West Virginia Air Quality Board

WEST VIRGINIA AIR QUALITY BOARD

CHARLESTON, WEST VIRGINIA

SIERRA CLUB, WEST VIRGINIA HIGHLANDS

CONSERVANCY, GREENBRIER RIVER

WATERSHED ASSOCIATION,

Appellants,

vs. APPEAL NO. 2006-03-AQB

DIRECTOR, DIVISION OF AIR QUALITY,

DEPARTMENT OF ENVIRONMENTAL PROTECTION,

Appellee,

And

WESTERN GREENBRIER CO-GENERATION, L.L.C.

Intervenor

APPELLANTS’ MEMORANDUM IN SUPPORT OF ITS PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW

INTRODUCTION

Appellants have petitioned the Board to review the Department of Environmental Protection’s (“DEP’s”) issuance of an air pollution permit to Western Greenbrier for its proposed coal-burning circulating fluidized bed (“CFB”) facility in Greenbrier County, WV. The Board should vacate DEP’s approval of Western Greenbrier’s permit application because Western Greenbrier and DEP cannot establish that proposed facility’s emission limits for SO2 and NOx are, in fact, Best Available Control Technology (BACT) for the proposed facility. That is, DEP has issued this permit allowing Western Greenbrier to use less stringent control technology than is required by law. That failure to comply with BACT would harm the citizens of Greenbrier County and beyond and thwart the intent of Congress and the Legislature to protect the public health of West Virginia citizens.

As was revealed at the evidentiary hearing before the Board, the control technologies that are to be applied to Western Greenbrier’s proposed facility were selected and committed to before the BACT analysis was ever performed. That is, Alstom, the vendor of the CFB, told Western Greenbrier that its CFB came prepackaged with Selective Non Catalytic Reduction technology to control NOx and with a Flash Dryer Absorber to control SO2. Transcript of 8/29/2006 at 243. Alstom also resisted any efforts to change those control technologies. Id. at 256, 321; Transcript of 8/30/2006 at 93-94, 122. As a result, the proposed facility was outfitted with the Best Alstom Control Technology, rather than the Best Available Control Technology.

The process used by Alstom and Western Greenbrier is analogous to a person walking into a car dealership to buy a car and telling the dealer that he has $20,000 to spend, that he has to spend it at this dealership, that he has to spend it today and that he cannot meaningfully ask questions about the car or the price. Remarkably, Western Greenbrier sent out no request for proposals for any of the pollution control technology proposed in the permit. Transcript of 8/29/2006 at 292. To compound the problem, it did not even get prices from Alstom. See, e.g., CR at 252. To this day, neither Western Greenbrier nor Alstom know how much Western Greenbrier is paying for any piece of pollution control technology. In fact, the total cost of the proposed pollution control technology is unknown because Alstom did not provide the costs to Western Greenbrier. It is irresponsible, to say the least, for Western Greenbrier to have bought such a pig in the poke. It was impermissible under the Clean Air Act for DEP to have accepted of this “pig” without having before it information sufficient to allow it to assess its value.

The record and the transcript of the evidentiary hearing are devoid of the type of evidence necessary to reject the more stringent control technologies that Western Greenbrier and DEP rejected. As explained below, the Administrator of EPA has interpreted the Clean Air Act (CAA) to place the burden on the applicant to demonstrate a significant justification for rejecting the most stringent control technologies. Western Greenbrier’s justifications in this case fall far short of the legal standard. That is, there is insufficient evidence in the record to support the assertions that Western Greenbrier has made about the more stringent technologies.

But that insufficiency, though unlawful, in not unsurprising considering who controlled the information—Alstom, the vendor. Alstom’s reluctance to use alternative technologies to those it had already designed forced Western Greenbrier to manipulate the BACT analysis, using numbers in some cases concocted by the vendor, in order to reject alternative—and more stringent—technologies. But, because Western Greenbrier cannot provide sufficient evidence to back up the bare assertions in the BACT analysis, and because DEP relied on Western Greenbrier’s submission, this permitting decision cannot stand. Accordingly, the Board must vacate DEP’s decision and remand it to the agency for further proceedings.

STANDARD OF REVIEW

W. Va. Code § 22B-1-7(e) provides that “[t]he [Air Quality Board] shall hear the appeal de novo. (Emphasis added.) In W. Va. Div. of Environmental Protection v. Kingwood Coal Co., 200 W.Va. 734, 745 (1997), the West Virginia Supreme Court interpreted the Board’s review in this way:

The term “de novo” means “‘[a]new; afresh; a second time.’ ” Frymier-Halloran v. Paige, 193 W.Va. 687, 693, 458 S.E.2d 780, 786 (1995) ( quoting Black's Law Dictionary 435 (6th ed. 1990)). The term “hearing de novo ” means “ [g]enerally, a new hearing or a hearing for the second time, contemplating an entire trial in same manner in which matter was originally heard and a review of previous hearing. Trying matter anew the same as if it had not been heard before and as if no decision had been previously rendered. On hearing ‘de novo’ court hears matter as court of original and not appellate jurisdiction.” Black's Law Dictionary 721 (6th ed.1990). (citations omitted and emphasis added).
. . . . .
Thus, in a de novo hearing, the Board “is not concerned with what took place below ··· [a]s no presumption of correctness attaches to the action of the [DEP].”

Relying on Kingwood Coal, the Board has explained its standard of review this way:

The Air Quality Board’s standard of review is de novo (W.Va. Code § 22B-1-7(e))[.] That [statute] requires the Board to hear the appeal and be the “ultimate finder of fact and to act independently on the evidence before it.” Accordingly, the Board does not afford deference to the decision of the Department of Environmental Protection.

Catenary Coal Co. v. Director, Div. of Air Quality, Dep’t of Environmental Protection, No. 2004-07 (AQB) at 11 (emphasis added).

In 1988, in a formal adjudication, the Administrator of EPA stated that he

interpreted the statutory definition of BACT as placing the burden on the applicant of "demonstrating that significant technical defects, or substantial local economic, energy, or environmental factors or other costs warrant a control technology less efficient than [the most stringent available technology]." Id. at 7, 6 n.9. This interpretation was disseminated in operational guidance for municipal waste combustors on June 26, 1987, and was further refined in general guidance issued by EPA's Assistant Administrator for Air and Radiation on December 1, 1987.

In the Matter of Pennsauken County, New Jersey, Resource Recovery Facility, PSD Appeal 88-8, 2 E.A.D. 667, 1988 WL 249035 (Remand Order, Nov. 10, 1988). The United States Supreme Court has explained that an agency interpretation of the statute that it is charged with implementing is due substantial deference, even if that interpretation is explained in an agency adjudication. SeeU.S. v. Mead Corp., 533 U.S. 218, 226-27, 121 S. Ct. 2164, 150 L. Ed. 2d 292 (2001). Thus, because he is charged with implementing the Clean Air Act, the Administrator’s interpretation of the statutory definition of BACT is authoritative and controlling.

Consequently, the question before the Board is whether the applicant has met its burden, through its permit application and at the evidentiary hearing, in “demonstrating that significant technical defects, or substantial local economic, energy, or environmental factors or other costs warrant a control technology less efficient than [the most stringent available technology].” In the Matter of Pennsauken County, PSD Appeal 88-8, 2 E.A.D. 667, 1988 WL 249035) (emphasis added). Because the Board conducts de novo review in the instant action, it stands in the shoes of the agency and reviews the record afresh to determine whether there the applicant has carried its burden to demonstrate the rejection of more stringent technologies is justified.[1] The applicant has fallen well short of this standard and had instead presented nothing other than self-interested, conclusory, and baseless statements to justify the low emission limits that it has selected for SO2 and NOx. EPA’s guidance makes it clear that “[r]ejection of a technology must have a rationale arrived at after full consideration of data determined in a consistent and sound manner.”[2] Further compounding the error, DEP conducted no meaningful analysis of its own to scrutinize or carefully consider the assertions in Western Greenbrier’s permit application. The Board must not make the same error and, for the reasons that follow, must vacate and remand this permit.

BACT IS TECHNOLOGY FORCING

Many of the errors committed by the applicant and the agency in this permit application stem from their view that BACT is not technology forcing. At the evidentiary hearing, the following exchange occurred between counsel for Western Greenbrier and DEP’s engineer:

QI just have one [question]. Mr. Kessler, you just told Mr. Huson that the BACT program was not designed to push development of innovative technology. And my question is, is that why the BACT process includes cost considerations and is done on a case by case basis?

AI would say that that is appropriate.

Transcript of 8/30/2006 at 11. That fundamental misunderstanding of the purpose of BACT doomed this permitting process from the start. Rather than conduct a top down BACT analysis as DEP purported to do, the process used by DEP was a “bottom up” analysis; rather than making an informed determination about the best available control technology, the DEP looked back at what had already been permitted by other states.[3] If this were the way that BACT worked, future plants would be no better than past ones.

The legislative history of the Clean Air Act is crystal clear on this point. About section 165 of the Clean Air Act (the provision of the statute that requires sources like the proposed facility to use the Best Available Control Technology for pollutants like NOx and SO2) the Senate Report explained:

This procedure to prevent significant deterioration requires and case-by-case determination by the States of best available control technology for any new major emitting facility that will be built in a clean air region. Thus, each State is free to — and encouraged to — examine and impose requirements for the use of the latest technological developments as a requirement in granting the permit. This approach should lead to rapid adoption of improvements in technology as new sources are built, not the stagnation that occurs when everyone works against a single national standard for new sources.

S. Rep. No. 95-252, 95 Cong. Senate Report 127, *18. In other words, Congress intended BACT to be technology forcing, DEP’s protests to the contrary notwithstanding. By failing to acknowledge the Congressional intent, Western Greenbrier and DEP have contributed to the stagnation of technology that Congress was trying to prevent.

The notion that BACT is not technology forcing is contrary to EPA’s understanding of the policy as well. In its 1989 Background Statement on the Environmental Protection Agency’s (EPA’s) Top-Down Policy, EPA explained that

EPA believes that the top-down approach to BACT is supported by the statutory definition in section 169(3) of the CAA. The legislative history is clear that Congress intended BACT to perform a technology forcing function. . . . This construction was reinforced by [two PSD appeals decided by Environmental Appeals Board]. In those cases the Administrator interpreted the BACT definition as requiring the PSD applicant to demonstrate to the permitting authority why the most stringent control technology “available” is not “achievable” in that case.

Memorandum From John Calcagni, Director of Air Quality Management Division to Regional Directors Regarding EPA’s Background Statement on “Top-Down” BACT Analyis at 5 (June 13, 1989).

By failing to acknowledge BACT’s technology forcing function, Western Greenbrier and DEP failed to properly identify the best achievable control technology for the proposed facility. This failure infected DEP’s entire permit review and led it to issue the permit without discharging its fundamental duty to assure that the applicant meets its burden to demonstrate that there are significant obstacles to using the most stringent technologies for the control of NOx and SO2.

THE BACT ANALYSIS FOR SO2 PERFORMED BY WESTERN GREENBRIER AND APPROVED BY DEP IS FATALLY FLAWED

As discussed above, SO2 is a criteria pollutant for which an NAAQS has been promulgated. Consequently, under state regulations, SO2 is a regulated NSR pollutant that the proposed facility, as a major source, must control using BACT. 45 C.S.R. §§ 14-2.66.a & 14-8.2. However, there is insufficient evidence in the record or in the evidence presented at the evidentiary hearing to allow a reasonable person to conclude that an emission limit of 0.14 lbs/mmBtu constitutes BACT for SO2 for the proposed facility.

The two alternative technologies to control SO2 emissions on a facility of this kind are a flash dryer absorber (“FDA”) and wet flue gas desulferization (“WGFD”), the latter being commonly referred to as a “wet scrubber.” Transcript of 8/29/2006 at 158-59. WGFDs achieve lower emissions of SO2 than do FDAs, but require the use of higher quantities of water than do FDAs. Id. at 158-59, 261.Western Greenbrier’s rejection of a WGFD and its selection of an FDA were based on a purported top-down BACT analysis that was flawed in several ways.

First, there is insufficient evidence in the record to support the fundamental factor in the cost calculation of using a wet scrubber to control SO2. In its BACT analysis, Western Greenbrier calculated its estimate of the cost of a wet scrubber based on an estimate of $300/kw. CR at 251. The documents in the record attribute that figure to an “engineering estimate” and asserted, without support or explanation, that the $300/kw was “estimated from similar sources.” CR at 251-52. Remarkably, nothing in the certified record identifies those similar sources or the person that provided the $300/kw estimate.

At the hearing before the board, it came to light through the testimony of Jerry Joseph that the source of the $300/kw figure was Alstom, the vendor for the CFB, who had already designed the CFB to include a dry scrubber to control SO2 and not to use a wet scrubber. Transcript of 8/29/06 at 260, 286. Beyond that bare assertion by the vendor of the project—that was committed to providing its prepackaged CFB and FDA without a wet scrubber—nothing in the record supports such a figure. In fact, appellants’ expert testified at the hearing before this board that the cost-per-kilowatt of wet scrubbers were typically in the neighborhood of $100/kw. Transcript of 8/29/06 at 154. Nonetheless, Western Greenbrier and DEP uncritically accepted the unsupported $300/kw figure from Alstom and did not even seek a price from another vendor.

Based on that $300/kw “estimate” for a wet scrubber from the vendor of the dry scrubber to which Western Greenbrier was committed, Western Greenbrier represented that the cost of a wet scrubber to control SO2 would exceed $30,000,000, and DEP unquestioningly and uncritically accepted that number. The $300/kw figure became the prime factor in every calculation that followed. In fact, the figure by which a technology’s cost effectiveness is ultimately measured—the dollars per ton of pollutant removed—stems directly from that $300/kw “estimate” by the vendor. The cost per ton is calculated by dividing the total annual cost of the technology by total tons of pollutant that will be removed by that technology in a year. CR at 252. As appellants’ expert testified, the $300/kw figure was directly or indirectly a factor in many of the crucial preliminary calculations that ultimately determined the total annual costs of the wet scrubber. Transcript of 8/29/2006 at 153-54; CR at 252. As a result, Western Greenbrier represented that the cost per ton for SO2 removal by a wet scrubber was $4,594 for both the CFB and the kiln. Id. Thus, because Alstom’s concocted figure of $300/kw overestimates the true equipment costs, Western Greenbrier greatly overestimated the cost per ton of pollutant removed and to underestimate the cost-effectiveness of the technology.

Second, even assuming the validity of the $300/kw figure, Western Greenbrier made a significant miscalculation in its determination of the cost per ton of SO2 removed. Had DEP closely scrutinized Western Greenbrier’s BACT analysis for SO2, it would have discovered that Western Greenbrier overestimated the total capital costs of a wet scrubber by nearly $5,000,000. Table C-11, located at page 252 in the Certified Record, establishes that the total capital cost of a technology includes a “contingency” value that is 20% of the sum of the total direct costs and total indirect costs. CR at 252. Table C-11 also represents that the total direct cost of a wet scrubber would be $30,000,000 and that the total indirect costs would be $8,400,000, the sum of which is $38,400,000. Id. By appellants’ calculations, 20% of $38,400,000 is $7,680,000. However, Western Greenbrier represented to DEP, and DEP accepted uncritically, that 20% of $38,400,000 is $12,000,000. Id. That is, Western Greenbrier overestimated the contingency factor and, hence, the total capital cost of the wet scrubber—the driving factor in the BACT analysis—by nearly $5,000,000. Not only is such sloppy math symptomatic of the overall carelessness with which the BACT analysis in this case was conducted, it also provided Western Greenbrier and DEP a fictional cost-effectiveness basis on which to rest their rejection of a higher BACT emission level for SO2.

Third, even assuming that $4,594/ton of SO2 removed is an accurate estimate of the cost effectiveness of the wet scrubber, the wet scrubber is still cost effective by DEP’s standards. As EPA noted in its comments on the permit application, $4,594 per ton falls below DEP’s nominal cost-effectiveness threshold of $5,000 per ton, and, therefore, Western Greenbrier selected a less efficient control unit. CR at 121. In its fact sheet describing the proposed facility, DEP observed that technologies with a cost effectiveness of under $5,000 are generally not economically prohibitive. CR at 218. Nonetheless, DEP permitted Western Greenbrier to reject the emission level achieved by a wet scrubber as BACT on a cost-effectiveness basis.[4]

Finally, DEP’s use of $5,000/ton as a threshold for cost effectiveness is neither reasoned nor justified. Although DEP’s engineer testified about an EPA guidance document from the 1990s in which EPA purportedly endorsed such a threshold, no such guidance was introduced as evidence at the hearing, nor is included in the Certified Record. In fact, a 1997 Executive Order directs EPA that a cost of $10,000 per ton removed of ozone or particulate matter—which are criteria pollutants just like SO2—is “at the high end of the range of reasonable cost.” 62 Fed. Reg. 38421, 38429. In view of that directive, a cost of $4,595 per ton appears to be within the range of reasonable cost, contrary to the conclusions of Western Greenbrier and DEP, especially when one takes inflation since 1997 into account. Furthermore, even if the guidance document including the $5,000 threshold referred to by DEP’s engineer were dated 1999 (as opposed to 1990, since we only know that it was from the 1990s), seven years have passed and an adjustment for inflation should be implemented at the very least.