Appendix F

RA BART Case History

Mt. Zirkel Wilderness Area in Northwest Colorado

Prepared for:

The WESTAR RA BART Working Group

Prepared by:

Dan Ely

Colorado Department of Public Health and Environment

Air Pollution Control Division

F-1

  1. Summary of Facts

The U.S.D.A. Forest Service (USFS) officially certified impairment in the Mt. Zirkel Wilderness Area (MZWA) to then Colorado Governor Roy Romer and the Director of the Colorado Air Pollution Control Division (Division) on July 14, 1993.[1] The USFS concluded in its letter that it was reasonable to believe visibility impairment existed in the MZWA and that local existing stationary sources, the Craig and Hayden power stations, contributed to the problem. The letter also expressed concern with acid deposition and possible damage to aquatic ecosystems in the Wilderness but noted that there were no legal mechanisms at the state or federal levels to address existing non-visibility Air Quality Related Value (AQRV) impairment issues. The certification letter contained a Technical Support Document[2] with data and evidence upon which the certification was made.

The Division considered the existing information available at the time of certification to determine if it could make a decision to “reasonably attribute” visibility impairment within the MZWA to the Hayden and/or Craig generating stations. Environmental groups believe the information in the Technical Support Document is sufficient to attribute. The Division concluded that existing information is insufficient to reasonably attribute given the variety of pollutants and multiple emitting units at different locations. The Division’s response was to collaboratively develop the Mt. Zirkel Visibility Study (MZVS) in order to collect additional information and avoid ‘dueling studies’ with the power plant owners. The study was funded by the owners of Craig and Hayden; jointly managed by the Division, the owners of Craig and Hayden, and the USFS; and chaired by the Division. EPA provided technical advice throughout the study process. Monitoring occurred during 1995 and the $3.5 million effort was completed July 15, 1996.[3]

Hayden Resolution. Prior to the finalization of the MZVS, the groundwork for resolution of the certification regarding the Hayden facility occurred when the Sierra Club won a lawsuit on July 21, 1995, against Hayden concerning opacity exceedances. After a failed appeal by the plant’s owners, global settlement talks began in early 1996 between the State of Colorado, Sierra Club, owners of Hayden Station, and Environmental Protection Agency/Department of Justice. After over 5 months of negotiations, the parties signed an agreement -- the Hayden Consent Decree – on May 21, 1996.[4] The Decree is intended to resolve a number of issues including the successful Sierra Club lawsuit, the State’s visibility regulatory proceedings regarding Hayden and the certification of visibility impairment, and an EPA opacity complaint against the facility. In addition, the Decree is intended to make progress toward reducing acid deposition in the MZWA. Emission limitations and construction and compliance schedules taken from the Hayden Consent Decree were incorporated into the Colorado Visibility SIP by the Colorado Air Quality Control Commission on August 15, 1996[5] and are summarized in Table 1.

The State indicated in the SIP amendment that it believes that these significant emission reductions effectively eliminate the visibility impairment in the MZWA that could be associated with the Hayden Station, constitute reasonable progress toward the national visibility goal, and resolve the certification of impairment brought by the USFS against Hayden Station. EPA approved this SIP amendment on January 16, 1997.[6]

Table 1

Hayden Units 1 and 2 Emission Limitations and Schedule per Consent Decree

Existing Control Equipment / New Control Equipment / Decree Emission Limit / Expected Reduction (%) / Schedule Unit 1 / Schedule Unit 2
SO2 / None / Lime spray dryer / 1) 0.16 lbs/Mmbtu on 30 day rolling average
2) 0.13 lbs/Mmbtu on 90 day rolling average
3) Minimum of 82% removal on 30 day rolling average / 85% from existing emissions / Decree: startup by 12/31/98 and in compliance within 180 days of startup / Decree: startup by 12/31/99 and in compliance within 180 days of startup
NOx
/ None / Low NOx burners and overfire air / Unit 1: 0.50 lbs/Mmbtu on calendar year annual average
Unit 2: 0.45 lbs/Mmbtu on calendar year annual average / 50-60% from existing emissions / Decree: startup by 12/31/98 and in compliance January 1, 1999 / Decree: startup by 12/31/99 and in compliance January 1, 2000
PM / ESPs / FFDC/bag-house / 1)0.03 lbs/Mmbtu averaged over 6 hours via EPA method 5
2)opacity of 20% as averaged each 6-minute period with COMs / NA / Decree: startup by 12/31/98 and in compliance within 90 days / Decree: startup by 12/31/99 and in compliance within 90 days

Craig Resolution. After Hayden was resolved, attention turned to Craig Units 1 and 2. Unit 3 is PSD permitted and not subject to Reasonable Attribution and Best Available Retrofit Technology (RA/BART). On October 9, 1996 Sierra Club filed a lawsuit against the owners of Craig alleging excess opacity emissions from Units 1 and 2. In late-1996, EPA, Craig, and the Division met together to discuss the findings of the MZVS and other studies as well as the alternatives available to address the certification of impairment regarding Craig station. At this meeting, EPA proposed a joint, collaborative study of possible enhancements to the existing SO2 emission control equipment on Units 1 and 2 that currently achieve approximately 65 percent SO2 removal as a way to “get out of the box” and move toward resolution. The State strongly supported this effort and believed it would be very useful to have the results of this study before addressing Craig Units 1 and 2, whether through the RA/BART process or some alternative resolution approach. The State indicated to Craig that it would put on hold the writing of its RA determination while the engineering study was completed in order to further the likelihood of a settlement. The study was initially projected to be complete by Fall 1997. Due to numerous contracting and fiscal challenges, the study was considerably delayed and a final product was not completed until August 1999.[7]

While the study was being completed between late-1997 and the first half of 1999, the Division met with the various parties separately and together while the pre-conditions for negotiations were completed (i.e., final engineering study report, opacity workshop with all parties, and an opacity proposal sent to Craig developed by the Division, Sierra Club, the USFS, and EPA). When all the pieces were in place (September 1999), EPA issued a SIP-call letter to Colorado with a deadline of one year to have an amended SIP resolving the certification.[8] A mediator was hired and the parties (i.e., EPA, Colorado, USFS, Sierra Club, Craig) began a mediation process in October 1999.

Negotiations continued during 2000. As the SIP-call deadline approached and negotiations did not yield an agreement, the State determined that although it still actively supported the negotiations the regulatory pathway had to be re-opened and a RA decision prepared.[9]

The RA determination was never officially released by the Division because of optimism regarding the likelihood of success emerging from the negotiating room. Even though the SIP deadline passed, the Division did not issue the RA decision. This decision was made in consultation with EPA Region 8. The Division shared the draft RA decision with EPA to demonstrate that the State was technically prepared to release the regulatory decision if settlement became hopeless.

On October 17, 2000 Craig and Sierra Club signed a Settlement Outline. The parties disclosed to the Division the emission limitations it contained and it appeared to the Division that these would resolve the certification of impairment. The Division began drafting a SIP amendment. The parties concluded negotiating a Consent Decree and filed it with the federal district court on January 10, 2001.[10] Table 2 contains the emission limitations and schedules as per the Decree.

At the time this report was written the current status of resolution is:

  • A SIP amendment was prepared by the Division and approved in a public hearing by the Colorado Air Quality Control Commission on April 19, 2001.
  • The SIP amendment contains the emission limitations and construction schedules from the Consent Decree and, in a manner similar to Hayden, resolves the certification of impairment in relation to Units 1 and 2 of Craig Station.
  • EPA proposed approval of the SIP amendment on May 1, 2001.[11] After a 30 day comment period, a final approval is expected.
II.Facilities Background and Ownership

Hayden Units 1 and 2 and Craig Units 1 and 2 are “existing stationary facilities” as defined by federal regulation and subject to a BART analysis if reasonably attributed.

Hayden. The Hayden station, near Hayden, CO, is a coal-fired facility located in Northwest Colorado approximately 20 miles west of the MZWA and 20 miles east of the Craig power plant. At the time of the certification Hayden’s Unit 1 had a 184 MW capacity and a 250 foot stack. Unit 2 had a 262 MW capacity with a 395 foot stack. Hot-side ESPs were used to

Table 2

Craig Units 1 and 2 Emission Limitations and Schedule per Consent Decree

Existing Control Equipment / New Control Equipment / Decree Emission Limit / Schedule Unit 1 / Schedule Unit 2
SO2 / Wet limestone flue gas de-sulfurization (FGD) system – 2/3s of flue gas scrubbed achieving an overall 65% removal / Enhancements to system to scrub 100% of flue gas / 1) 0.16 lbs/Mmbtu on 30 day rolling average
2) 0.13 lbs/Mmbtu on 90 day rolling average
3) 90% reduction on 90 day rolling average / Decree: construction complete by 12/31/03 and in compliance with 85% removal within 180 days of startup and 90% removal within an additional 270 days. / Decree: construction complete by 6/30/04 and in compliance with 85% removal within 180 days of startup and 90% removal within an additional 270 days.
NOx
/ Older generation low NOx burners / State-of-the-art low NOx burners and overfire air / 0.30 lbs/Mmbtu on calendar year annual average / Decree: construction complete by 12/31/03. First compliance year is 2005. / Decree: construction complete by 6/30/04. First compliance year is 2005.
PM / ESPs / FFDC/bag-house / 1) 0.03 lbs/Mmbtu
2) opacity of 20% as averaged over each separate 6-minute period with COMs / Decree: construction complete by 12/31/03 and in compliance within 180 days / Decree: construction complete by 6/30/04 and in compliance within 180 days

control particulate emissions. Bituminous coal with a nominal average sulfur content of 0.46% is obtained from a nearby mine. Hayden Unit 1 began operation in 1965 and emitted 4,856 tons of SO2 in 1990. Hayden Unit 2 began operation in 1976 and emitted 6,420 tons of SO2 in 1990.

Hayden’s ownership is split between several private and public companies. Public Service Company of Colorado is the operator of Hayden station and owns 75.5% of Unit 1 and 37.4% of Unit 2. Salt River Project owns 50% of Unit 2. PacifiCorp owns 24.5% of Unit 1 and 12.6% of Unit 2.

Craig. The Craig generating station is located in Northwest Colorado near the town of Craig approximately 40 miles west of the MZWA. The station consists of three units, each with a 600 foot stack. Units 1 and 2 are 428 MW units. These two units, also known as Yampa Project, are equipped with electrostatic precipitators, low NOx burners, and wet limestone scrubbers that remove sulfur dioxide with about 65% efficiency. Unit 3 is a 408 MW unit and is equipped with a dry scrubber that removes SO2 with about 80% efficiency, low NOx burners, and a baghouse. Craig Units 1 and 2 burn bituminous coal from the nearby Trapper Mine with a sulfur content of 0.24% to 0.65%. Craig Units 1 and 2 began construction in 1974 and were in operation in 1980. In 1990, Unit 1 emitted 4,608 tons of SO2 and Unit 2 emitted 5,062 tons of SO2.

Craig Units 1 and 2 are owned by a group of private and public power supply and transmission organizations. Tri-State Generation and Transmission Association, Inc. (Tri-State) is the operator of all three units at the Craig Station. Tri-State owns 24% of Units 1 and 2 and 100% of Unit 3. Public Service Company of Colorado owns 9.72% of Units 1 and 2. Salt River Project owns 29% of Unit 1 and Unit 2. PacifiCorp owns 19.28% of Units 1 and 2. Platte River Power Authority owns 18% of Units 1 and 2.

III.Activities Leading to the Certification
Initial Indications. During the 1980's the USFS and the Division were aware of anecdotal accounts regarding localized plumes and hazes being observed in and around the Yampa River Valley and the MZWA in northwest Colorado. The Yampa River Valley is home to the Hayden and Craig power stations and is upwind of the MZWA.

First Camera System. On October 23, 1990, the USFS, in consultation with the Division, installed an automatic 35mm camera system on the top of the Storm Peak Lab (owned by Desert Research Institute, University of Nevada) viewing Hahns Peak along the western boundary of the MZWA in order to begin systematic monitoring of visibility conditions in the area. At the time the site was selected, the Division agreed with the USFS that the view incorporated by this camera system would represent visibility conditions within the Wilderness since there were no available or accessible camera sites and views that looked directly in to the MZWA.

Review of 1st Year of Slides and More Indicators of a Local Source(s) Influence. In November 1991, Air Resource Specialists, Inc. (ARS), the USFS visibility monitoring contractor and slide archivist, contacted the USFS concerning possible plumes in the monitored view. The slides were then viewed at ARS' offices in Fort Collins, Colorado, by representatives of the USFS and the Division. A number of slides showed apparent elevated layered haze as well as ground-based layered haze. These types of hazes are often suggestive of impairment from a local source(s) and are not generally associated with long-range transport of regional haze.

Two More Cameras Added. Due to heightened concerns raised by the first year of slides, the Division offered two additional camera systems and film to the USFS to document visibility conditions in another view near the Wilderness and a third view to the west down the Yampa Valley toward the power stations. These two systems were installed by the Division and USFS in mid-June 1992.

Snowpack Sampling Occurs and Provides Some Evidence that There May Be Another Problem. Precipitation chemistry sites of the National Atmospheric Deposition Program/National Trends Network that are located at high altitude often have many samples invalidated during wintertime due to extreme weather conditions affecting the sampling equipment and integrity of the samples. Because of these problems, in 1990 the U.S. Geological Survey (USGS) proposed to the Division, the USFS, and EPA/Region 8 a feasibility study of synoptic sampling of the high altitude seasonal snowpack in Colorado through simply digging snowpits and analyzing the snow’s chemistry. The data[12] from 1991 indicated an increase in acidity in the snowpack downwind of the Yampa River Valley in northwest Colorado relative to all other high elevation snowpack sampled in Colorado (approximately 20 sites). The snowpack downwind of the Yampa River Valley contained about twice the concentration of sulfate and nitrate and about 250% of the acidity compared to the remainder of the high-elevation snowpack in Colorado. This is the highest acid deposition measured in the Western U.S. Lakes in the MZWA and just south of its borders also have the greatest concentrations of sulfate in the vicinity of the greatest concentrations of sulfate in the snowpack, indicating a direct link between increased deposition in snow and effect on lake chemistry. Additionally, the chemical data are verified by stable-sulfur isotopes, which indicate that the increased deposition of sulfate to snow and increased concentration of sulfate in lakes is caused by an upwind source(s) that adds isotopically distinct sulfate to the regional ambient sulfate. These data provided further indicators to the Division and the land manager that there is a local source(s) of air pollution in addition to the regional background.

ZAQ Collaborative Formed. In February 1993, the Division formed the Zirkel Air Quality Collaborative (ZAQ Collaborative) as an advisory group to the Division and to provide an opportunity for interested parties to work together on air quality issues concerning MZWA. Because visibility studies and Class I issues can be so factious, the Division's intent was to utilize the ZAQ Collaborative in a manner to work toward agreement or at least mutual understanding, on the certification and then on the scope and nature of additional data collection to support regulatory decision making. Since its first informal meeting in December 1992, the Division convened this advisory group approximately a dozen times to further information exchange, communication, and to explore opportunities for cooperation.[13] ZAQ meetings were facilitated by a professional meeting facilitator hired by the Division.

EPA, State, and USFS Coordinate on Assembling Technical Information. The Division coordinated the preparation of a summary of the technical information supporting a possible certification. The Division chose to do this so it would be familiar with the evidence upon which the USFS indicated it would certify. During the first few months of 1993, the information was assembled and included: selected slides, emission inventories, meteorological data, VISCREEN plume modeling, PLUVUE plume modeling, and information about acid deposition/aquatic ecosystems. A draft of the Technical Background Document (TBD) was sent in April 1993 to the ZAQ Collaborative, including the owners of the Craig and Hayden power stations, for comment and input.