Inter-Power (PSC) - Recommended Decision June 1, 1992
Recommended Decision, June 1, 1992
STATE OF NEW YORK : BOARD ON ELECTRIC GENERATION SITING AND THE ENVIRONMENT
CASE 80010 - Application by Inter-Power of New York, Inc. for a Certificate of
Environmental Compatibility and Public Need to Construct and Operate a 200 MW Fluidized Bed,
Coal-Fired Cogeneration Facility in the Town of Halfmoon, Saratoga County.
RECOMMENDED DECISION ON REMAND
PRESIDING EXAMINER JANET HAND DEIXLER
- PROCEDURAL HISTORY
- PROCEDURAL COMMENTS
- THE INVENTORY
- DEC's Presentation
- Parties' Positions
- Massachusetts/Sierra Club
- Concerned Citizens for the Environment
- DEC's and Inter-Power's Responses
- Conclusion - Inventory
- THE MODELING
- Inter-Power's Presentation
- Class II PSD Modeling
- Class I PSD Modeling
- NAAQS Modeling
- Modeling Issues
- Modeling with Downwash
- CCE's Contention that Article VIII Does Not Permit Modeling Using the "Nearby/Other" Distinction
- Appropriate Sulfur in Fuel Content
- Appropriate Background Concentration
- Merge Methodology
- Correct Emissions Rate for Modeling the HCP; Vermont's Contention that HCP Should be Held to Lower Rate
- Assuredness of Impact Credit
- Emissions of NOx
- Minimum Adverse Environmental Impact
- Conclusion - Modeling
- Inter-Power's Presentation
- HEALTH RISK ASSESSMENT
- Inter-Power's Revised Health Risk Analysis Presentation
- Department of Health's Conclusions
- CCE's Position
- Appropriateness of HRA Inventory
- SO2 Emission Rate
- Conclusion - Health Risk Assessment
APPENDIX A - List of Appearances
APPENDIX B - List of Witnesses
STATE OF NEW YORK
BOARD ON ELECTRIC GENERATION SITING AND THE ENVIRONMENT
CASE 80010 - Application by Inter-Power of New York, Inc. for a Certificate of
Environmental Compatibility and Public Need to Construct and Operate a 200 MW Fluidized Bed,
Coal-Fired Cogeneration Facility in the Town of Halfmoon, Saratoga County.
APPEARANCES: SEE APPENDIX A
JANET HAND DEIXLER, Presiding Examiner: Associate Examiner Daniel E. Louis has prepared a separate report.
I. PROCEDURAL HISTORY
This proceeding was remanded by order of the Board on Electric Generation Siting and the Environment (Siting Board) issued June 27, 1991 Case 80010, Order Directing Further Proceedings (issued June 27, 1991)., primarily for further exploration of the air quality impacts of the proposed Halfmoon Cogeneration Project (HCP).
After a session, in May 1991, at which the Siting Board had voted to certify the facility, the Siting Board was advised by the Department of Environmental Conservation (DEC) of two major developments affecting the modeled air quality impacts of the proposed project. The first was that DEC had recently issued a permit to General Electric Company to replace two of its existing fuel-oil-fired boilers with a single boiler fueled by natural gas or oil, the effect of which was to reduce the existing General Electric Silicone Products Division (GESPD) plant air pollutant emissions. The second, more controversial, advice was that DEC had identified errors in the base line inventory of emissions from nearby sources which had been used by Inter-Power of New York, Inc. (Inter-Power) to model the air quality impacts of its project. The Siting Board noted, in its remand order, that DEC had corrected the data and supplied the corrections to Inter-Power, but that the significance of the error could not be determined until the model had been rerun with the corrected data.
After considering various motions filed by parties to the proceeding concerning action the Siting Board should take in light of the modifications, the Siting Board determined that the record in the proceeding should be reopened for supplementation in certain limited ways. The Examiners were directed to take the following steps:
"1. The permits for the new General Electric boilers should be incorporated into the record, and all previous analyses that used emission data (actual or permitted) from General Electric's Waterford facility should be redone in light of these permits and any other new or modified installations at the facility. [footnote omitted]
"2. The baseline inventory of relevant sources of SO2 should be corrected, and new modeling of the short-term impacts of SO2 should be performed on the basis of corrected inputs.
"3. The data used to model annual impacts of SO2 and NO2 should be corrected, and new modeling of those impacts should be performed on the basis of corrected inputs.
"4. To the extent the Health Risk Assessment relied on any of the foregoing data requiring correction, the corrected data should be used to reevaluate the health risk.
"5. Any other underlying data pertinent to the Board's decision discovered by its proponent to be in error should be corrected and studies based on the data redone as necessary." Case 80010, Order Directing Further Proceedings, pp. 3-4.
The Siting Board further directed that an expedited hearing procedure be developed for considering these matters, and it provided that at the conclusion of the hearing process, the Examiners would report and make recommendations directly to the Siting Board, without receipt of exceptions. The Siting Board expressed its intention that the entire process should be completed before the end of August, and it directed the Examiners to submit a status report if the proceeding could not comply with that deadline.
On July 9, 1991, a pre-hearing conference was held before Presiding Examiner Janet Hand Deixler. Attending the conference and indicating an intention to participate actively in the proceeding were the applicant, DEC, Department of Public Service staff (DPS), the State Department of Health (DOH), the State Energy Office (SEO), the Adirondack Park Agency (APA), Concerned Citizens for the Environment (CCE), Niagara Mohawk Power Corporation (Niagara Mohawk), a group of private intervenors represented by Henrietta O'Grady (Joint Intervenors), the Town of Halfmoon (Halfmoon), the Hudson-Mohawk Group of the Sierra Club (Sierra Club), the Commonwealth of Massachusetts (Massachusetts), the State of Vermont (Vermont), and the Bennington County Regional Commission. Members of the New York State Assembly Maurice D. Hinchey and Neil Kelleher later sought and were granted intervenor status in the proceeding.
DEC had requested the opportunity at the conference to present a technical briefing explaining the changes in the emissions source inventory to the parties. At the conference, Inter-Power stated its intention to complete the modeling of the new inventory data by August 2, 1991. Subsequent to the conference, however, the Examiners were notified that additional changes to the inventory required delay in completion of the modeling. Ultimately, an inventory which was considered complete by DEC was submitted to Inter-Power and the parties in mid-July, and the modeling, further delayed by modeling complexities, was completed and submitted with testimony on September 5, 1991.
At the conference, as well, DEC reported on its intention to retain, through the United States Environmental Protection Agency (USEPA), an independent consultant to review the emissions inventory and its development, and the modeling protocol and results. As a result of this proposal, as modified by a number of rulings, an independent audit was performed by Systems Applications, Inc. (SAI). The audit was primarily concerned with the emissions inventory, and, to a lesser degree, with the question of whether the modeling protocol was consistent with USEPA guidance. At the request of the Examiners, the SAI auditors appeared as witnesses to respond to questions about their review.
After Inter-Power had filed its testimony, dates were established for the rest of the testimonial submissions. The state agencies (DEC, DPS, and DOH) were directed to file their responsive testimony by October 18, 1991, all intervenors This included the independent SAI auditors and a representative of General Electric, both of whom were testifying at the request of the examiners. were to file by November 1, 1991, and limited rebuttal was to be filed by November 12, 1991. Hearings to adopt and cross-examine all testimony were scheduled to begin on November 19, 1991.
Hearings began on the appointed date and ran daily through November 26, 1991. However, from the start it was obvious that the July emissions inventory had still to be modified substantially. In response to the results of its own internal review process, as well as following suggestions made by intervenors and the SAI auditors, DEC had made abundant changes to the inventory by memoranda submitted on November 8, 1991, and it continued to make changes during the hearings. Inter-Power indicated its intention to remodel the further-revised emissions inventory. Intervenors strenuously objected to what they characterized as having to aim at a "moving target."
The hearings adjourned on November 26, 1991 with the expectation that no further revisions to the emissions inventory would be permitted, and that Inter-Power would file its new modeling on December 20, 1991.
After the close of hearings, numerous motions were filed by intervenors, seeking additional funding and alleging inability to continue to participate in the hearing process. On January 16, 1992, the Presiding Examiner issued a ruling which, while declining to recommend further funding for the intervenors, found that they had been disadvantaged by the continual modifications in the inventory and subsequent modeling, and directed that the December modeling should not be accepted into the record. Case 80010, Ruling on Motions for Further Funding (issued January 16, 1992). The ruling further directed that hearings previously scheduled to begin on February 19, 1992, should be held for the purpose of completing the record with respect to the matters already in the record: the July and November inventories and the September modeling.
Interlocutory appeals of the ruling were filed by DEC and Inter-Power. Pending a determination by the Siting Board on the interlocutory appeals, the February hearings were held, over two days.
At a session held on March 13, 1992, the Siting Board determined that the December modeling was necessary to complete the record, and, further, that information necessary to respond to air quality issues raised by USEPA in letters dated December 12, 1991 and December 23, 1991 could also be submitted. No further supplementation was to be permitted. The Siting Board also stated that a staggered hearing schedule should be adopted, so that intervenors were not required to present any evidence until Inter-Power and the state agencies had presented and been cross-examined on their evidence, that hearings should begin before the end of March and be completed in time to permit a final Siting Board decision by the end of June 1992, and that one round of exceptions to the recommended decision would be permitted. Case 80010, Order Granting Interlocutory Appeal in Part (issued March 16, 1992).
Hearings to cross-examine the filings made on March 18, 1992 by Inter-Power, DEC, and DOH were held on March 30, 1992 through April 1, 1992. The only responsive testimony was filed by CCE, to which DOH filed rebuttal testimony. Hearings on these filings began and concluded on April 27, 1992.
Initial Briefs were filed on May 11, 1992 by Inter-Power, DPS, DEC, DOH, Massachusetts, Sierra Club, CCE, Joint Intervenors and Vermont. Reply briefs were filed by Inter-Power, DEC, DOH, DPS, Massachusetts, CCE and Sierra Club on May 18, 1992.
The record on remand consists of approximately 3,420 pages of stenographic minutes (S.M.) and exhibits 263 through 423, including subparts, and excluding Nos. 330, 369, and 370. Seventeen witnesses testified in this phase of the proceeding: three for Inter-Power, three for DEC, one for DPS, two for DOH, two for CCE, two for Massachusetts and the Sierra Club, jointly, one for the Town of Halfmoon, and, at the request of the bench, one from General Electric and two for SAI. A list of witnesses is attached as Appendix B.
II. PROCEDURAL COMMENTS
Most of the Sierra Club's Initial Brief on Remand is devoted to a recitation of prejudice it alleges it and the other intervenors have suffered in the course of this proceeding. All of the intervenors, to a greater or lesser extent, assert that they have been prevented, by the procedural course of this case and by financial constraint, from intervening effectively. These parties claim, as well, that it was erroneous not to allow them to pursue their inquiry into responsibility for the miscalculations which led to the remand. In addition, they assert that it would be unfair for the Examiners and Siting Board now to consider allegations of intervenor witnesses' lack of expertise because the intervenors have not been permitted to demonstrate incompetence by the asserted experts who made the original errors.
DEC and Inter-Power assert that the intervenors have failed to demonstrate that their resources are depleted and argue that the applicant has been disadvantaged by the extra expense and effort required to produce air quality modeling which purportedly exceeds the normal expectations for detail and conservatism. They focus on the asserted lack of expertise of the witnesses offered by the intervenors, and ignore the question of responsibility for the remand.
The issue of responsibility for the earlier incorrect modeling is clearly beyond the scope of this remand, and I have refused to take evidence on the matter into the record. Case 80010, Ruling on Pending Motions (issued October 8, 1991). Case 80010, Ruling on CCE Motion to Compel and Information on Hearing Procedures (issued November 4, 1991). For this reason, I have not considered the issues raised concerning witnesses' expertise. The Department of Environmental Conservation has performed an internal investigation and, for purposes of this proceeding, DEC has assumed the responsibility to prepare the new inventory; Inter-Power, it is clear from the record, has merely accepted what DEC has provided and modeled it. The record compiled here suggests that, ordinarily, an applicant has responsibility to review the inventory information provided by a state agency. For purposes of this remand, particularly in view of the attention focused on the inventory by all parties, that requirement has no decisional consequence.
It is clear that all parties, including the applicant, have been adversely affected by the remand. Nonetheless, Inter-Power is the applicant and the burden of proof rests properly on it.
There can be no dispute, as the Siting Board has recognized on three separate occasions, Case 80010, Order Directing Further Proceedings (issued June 27, 1991); Case 80010, Order Denying Interlocutory Appeal (issued October 1, 1991) and Case 80010, Order Granting Interlocutory Appeal in Part (issued March 16, 1992). that the intervenors, who were not arguably responsible for the remand, have been disadvantaged by the depletion of the funds provided for Article VIII intervention on analysis of the inaccurate inventory and modeling. Lacking any authority to order additional funding, however, the Siting Board directed that the applicant and state agencies cooperate fully with the intervenors in order to assist their participation, and it provided, in a recent order, that proponent cases be completed before the intervenors were required to make theirs.
Despite these admonitions by the Siting Board, such cooperation was not forthcoming, and, as I stated in an earlier ruling, Case 80010, Ruling on Motions for Further Funding (issued January 16, 1992). there was active resistance to discovery. It is also true that, because of the schedule in the proceeding, intervenors focused time and resources on an inventory and modeling later abandoned by the proponents.
Ultimately, to some extent the November inventory, and to a greater extent the December modeling, have not been fully examined or challenged on the record. Whether this is the result of strategy or necessity is unclear and immaterial. Consistent with the directives of the Siting Board the information has been taken into the record, adopted under oath, and subject to limited cross-examination. In the course of my review, I have kept this history in mind. I have also considered, as discussed further herein, that much of the debate in this case has centered on the compliance of the modeling and inventory with USEPA Guidelines, but that the question of such compliance is, at bottom, within the province of USEPA.
I am convinced, as I discuss in more detail herein, that the record in this proceeding is sufficient for the Siting Board to make a determination on the air quality impact of this proposed plant for purposes of the balancing required by Article VIII. While some questions remain about compliance with USEPA guidelines, that agency will make those determinations in the course of its own proceeding; the Siting Board's determinations here should be made contingent, in any event, on a showing of such compliance acceptable to USEPA.
III. THE INVENTORY
DEC defines the baseline source inventory as a database of operating parameters for the major sources surrounding a proposed facility, . . . [which] is used in modeling the contribution to total air quality impacts of a proposed source. DEC's Initial Brief on Remand, pp. 1-2.
The original air quality modeling performed for this plant used ata that were assumed to represent the maximum permitted emissions of SO2 and NOx of each major source in the area affected by the proposed facility. Instead, as DEC informed the Siting Board in June 1991, the modeling had been performed using actual emissions.
Ultimately, the revised inventory used for air quality modeling in the remand phase of this proceeding included changes to the maximum permitted emission rates for the sources contained in the 1988 inventory, changes to the number of sources included in the inventory, and revisions to the stack parameters and locations of the sources. The chronology of the various iterations of the emissions inventory is summarized in Inter-Power's Initial Brief on Remand, pp. 9-10. These modifications were, as noted earlier, prompted by DEC's internal review, the independent review of the SAI auditors, and the criticisms of parties to the proceeding.
The inventory developed by DEC, as ultimately revised, was used for several different air quality analyses: compliance with National Ambient Air Quality Standards (NAAQS) and Prevention of Significant Deterioration (PSD) increment consumption. In addition, it provided input to the Health Risk Assessment (HRA). And, while much of the proceeding focused on the results of these analyses as they relate to compliance with these federal air quality standards, it is urged by some parties that such compliance is not synonymous with satisfaction of the requirements of Article VIII.
A. DEC's Presentation
DEC's testimony and brief describe in detail the development of the inventory, from the identification of major sources, to the various procedures followed to verify the source classifications and emissions parameters. At each step, DEC asserts, it took a conservative approach, ensuring that its inventory exceeded the requirements for air quality modeling, and contributing to what DEC views as conservative results.
Existing sources which DEC identified were classified as major or minor, using a threshold emission level of 100 tons per year (tpy) of either SO2 or NOx although 250 tpy is the usual NAAQS cutoff. According to DEC, this led to the inclusion of between ten and 16 emission points in the inventory which would otherwise have been eliminated.
According to DEC, the USEPA Guideline USEPA, Guideline on Air Quality Models (Revised 1986). provides that short-term emissions from existing point sources are to be modeled at maximum allowable emission limits, using design capacity operating level and assuming continuous operation. Annual emissions, however, may be modeled in what DEC terms a more realistic fashion, which distinguishes between "nearby" and "other" existing sources for NAAQS compliance modeling. "Nearby" sources may be modeled using maximum permitted emission limits at design capacity operating levels, using actual hours of operation. "Other" sources are modeled at maximum permitted emission levels, using actual capacity operating levels, and actual hours of operation. DEC states that it calculated maximum and annualized hourly SO2 and NOx emission rates for all major sources.
The Guideline also provides, according to DEC, that major sources should be considered "nearby" if they "caus[e] a significant concentration gradient in the vicinity of the source under review." DEC's Initial Brief on Remand, p. 8. Distant major sources can, in DEC's view, be assumed to be included in background levels or can be modeled as "other" sources. DEC considered as "nearby" all major sources which produced a modeled impact gradient of greater than 10% near the actual significant impact area (SIA) of the HCP. The actual significant impact area are those locations where the proposed facility has a significant impact; the circular significant impact area is a circular area with a radius extending from the source to the most distant point where modeling indicates a significant impact. By definition, the latter is larger than the former. USEPA Region II has asked for further "nearby/other" analysis using not only the circular rather than actual SIA, but including sources located up to an area 50 km beyond the circular SIA which have a significant concentration gradient within the SIA. This issue is discussed further infra.
DEC also provided inventory data for the PSD increment analysis. DEC explained that existing and proposed sources can be increment consuming, increment expanding, or increment offsetting. Pursuant to USEPA guidance, as DEC defines it, the PSD increment analysis uses actual emissions for increment expanding sources and allowable emissions for increment consuming sources. Citing as his basis USEPA's New Source Review Workshop Manual, New Source Review Workshop Manual, Draft, October 1990. DEC's witness Leon Sedefian defined actual to mean "maximum actual" source operations in the case of short-term impacts; in the case of annual impacts, he considered actual to mean "average actual" emissions. Mr. Sedefian explained that, therefore, short-term impacts were based on permitted maximum allowable emissions, while for annual impacts, he used annual average emissions data based on the most recent two years.
B. Parties' Positions
One of the primary arguments of the intervenors in this proceeding has been that the later iterations of the inventory, with changes made subsequent to the SAI audit and the intervenors' review, should, like the later modeling exercises, not be accepted because of the inability of the intervenors to review them thoroughly. These arguments are not considered in this Recommended Decision, because the Siting Board, in its March 16, 1992 Order, considered and rejected them, and determined that this information was necessary to a complete record.
1. Massachusetts/Sierra Club
Massachusetts and Sierra Club jointly sponsored testimony by two witnesses, Dr. Colin High and Mr. Kenneth Kaliski. They assert that the inventory is flawed in a number of respects. The High/Kaliski panel addressed the July inventory and September modeling. Assertedly because of financial constraint, they were not asked to review either the November inventory or the December modeling.
Dr. High and Mr. Kaliski allege that the screening for "nearby" and "other" sources was inconsistent with USEPA guidance, which requires that the sources reviewed be located in the circular SIA, or in an area within 50 km of the circular SIA if they have a significant concentration gradient within the SIA. They criticized DEC for considering only the smaller, actual SIA. Massachusetts contends that this deficiency eliminated sources which should have been included in the inventory; 43 sources should assertedly have been screened within the 47 km circular SIA and more sources, which are not enumerated, allegedly exist in the outer 50 km band.
In its Reply Brief on Remand, Massachusetts notes that, for the November inventory, DEC screened sources located within the 97 km band around the HCP, but it argues that the screening methodology was not accepted by USEPA. In Massachusetts' view, its witnesses' position continues to be the view of USEPA, despite the attempts of DEC to convince the federal agency to adopt its position.
The Massachusetts/Sierra Club witnesses contend, furthermore, that minor sources should have been included in the inventory for PSD increment purposes. Failure to include these sources, according to Massachusetts/Sierra Club, understates cumulative air quality impacts for the HCP. Indeed, in Sierra Club's view, DEC should have developed separate inventories for the NAAQS and PSD analyses, and asserts that these are required by the New Source Review Workshop Manual, supra.
2. Concerned Citizens for the Environment
CCE similarly insists that the inventory excludes sources which should have been included. On brief, CCE argues that DEC failed to consider Owens Corning (a major source for NOx), should have included Atlantic Cement (which potentially emits over 8,000 tpy of SO2), and did not include Seagroatt Florist and Williams College. Most flagrant, in CCE's opinion, was the omission of two fixed-box incinerators at General Electric's Waterford site (GESPD), which are assertedly capable of emitting more than 100 tpy of SO2 and which are not subject to any agreement to keep their emissions below that level.
CCE claims, as well, that several of the parameters of inventory sources are incorrect, and that this led to other errors. For example, CCE's witness Ziegler contends that a number of source stacks are considerably shorter than suggested by Good Engineering Practice, but were not, as they should have been, modeled with "downwash." "Downwash" is a phenomenon of increased impacts in the vicinity of a source having short stacks relative to the height of surrounding buildings.
3. DEC's and Inter-Power's Responses
Inter-Power and DEC both contend that DEC's inventory, including its method of distinguishing between "nearby" and "other" sources, is acceptable in all respects. Inter-Power notes, however, that USEPA's concerns pertaining to the screening method and area considered in the nearby/other determination have effectively been mooted. In subsequent modeling, submitted to USEPA in February and March, Inter-Power treated as "nearby" all sources located in the circular SIA or located in the 50 km ring with a significant gradient at the edge of the circular SIA; Inter-Power states that there were no accedences. Inter-Power's Initial Brief on Remand, pp. 52-53; Exhibits 405 and 406. In any case, according to Inter-Power, contrary to the claims of Dr. High and Mr. Kaliski, the inventory includes all sources which should have been included.
DEC, similarly, asserts that its witness, Mr. Sedefian, has responded adequately to USEPA's concerns about the division of sources into "nearby" and "other." While acknowledging that it has a continuing dialogue with USEPA about suitable screening methods for future air quality modeling, for purposes of this proceeding, it has complied with USEPA's procedures. Mr. Sedefian testified that the federal agency had been informed of this by letter dated February 10, 1992 Exhibit 416. and that USEPA had indicated it was satisfied.
As concerns Massachusetts'/Sierra Club's claim that the PSD inventory was incomplete, DEC and Inter-Power contend that sources identified by Dr. High and Mr. Kaliski as not included in the inventory were appropriately excluded, based on proper information about start dates and emission rates. DEC's witnesses testified that they contacted the identified sources and found that most predated the applicable PSD baseline dates. In any case, Inter-Power argues, the PSD inventory overestimates PSD increment consumption because it does not include a number of identified increment-expanding sources which discontinued operations subsequent to the development of the PSD baseline or because there is data available to demonstrate that the sources, in practice, burned a lower sulfur content fuel than the permitted content, which was assumed in the PSD baseline. According to Inter-Power, none of these increment expansions is included in the inventory, providing "even greater confidence that the operation of the HCP will fully comply with the PSD increment consumption regulations and guidelines." Inter-Power's Initial Brief on Remand, p. 20. Mr. Sedefian, for DEC, testified that, while both major and minor sources may affect PSD increment, it is unreasonable to expect that "every change at every source in the Capital District should be accounted in a source specific permit application." S.M. 9384. Although there are, in fact, numerous minor sources in the inventory, DEC asserts, it is not unreasonable to limit the set of PSD-affected stacks to those in the inventory. Indeed, in DEC's view, the major reduction in the areawide sulfur content of fuel compelled by the State Acid Deposition Control Act would more than offset any increment consumption increase. Mr. Sedefian noted that states are obligated, under the Clean Air Act, to review changes in increment consumption and expansion every five years.
Furthermore, DEC contests Sierra Club's contention that any guidance requires the establishment of two completely separate inventories, noting that there is no such requirement in the New Source Review Workshop Manual, supra.
With regard to CCE's contentions that major sources of SO2 and NOx within the SIA were improperly excluded from the inventory, DEC replies that all sources identified by CCE were appropriately evaluated and excluded by its witnesses. In particular, DEC observes that the fixed-box incinerators at General Electric will soon be restricted to emissions below 100 tpy, and that General Electric has committed itself to accepting these permit reductions. Attached to DEC's Reply Brief on Remand is an internal DEC memorandum demonstrating that the department is in the process of incorporating the restrictions into the permits.
Inter-Power, seconding DEC's contention that the identified sources were properly excluded, states that it performed limited source-specific modeling of three identified sources, and found that, at a maximum, they changed cumulative concentrations by 0.04%. Therefore, in Inter-Power's view, any omitted source is unlikely to affect its modeling results significantly. Inter-Power's testimony on this point is apparently no longer questioned by USEPA.
C. Conclusion - Inventory
The Siting Board directed that the emissions inventory submitted in this proceeding should be "corrected" so that air quality modeling of the proposed facility could be performed. My review of the record in this remand reveals substantial compliance with that directive.
The base line emissions inventory presented by DEC and supported by the applicant, through undergoing numerous iterations, has been exhaustively reviewed by numerous parties, including an independent auditor. Undoubtedly, errors continue to exist in the inventory, but, as I have previously ruled, Case 80010, Ruling on Motion for Further Funding (issued January 16, 1992). and as the Siting Board has found, Case 80010, Order Granting Interlocutory Appeal in Part (issued March 16, 1992). 100% accuracy is probably neither attainable nor necessary.
It is not precisely clear why accuracy appears impossible, but the record shows that even site visits to the same facility can yield different information depending on the person interviewed at the site and the exercise of judgment by the person compiling the inventory.
There has been a great deal of controversy in this proceeding pertaining to compliance of this inventory with USEPA's guidelines for inventory compilation. There have been many letters to and from USEPA marked as exhibits intended to demonstrate that some procedure or other did not conform with USEPA requirements. But USEPA will, ultimately, make its own decision as to whether its rules have been followed, and Inter-Power, whether or not a certificate is granted here, will need USEPA's approval to operate the HCP. Therefore, while the question of precise compliance with USEPA guidance is relevant to this proceeding, it is not necessarily determinative of it.
In this instance, most of the omissions from the inventory challenged by Massachusetts/Sierra Club witnesses or CCE witnesses have been shown to be properly excluded from the inventory. The fixed-box incinerators at General Electric (GE), in particular, have been demonstrated to have SO2 and NOx emissions below the major source threshold. Other flaws alleged by intervenors or the SAI auditor have been rectified or mooted by later iterations of the inventory or modeling. And while, as noted above, the inventory is probably still not perfect, there is no evidence that the alleged remaining inconsistencies with USEPA requirements are material or would cause a significant difference in the modeling outcome. Indeed, evidence submitted by Inter-Power and DEC suggest that the proposed changes would have a minimal impact at most.
Moreover, the record is convincing that, in compiling the inventory, a number of conservative assumptions were made. For example, although USEPA considers 250 tpy the cutoff for a major source, sources were considered major if their emissions exceeded 100 tpy of either SO2 or NOx. This conservatism would tend to offset sources erroneously omitted, if indeed there were any.
IV. THE MODELING
The air quality modeling protocol in this proceeding is asserted to be essentially the same as submitted and approved in the first phase of this proceeding. The actual air quality modeling in this phase, as distinguished from the unchanged protocol, is considerably more complex. This is because the original modeling, based on emissions figures which were too low, did not trigger the need for more refined grid modeling, or the division of sources into "nearby" and "other."
A. Inter-Power's Presentation
Inter-Power performed at least two sets of modeling for this proceeding, one submitted in September 1991 (the September modeling), and one performed in December 1991 but submitted with its testimony in March 1992 (the December modeling). In addition, various supplemental modeling exercises designed to demonstrate contentions to USEPA or to moot issues raised by USEPA or a party were performed. Most of these are contained in Exhibits 405 and 406, the February and March 1992 submissions to USEPA, respectively.
Inter-Power urges the Siting Board to rely on the December modeling (based on the November inventory), but it contends that either modeling result demonstrates that the HCP will meet applicable air quality standards provided that appropriate offsets are obtained from GESPD.
The applicant characterizes the air quality modeling as, "in effect . . . a 'black box' which is unchanged between 1989 and 1991." Inter-Power's Initial Brief on Remand, p. 22. In Inter-Power's view, the aspects of the modeling which opponents contend are modeling modifications were, in fact, anticipated by the earlier protocol, and hence consistent with it. Examples of such changes are lowered background levels resulting from use of average observed background values, rather than the "highest second-highest" maximum background value; the division of sources into "nearby" or "other" for modeling annual impacts. All sources are modeled at maximum permitted emission rates for short-term impacts; the inclusion of increment expansion sources in the modeling; the use of an updated ISC model, required by USEPA for PSD modeling; and the reduction from six to two of the sources modeled with downwash. All of these changes, with the exception of the mandated and therefore unavoidable update in ISC model, Inter-Power contends that, in any event, the change in model had little impact on the results; are, assertedly, part of the "black box" modeling protocol.
The estimated modeled cumulative air quality impacts resulting from the December modeling, according to Inter-Power, demonstrate compliance with three regulatory standards: (a) modeled PSD increment consumption limits in the vicinity of the HCP (a class II PSD area), (b) modeled PSD increment consumption at the Lye Brook Wilderness Area in Vermont (Lye Brook; a class I PSD area), and (c) compliance of the HCP with NAAQS.
1. Class II PSD Modeling
The applicable SO2 PSD increments in a Class II area are:
512 micrograms per cubic meter (g/m3) for the 3-hour averaging period;
91 g/m3 for the 24-hour averaging period; and
20 g/m3 for the annual averaging period.
For NO2, there is no short-term increment consumption limit. The annual increment limit is 25 g/m3.
Significant impact levels (SIL) for a single source are 25 g/m3, 5 g/m3, and 1 g/m3 for the 3-hour, 24-hour, and annual The annual NO2 SIL is also 1 g/m3. SO2 averaging periods, respectively.
Inter-Power observes that there is at least one receptor/time period combination where the PSD inventory consumes more than the allowable short-term SO2 increment. However, the HCP impact at those points is below the SIL; Inter-Power asserts, therefore, that these receptor/time periods are not of concern in determining the compliance of the HCP with Class II PSD increment consumption limits. Similarly, Inter-Power has identified other times and receptor locations where total cumulative impact is below the maximum, but still above the allowable increment and, similarly, the HCP impact at those points is below SIL.
Finally, Inter-Power asserts that the HCP impacts fall below the allowed percentage of available increment.
2. Class I PSD Modeling
The Class I PSD area close to HCP, Lye Brook, is about 56 km from the proposed site. Modeling for Class I PSD areas, according to Inter-Power, uses the same PSD source inventory and modeling as Class II PSD modeling, but the applicable increments vary. For SO2 the 3-hour, 24-hour, and annual Class I increments are 25 g/m3, 5 g/m3, and 2 g/m3, respectively. For the annual NO2 period, the applicable increment is 2.5 g/m3.
Inter-Power asserts that its modeling demonstrates PSD compliance; it notes that the highest 3-hour SO2 value exceeds the applicable increment, but observes that PSD regulations permit one such exceedence at any given receptor per year. The "highest second-highest" value for the 3-hour averaging period is 21.35 g/m3, below the allowable increment, so that, it is asserted, compliance is demonstrated. In the Class I modeling, as in the Class II modeling, Inter-Power alleges that it does not consume more than the allowed percentage of available increment.
3. NAAQS Modeling
NAAQS modeling is intended to "accumulate the modeled air quality impacts at each receptor/time period combination for the NAAQS inventory sources, [add] this cumulative impact to a measured background value, and [compare] this to the NAAQS to assess compliance." Inter-Power's Initial Brief on Remand, p. 33. For the 3-hour, 24-hour, and annual SO2 averaging periods, the standards are 1300 g/m3, 365 g/m3, and 80 g/m3, respectively, and for the annual NO2 averaging period, the standard is 100 g/m3.
The December modeling identified a number of receptor/time periods where the relevant standard was exceeded. In Exhibit 394, Inter-Power identified the receptor/time periods at which an exceedence occurred simultaneously with a modeled HCP concentration in excess of the SIL.
Annual NAAQS compliance, according to Inter-Power, is demonstrated by modeling each source at its annual emission rate, calculating the "average annual impact . . . for each source at each receptor over the 8760 hour meteorological year, adding the calculated impacts from each source (including the HCP) and a representative background together, and comparing the cumulative all sources totals to the annual standards . . . ." Id., p. 36. Furthermore, Inter-Power states that the annual modeling uses the same inventory, meteorological year, air quality impact models and "merge" methodology as the short-term modeling and as was used in its original modeling. The emission rates used in the annual NAAQS modeling, which witnesses for Inter-Power and DEC testified were consistent with USEPA guidance, were based on the classification of sources as "nearby" and "other," as discussed earlier in this Recommended Decision. "Nearby" sources are modeled at emission rates which assume the worst case fuel and maximum heat input, but, instead of assuming that the source operated at all 8760 meteorological hours, actual data from the most recent two years is used to prorate the maximum permitted emission rate. "Other" sources are modeled similarly to "nearby" sources, but the emission rate is further modified to reflect actual hourly heat input rather than assuming maximum potential heat input. Using this method, Inter-Power concluded that there are no accedences of annual NAAQS.
Short-term impacts are modeled using short-term permitted emission rates. This generally means that sources are assumed to operate using the worst case fuel at maximum heat input for all 8760 hours in the meteorological year; this would not be the case if there were limiting permit conditions. Inter-Power's modeling witness testified that he first modeled short-term impacts using coarse inner and outer grids. A coarse grid is based on relatively large grid spacing, and receptor elevations are defined by the highest elevation in a grid square. As an example, Inter-Power notes, "if the coarse grid spacing is one kilometer, the receptor would be located in the center of the one kilometer square, but the elevation chosen for that receptor would be the highest elevation within the one kilometer square area. The coarse grid, therefore, could 'move' the highest elevation terrain up to approximately one half a kilometer closer to a source than the actual geography would place it." (Inter-Power's Initial Brief on Remand, p. 40.) He then identified receptors where coarse grid results suggest all- sources impacts are above NAAQS when HCP's impact is above SIL, and, based on coarse grid results, ran a series of refined grids to determine precisely, at the point of an all-sources NAAQS exceedence, where and when the HCP has an impact in excess of the SIL. Based upon this analysis, Inter-Power states that the receptors can be identified at which modeled HCP impact is above the applicable SIL when the cumulative all-sources impact is above the applicable standard; for all other receptor/time period combinations, NAAQS compliance is demonstrated by the modeling to this point.
Inter-Power states that most of the areas of concern are located in a refined grid south of the HCP, with nine such areas of concern in a smaller refined grid to the north of the HCP, and five such areas in a smaller refined grid north northeast of the HCP. It is asserted by Inter-Power that compliance with NAAQS can still be demonstrated, even where such accedences have been modeled, if an emission reduction can be achieved at an existing source which reduced the impact of the proposed source to below the applicable SIL. Inter-Power claims it has negotiated an impact credit with General Electric by which the maximum sulfur content of the fuel burned at GESPD emission points 2 and 3 would be reduced from 1.5% to 1.3%.
The impact credit at the GESPD is sufficient to achieve NAAQS compliance, according to Inter-Power, at all but one receptor/time period combination. That one receptor/time period is located in a refined grid located to the east, northeast, and southeast of the HCP. Specifically, this is at receptor 124 during hours 4, 5, and 6 of day 297. At those points, the cumulative all-sources impacts (including background) are modeled at 1309 g/m3 (versus the 3-hour NAAQS of 1300 g/m3) and the HCP impact is 26.37 g/m3 (versus the 3-hour SIL of 25 g/m3).
Although during the course of the proceeding Inter-Power suggested a number of alternatives for eliminating that exceedence, including reducing the level of emissions at the HCP from 0.22 lb/MMBtu to 0.208 lb/MMBtu, at the conclusion of hearings and in brief, Inter-Power contends that the modeled exceedence is eliminated by the reduced emissions rate at Norlite, a major contributor to the all-sources violation.
B. Modeling Issues
Because testimony responsive to the December modeling was extremely limited, modeling issues discussed herein were actually raised with reference to the September modeling. Some of the issues were effectively mooted by the December modeling, but others must still be considered. Issues concerning the identification of sources as "nearby/other" have been discussed in the section of this Recommended Decision on the inventory. In addition, some issues raised in the course of the proceeding were not discussed in detail on brief; I have assumed, therefore, that they are no longer pressed by those parties, and they are not discussed in this Recommended Decision.
1. Modeling with Downwash
The intervenors have asserted that Inter-Power failed to model enough sources with downwash. Inter-Power observes that, while it modeled six sources with downwash in the 1988 air quality modeling, only two have been so modeled in its most recent modeling. This, as it has explained to USEPA, is because the phenomenon of downwash can only be relevant if it occurs at a source close to the location where the HCP exceeds the USEPA significant impact level (SIL). Since the new modeling relies on an HCP emissions rate of 0.22 lbs/MMBtu, rather than the 0.26 lbs/MMBtu of the earlier modeling, and the area where the HCP has an impact above the SIL is consequently reduced, the number of sources located close to that area is also reduced.
Inter-Power's explanation is reasonable; there is no reason to require additional downwash modeling.
2. CCE's Contention that Article VIII Does Not Permit Modeling Using the "Nearby/Other" Distinction
CCE contends that 806.4(c) of the Article VIII regulations requires aerometric modeling which is "based upon both maximum expected and maximum allowable mass emission rates for each pollutant, where appropriate (emphasis added)." CCE's Initial Brief on Remand, p. 10, citing 16 NYCRR 806.4(c)(11). In CCE's view, this requires Inter-Power to model only a "worst case scenario."
CCE states that the original NAAQS analysis presented by Inter-Power was labeled a "worst case scenario" because it supposedly modeled existing sources with maximum allowable emissions; it found no significant air quality impact violations. CCE argues that Inter-Power has departed from this "worst case" by such techniques as the denomination of sources as "nearby" and "other" and refined grid modeling, and it suggests that this departure is contrary to the intent and requirements of Article VIII. Another departure from a conservative posture cited by CCE is the reduction in background emissions utilized by Inter-Power in the more recent modeling attempts (26 g/m3 v. 160.9 g/m3). In CCE's view, it was inappropriate for Inter-Power to depart from a "worst case" analysis of the air quality impacts of the HCP.
Inter-Power insists that the modeling it has performed contains a number of conservative assumptions, such as the modeling of all PSD increment-consuming sources as "nearby," and the selection of 100 tpy (rather than 250 tpy specified by USEPA) as the cutoff for major source categorization. It contends its modeling is consistent with Article VIII, noting that 806.4 requires modeling with maximum expected and allowable emission rates where appropriate; all regulatory agencies, it observes, have endorsed the "nearby/other" distinction as appropriate.
DEC points out that "worst case" air quality modeling is acceptable to USEPA as a "labor saving device to determine if more refined modeling is required, with the latter being used for final air quality decisions." DEC's Reply Brief on Remand, p. 8. DEC notes that the section of Article VIII (806.4) on which Inter-Power relies is actually pertinent to the source seeking certification, and it contends the HCP modeling is consistent with that regulation.
The modeling refinements applied by Inter-Power do, indeed, abandon the "worst case" analysis originally presented by the applicant. In this instance, such a "worst case" approach would show failure to comply with applicable air quality requirements.
However, the procedures adopted by Inter-Power are acceptable to the state and federal agencies charged with evaluating air quality. And, it must be remembered, the worst case bears little relationship to air quality conditions which can be expected to occur.
Even the refined modeling assumes events unlikely to occur; and the violations encountered are not anticipated to occur, but are "modeled" violations based on the particular assumptions of the model. Moreover, as discussed earlier, the modeling here contains a number of conservative assumptions. Contrary to CCE's contentions, the modeling protocol is not inconsistent with Article VIII, which does not require a worst-case analysis of the sort CCE advocates.
3. Appropriate Sulfur in Fuel Content
The NAAQS inventory was modeled assuming a sulfur in fuel content of 1.5% because this is the maximum sulfur content permitted by state regulation. 6 NYCRR Part 225-1. Because this regulation has not been incorporated in the USEPA State Implementation Plan (SIP) for New York, several parties have urged that the sources should have been modeled using the SIP sulfur content of 2.0%. CCE, Massachusetts, and Sierra Club argue that if Inter-Power had modeled the sources using the 2.0% sulfur content, many more accedences would have been modeled.
CCE contends that a conservative modeling approach would have adopted consistency between the NAAQS and PSD inventories with respect to the sulfur content of fuel. CCE points to a letter from USEPA which refused to accept, for PSD increment expansion purposes, the "blanket assumption" that an expansion credit was due for sources in operation when the state sulfur limit was decreased (in 1988) from 2.0% to 1.5%. Exhibit 280. In that letter, USEPA stated that the reduction would have to be demonstrated on a case-by-case basis. For this reason also, CCE would discount Inter-Power's assertion that its modeling is conservative.
Massachusetts makes a similar argument, noting that a number of inventory sources have been demonstrated to be operating in excess of permit restrictions. Exhibit 413. Sierra Club also mentions this issue.
Inter-Power contends, in reply, that the enforceable limit on the sulfur content of fuel in New York is 1.5%, and that reliance on the superseded 2.0% limit would not be, as CCE claims, conservative, but simply engaging in a "game of 'let's pretend.'" Inter-Power's Reply Brief on Remand, p. 8. The law in New York, as Inter-Power contends CCE admits, has, for a number of years, provided that the maximum sulfur content is 1.5%. Inter-Power states:
Whatever the impact in this proceeding, if applied uniformly to permitting in New York State, the CCE theory would result in sources failing to meet regulatory requirements when their impacts could not, as a matter of law contribute to NAAQS violations. Id.
DEC contends that CCE and Massachusetts have confused USEPA's position on PSD increment expansion with its requirements for NAAQS modeling. For the former, DEC says it now agrees that a case-by-case showing must be made that sources previously permitted to use 2.0% sulfur fuel were, in fact, using it when the limit changed. In the case of the latter, DEC states that the Guideline on Air Quality Modeling provides that the emission rate should reflect the lower of the maximum allowable rate or federally enforceable limits; DEC asserts that USEPA has agreed that the state limit should be used.
The opponents' arguments are not persuasive. No compelling reason has been offered to assume the higher federal sulfur limit for use in the air quality modeling, and it does appear that the intervenors have confused the need to prove actual usage when determining the existence of a PSD increment expansion credit and the logic of using enforceable limits to model air quality impacts.
CCE suggests that the higher limit should be used because the record has revealed a number of emission sources in New York which burned fuel with a sulfur content in excess of 1.5% in 1989 and 1990. If the information submitted by DOH in Exhibit 355 is typical of New York State sources, CCE contends, we can have no assurance that the 1.5% limit is being observed. Massachusetts similarly argues that Exhibit 413 shows that a number of inventory sources are operating in excess of permit limitations. But the fact that some sources have at times exceeded emission limits does not mean that the limits are unenforceable or should not be used. Indeed, Exhibit 413 details the steps being taken by DEC to enforce applicable restrictions.
Again, USEPA will ultimately determine for itself what data conform to its requirements, and it is not necessary for that issue to be resolved here. For Article VIII purposes, the proponents are convincing that the sulfur content required by state regulation is the appropriate limit to be reflected in the modeling.
4. Appropriate Background Concentration
As discussed earlier, Inter-Power used an average background concentration developed from hourly observed background values in lieu of the "highest second-highest" maximum background value used in the earlier modeling. The effect of this was to lower the background value to be added to cumulative modeled impacts from 160 g/m3 to 26 g/m3. Inter-Power contends that this change in background value was contemplated by its original modeling protocol, which anticipated that the average concentration would be used if the combination of cumulative impacts and "highest second-highest" maximum background concentrations reached 95% of the applicable standard.
CCE considers this reduction "sleight-of-hand" and arbitrary CCE's Reply Brief on Remand, p. 8. and asserts that the background values used by Inter-Power are another example of the abandonment of conservatism. Because air quality modeling, in CCE's view, is not precise, CCE contends that "conservatism must rule". Id. CCE asserts that this is precisely the same type of error as led to the remodeling initially, because it uses an annual average to model short-term impacts. Moreover, CCE argues, it is immaterial that the background refinement used here was anticipated in the 1989 protocol. In CCE's view, the 1989 modeling was erroneous, resulting in the remand, and consistency with it is not persuasive.
Inter-Power responds that CCE's argument amounts to no more than a complaint that the background values utilized differ from those used in the 1989 modeling. In Inter-Power's view, however, it has approached the development of a background value appropriately. Essentially, according to Inter-Power, the background used in the December modeling is "the background associated with the specific set of meteorological conditions responsible for the cumulative impacts resulting in a violation at the particular receptor/time period of concern to which the background value is being applied." Inter-Power's Initial Brief on Remand, pp. 50-51.
Massachusetts and Sierra Club also contend that the background value utilized in the modeling was too low. Inter-Power and DEC, however, insist that the background value and method of calculating it have both been accepted by USEPA. Indeed, according to Inter-Power, pursuant to USEPA guidance, "other" sources can be assumed to be represented by background values and not explicitly modeled; in its modeling, however, this assumption was not made and "other" sources were explicitly modeled, leading to some double counting of these sources. Id., p. 24.
The fact that the background value computation method is consistent with the 1989 protocol does not automatically establish its acceptability although it should be remembered that the scope of the remand involves primarily the inventory and its remodeling, not the modeling protocol. On the other hand, the mere fact that the background value is considerably reduced from the 1989 modeling does not provide confirmation that the level is too low.
While this again is an area where USEPA will make its own determinations, it is of some interest that the federal agency has apparently not questioned the background value used in the December modeling.
Inter-Power has provided a reasonable basis for the procedure it followed, while CCE's argument amounts primarily to a contention that a higher number would have been more conservative. In this instance, there does not appear to be a compelling reason to use the higher value, regardless of its use in the earlier modeling.
5. Merge Methodology
In modeling short-term impacts, Inter-Power modeled each receptor for each time period with both a simple terrain model (ISC model) and a complex terrain model (COMPLEX I for all sources except the HCP and GESPD, and RTDM for HCP and GESPD), and, after comparing the two results (by means of another computer program termed a "merge" program), selected the higher of the two values for use in the cumulative analysis. USEPA guidance provides that this "merge methodology" should be used in intermediate terrain (where the receptor is located above the stack height, but below the plume centerline for a source); for simple terrain (where the receptor is located below stack top) USEPA guidance directs use of the simple terrain model and for complex terrain (where the receptor is located above the centerline of the plume) it directs use of the complex terrain model. Inter-Power contended that its procedure yielded a more conservative result by always using the higher impact value.
The independent auditors pointed out that this method was not consistent with USEPA guidance. They agreed that it provided a more conservative result for cumulative impacts and for PSD increment consumption. Under cross-examination by Vermont, the SAI auditors agreed that use of the merge procedure for calculating PSD increment expansion and the GE offset could possibly result in their overstatement.
In its Initial Brief, Vermont asserts that, although this method was originally approved by USEPA, the approval is not dispositive, because in neither the 1989 modeling nor the September 1991 modeling was increment expansion discussed.
Inter-Power contends that separate modeling of increment expanding and consuming sources, such as it asserts Vermont is suggesting, would be both contrary to USEPA guidance and analytically unsound because the modeling protocol would be internally inconsistent. In any case, Inter-Power asserts, it provided additional analysis of the Lye Brook Wilderness Area to test Vermont's theory that its modeling procedure overstates increment expansion. This additional analysis, according to Inter-Power, demonstrates that the merge methodology does not overstate the net PSD increment.
DEC agrees with Inter-Power that the merge methodology followed by the applicant was a reasonable procedure. DEC's witness Sedefian contended that all currently approved modeling analysis is consistent with the approach followed by Inter-Power. Furthermore, DEC is in accord that the additional modeling performed by Inter-Power demonstrates that the merge methodology does not overstate available increment by overstating increment expansion.
In reply to Inter-Power's additional modeling, Vermont insists that the result is applicable only to Lye Brook, and that the validity of the modeling had thus not been demonstrated outside the confines of Lye Brook. Vermont alleges that the merge methodology is "a mistaken approach which departs from EPA guidance and which may have overstated PSD emission expansion figures and NAAQS offsets from GE Silicone." Vermont's Initial Brief on Remand, p. 13.
It appears that the use of the merge methodology for all receptors, rather than only for intermediate terrain receptors, was not precisely in conformance with USEPA guidance, although there is no dispute that the proposed protocol was approved by USEPA and DEC. Again, precise conformance with USEPA guidance is yet to be determined by USEPA. Vermont notes that there was no discussion of increment expansion when USEPA reviewed the modeling protocol in 1988 and 1991, but that does not mean that the possibility of increment expansion was overlooked. Moreover, as DEC's witnesses testified, the merge methodology is consistently used for intermediate terrain receptors, and the issue raised by Vermont applies to the merge methodology in general, and not simply to its untraditional use at complex terrain receptors. Inter-Power and DEC state that the use of the merge methodology at simple terrain receptors is not likely to cause a problem, because the simple terrain model will almost always be the higher value chosen.
Inter-Power hypothesized that the effect on net PSD increment of using the merge methodology would be neutral, because if increment expansion were thereby overstated, increment consumption would similarly be overstated. Its additional analysis demonstrates that, at least in the Lye Brook Wilderness area, its hypothesis is reasonable. While Vermont suggests that the analysis might not be valid for other geographical areas, its contention is merely speculation, and is not supported by any evidence in this record. And even if Vermont were correct, it is not clear that such overstatement of available increment would be material to the HCP; the Lye Brook Wilderness area is the most sensitive PSD area involved in the application.
6. Correct Emissions Rate for Modeling the HCP; Vermont's Contention that HCP Should be Held to Lower Rate
HCP emissions were modeled at an SO2 emission rate of 0.22 lbs/MMBtu, the rate approved by the Siting Board at the May 1991 session when it tentatively approved the HCP, and contained in the draft USEPA permit issued last year. DEC states that the rate will be verified by stack tests and will be "complied with on a 3-hour average basis." DEC's Reply Brief on Remand, p. 10. Previous modeling for the HCP was performed using a maximum emissions rate of 0.26 lbs/MMBtu, which had been proposed by Inter-Power.
CCE contends that, since the HCP can, on a one-hour basis, emit SO2 at a rate of 0.26 lbs/MMBtu, the modeling should have been done on that basis. CCE contends that Article VIII requires that aerometric analysis be based "upon maximum facility load and any other loading that might produce a larger air pollutant concentration for averaging periods of 24-hours or less" and be based upon "maximum expected and maximum allowable mass emission rates for each pollutant." 16 NYCRR 806.4(c)(6) and (11). Article VIII also requires that meteorological data be based on "one hour periods centered on the half hour." 16 NYCRR 806.2(f). CCE acknowledges that the use of the 3-hour average may not be inconsistent with USEPA requirements for PSD modeling, but it contends that it should not be allowed for Article VIII purposes.
DEC responds that it shares USEPA's view that, although modeling is performed for each hour of meteorological data, the three-hour emission rate is the appropriate rate for modeling purposes. This is reasonable, according to DEC, because there is no one-hour standard or permit restriction which requires a less than three-hour compliance period. Modeling for SO2 has always been consistent, it claims, with the shortest SO2 NAAQS, which is three hours.
Inter-Power similarly argues that the three-hour emission rate is the appropriate rate to use in its modeling, and it notes that this is consistent with the 1989 modeling, in which emissions could have varied on an hourly basis considerably more than they now could. In fact, Inter-Power claims, because it must average no more than 0.22 lbs/MMBtu in any three-hour period, if it emits its maximum 0.26 lbs/MMBtu in any one hour as it is allowed, its emission rate during at least one of the other two hours in the averaging period must be substantially below 0.22 lbs/MMBtu.
CCE, however, asks why, if for practical purposes the 0.26 lbs/MMBtu rate is not realistic, Inter-Power did not accept the lower standard for the shorter period.
Vermont, in a related issue, goes further. Vermont notes that, earlier in the proceeding, as an alternative to resolve an exceedence not resolved by the GESPD offsets, Inter-Power had considered reducing its SO2 emissions to 0.208 lbs/MMBtu. Instead, the exceedence was resolved by a reduction of emissions at the Norlite source; but Vermont proposes that the HCP be held to the lower level anyway. Vermont contends that Inter-Power's witness, Mr. Pope, testified that the lower level was achievable. If the lower emissions level is achievable, Vermont argues, then the minimum adverse environmental impact is not represented by the 0.22 lbs/MMBtu previously approved.
Inter-Power denies that the reduction in permitted emissions rate is required. Inter-Power acknowledges that the reduction is feasible, but it asserts that the modification is "irrelevant." Inter-Power's Reply Brief on Remand, pp. 14-16. The exceedence in question was eliminated by an emissions reduction of about 90% at the Norlite facility, a source which was solely responsible for substantial NAAQS accedences. Inter-Power contends that if the Siting Board recognizes the reduction in Norlite emissions, as it should do, there is no basis for compelling Inter-Power to reduce its emission rate as well.
The 0.22 lbs/MMBtu emission rate, Inter-Power argues, was established by USEPA as a result of the Best Available Control Technology (BACT) demonstration for the PSD permit. The BACT demonstration assertedly weighed the costs of further emission control against the benefits of the resulting reduced impacts and found that 0.22 lbs/MMBtu was BACT. Nothing in this proceeding, in Inter-Power's view, can or should modify that determination. Nor, Inter-Power argues, does Article VIII compel the proposed reduction, because the reduction in SO2 emissions resulting from the change is so insignificant that it will not produce measurably different air quality or change the determination of "minimal environmental impact."
DEC and DPS argue similarly. DEC contends the change is not necessary because it would provide only "incremental benefit". DEC's Reply Brief on Remand, p. 16. DPS, like Inter-Power, asserts that the reduction in emission rate would not have an appreciable impact on air quality, and so should not be ordered.
To demonstrate air quality compliance in this proceeding, Inter-Power has consistently submitted the same air quality modeling performed to show compliance with NAAQS and PSD increment requirements. Use of one-hour SO2 emissions would be inconsistent with those modeling requirements, and the record does not provide a convincing reason for such deviation, particularly in light of DEC's statement that the three-hour emission limit is the standard for state air quality modeling as well.
Although it may reasonably be argued that Article VIII contemplates air quality modeling based on one-hour maximum emissions, the assumption of 0.26 lbs/MMBtu emissions for modeling purposes is not reasonable, because it is not likely that the HCP will operate with such emissions for any significant period. Indeed, Inter-Power's presentation suggests that, for practical purposes, it would always attempt to achieve no more than 0.22 lbs/MMBtu emissions of SO2 because of the requirement not to exceed that rate on a three-hour average basis.
Furthermore, the reduced impacts of the HCP in this phase of the proceeding vis-a-vis the earlier modeling are the result of both the increased stack height and the lowered SO2 emission rate; it is not clear that use of the one-hour emissions limit would have any significant impact on the modeling results.
The Siting Board might wish to consider, based on the record, reducing the one-hour maximum limit. Inter-Power has not argued that it is unable to comply with a lower short-term limit or explained why one should not be imposed.
Similarly, with respect to Vermont's proposal, the Siting Board might reasonably decide to lower the three-hour maximum, lowering, even if only marginally, the impact of SO2 emissions on air quality. The BACT demonstration upon which Inter-Power relies is not part of this record; on this record, however, the evidence is not refuted that the lower emissions rate is reasonably achievable. The fact that it is no longer required to eliminate an exceedence does not make it irrelevant, as Inter-Power argues. The only issue, which was not a major focus of the proceeding, is whether the reduction would have any material impact on air quality.
7. Assuredness of Impact Credit
Several parties--Massachusetts, CCE, and Sierra Club--argue that the HCP may not be permitted because, without the impact credits it plans to get from GESPD, there are modeled NAAQS violations. And, these parties assert, the credits cannot be relied on, for the emission reductions to be taken by GESPD are not federally enforceable, and GE has not fully committed itself to them. They argue further that the record does not support the assumption that air quality will be improved by the replacement of steam production at GESPD with steam production at the HCP. They urge this conclusion on two primary bases. First, they assert, the record is not clear as to what percentage of its steam operations will be continued by GESPD after the HCP begins operations. In addition, they maintain the assumption that HCP steam production operations will be substantially cleaner than GESPD steam production operations has been undermined by the recent permitting of a new gas boiler at GESPD to replace a soon-to-be-retired oil-fired boiler. Clarification of the status of the new boiler at GESPD and of the terms of the steam sales contract were two of the purposes of the remand ordered by the Siting Board. The steam sales contract has been marked as an exhibit in this proceeding (Exhibit 293), as directed by the Siting Board.
At my request, a representative from General Electric, Mr. H. F. Lindner, testified early in this proceeding, in an attempt to resolve questions about General Electric's operations when the HCP begins to supply GESPD's steam needs. CCE and Massachusetts cite Mr. Lindner's testimony that General Electric has not determined how precisely it will operate once the HCP is constructed. Mr. Lindner stated:
I would assume that all boilers would be utilized because we must maintain the capacity should Inter-Power not be able to supply steam, but the exact operation situation, the exact administrative details, have not been worked out. We have never received steam from another source before, and until we can experience what kinds of reliability the HCP will have, we are not prepared to lock ourselves into a specific standby situation. S.M. 10488.
And, they note, despite estimates by Inter-Power and DEC that after the HCP begins operations GE boiler operations would be reduced by 90-95%, Mr. Lindner could not specify what level of hot standby would be maintained.
Massachusetts argues that this lack of certainty undermines a basic assumption earlier relied upon by the Siting Board, that "air quality would improve, or improve in any significant manner, if GE relied upon the HCP while maintaining some unspecified level of hot standby as compared with GE simply meeting its steam needs through its own complement of boilers." Massachusetts' Initial Brief on Remand, p. 7.
Furthermore, Massachusetts argues, this uncertainty means the other major assumption made earlier by the Siting Board--that the HCP would be substantially cleaner than the GE boilers it would replace--is no longer supportable, because of the presence of the new gas-fired boiler at GESPD. Massachusetts claims that "if certified, the HCP potentially would operate at the expense of GE's cleaner energy source." Id. Massachusetts contends that to the extent that GESPD would have relied on the expanded capacity from its new boiler, earlier estimates of expected emissions from GESPD were overstated and earlier estimates of air quality improvement from HCP operations were similarly overstated. Massachusetts points out, as well, that Mr. Lindner testified that GESPD could meet its steam needs with its internal steam generation and increased future needs could be met with upgrade and replacement of existing older boilers.
CCE asserts that it would be inappropriate to consider the offsets from GE and Norlite without considering increased future use from sources whose operators indicated that their current full use was understated.
Inter-Power, in reply, argues that despite uncertainty on the part of GE as to exactly how it would scale back operations in the event of HCP's commencement of steam production, the record is clear that the intent of the steam sales agreement is that the HCP supply all of GESPD's steam needs, and that GESPD's boilers be reduced only to the level of hot standby necessary to allow it to produce steam should the HCP be unable to do so at some particular time. Indeed, Inter-Power states, beyond the contractual obligation, GESPD will reap substantial economic benefits by taking its steam needs from HCP. Even if GESPD is as yet uncertain as to the precise order in which its boilers would operate on hot standby, argues Inter-Power, that does not suggest, as Massachusetts claims, that GESPD and HCP would operate at full emissions simultaneously.
In Inter-Power's view, the provisions of the contract assure that the HCP and GESPD will not both operate at the same time, because "pursuant to the agreement either the HCP is operating to meet the GESPD steam load and GESPD is not, or the HCP is off-line and the GESPD boilers are meeting the load." Inter-Power's Initial Brief on Remand, p. 6. Inter-Power acknowledges that GESPD is authorized to maintain hot standby capability, and it estimates that about 90 million pounds of steam (or about 4.5% of annual steam requirement) annually is required to do this.
DEC agrees with Inter-Power that the record supports the assumption that GE's hot standby status will be less than 10%. It asserts, as well, that, despite the change in emissions profile represented by the new boiler, the reductions in emissions once the HCP begins operations are essentially the same as earlier predicted for SO2 and increased for NOx. Inter-Power argues that even if pre-HCP emissions were lower with the new boiler than without (a point which it apparently contests), that would "simply [reduce] the base from which GE Silicone will be cut back." Inter-Power's Reply Brief on Remand, p. 10. In response to CCE, DEC states that its fuel use survey, which revealed sources who claimed current usage was not representative of future operations, and on which CCE relies, revealed many more sources which considered their current usage overstated than the converse.
The parties who argue that HCP may not be granted a permit in the absence of enforceable offsets are correct, but that does not mean that the application must be dismissed. CCE contends that compliance with air quality standards cannot be demonstrated by seeking offsets, but that argument is simply misplaced. It is clear that such procedures are generally contemplated in air quality modeling. It simply means, as DPS and Inter-Power have recognized, that the certificate, if granted, must contain a restriction on operation without appropriate, enforceable operating reductions. Similarly, the certificate can be conditioned on any other restrictions the Siting Board deems necessary. It would seem prudent, for example, to require a commitment not to operate both HCP and GESPD boilers in tandem.
There is some merit to the opponents' contention that the assumption of improved air quality from the operation of Inter-Power's rather than GESPD's boilers has been undermined by the replacement of two emission points at GESPD with the new boiler. To the extent that anticipated air quality improvement is a balancing factor considered by the Siting Board, it should be recognized that the record here shows that the expected amelioration is considerably reduced.
According to the record in this proceeding, both potential and actual emissions at GESPD of both SO2 and NOx will be substantially reduced by the operation of the new boiler, and either will be significantly reduced after HCP operation. Exhibits 341 and 342. But the GESPD pre-HCP emissions, according to DEC witness Edward Bennett, assume the replacement of the retired boilers by emission point 6 and do not adjust any other boiler output for the increased capacity of the new boiler. Since the new boiler has a capacity 50% greater than the emission points it replaced, it is reasonable to assume, despite the estimated 10% growth in needs anticipated, that there would be some reduction in output from the other existing boilers. Hence, it is likely that the pre-HCP emissions are overstated, and post-HCP improvement similarly optimistic.
The intervenors are off the mark, however, when they suggest the post-HCP operations at GESPD are likely to approach pre-HCP levels. While Mr. Lindner could not state precisely how GESPD would approach readiness to serve capability, and he indicated that all boilers would likely be utilized in some way, that does not imply that GESPD is considering maintaining anything more than the 10% capacity anticipated by Inter-Power and DEC. The steam sales agreement and all testimony by the contracting parties, including Mr. Lindner, confirm the stated intention to operate the facilities alternately, and not contemporaneously. Inter-Power's estimate of about 5% capacity to maintain hot standby appears reasonable; in any event, there is no basis to impeach the 10% outside figure supported by DEC.
8. Emissions of NOx
CCE complains that NOx emissions have not been appropriately analyzed in this proceeding. The intervenor states that NOx is of concern primarily because it is an ozone precursor, and it asserts that natural gas combustion results in higher NOx emissions than does oil combustion. Because a number of sources have, in this proceeding, switched from oil to natural gas to comply with permit restrictions, CCE asserts these sources should have been modeled with the "worst case fuel" for NOx, which is assertedly natural gas. Moreover, CCE is concerned that NOx has not been modeled for short averaging periods, but only for an annual period, leaving no basis, it is contended, to determine whether air quality health-related problems related to that pollutant will be aggravated. CCE argues, as well, that the HCP's NOx emission rate is not the lowest achievable.
DEC replies that there is no short-term NAAQS standard for NOx, but it emphasizes that such short-term impacts, have, indeed, been analyzed in the HRA. It contradicts CCE's contention that natural gas combustion results in higher NOx emissions, and submits that CCE has misunderstood a table compiled by the SAI consultants. In fact, the emission rate for natural gas is about six times lower than the rate for combustion of fuel oil.
DOH also states that the HRA considered NOx emissions.
CCE's arguments are misplaced. The record in this remand appropriately models NOx.
9. Minimum Adverse Environmental Impact
Massachusetts and Sierra Club contend that, whether or not Inter-Power can demonstrate that the HCP complies with federal air quality requirements, it should not be certified because, pursuant to Article VIII, it does not represent the minimum adverse environmental impact. CCE also argues that, particularly because modeling is inexact, a decision based strictly on the compliance demonstration by the modeling would be unwise.
Massachusetts points out that the focus of the remand has been primarily on whether the HCP met the "threshold emission standards established by EPA guidance." Massachusetts' Initial Brief on Remand, p. 9. This, according to Massachusetts, is a useful approach, because failure to satisfy federal standards would, as a matter of law, preclude federal approval and, necessarily therefore preclude Article VIII certification.
Massachusetts believes that it proved that the September modeling did not meet the minimum standards established by USEPA; if the record had ended there, Massachusetts contends it would not have been necessary to apply Article VIII's "more stringent standard". Id. However, it goes on, because the intervenors have been prevented from meaningfully contesting the December modeling, its results have not been reviewed against the same USEPA empirical standards. Therefore, Massachusetts focuses, in its Initial Brief, on the issue of whether the HCP presents the minimum adverse environmental impact, balanced against available alternatives. Massachusetts argues that, because "the HCP, as proposed, will significantly contribute to multiple violations of National Ambient Air Quality Standards," Id. it fails the Article VIII test. Massachusetts contends that it is not sufficient for accedences to be resolved by reduction of emissions at other sources. It argues that the fact that DEC has belatedly taken enforcement action against Norlite or other sources does not, as assumed by the project proponents, necessarily mean that the HCP is entitled to an Article VIII certificate to refoul the air. The federal thresholds merely establish the regulatory floor that all projects must meet--but do not replace Article VIII's stricter requirement to further reduce impacts where reasonable alternatives are available. Id., pp. 12-13.
Thus, for example, Massachusetts asserts, as did Vermont, that Inter-Power should be compelled to reduce its SO2 emissions to 0.208 lbs/MMBtu, despite the reduction of emissions at the Norlite facility. And, Massachusetts suggests, where there is a cumulative-impacts-NAAQS-modeled violation, but Inter-Power's emissions do not exceed federal SIL, Inter-Power should be required nonetheless to offset its addition to the modeled unacceptable air quality.
Sierra Club also contends that the HCP does not represent the minimum adverse environmental impact. Sierra Club states the findings required by Article VIII are stricter than the findings required under other federal and state permits. The fact that a facility may be able to obtain necessary air quality permits does not mean that the facility is entitled to a certificate under Article VIII. Sierra Club's Initial Brief on Remand, p. 49.
Inter-Power vigorously contests Massachusetts' contention that the HCP will significantly contribute to multiple NAAQS violations; to the contrary, Inter-Power argues, the HCP is a "low-emission source with very low impacts in this region." Inter-Power's Reply Brief on Remand, p. 10. And, it says, where the modeling of other sources in the NAAQS inventory shows an exceedence, the HCP generally has a very minor role in the total modeled concentration. Even where the HCP exceeds SIL, it argues, it does so only minimally. Inter-Power accuses Massachusetts of arguing a play on words, when it equates a modeled concentration in excess of the SIL to a significant adverse impact. Nor, in Inter-Power's view, is there any basis for the suggestion that it reduce its impacts at a modeled violation when it does not exceed the SIL; it claims that there is no evidence that "such further reductions could have any meaningful impact on the NAAQS compliance of any receptor/time period combinations of concern." Id., pp. 11-12. This is logical, Inter-Power asserts, because the reason for defining the SIL is to distinguish modeled impacts which are not likely to affect NAAQS compliance from those which could.
DEC agrees with Inter-Power, arguing that the facility's impacts, due to the higher stack height and lower SO2 emissions, have decreased since the original modeling, and that it therefore meets the Article VIII standard.
DPS urges the Examiners to recommend to the Siting Board that it adopt the statutory finding that the proposed facility represents the minimum adverse environmental impact as compared with alternatives.
To begin, the findings on minimum adverse environmental impact urged by both proponents and opponents of the facility are beyond the scope of this remand. It is for the Siting Board to evaluate the information for which this proceeding was remanded, including the results of the corrected inventory and modeling and the revised HRA, and to balance that evaluation with the other factors which must be considered under Article VIII.
With respect to the specific proposals offered by Massachusetts, I have already concluded that, on this record, the SO2 emissions reduction (to 0.208 lbs/MMBtu) is not unreasonable. But it offers no acceptable rationale for its contention that an offset should be required for HCP emissions, below the SIL, that coincide with a modeled NAAQS violation. That contention should be rejected.
C. Conclusion - Modeling
In remanding this proceeding, the Siting Board directed that modeling be performed, using the corrected inventory, of the short-term impacts of SO2, and the annual impacts of SO2 and NOx. My review of the record reveals that there has been substantial compliance with this directive as well.
The modeling performed by Inter-Power, specifically including the December modeling of the inventory revised in November, demonstrates that, provided the appropriate impact credits and modifications to other sources' operations are realized, the HCP will comply with all relevant air quality standards.
Because of the reduced SO2 emissions and increased stack height modeled for the HCP in this phase of the proceeding, as compared with the earlier modeling, the air quality impacts of the HCP individually are generally reduced, although that is not true for all receptors at all time periods. The cumulative air quality impacts of the source inventory are, however, considerably increased as compared with the earlier modeling, and modeled accedences of NAAQS are demonstrated where earlier there were none. Not all of these interact with HCP. HCP exceeds SO2 SIL only at the identified receptors of concern for which offsets are proposed or where Norlite emissions will be reduced. It should also be noted DEC is pursuing elimination of these accedences by permit limitations at the ten affected sources (Exh. 413). These air quality impacts modeled in the remand should be weighed by the Siting Board in determining whether the facility meets the Article VIII balancing test.
In addition, if the Siting Board determines that a certificate should be granted, the certificate should be conditioned as recommended herein, on the acquisition of enforceable offsets, and on the non-contemporaneous operation of the HCP and GESPD, with the exception of minimum operation necessary to retain a readiness to serve at GESPD. It should also be conditioned on grant of the necessary federal permits.
It is recommended, further, that, in order to comply with Article VIII modeling requirements, the one-hour SO2 emissions limit should be reduced. Finally, the record does not reveal any impediment to the imposition of a more stringent three-hour SO2 emissions limit as originally contemplated by Inter-Power and recommended by Vermont.
V. HEALTH RISK ASSESSMENT
In the first part of this proceeding, Appendices C and F of the HRA addressed the health effects related to air quality impacts from the HCP. The analysis was not based upon the NAAQS or PSD modeling, in part, according to DOH witness Dr. Hawley, because that modeling revealed only minimal air quality impacts. Instead, the HRA assessed "exposures to criteria air pollutants at concentrations below applicable federal and state ambient air quality standards" by "evaluating [in Appendix C] short and long-term exposures to SO2 and NOx concentrations attributable to HCP stack emissions alone and HCP stack emissions plus representative background concentrations and emissions of seven 'planned' sources." Inter-Power's Initial Brief on Remand, p. 56. In Appendix F, there was a related evaluation which incorporated the significance of emissions from GESPD with the HCP plus background plus planned source analysis.
A. Inter-Power's Revised Health Risk Analysis Presentation
Inter-Power revised the HRA based upon information from DEC and the July inventory by modifying emission parameters for the HCP, GESPD, and five of the seven planned sources. Inter-Power states, in its Initial Brief on Remand, that parameters for GESPD emission points 2 and 3 and two of the planned sources changed as a result of the November inventory; these changes were not remodeled in the HRA, because it was the opinion of Inter-Power's consultant that the changes in the emission parameters of the planned sources would not materially affect the HRA and that the modeled increased impacts of GESPD emission points 2 and 3 (with their consequent increased impacts) could only mean that the public health benefit of the shutdown of GESPD was increased. DOH agreed that the changes in the November inventory did not affect the HRA. The new modeling showed that areas of predicted maximum impact of HCP, background, and planned sources, combined, had shifted from Rice Mountain (the area of maximum HCP impact) to Alcove Reservoir, in the Town of Coeymans, (where, Inter-Power claims, the major influence on most receptors is not the HCP, but other cogeneration facilities). Inter-Power's witness, Dr. Mendez, concluded, on the basis of the remodeling, that there are a reduced number of receptor locations with maximum one-hour SO2 concentrations above 400 g/m3, as compared with nine locations in the previous modeling results; all areas near the HCP experience SO2 impacts below levels of concern for adverse effects on the most sensitive individuals, for all averaging periods.
With respect to NOx, total concentrations decreased at all but three of the 42 modeled receptor locations. At two of the three locations where concentrations increased, the revised impacts are still below the highest predicted impacts in the original modeling.
For Appendix F, which addressed GESPD in combination with HCP, background, and other planned sources, the new modeling showed substantial impacts in GESPD predicted impacts. Inter-Power asserts that the significance of the revised Appendix F is that it "indicate[s] the improvement in regional air quality (particularly with respect to predicted maximum short-term impacts) that will occur once GESPD boilers are substantially shut down." Inter-Power's Initial Brief on Remand, p. 61.
Recognizing that DOH witness, Dr. Hawley, disagrees with Inter-Power's conclusion that there will be no adverse health impact even on sensitive individuals at any location, Inter-Power concludes that, at any rate, once the GESPD boilers are replaced by HCP operation, air contaminant levels will be "far below levels associated with adverse effects, irrespective of interpretive differences, even for the most sensitive individuals." Id., p. 62 fn.
B. Department of Health's Conclusions
DOH states that it reviewed the health risk analysis of the HCP using two data sets, the NAAQS analysis and an HCP plus background, plus planned sources analysis similar to the one presented by Inter-Power. DOH states that the NAAQS analysis data predict frequent and widespread accedences of ambient air quality standards due primarily to sources other than the HCP, and that it found the NAAQS modeled data much higher in SO2 and NOx concentrations than actual ambient air monitored results; furthermore, DOH stated that, because DEC is pursuing steps to alleviate modeled violations, the data may not be representative of future conditions. For these reasons, DOH did not believe that the NAAQS data should be used to evaluate the health impacts of the HCP.
DOH's witness, Dr. Hawley, found that
the highest projected levels of short-term exposure to SO2 and NO2 are above levels found by some authorities to cause increased bronchial constriction in a small percentage of exercising asthmatics, but below the level found by other authorities to be associated with sensitive or healthy individuals.
the calculated concentrations of these pollutants are close to the levels presented in the original HRA at all receptor locations except Alcove Reservoir; DOH's evaluation of the significance of the health risks described above was unchanged.
projected maximum levels of these pollutants near the Alcove Reservoir are highest of all receptor locations and indicate a higher likelihood of effects on sensitive subjects here than elsewhere in the Capital District, but the contribution of the HCP to these short-term maximum concentrations is small.
Dr. Hawley concluded that, at most locations, the reduced emission rates and increased stack height at the HCP reduce its projected impact on short-term and annual SO2 and NO2 levels. He stated, as well, that modeling based on permitted emission rates demonstrates "the potential for widespread and serious accedences of the NAAQS for SO2 . . . . (At maximum impact points, the 3-hour and 24-hour standards are exceeded by a factor of four and the annual standard by a factor of one and one-half.)" DOH's Initial Brief on Remand, p. 6. However, he noted, the HCP impact at the modeled accedences is less than 3% of the total SO2 levels.
DOH's review of the December modeling showed differences from the September modeling, but those changes did not affect its ultimate conclusions. For example, DOH noted that the December modeling showed a different distribution of air contaminant levels in the Capital District region and more widespread and frequent NAAQS accedences, but the HCP contribution to the NAAQS accedences continued to be small. Days on which HCP emissions in excess of SIL coincide with modeled cumulative NAAQS violations are more frequent in the December modeling; DOH states there were six days on which HCP emissions exceed SIL when the 24-hour NAAQS is exceeded and two days when the 3-hour NAAQS is exceeded, as compared with four days involving the 24-hour NAAQS in the September modeling. DOH concluded, however, that this was because of increased modeled impacts from other sources while HCP emissions remained unchanged. Furthermore, changes in stack height for two of the other planned sources were judged by DOH unlikely to affect contaminant levels, because the emission rates of the sources are unaffected.
Finally, DOH, emphasizing its disagreement with Inter-Power about the modeled health risks to sensitive individuals, states its further disagreement that the discontinuance of operations at GESPD is certain to cause air contaminant levels to decrease below levels associated by any research with adverse health effects on sensitive individuals. In DOH's view, the record does not permit that conclusion.
C. CCE's Position
CCE contends that the HRA should have been based upon the NAAQS compliance analysis, rather than the HCP plus background plus planned sources protocol followed in the earlier phase of the proceeding. CCE asserts, as well, that both Inter-Power and DOH have used an incorrect one-hour SO2 emission rate (0.22 lbs/MMBtu when the maximum one-hour limit is 0.26 lbs/MMBtu), and have erroneously converted the one-hour limited into shorter-term (two-minute and 30-minute concentrations) by an averaging technique, which CCE contends is prohibited by USEPA guidance.
CCE terms it irresponsible of DOH not to use the NAAQS modeling results to determine the compatibility of the proposed project with public health. It states:
Discounting NAAQS modeling results in drawing public health impact conclusions for the HCP is a bald violation of Article VIII intent. The increased impacts from the December modeling make it even more insupportable than the first time the agencies and the Board allowed this abuse. CCE's Reply Brief on Remand, pp. 31-32.
1. Appropriateness of HRA Inventory
Inter-Power argues, in response to CCE, that its use of the Appendix C protocol is "consistent with the Siting Board's carefully defined remand directive." Inter-Power's Initial Brief on Remand, p. 63. Moreover, in Inter-Power's view, analysis of health risks using the NAAQS results would be misleading and unhelpful, because it would yield misleading results, particularly in view of the commitment DEC has made to eliminate modeled accedences in the future.
DOH states that, contrary to CCE's suggestion, Dr. Hawley did testify on the health risks for the NAAQS analysis, and found that the HCP is only a minor contributor to NAAQS accedences. In any case, DOH argues, Dr. Hawley did not ignore the NAAQS results, but considered that assessment procedure inferior to the Appendix C approach. Dr. Hawley testified:
I've analyzed the existing air quality for the Capital District and compared the modeling approaches using these various approaches to those [data] to see just what is the best characterizations of what's likely to ensue if the plant operates. And I think that the NAAQS data set is not a realistic representation of the air quality results. S.M. 11016.
In the earlier phase of this proceeding, when air quality modeling did not reveal pollutant concentrations in excess of NAAQS, the HRA followed the protocol now utilized by Inter-Power and DOH in the remand. In a very technical sense, Inter-Power is correct that continuation of the protocol was anticipated by the remand order, which directed that the HRA be reexamined to the extent it relied on air quality data which was to be corrected in the remand proceeding. In my view, however, that would not preclude use of a different methodology if that protocol could no longer provide reliable health impact information.
DOH, however, is convincing that, whatever the rationale for the air quality modeling required to show compliance with NAAQS or PSD increment consumption, the assessment of health risk should reasonably be based upon a realistic anticipation of contaminant concentrations; the NAAQS modeling does not provide such a basis.
Furthermore, DOH has examined the results of the NAAQS modeling in the context of health risk. Although Dr. Hawley expects that the modeled accedences would result in health impacts, he testified that the HCP is only a minor contributor to the accedences and consequent health risks.
The record contains an adequate basis on which to assess the health risks of HCP operation.
2. SO2 Emission Rate
In response to CCE's assertion that the emission rate used to calculate maximum one-hour impacts for the HCP should have been 0.26 lbs/MMBtu, the maximum one-hour limit, DOH estimated the effect of using the higher rate. DOH states that the higher emission rate would increase maximum one-hour SO2 impacts by 18%, and would be comparable to the results calculated in the original application. They would not be identical because of the change in stack height from 350 to 400 feet. DOH states that the one-hour maximum SO2 levels increase at 11 of the 41 receptor locations. All increases are less than 5 g/m3 except for four receptors (Rice Mountain, from 448.2 g/m3 to 491.1 g/m3; Troy Reservoir, from 344.6 to 368.7; Rice Mountain Flats, from 305.8 to 322.8; and GESPD Fenceline, south, from 273.3 to 283.9). While these increases would correspondingly increase the risk to sensitive individuals from short-term peak SO2 exposure, short-term levels at the receptors are still below the level calculated by Dr. Hawley in his rebuttal testimony for Alcove Reservoir.
DOH states that the higher maximum projected short-term exposure levels thus calculated do not affect its conclusion that the facility could be viewed as compatible with public health and safety for purposes of Public Service Law 142(2)(c)(ii).
DOH's response is persuasive that using the higher SO2 emission rate causes only modest increases in concentration levels and health impacts. Further, there is no evidence that these changes should affect the acceptability of the project from the health risk perspective, and DOH states that its conclusions have not changed.
With respect to the appropriate emission rate to be used, CCE is correct that the one-hour limit should be used. However, it should be noted that I have elsewhere recommended that the one-hour limit be lowered.
No party has responded to CCE's contention that it is incorrect to prorate concentration levels to develop shorter term concentrations. It may well be that this would not be acceptable in a NAAQS analysis, as CCE has claimed, but this analysis does not follow that format. Dr. Hawley testified that such conversions have routinely been made in health risk assessments and the same procedure was apparently followed without objection in the first phase of the proceeding. Indeed, the Recommended Decision of Administrative Law Judge Boschwitz notes that concentrations were apparently "calculated over the different time periods to match the exposure times reported in the scientific literature." Case 80010, Recommended Decision, p. 219. There does not appear to be any reason to depart from that approach.
D. Conclusion - Health Risk Assessment
In its remand order, the Siting Board directed that the Health Risk Assessment should be reexamined to the extent that it had relied on any of the information which was to be corrected on remand. Again, a review of the record reveals that there has been substantial compliance with this directive.
CCE has asserted in the course of this proceeding that other issues involving the public health and safety of the proposed facility, such as ozone levels or emission of volatile organic compounds (VOCs), should be considered by the Siting Board. These issues are beyond the scope of this remand and arguments concerning them have not been considered.
On the basis of the HRA performed by Inter-Power and analyzed by DOH, it appears that the potential health impacts of the HCP have changed somewhat based on the corrected information. Inter-Power contends that the HCP will have no health impact on even the most sensitive individuals. DOH disagrees, and concludes that the highest levels of SO2 and NO2 short-term exposures are above levels some investigators have found to cause increased bronchial constriction in a small percentage of exercising asthmatics, but less than the levels found by most studies associated with any effects on healthy individuals. The highest likelihood of such effects on sensitive subjects has shifted from Rice Mountain to Alcove Reservoir. DOH still concludes, as it did in the earlier phase of the proceeding, that the HCP may be considered to be "compatible with public health and safety" for purposes of PSL 146(2)(c)(ii), which it adds, does not necessarily mean that the facility is the preferable option from a public health perspective.
On the basis of this record, accepting for its purposes the more conservative interpretation of scientific literature employed by DOH, the compatibility of this facility with public health and safety is unaffected by the corrections made in this remand.
The purposes for which this proceeding was remanded have been substantially achieved. To the extent possible and necessary, the emissions inventory has been corrected and remodeled, the effect of those corrections on the Health Risk Assessment has been examined, and the information sought with respect to General Electric operations and its contract with Inter-Power have been supplied and examined.
The record on remand demonstrates that, provided the appropriate impact credits and modifications of operations at other sources are acquired, the HCP complies with applicable state and federal air quality requirements, and that the evidence adduced during the remand does not affect the compatibility of the facility with the public health and safety. Additional conditions on the operation of the HCP, if certified, are recommended in this Recommended Decision. The conditions include:
(1) acquisition of enforceable offsets at GESPD,
(2) certain and enforceable reduced emissions at the Norlite facility,
(3) grant by USEPA of a PSD permit and determination by USEPA of compliance with NAAQS,
(4) reduction of the one-hour SO2 emissions limit at the HCP to 0.22 lbs/MMBtu and/or of the three-hour limit to 0.208 lbs/MMBtu.
The record also demonstrates that, although HCP emissions and air quality impacts are generally (with a few exceptions) lower than in the previous modeling, the air quality in the Capital District is significantly lower than it was believed to be at the time of the Siting Board's earlier deliberations; there are ten cumulative air quality accedences shown, as opposed to the earlier modeling, which showed none. It is recommended that the Siting Board take these findings into consideration when balancing the various factors required to be considered by Article VIII.
June 1, 1992
LIST OF APPEARANCES
FOR DEPARTMENT OF PUBLIC SERVICE STAFF:
Steven Blow, Staff Counsel,
3 Empire State Plaza,
Albany, New York 12223.
FOR INTER-POWER OF NEW YORK, INC.:
COHEN & DAX (by Jeffrey C. Cohen and Ben Wiles, Esqs.),
126 State Street,
Albany, New York 12207.
FOR DEPARTMENT OF ENVIRONMENTAL CONSERVATION:
Marc Gerstman, General Counsel
(by Keith G. Silliman, Assistant Counsel),
50 Wolf Road,
Albany, New York 12233.
FOR STATE ENERGY OFFICE:
Howard Fromer, General Counsel
(by Mollie Lampi, Assistant Counsel),
2 Empire State Plaza,
Albany, New York 12223.
FOR NYS DEPARTMENT OF HEALTH:
Peter Millock, General Counsel
(by Seth Abrams, Senior Attorney),
Corning Tower, Empire State Plaza,
Albany, New York 12237.
FOR ADIRONDACK PARK AGENCY:
David W. Quist, Assistant Attorney II,
P.O. Box 99, Ray Brook,
New York 12977.
FOR NIAGARA MOHAWK POWER CORPORATION:
Gary J. Lavine,
Vice President and General Counsel,
300 Erie Boulevard West,
Syracuse, New York 13202.
LeBOEUF, LAMB, LEIBY & MacRAE
(by Lisa G. Bradley, Esq.),
121 State Street,
Albany, New York 12207.
FOR CONCERNED CITIZENS FOR THE ENVIRONMENT:
Kenneth Dufty, President,
RD 2 - Box 124,
Schaghticoke, New York 12154.
FOR JOINT INTERVENORS OF HALFMOON:
Henrietta O'Grady, Private Intervenor,
44 Church Hill Road,
Waterford, New York 12188.
FOR TOWN OF HALFMOON:
Robert J. Chauvin, Town Attorney,
1691 Route 9 - P.O. Box 1258,
Clifton Park, New York 12065.
FOR HUDSON-MOHAWK GROUP OF THE SIERRA CLUB:
BERNSTEIN, CUSHNER & KIMMELL
(by Jeffrey Bernstein, Esq.),
One Court Street, Suite 700,
Boston, Massachusetts 02108.
John Seakwood and Grace Morton,
Members Inter-Power Opposition Committee.
FOR COMMONWEALTH OF MASSACHUSETTS:
Office of the Attorney General
(by Matthew T. Brock, Assistant Attorney General
Environmental Protection Division, Lisa Heinzerling),
One Ashburton Place,
Boston, Massachusetts 02108.
FOR STATE OF VERMONT, AGENCY OF NATURAL RESOURCES:
Office of the Attorney General
(by J. Wallace Malley, Jr., Assistant Attorney General,
Chief - Public Protection Division),
Pavilion Office Building,
109 State Street,
Montpelier, Vermont 05602.
FOR BENNINGTON COUNTY REGIONAL COMMISSION:
Gregory Burke, Esq.,
Box 342, Arlington,
FOR ASSEMBLYMAN NEIL KELLEHER - 100th ASSEMBLY DISTRICT:
David Little, Counsel, New York State
Legislature, Legislative Office
Building, Room 521, Albany, New York 12248.
FOR ASSEMBLYMAN MAURICE D. HINCHEY - 101st ASSEMBLY DISTRICT:
CO-OP RESOURCES (by John J. Mavretich, Representative),
Burroughs Drive, Box 36,
West Park, New York 12493.
Ralph Kulseng, Counsel,
New York State Legislature,
Legislative Office Building,
Albany, New York 12248.
FOR GENERAL ELECTRIC COMPANY:
READ & LANIADO (by Howard Read, Esq.),
25 Eagle Street,
Albany, New York 12207.
Appendix B available upon request.