 5; SURCCo Petition at 5.
Petitioners point out that AES’s modeling results indicate SO2 impacts that are “minutely below” the SILs. SURCCo Petition at 10. They are particu- larly concerned about the predicted 24-hour SO2 impacts, estimated at 4.97 µg/m3. The 24-hour SIL for SO2 is 5.0 µg/m3. See supra Table 1.

The SILs are a tool used by EPA to screen emissions for projects such as the proposed AES facility. Instead of requiring every PSD applicant to perform costly and time-consuming full impact air quality analyses, the SILs allow EPA to readily identify those projects whose air quality impacts will be less than significant.18 In this case, the modeled SO2 impacts from

18
Note that there is a difference between a significant emission rate from a facility and a significant air quality impact once those emissions have dispersed in the air. A facility that will have significant emissions, as defined at 40 C.F.R. § 52.21(b)(23), may or may not have a significant air quality impact when modeling results are compared to the SILs. In this case, the SO2 emission rate for the proposed AES facility is significant, but the pre- dicted air quality impact for SO2 is not.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 340

340
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

the proposed AES facility are less than significant, using the SILs as the basis for that determination. If predicted air quality impacts are less than significant, EPA guidance allows a permit applicant to forego the full impact air quality analysis, including multi-source modeling. NSR Manual at C.24.

We acknowledge petitioners’ concern that the predicted 24-hour SO2 impact from the proposed facility is very close to the corresponding SIL.
It may seem that a difference of only 0.03 µg/m3 between AES’s predict- ed impact and the SIL is an insufficient basis upon which to decide that AES need not conduct multi-source modeling. Notably, the PSD regula- tions do not specifically mandate multi-source modeling. EcoEléctrica, 7 E.A.D. at 65. Rather, the regulations contain a general requirement that permit applicants demonstrate that a proposed source will not cause or contribute to a violation of the NAAQS or a PSD increment. 40 C.F.R.
§ 52.21(k); EcoEléctrica, 7 E.A.D. at 65. The “requirement” for multi- source modeling comes from EPA guidance, and a permitting authority may insist on such modeling as a matter of discretion. NSR Manual at C.25 (multi-source modeling referred to as full impact analysis). In exer- cising that discretion, it is reasonable for the permitting authority to be guided by generally applicable thresholds in determining how detailed an air quality analysis needs to be.

In this case, the Region’s decision to abide by the SIL threshold is also supported by the quality of the modeling that produced the 4.97 µg/m3 estimate of AES’s 24-hour SO2 impacts. The modeling made use of EPA’s preferred air quality models for the particular terrain and pollutants in question. Region’s Response at 22–23; RS at 27. In addition, AES collected a year’s worth of on-site meteorological data that were used as an input to the models. RS at 26–27. The meteorological data were subject to quality assurance and controls. Id. at 26. The characteris- tics of the modeling effort in this case provide additional assurance that the estimated 24-hour SO2 impact of 4.97 µg/m3 is valid.

Thus, even though the predicted 24-hour SO2 impact from the pro- posed AES facility approaches the 24-hour SO2 SIL, the Region’s decision not to require multi-source modeling is supported both by established policy regarding SIL thresholds and the quality of the modeling that pro- duced the impact estimate. Having denied review of the petitioners’ spe- cific challenges to the air quality analysis for SO2, we also deny review of the challenge to the Region’s decision not to require a multi-source air quality analysis for SO2 because predicted SO2 impacts from AES fall

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 341

AES PUERTO RICO L.P.
341

below the SILs.19 Petitioners have not shown that this decision was clear- ly erroneous or otherwise warrants review.

4. Questions About the Attainment Status of Guayama and the Need for Preconstruction Monitoring

In addition to concerns about the accuracy of the SO2 modeling for the proposed facility, petitioners raise questions about the SO2 attainment status of Guayama. SURCCo Petition at 18–26. Petitioners believe that their concerns and a general dearth of data about current ambient air quality in Guayama should have caused the Region to require precon- struction monitoring from AES. Id. at 22. Petitioners acknowledge that AES’s predicted SO2 impacts are below the de minimis monitoring levels but believe that the Region should have used its discretion to nonethe- less require preconstruction monitoring. Id.

The Region points out that a 1983 study of Puerto Rico air quality, performed in conjunction with Puerto Rico’s State Implementation Plan (“SIP”), determined that Guayama was in attainment of the SO2 NAAQS.
RS at 31, 58; Region’s Response at 31. The Region believes that the 1983 attainment demonstration is still valid for Guayama because there has been no major source construction in Guayama since that date. Region’s Response at 31; see also RS at 31, 58. The Region further states that there are “no data to suggest that there is an existing exceedance of any of the NAAQS.” RS at 58. During this permit process, the Region obtained an estimate of SO2 concentrations in ambient air by examining data from a monitoring facility at Cerro Modesto, located approximately seventeen kilometers from the proposed site. Id. at 20. The Cerro Modesto data showed SO2 concentrations well below the NAAQS. Id. Although the Region concedes that conditions at Cerro Modesto may not be identical to Guayama, it believes that the data are useful for estimating background conditions. Id.; Region’s Response at 44. Thus, the Region has offered several rationales for concluding that Guayama’s SO2 attainment status is valid. However, in the interest of verifying ambient air conditions in Guayama, the Region has required AES, as a condition of its permit, to perform ambient air monitoring for SO2 after the facility commences operations. Permit ¶ XVI.5; RS at 61.

19
We note that the Region included a permit condition requiring AES to conduct a multi-source air quality analysis as a confirmatory measure within six months of the effec- tive date of the permit. Permit ¶ XVI.4. This condition was included as part of the Region’s environmental justice efforts on behalf of the surrounding community. RS at 61; see infra Section II.E.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 342

342
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

Petitioners are skeptical of the Region’s position regarding ambient SO2 levels in Guayama. They believe that a variety of other factors mili- tate in favor of requiring preconstruction monitoring. Petitioners first point out that the 1983 SIP attainment demonstration was based on mod- eling, not actual ambient air quality data, and that little ambient air qual- ity data have been collected in the interim. SURCCo Petition at 19.
Petitioners also counter the Region’s assertion that no data suggest an existing exceedance of the NAAQS. Petitioners offer the results of SO2 modeling performed by the Puerto Rico Environmental Quality Board (“PREQB”) in 1990. According to SURCCo, the PREQB modeling predicts SO2 concentrations in the Guayama area that exceed the NAAQS.
SURCCo Petition at 20–21. Third, petitioners question the representative- ness of the Cerro Modesto data. Id. at 23. Finally, petitioners claim that existing sources already emit large quantities of SO2 and that certain sources may be violating the terms of their permits, thus compounding problems with SO2 air quality. Id. at 25. In sum, petitioners do not believe that the Region’s plan to obtain SO2 ambient air data after the AES facil- ity commences operation meets the spirit of the Clean Air Act and its reg- ulations. Id. at 17.

Even if we were convinced by each of the factors mentioned in SURCCO’s petition,20 we would nonetheless uphold the Region’s decision

20
There are several weaknesses in petitioners’ arguments regarding Guayama’s attain- ment status for SO2, not least of which is that certain arguments were not brought to the Region’s attention during the public comment period. With regard to the 1990 PREQB study, SURCCo claims that the study was not reasonably ascertainable during the public comment period because SURCCo only discovered the study when preparing a response to the Region’s Responsiveness Summary. SURCCo Petition at 21. Although SURCCo did not learn of the study until after the close of the public comment period, that does not mean that the study was not reasonably ascertainable at an earlier date. Obviously, based on the date of the study, the study results were available long before the public comment period on the proposed AES permit. SURCCo only discovered the 1990 PREQB study when it asked PREQB for SO2 modeling data for Puerto Rico after reviewing the Responsiveness Summary.
Id. at 20. In light of SURCCo’s admission that its concern about SO2 air quality “has always been an issue,” we see no reason why SURCCo could not have made an inquiry of PREQB much earlier in this permit process. Thus, we find that the 1990 PREQB study was reason- ably ascertainable, but it was not brought to the attention of the Region during the public comment period. Therefore, the study is not properly before us on appeal.

In addition, to briefly address SURCCo’s principal arguments regarding SO2 attainment in Guayama, we note the following:

• The PREQB study is not necessarily an accurate predictor of Guayama condi- tions, as it is based upon meteorological data from a different part of Puerto Rico.
Region’s Response at 35. In contrast, AES collected a year of on-site meteoro- logical data for use in its air quality analysis. See supra note 15.
Continued

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 343

AES Puerto Rico L.P.
343

not to require preconstruction monitoring in this case. SURCCo’s petition raises questions about actual SO2 levels in the Guayama area. According to SURCCo’s information, there may or may not be current exceedances of the SO2 NAAQS. Although additional monitoring data may help clari- fy the situation, for the reasons explained below, the Region’s decision to obtain those data after commencement of operation of the proposed AES facility is not clearly erroneous from a legal point of view, nor is it an important matter of policy that warrants our intervention.

Legally, the SO2 attainment designation for Puerto Rico is still in effect, even if the underlying study in support of that designation dates from 1983. See 40 C.F.R. § 81.355 (Puerto Rico SO2 air quality rated as “better than national standards”). That designation may not be challenged in this proceeding.21 Because Guayama is officially an SO2 attainment area, the PSD regulations govern the preconstruction permitting process.
Thus, the PSD regulation providing for an exemption to preconstruction monitoring may be validly applied in this case. See 40 C.F.R. § 52.21(i)(8).
The Region has established that such an exemption was justified in this case because the predicted SO2 impacts from the proposed facility are lower than the de minimis monitoring levels. See 40 C.F.R. § 52.21(i)(8)(i);
EcoEléctrica, 7 E.A.D. at 64.

Moreover, a showing that the predicted impacts are also below the SILs generally constitutes an acceptable demonstration of compliance with the NAAQS. NSR Manual at C.51. If a proposed facility has modeled impacts that are below the SILs, that facility is not considered to cause or contribute to a violation of an air quality standard. See Memorandum from Gerald A. Emison, OAQPS, U.S. EPA, to Thomas J. Maslany, Air Management Division, U.S. EPA at 1 (July 5, 1988) (included in adminis- trative record as item #189C); Region’s Response at 39. The Agency has

• The Cerro Modesto monitoring data are not subject to representativeness requirements because they were not being offered as official SO2 background data for use in a full impact analysis. The Cerro Modesto data were merely offered as an estimate of background conditions. RS at 38.

• Potential violations at existing sources do not necessarily indicate that an area is not meeting the NAAQS. Moreover, we have repeatedly held that potential per- mit violations by other sources are primarily an enforcement issue, and not grounds for challenging a permit decision. See EcoEléctrica, 7 E.A.D. at 70;
Federated Oil & Gas, 6 E.A.D. at 730.
21
Attainment and non-attainment designations are made pursuant to specific statutory procedures that include provisions for redesignation. CAA § 107(d), 42 U.S.C. § 7407(d).

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 344

344
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

made a judgment that, as a general matter, proposed facilities with insignificant air quality impacts (i.e., impacts below the SILs) do not cause or contribute to air quality violations. Thus, even if SURCCo’s infor- mation conclusively supported a finding that there are current NAAQS violations in Guayama, the AES facility as proposed would not be con- sidered a cause of or contributor to such violations. AES could still there- fore obtain a PSD permit.

With regard to whether this issue presents a reviewable matter of policy under 40 C.F.R. § 124.19, we believe that the Region’s decision to proceed with permit issuance and later examine SO2 air quality in Guayama more closely is a legitimate exercise of its discretion. If SURCCo’s suspicions about existing ambient SO2 levels are correct, the SO2 NAAQS may be threatened even without the emissions from the AES facility. Although the Region could have required AES to conduct pre- construction monitoring, even the worst case results of such monitoring would not prevent AES from being eligible for a PSD permit because as discussed above, AES’s own predicted SO2 impact is below the threshold for causing or contributing to air quality violations. Under the petitioners’ approach, monitoring data would become available sooner, but it is entirely possible that it would neither affect the AES permit nor be accompanied by an effort to control ambient SO2 levels. Under the Region’s approach, AES is required to install and operate an ambient SO2 monitor. If violations of the SO2 NAAQS are discovered, the Region and PREQB have committed to undertake corrective action, including a pos- sible revision of Puerto Rico’s SIP on an expedited basis. RS at 25, 61–62;
Region’s Response at 40. The Region’s approach has the potential bene- fit of efficiently and permanently achieving lower SO2 levels in the ambi- ent air by addressing SO2 emissions from multiple facilities rather than focusing only on the AES project.

Thus, we find that the Region’s decision not to require AES to con- duct preconstruction monitoring for SO2 is not clearly erroneous or oth- erwise warranting of our review. Review is denied.

D. Issues Related to Fine Particulate Matter (PM10)

The proposed AES facility has a potential to emit PM10 at a rate of 316 tons/year. This level of PM10 emissions subjects the project to the air qual- ity analysis and BACT requirements of the PSD regulations. Petitioners challenge elements of the Region’s decisions in both of these areas.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 345

AES PUERTO RICO L.P.
345

1. PM10 Air Quality Analysis

Unlike the predicted impact of SO2 emissions from the proposed facility, the predicted impact of PM10 emissions on air quality is significant.
It exceeds both the de minimis monitoring levels and the SILs. Table 2 presents PM10 modeling results for the proposed AES facility along with the corresponding de minimis monitoring levels and SILs:
TABLE 2
PM10 Air Quality Impacts from AES*
Averaging AES De Minimis Significant Impact
Time Modeling Monitoring Level23 Level24
Results22
Annual 2.95 N/A 1
24-hour 20.4 10 5

• All concentrations presented in µg/m3

Because the predicted PM10 impacts exceeded the de minimis moni- toring level, AES was required to obtain background ambient air moni- toring data for PM10. PREQB maintains a PM10 monitor at a school located 1.8 kilometers from the site of the proposed facility. The Region deter- mined that the data from the PREQB monitor were representative of the site and could provide the required background monitoring data on PM10.
See RS at 98.

AES was also required to conduct a full impact analysis for PM10 because predicted impacts exceeded the PM10 SILs. The full impact analysis involves modeling the cumulative impacts of certain existing PM10 sources plus the impacts from the proposed facility as well as back- ground concentrations obtained from the monitoring data. The results of a full impact analysis are compared to two air quality standards: 1) the NAAQS,25 and 2) the PSD increment.26 The PSD regulations require a

22
See RS at 102.
23
See 40 C.F.R. § 52.21(i)(8)(i).
24
See NSR Manual at C.28.
25
The PM10 NAAQS are maximum ambient air concentrations. See 40 C.F.R. § 50.6. The standards are set at levels that EPA has determined are necessary to protect the public health and welfare. 40 C.F.R. § 50.2(b).
26
A PSD increment is the maximum allowable increase in pollutant concentration over a baseline concentration. See 40 C.F.R. § 52.21(c).

VOLUME 8
187-274/Sections16-18
10/15/01 3:32 PM Page 346

346
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

permit applicant to demonstrate that the proposed project will not cause or contribute to a violation of either the NAAQS or the PSD increment.
40 C.F.R. § 52.21(k). As shown in Table 3, AES’s full impact analysis for PM10 demonstrated that both standards will be protected:
TABLE 3
PM10 NAAQS and PSD Increment Compliance Demonstration*
Averaging AES PM10 PSD AES PM10 PM10
Time Impact Plus Increment Impact Plus NAAQS30
Other for PM1028 Other Sources
Sources
and
Consuming
Background
Increment27
Concentration29

        Annual                   3.3           17                 44.5                50
        24-hour                 16.631         30                112.1               150

• All concentrations presented in µg/m3.

Petitioners claim that the Region did not require use of the most up- to-date background monitoring data in the full impact analysis. SURCCo Petition at 27. Petitioners believe that 1997 data from the PREQB monitor at the local school should have been used because the 1997 data were available before the Region issued its final permit decision. Id. Petitioners identify three data points from the 1997 data set that, when added to the modeled impact of AES’s proposed facility and other existing sources, result in exceedances of the PM10 (24-hour) NAAQS. Id. at 28.

AES points out that the PSD regulations require background moni- toring data to “represent at least one year preceding receipt of the appli- cation,” 40 C.F.R. § 52.21(m)(1)(iv), and that they submitted data from the five years immediately previous to the date of their permit application.

27
See RS at 102.
28
See 40 C.F.R. § 52.21(c).
29
See RS at 102.
30
See 40 C.F.R. § 50.6.

The value listed in Table 3 representing the predicted 24-hour PM10 impact from AES 31
plus other increment consuming sources is lower than the value shown in Table 2 for the 24-hour impact of the AES facility by itself. The value used in Table 3 reflects the “highest, second highest estimated concentration” which is to be used in NAAQS and PSD increment compliance demonstrations for averaging times of 24 hours or less. 40 C.F.R. pt. 51 app.
W §§ 11.2.3.2, 11.2.3.3.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 347

AES PUERTO RICO L.P.
347

AES Response at 32–33. The 1997 data discussed by the petitioners dates from after AES submitted its permit application (in December 1995), and from after the Region determined that the permit application was com- plete (in September 1996).

The Region and AES also note that the three 1997 data points high- lighted by the petitioners have been flagged for quality assurance/quali- ty control purposes. Region’s Response at 55; AES Response at 34. The flags on these data points indicate that the values were influenced by Sahara dust and are therefore not considered representative of normal conditions. AES Response at 34–35. EPA corrects such flagged data before using it in compliance demonstrations. Further, even though 1997 data were not required to be used in the NAAQS and PSD increment compli- ance demonstration because they post-date the permit application, the Region notes that using the corrected 1997 data in a compliance demon- stration does not yield a NAAQS exceedance. Region’s Response at 56.

The PM10 compliance demonstration for the proposed AES facility is well supported by the background monitoring data. Petitioners have not established that the Region’s choice of PM10 monitoring data and the resulting air quality analysis were clearly erroneous or involve an impor- tant policy matter that warrants a grant of review. Review is therefore denied.32

2. BACT Determination for PM10

Petitioners challenge the final PM10 permit limit for the CFB boilers.
PM10 consists of two types of particulate, condensibles and non-condensi- bles. See 55 Fed. Reg. 14,246 (Apr. 17, 1990) (“emissions that contribute to ambient PM10 concentrations are the sum of in-stack [non-condensible] PM10 * * * and condensible emissions”). The permit is designed to address both types of PM10. There are two PM10 emission limits applicable to the boilers. The first limit states, “emissions of PM10 (condensible and noncon- densible) shall not exceed 0.015 lb/MMBTU * * *.” Permit ¶ VIII.1–CFB.a.2.
The second limit notes that AES may not be able to meet the 0.015 lb/MMBTU limit due to the condensible portion of PM10 and permits EPA to “adjust the PM10 emission rate to a level not to exceed 0.05 lb/MMBTU.” Id. ¶ VIII.1–CFB.a.3. The two-limit approach involved a change from the draft permit and petitioners claim that it violates the New Source

32
Review is also denied on petitioners’ objection to the multi-source component of the PM10 air quality analysis. Petitioners have not shown how their concern regarding benzene emissions from a neighboring facility relates to the full impact analysis for PM10.
See SURCCo Petition at 28–29.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 348

348
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

Performance Standard (“NSPS”) for this type of facility. SURCCo Petition at 29.

In the draft permit, the Region proposed a PM10 emission limit for the CFB boilers of 0.015 lb/MMBTU. Draft Permit ¶ VIII.1–CFB.a. The draft permit also required that a particular test method, i.e., Method 202, be used to measure PM10 emissions. Id. ¶ XV.4.e. Method 202 is used to quantify the condensible fraction of total PM10 emissions.

AES submitted comments on the pairing of the specified emission limit and test method. AES pointed out that the limit of 0.015 lb/MMBTU had been derived from BACT determinations for other coal-fired CFB boilers that did not take into consideration the condensible fraction of PM10. See RS at 10. Instead of requiring use of Method 202, the emissions from the other facilities were to be tested using an “in-stack” method which is designed to measure non-condensibles. AES requested that the Region require testing only by an in-stack test method and further requested that the Region not impose a limit for the condensible fraction of PM10. RS at 11.

The Region recognized that the permit limit for PM10 needed to be reconsidered in light of the mismatched emission limit and test method.
Region’s Response at 57. However, the Region insisted on retaining a limit for PM10 that included condensibles. See RS at 13. The Region noted that EPA considers condensible particulate matter to be included in PM10.
Id; see also 56 Fed. Reg. 65,433 (Dec. 17, 1991) (promulgation of test method 202 for measuring condensible particulate matter from stationary sources); 55 Fed. Reg. at 14,246 (noting need for special test method for condensible PM10 emissions). The Agency has also stated that it is partic- ularly important to account for condensible particulate matter at sources where condensibles constitute a significant fraction of the total PM10 because otherwise, the PM10 impact will be underestimated. Letter from Thompson G. Pace, U.S. EPA, to Sean Fitzsimmons, Iowa Department of Natural Resources (Mar. 31, 1994) (included in administrative record as item # 189E). The same guidance letter notes that Method 202 is the rec- ommended method for measuring the condensible fraction of PM10 and that it is generally not acceptable to waive Method 202 testing. Id. at 2.

In the course of processing AES’s permit application and responding to comments, the Region discovered that there was very little information available on which to base an emission limit that includes condensible PM10 from CFB boilers. See RS at 13. Most other CFB boilers had PM10 emission limits that were not designed to control the condensible fraction of PM10. AES surveyed fifteen state environmental agencies and found that

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 349

AES PUERTO RICO L.P.
349

only two of the agencies included the condensible fraction of PM10 when setting emission limits under their PSD programs. RS at 11.33 Further, AES found only one example of PM10 emission rates that included condensi- bles from a coal-fired CFB boiler, and that facility used a test method other than Method 202.

Thus, with little guidance regarding achievability, the Region was left to derive a PM10 limit that included both condensible and non-condensi- ble particulate matter. The Region set a limit of 0.015 lb/MMBTU but pro- vided that the limit may be adjusted upward after obtaining actual stack test data from AES. Region’s Response at 57–58. Nonetheless, the Region set a cap on the upward adjustment at 0.05 lb/MMBTU. AES performed an air quality analysis using the upper limit of 0.05 lb/MMBTU and found that even at the higher limit the NAAQS and PSD increment for PM10 will be protected. Id. at 58.

The Region’s approach to setting a PM10 limit for the CFB boilers is similar to an approach that the Board upheld in In re Hadson Power 14, 4 E.A.D. 258 (EAB 1992). In Hadson Power, the Board denied review of a NOx emission limit that involved the first time a control technology was applied to a particular type of coal-fired boiler. The petitioner in Hadson Power objected to the NOx limit as being too high, but the permitting authority had included a permit provision that allowed the NOx limit to be adjusted downward after the facility commenced operation. Id. at 291.
The Hadson Power approach began with a high emission limit and included the potential for downward adjustments. The Region’s approach here begins with a low emission limit and allows for upward adjustments, if necessary, subject to a cap. Both cases involve a situation where the permitting authority was faced with some uncertainty as to what emission limit was achievable. In the circumstances of this case, as in Hadson Power, the use of an adjustable limit, constrained by certain parameters, and backed by a worst case air quality analysis, is a reasonable approach.

We recognize that the potential PM10 emission limit of 0.05 lb/MMBTU is significantly higher than the original limit of 0.015 lb/MMBTU. The rea- son for the potentially higher limit, however, is to control a type of pol- lution (i.e., condensible particulate matter) that might otherwise go uncon- trolled. Through this permit, the Region is committed to addressing con- densible particulate matter. This approach will ultimately yield a more accurate picture of PM10 emissions from the AES facility. By including these permit limits in EPA’s database of BACT determinations, the Region also

33
The two states that do include condensible particulate matter are New York and New Jersey, and they do so at the direction of Region II. Letter from Rebecca Cranna, AES Puerto Rico, to Steven Riva, U.S. EPA Region II at 2 (June 19, 1997).

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 350

350
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

increases the likelihood that the next facility subject to BACT for PM10 will also include a limit and test method designed to take into account the con- densible fraction of PM10. Upon close examination, petitioners’ objection to the PM10 emission limit for the CFB boilers in the Region’s final permit decision does not indicate clear error or an important policy matter that warrants our review.34

E. Environmental Justice

Petitioners’ final basis for appeal of the Region’s PSD permit decision for AES invokes President Clinton’s Executive Order on environmental justice. SURCCo Petition at 30–33; see Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations, Exec. Order No. 12,898, 59 Fed. Reg. 7629 (Feb.16, 1994) (“Executive Order”). Petitioners’ environmental justice arguments rest in large part on their technical arguments, but with a slightly different emphasis.

Petitioners state that the proposed location for the AES facility is a low-income community and therefore the Region should have taken additional safeguards to protect this community on the basis of environ- mental justice. SURCCo Petition at 30. Petitioners believe that the Region should not have strictly adhered to its “significant impact” levels, should have required additional air monitoring and modeling prior to permit issuance, and should not have relied on the 1983 attainment demonstra- tion. In addition, petitioners claim that the Region did not consider pub- lic comment and testimony regarding health problems in the Guayama area. Id. at 31. They also suggest that the Region could have done more to enhance public participation and comment. Id. at 32.

The Region performed an environmental justice analysis that is reproduced in the Responsiveness Summary. See RS at 52–62. The Region’s environmental justice analysis begins by analyzing income lev- els in the areas of maximum potential impact from AES emissions. Id. at 54–55. The Region concluded that these areas are low-income and there- fore proceeded to assess whether the proposed AES project would result

34
Petitioners’ argument that the new PM10 limit is inconsistent with the NSPS is not an accurate reading of the NSPS. The applicable NSPS is Standards of Performance for Electric Utility Steam Generating Units for Which Construction is Commenced After September 18, 1978, 40 C.F.R. pt. 60 subpt. Da. This NSPS only contains a standard for PM emissions (i.e., total particulates), not PM10. The PM limit in the NSPS is 0.03 lb/MMBTU, to be determined by an in-stack test method. 40 C.F.R. §§ 60.42a(a)(1), 60.48a(b)(2). The AES permit also con- tains a limit on PM, distinct from its limit on PM10. The PM limit is 0.015 lb/MMBTU. Permit ¶ VIII.1–CFB.a.1. Thus, the permit is fully consistent with the applicable NSPS.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 351

AES PUERTO RICO L.P.
351

in a disproportionately high and adverse effect on human health or the environment in these areas. Id. at 55.

The Region prepared a thorough assessment of the potential impacts of air emissions from the proposed AES facility. The Region looked at maximum short and long-term impacts of carbon monoxide, sulfur diox- ide, nitrogen dioxide, and fine particulate matter. Id. at 55–57. Not only were all maximum predicted concentrations of these pollutants below the corresponding NAAQS, the maximum predicted concentrations of carbon monoxide, sulfur dioxide, and nitrogen dioxide were all below the SILs as well. The Region pointed out that NAAQS are health based standards, designed to protect public health with an adequate margin of safety, including sensitive populations such as children, the elderly, and asth- matics. Id. at 57.

In support of environmental justice for this community, the Region took steps to require that many elements of the air quality analyses per- formed during the permit process be reconfirmed after the permit is issued. As conditions of the permit, AES is required to conduct ambient SO2 monitoring and to perform a multi-source air quality analysis for SO2.35 Permit ¶¶ XVI.4, XVI.5; see RS at 61. These permit conditions are a testament to the role of public participation in the permit process.
Because of the concerns raised during the public comment period, this permit contains additional conditions that are not mandated by the PSD regulations but are within the Region’s discretion to require. The Region incorporated the conditions into the permit as a tangible response to the community’s concerns about air quality and to fulfill the goals of the Executive Order.

The Region also addressed information submitted during the public comment period regarding adverse health impacts and studies of health impacts that were previously conducted in the community. RS at 58–59.
The Region further analyzed the distribution of Toxic Release Inventory (“TRI”) facilities on the island of Puerto Rico and the quantity of toxic chemical releases reported by those facilities. The information from the health studies and the TRI analysis pertains primarily to toxic chemicals rather than criteria pollutants (which are the focus of the PSD program), but the Region’s effort to provide meaningful responses on these issues contributes to environmental justice for the Guayama community.

Finally, with regard to opportunities for the public to participate and provide comments on the AES permit decision, the Region points out that 35
In addition to the SO2 monitoring, the Region points out that PREQB has committed to indefinitely continue the PM10 monitoring at the local school. RS at 62.

VOLUME 8
187-274/Sections16-18 10/15/01 3:32 PM Page 352

352
ENVIRONMENTAL ADMINISTRATIVE DECISIONS

it provided expanded public comment opportunities and engaged in extensive correspondence with petitioner SURCCo over the course of the permit process. Region’s Response at 70. The Region also took several steps to ensure that comments could be received in either English or Spanish and all comments were granted equal consideration. RS at 69.

In light of the Region’s thorough environmental justice analysis and incorporation of environmental justice elements into the permit decision, we find that petitioners have not demonstrated that the Region commit- ted clear error on issues of environmental justice. See In re EcoEléctrica, L.P., 7 E.A.D. 56, 69 (EAB 1997). Neither do we see an important matter of policy that warrants our review. The Region accommodated several of petitioners’ concerns (e.g., concerns about SO2 ambient air levels and multi-source modeling) albeit not in the precise manner that petitioners desired. On the basis of the administrative record, the Region’s environ- mental justice efforts appear satisfactory, and the petitioners have not demonstrated otherwise.

III. CONCLUSION

The petitions for review of Region II’s permit decision for AES are denied. First, the Region’s decisions with regard to modeling AES’s SO2 impacts, multi-source modeling for SO2, and preconstruction monitoring for SO2 were made in accordance with PSD regulations and established guidance. Second, the Region’s choice of a PM10 emission limit for the CFB boilers is a creative yet justifiable approach to ensuring that the per- mit contains effective control of condensible particulate matter. Further, the PM10 air quality compliance demonstration is well supported by the background monitoring data. Finally, petitioners have failed to show, in light of the Region’s thorough environmental justice analysis, that an issue of environmental justice meets the standard for Board review.

So ordered.36

36
This decision constitutes final agency action for purposes of judicial review. See 40 C.F.R. § 124.19(f)(1)(i). The Region shall see that notice of this decision is published in the Federal Register in accordance with 40 C.F.R. § 124.19(f)(2).

VOLUME 8