WPCL 2 ZBg 0 XX P7XP"*t _ @6'LSmall CircleT/==(&*&?*t _ .E"ЫX P7XP),h  Z (CG Times RegularXx P7P),h  Z (CG Times RegularX P7XP),h  Z (CG Times RegularX2BJ3|xMDRAFT DO NOT QUOTE OR CITEMՊThis is a preliminary document prepared by a subgroup of members of the National and Regional Strategies Work Group as part of the Federal Advisory Committee Act (FACA) Subcommittee process. It is not an EPA document. This document has been discussed by the subgroup, the National and Regional Strategies Work Group, and the Science and Technical Support Work Group. Interested people should forward their comments to a member of the National and Regional Strategies Work Group. Regional Haze Issue Paper Table of Contents      I. INTRODUCTIONp"(#J2  II. THE NATIONAL GOALp"(#J6  III. IMPLEMENTATION OF A REGIONAL HAZE PROGRAM TO MAKE REASONABLE PROGRESS TOWARD NATIONAL GOALp"(#J6 ` ` ` A. Regional Haze in the Context of PM and Ozonep"(#J7 ` ` ` B. Developing Quantitative Objectives to be Addressed in State and Tribal Plansp"(#J9 ` ` ` C. Commissions and Alternative Forms of Institutionsp`"(#I10 ` ` ` D. Should Class I Areas be Addressed Individually or in Groups?p`"(#I12 ` ` ` E. Issues and Strategies to be Consideredp`"(#I14 ` ` ` F. How Long Should the SIP/TIP Planning Take?p`"(#I16 ` ` ` G. How Often Should Implementation Followup Progress be Reported?p`"(#I16  IV. METHODS FOR MEASURING REASONABLE PROGRESSp`"(#I17 ` ` ` A. Aerosol and Visibility Monitoringp`"(#I17 ` ` ` B. Other Scientific and Technical Data Issuesp`"(#I19  V. CONTROLSp`"(#I19 ` ` ` A. Longterm Strategiesp`"(#I19 ` ` ` B. Best Available Retrofit Technology and/or Alternativesp`"(#I20    Regional Haze Issue Paper  DATE: November 12, 1996 CONTACT PERSON: Donna Lamb, Telephone 202-205-0800, Fax 202-205-1096 I. INTRODUCTION This paper examines visibility protection issues in the context of EPAs current review of the PM and Ozone National Ambient Air Quality Standards (NAAQS) and the pending response to the recommendations of the Grand Canyon Visibility Transport Commission (GCVTC). State of the Science: Various commentors have questioned the understanding of the science of visibility. The National Research Council completed a threeyear study of the state of the science. This is a quick overview of their findings.  Visibility degradation in parklands is a consequence of broader regional-scale visibility impairment. The causes of this impairment are well understood. Most impairment is caused by fine particles that absorb or scatter light. Some of these particles (primary particles) are emitted directly to the atmosphere; others (secondary particles) are formed in the atmosphere from gaseous precursors. Visibility-reducing particles and their precursors can remain in the atmosphere for several days and can be carried tens, hundreds, or thousands of kilometers downwind from their sources to remote locations, such as national parks and wilderness areas. During transport, the emissions from many sources mix together to form a uniform, widespread haze known as regional haze.  Current scientific knowledge is adequate and control technologies are available for taking regulatory action to improve and protect visibility...Continued national progress toward this goal will require a greater commitment toward atmospheric research, monitoring, and emissions control research and development. Efforts to improve visibility in Class I areas also would benefit visibility outside these areas. Reducing emissions for visibility improvement could help alleviate other air-quality problems, just as other types of air-quality improvements could help visibility.  Policy makers should consider the linkages between visibility and other air-quality problems when designing and assessing control strategies. Strategies should be adopted that consider many sources simultaneously on a regional basis, although assessment of the effect of individual sources will remain important in some situations. Visibility policy and control strategies might need to be different in the West than in the East. National Research Council, 1993.  The National Research Council study was very clear that there is an adequate scientific basis for EPA to establish a framework for a regional haze regulatory program. It is likely, especially with an integrated approach to Ozone, PMfine, and regional haze, that this information will need to be updated and improved. The EPA regulations could start a process to add any information needed for implementing a regional haze program in an integrated environment. (The Tools and Resources Timeline paper will identify in more detail some of the research and development needs.)  The East and the West: There are differences between the East and the West in terms of visual range or scattering coefficients, the relative contributions of sulfates to the problem, the involvement of relative humidity, and the amounts of existing anthropogenic contributions to haze. Numerous commentors have indicated that we know less about visibility in the East than in the West. Numerous other commentors have indicated that we know more about visibility and are doing more about visibility in the East than in the West. It seems appropriate in introducing this paper to summarize what we do know. Table 1 lists several parameters which demonstrate the difference in visibility conditions between the eastern and western U.S. Table 1. Differences in visibility conditions between the eastern and western U.S. T TTT-e TTT-e T P P PARAMETEREASTWESTP P Average Visual Range 31 km125 kmP P Average Visual Range with a 10% Sulfate Reduction33 km128 kmP P Average Visual Range with 0.2 ug/m3 Sulfate Reduction31 km*130 km*P P Background (with minimum anthropogenic contribution)150 km230 kmP   P   Sulfate Contribution to Particle Extinction70%30% #x P7P#* Data representative of the cluster of Class I areas in the midAppalachians for the East and the Colorado Plateau for the West. The point of this line is to show that the West is more sensitive to change than the East. Sources: Background levels were taken from the NAPAP State of the Science Report, No. 24, 1991. All other data was taken from Spatial and Seasonal Patterns and Long Term Variability of the Composition of the Haze in the United States: An Analysis of Data from the IMPROVE Network, July 1996. #X P7XP#ь Monitoring: Several commentors have suggested that the visibility in Class I areas be inventoried. Initial inventories of Class I areas were done with cameras beginning in 1979. (There is some visibility data for some areas from the 1920s - more like the airport visibility observations.) Some Class I areas are still monitored with cameras taking one to three pictures per day with special film and archival techniques to allow the estimation of standard visual range. The need to prove source attribution and the review done by the National Research Council has led to a reevaluation of these inventory types of monitors in favor of the Interagency Monitoring of Protected Visual Environments (IMPROVE) protocol that includes meteorological data, transmissometer and/or nephelometers, three aerosol modules that capture PM2.5 data on filters that can be analyzed for elemental concentrations (Na through Pb), absorption, particulate Hydrogen, nitrates (NO3), sulfates (SO4), other ions, and organic and elemental carbon, and a 4th module to collect PM10. IMPROVE differs from the draft Federal Reference Method (FRM) for PM2.5 monitoring in two ways. The IMPROVE intake does not give as sharp a cutoff as the FRM, so there may be some coarse particles that get through or some fines that may not be captured. IMPROVE also provides speciation. IMPROVE provides some information on ammonium nitrate (better than the FRM for PM2.5) but needs improvement. Neither method adequately characterizes volatile particulate organic compounds. There is some confusion over the difference between IMPROVE sites and IMPROVE-protocol sites. The differences are in funding and in funding commitment. There are a few sites without optical monitoring equipment and a few sites with a subset of aerosol modules, but the protocol and the data handling is the same across all stations. For the purposes of this paper, they will all be called IMPROVE sites. Many commentors have noted that the IMPROVE network does not uniformly cover the country. It was designed to measure regional haze at sites representative of the Class I areas. There are currently 70 to 80 IMPROVE sites in operation. In terms of monitoring the Class I areas in the East and the West, the network, except for having fewer years of monitoring in the East, is more intense in the East on a per Class I area basis. In addition, there are other studies, such as the Southeastern Aerosol Visibility Study (SEAVS), that include comparable kinds of monitoring information. Table 2 describes IMPROVE network coverage of the East, West, Alaska, Hawaii, and Virgin Islands. Table 2. IMPROVE coverage of the eastern and western U.S. and Alaska and Hawaii. ^ TTT,994 TTT,994 ^ P P EASTWESTALASKA, HAWAII, VIRGIN ISLANDSP P Number of Federally Mandatory Class I Areas34117 7P P Number of IMPROVE Sites2635 4P P No. IMPROVE Sites / No. Class I Areas 0.76 0.30 0.57P    P    Number of Non-Class I IMPROVE Sites1410 0 Length of Record: Thirtyone of the IMPROVE sites have additional years of stackedfilter fine particle data, and some sites have data as far back as 1979. Fortythree IMPROVE sites have three years of data or more as of February 1995. Fifteen others have operated since 1993 or are not at Class I areas. Modeling: The PM or Ozone models that calculate speciated precursor concentrations can be used for haze analysis. The visibility effects of primary and secondary particles can be estimated by multiplying concentration estimates by known extinction values and then summing these quantities to get the total extinction and standard visual range. Extinction values have been developed for the following:  " "Ammonium sulfate ((NH4)2SO4)  " "Ammonium nitrate (NH4NO3)  " "Organic Carbon  " "Soil  " "Coarse mass  " "Effects of relative humidityhh#  " "Rayleigh scattering Emissions Inventories: It is known that there are data gaps in emissions inventories. These gaps differ between the East and the West, in part, because sources are different. For both regions, inventories from area sources and attainment areas need to be improved.  II. THE NATIONAL GOAL In section 169A of the 1977 Clean Air Act Amendments (CAAA), Congress declared as a national goal the prevention of any future, and the remedying of any existing, impairment of visibility in mandatory Class I Federal areas which impairment results from manmade air pollution. The Act specifies that visibility impairment consists of reduction in visual range and atmospheric discoloration. The Act charged the EPA with promulgating regulations to assure reasonable progress toward meeting the national goal. III. IMPLEMENTATION OF A REGIONAL HAZE PROGRAM TO MAKE REASONABLE PROGRESS TOWARD NATIONAL GOAL Reasonable progress refers to progress in reducing human-caused haze in mandatory Class I areas and associated areas of influence under the national visibility goal. (The objective is to make reasonable progress in Class I areas. However, when it is measured, progress is evident in other areas as well.) Depending on the final form of the standards for PM the visibility protection may extend into other areas. Some areas (e.g., urban areas) may require measurement objectives that differ from those in Class I areas. The overall objective is to improve visibility in Class I areas on the worst days (as applicable by region) and to preserve existing visibility on the best days. Reasonable progress toward the national visibility goal needs to be defined. For example, GCVTC defined it for its region as: achieving the continuous emission reductions necessary to reduce existing impairment and attain steady improvement of visibility in mandatory Class I and other areas, and managing emissions growth so as to prevent perceptible degradation of clean air days. (If seasonal reductions are a concern, the continuous reduction requirement could be satisfied on a seasonal basis.) As recognized by GCVTC, achievement of reasonable progress includes distribution of the burden of improving visibility in an equitable manner. No single source or geographic region should be unfairly required to bear costs or limit expansion while others are not. Equity considerations indicate and the Clean Air Act (CAA) specifies that past emissions reductions, current levels of control, technical feasibility, cost of controls, the time necessary for compliance, the energy and non-air quality environmental impacts, the remaining useful life of existing emissions sources, and other relevant equity factors should be considered in achieving reasonable progress. It is also recognized that both equity and other relevant issues may vary significantly from region to region. Among other things, these relevant issues should include giving priority to accounting for progress made in improving visibility as a result of strategies to attain the health-based NAAQS or other CAA requirements. This is especially true in the East where Title IV provisions are expected to have important benefit to visibility. These differences must be considered in developing appropriate regional implementation strategies for achieving reasonable progress in improving visibility and may require other regional consensusbuilding mechanisms (e.g., like GCVTC) to gain public acceptance for implementing measures required to achieve welfare-based visibility objectives beyond those measures required to attain healthbased NAAQS. A. Regional Haze in the Context of PM and Ozone The regional haze program does not have the same emphasis as attainment of the NAAQS. The authority exists in sections 169A and 169B of the CAA to provide the regulatory framework for achieving the national visibility goal. What would give regulators the political will to ask for more controls, if needed, to make reasonable progress toward improving regional haze, and what would give industry the incentive to consider additional controls? Does adequate public support exist to improve visibility protection and to set and meet quantifiable objectives? These are key questions often raised by people who have not been involved in the GCVTC process. Participants believe the GCVTC process generated specific objectives and the will to attain those objectives. For others, there are some considerations that are not technically part of the Ozone, PM, and regional haze Implementation FACA charge, but are reasonable to discuss here to increase the understanding. Background on Regional Haze Emphasis: The 1977 CAAA, which added section 169A to the CAA, gave EPA authority to promulgate regulations to guide the States and Tribes in their determination of what emission management programs constitute reasonable progress toward the national visibility goal. The national visibility goal encompasses protection of all types of man-made visibility impairment in mandatory Class I Federal areas. When EPA promulgated its initial visibility protection regulations in 1980, EPA deferred addressing regional haze. Congress in the 1990 CAAA established additional mechanisms to protect visual air quality. The acid deposition control program is expected to also yield visibility benefits in the eastern United States. Congress also added section 169B to the statute that authorizes EPA to establish visibility transport regions and associated commissions to assess technical information and recommend regional haze measures, and specifically calls for the establishment of a commission to protect visibility in the Grand Canyon region. The resulting GCVTC issued recommendations to EPA in June, 1996. The CAA provides for EPA, within 18 months of receiving the recommendations, to carry out its regulatory responsibilities under section 169A to ensure reasonable progress toward the national goal. The statute also calls for EPA's regulations to include criteria for measuring reasonable progress toward the national visibility protection goal. There are some other considerations that may have a bearing on the contents of the regional haze rule, or how it is integrated with PM and Ozone. These are presented here as a discussion, not as options for this FACA. Regional haze can be integrated with PM and Ozone as is. The common precursors, the transport issues, and the potential benefits of control measures that address the needs of all three programs are all reasons to move forward with integration. That the areas of violation of the health standards and the impairment of visibility are different may complicate the planning and modeling, but that is to be expected. Some commentors have suggested that visibility should be protected through a set of specific standards with attainment dates. This would be like a fine particle secondary NAAQS. Other commentors have indicated that such an approach would restrict a regions ability to establish the timing and levels of improvement for the regional haze program. In order to have a national standard, it must be the same across the country. A standard based on the conditions in the East would afford no protection for an area like the Grand Canyon. Similarly, a standard based on conditions in the Grand Canyon would impose unattainable requirements on the East. This paper examines visibility protection issues in the context of EPA's current review of the PM and ozone NAAQS and the pending recommendations of the GCVTC. While EPA is considering visibility welfare effects in reviewing the NAAQS, this paper assumes that EPA will protect visibility by establishing a regional haze program under sections 169A and 169B of the CAA in addition to the NAAQS. The regional haze program will allow for State and regional differences in implementation steps taken toward the national visibility goal for Federal Class I areas. The issue of this paper is to address if and how the implementation of the possible new PM and ozone primary NAAQS could be integrated with a program to address regional haze under sections 169A and 169B. Planning Sequence:  The sequence of when to plan regional haze objectives and controls - before, during, or after deciding the controls needed to address health issues is another presumption of the Implementation FACA that has not been resolved. Some commentors argue that regional haze should be addressed everywhere, and, in doing so, the primary NAAQS violations will also be attained. Other commentors want to see what the controls for the health standards can do first and then figure out what additional controls, if any, are needed to make reasonable progress toward the national visibility goal. The middle ground, which consists of planning the regional haze objectives at the same time as planning for PM and Ozone, has the following distinct advantages.  " "The progress made to meet the health standards in urban areas would count toward making progress in the Class I areas. Ō " "The longterm strategies for regional haze (See section V.A.) could be designed to make reasonable progress toward the agreedupon visibility objectives, as well as toward the NAAQS. This becomes important in the design of market-based or other innovative systems, if the end point is a declining cap instead of a fixed one.  " "There is a key overriding link between the very reactive VOCs and the formation of Ozone, PM and regional haze.  " "The sources involved are similar. (The sulfate contribution needs to be addressed, but the range of sources will be less and the chemistry is more linear.) Several different pathways toward achieving regional haze objectives and PM and Ozone NAAQS are possible. For instance, areas in NAAQS attainment that contribute to visibility impairment may focus on regional haze objectives, while other areas violating the healthbased standards may initially concentrate on NAAQS. In addition, many areas in NAAQS attainment may also be in NAAQS AOIs and, therefore, may be subject to controls for NAAQS as well as for regional haze. Depending on an areas immediate needs, the regional haze program may or may not be an important influence on that areas initial control efforts. Revised State and Tribal Implementation Plans (SIP/TIP) for regional haze are due 12 months after the promulgation of the regional haze rule and before Ozone and PM designations are due. Thus, if Ozone, PM, and regional haze planning is integrated, EPA should consider allowing noncontrol measures (e.g., developing quantitative objectives, updating emissions inventories, and expanding monitoring networks, as needed) in revised SIPs. B. Developing Quantitative Objectives to be Addressed in State and Tribal Plans In developing quantitative objectives for regional haze, two main questions should be examined: (1) What is the target (e.g., emissions reductions, visibility, or others)? (2) How quickly should the target be met? The GCVTC, for example, developed quantitative objectives for visibility at Class I areas, emissions reductions, and the reductions expected each decade.   OPTIONS Option 1: EPA could specify the improvement needed for all Class I areas, by region, or on a Class I by Class I basis in the regional haze rule. (Something like one deciview improvement in five years.) Pros: The objective is known from the beginning. Cons: Even though the objective could vary by region, it would be set through a top-down process. Option 2: The Federal Land Managers (FLMs) could provide information on the current visibility impairment for their areas and the target objective. EPA might have to issue guidance to the FLMs on what the objectives would contain. Pros: The objective would be known from the beginning. Cons: There is a perception that FLMs might ask for zero anthropogenic impairment too quickly. Option 3: An institutional mechanism, which would include State(s), Tribe(s), FLMs, broadbased public representation, and other stakeholders would evaluate the available data, quantify the objectives, and determine how quickly they can be achieved. Pros: The regional haze improvement needs are addressed by planning what is achievable in each five or tenyear interval. By including the public, the objectives would account for what people want to see. If the Area of Influence (AOI) is limited in size, this will be a smaller process. (The drawback to a smaller planning process developing objectives State by State or Tribe by Tribe, for instance is that affected sources could be faced with multiple targets.) This approach also provides the institutional mechanism with the flexibility to choose who will participate. Cons: Requiring the planning body to wrestle with determining quantitative objectives could prolong the planning process.    RECOMMENDATION: Option 3. Several processes could be used to assess quantitative objectives. The institution could choose which process meets its needs. Furthermore, option3 could provide the flexibility to coordinate planning and controls for regional haze with Ozone and PM planning and controls. EPA would probably have to develop a list of criteria that should be considered in developing the quantitative objectives. (The FLMs are included in option 3, because they have to comment at some point.) (States and Tribes have the authority to develop SIP provisions. The FLMs may comment on the proposal.) C. Commissions and Alternative Forms of Institutions There are a lot of institutional issues associated with the regional haze program that should be discussed, but the institutional mechanism(s) that handle the planning will address the details. The role of these institutions will be planning, analyzing, and implementing the programs. The role may also include the development of the quantitative objectives that must be achieved. The institutions role will influence the direction of the regional haze rule.   OPTIONS Option 1: Visibility Transport Commissions (VTC). These are specified in the CAA. Composed of Governors, FLMs, and Tribes, the GCVTC  was established to develop and make recommendations on what should be in a regional haze rule to protect the Grand Canyon. They chose to look at 15 Class I areas together. They had a charge and a deadline, and they met both. Through the process, the GCVTC generated the political will to carry out the recommendations, which they had forged into an agreement. Option 2: AOI Planning Body. In order to establish the planning body, AOIs would have to be identified and the participating States, Tribes, and other stakeholders would have to be agreed upon. It could be a commissionlike planning group which assesses the best way to handle the Class I area(s) (individually or grouped). Instead of reporting recommendations to the EPA (as the CAA requires VTCs to do), this body would handle the planning for regional haze, PM, and Ozone for the AOI. Option 3: Multi-State Memorandum of Understanding (MOU) or informal agreements to work together. Several States and/or Tribes could get together and work through the planning issues on a more informational basis. Would there need to be expectations of such a group by some time? Option 4: State by State. For some Class I areas, individual States or Tribes might be the mechanism. This will likely end up being essentially a multiState mechanism when the appropriate Tribes are included. Option 5: Others or combination of the above.    RECOMMENDATION: Any recommendation should be developed in conjunction with the Institutional Mechanisms group. Any recommendation should reflect the fact that States and Tribes have the authority to develop SIP provisions where needed. KEY FACTORS: The important issue with respect to regional haze is whether the chosen institutional mechanism can handle the additional role of developing quantitative objectives. If a broad, regional planning level is established in addition to an AOI planning level, the responsibilities of the institutional mechanism handling the different layers of the planning process should be specified. NOTE: One commentor has supported a stakeholder process, but has suggested the need for upperlevel management to be involved. FACA or the EPA should consider creating or assuring that there are incentives to encourage participation in the planning process. D. Should Class I Areas be Addressed Individually or in Groups ? Class I areas could be grouped based on a number of criteria. Class I areas with similar visibility conditions, contributing sources, or ecosystem processes (e.g., biogenic sources, fire regimes, geographic proximity, relative humidity, meteorology) could be considered together. Another approach is to group Class I areas where data representative of several Class I areas is collected. Finally, Class I areas could be grouped based simply on geographic proximity. The GCVTC, for instance, evaluated 25 Class I areas and ultimately decided to consider a cluster of 15 on the Colorado Plateau, because they had similar visibility conditions and ecological processes. In the context of PM and Ozone problems, Class I areas should be considered within the AOI/Area of Violation (AOV) construct. In doing so, the question of addressing Class I areas individually or in groups may be less crucial.    OPTIONS Option 1: Consider the Class I areas one by one and develop AOIs for each. Pros: This option can account for special ecological processes and/or distance considerations. Since Class I areas are known, the process for determining regional haze AOIs can begin immediately. Cons: This approach would not be beneficial in cases where data representative of more than one Class I area is collected. In these cases, it would save time and work to consider them together. Option 2: Group Class I areas first and then identify regional haze AOIs. Criteria for the grouping would be specified. Pros: This would help identify AOIs within the large planning regions, or it might result in more than two regional planning areas for the U.S. The process for determining regional haze AOIs can begin as soon as the Class I areas are grouped. Cons: They might be grouped based on similar ecosystems, but the group might be better split up to address two different airsheds. Option 3: Identify the Ozone and PM AOIs first. Then, modify the AOIs accordingly depending on which Class I areas are in or near them. Pros: This approach begins by addressing health problems and then adds a second level of AOI determination to deal with remaining regional haze. Cons: Because AOIs for Class I areas are added after the initial designations, important participants may be left out at the beginning of the planning process. PM and Ozone AOIs may not be designated for some time, and, thus, the regional haze program would be delayed. Although this approach might work for nonattainment areas, it does not recognize that all Class I areas are affected by visibility impairment. There are large areas currently in NAAQS attainment which will be in AOIs and will be required to make reasonable progress toward the national visibility goal. Option 4: Assuming that multiState transport can contribute to PM, Ozone, and regional haze problems, every State and Tribe would be placed into a broad regional planning area. The regional planning areas would identify PM, Ozone, and regional haze AOVs. It would then determine AOIs by considering the AOVs for all three pollutants together. Since some AOIs could overlap, AOIs would be grouped to optimize the planning and should also recognize localized problems.  Pros: This does a better job of getting all of the PM, Ozone, and regional haze areas into the planning cycle at the same time and will address some of the issues related to areas with potential to impact. The approach allows problems to be addressed at the lowest level possible. For instance, the planning area could identify localized regional haze problems which could be dealt with at the State or Tribe level. Cons: As in option 3, this approach could potentially delay the regional haze program. However, since the AOVs for regional haze are known (the Class I areas), the large planning area could begin regional haze planning before the PM and Ozone designations.    RECOMMENDATION: Option 4. The September 19, 1996 How Should AOIs be Determined? paper and the concept of integration support this option. The regional haze program may include provisions for looking at clean air corridors or clear days as well as hazy days. This means that the AOIs may need to include areas with the potential to impact AOVs. Consistent with the designations paper, this does not mean that the clean areas get controls, but that they are included in the planning. KEY FACTORS: Identification of Class I areas as areas of concern (AOVs) with respect to regional haze is key. Areas with the potential to emit should probably be included in AOIs to address regional haze concerns. (One commentor has suggested that, from a regional haze point of view, a particular quadrant may be a clean air corridor for one Class I area but a source of dirty air for another. Thus, emissions reductions across the region may need to be considered.) (One commentor recommended that the distinction about what is being done in clean areas not be lost, because cost increments for visibility improvement will need to be examined as part of the costeffectiveness test.) NOTE: The institutional mechanism subgroup should consider developing an up front offramp where a single State or Tribe or a few States and/or Tribes could define the boundaries for a localized problem. Approval for them to proceed with their planning may need to come from the higherlevel planning group. The reason is that a situation may occur where State X and State Y decide to solve a local problem, but State Z wants to be at the table to ensure that its concerns are addressed. E. Issues and Strategies to be Considered Once the chosen institutional body has been convened (See section III.C.), it will need to adopt a set of principles to guide the regional haze planning process. Based on the experiences of the GCVTC, two guiding principles which should be considered are: (1) Emissions controls should be equitably implemented across all source categories, so that any single source group is not being disproportionately burdened; and (2) Cost-effectiveness of emissions controls relative to impact on visibility should be maintained. Criteria for Determining If Reasonable Progress is Being Achieved What EPA needs to include in its rule are the criteria for determining if reasonable progress can be measured with the plan (Spatially Integrated Plan [SPIP], SIP, TIP, etc.). The following are some suggested criteria:  " "Reductions in man-made impairment can be verified by emissions tracking and visibility monitoring data.  " "Steady progress is sustained and documented at each interval of review.  " "Program adjustment is made, as appropriate, incorporating periodic review of progress from visibility and non-visibility programs.  " "Continuing improvement is made to both remedy existing and prevent future impairment.  " "Costs are taken into account. (Costeffectiveness of additional controls are evaluated in relation to visibility improvement.)  " "Unintended adverse impacts of the program on energy, environmental, and other secondary factors are taken into account.  " "Unintended beneficial impacts of the program on energy, environmental, and other secondary factors are taken into account.  " "Enhancements are made to the accuracy of monitoring data and visibility modeling to ensure controls are effectively achieving visibility improvement.  " "Wellcoordinated monitoring program(s), administrative systems, funding, and other support mechanisms are in place to implement the program. This list of criteria is recommended as the basic set to be included in the regional haze rule. The planning bodies could modify the list as necessary. (One commentor suggested the following as an additional criterion: baseline emissions reflecting compliance with CAA should be projected so that the ability of nonvisibility controls to produce reasonable progress can be addressed.) Lessons from the GCVTC EPA would develop a definition of reasonable progress. The GCVTC developed a definition for reasonable progress: achieving the continuous emission reductions necessary to reduce existing impairment and attain steady improvement of visibility in mandatory Class I and other areas, and managing emissions growth so as to prevent perceptible degradation of clean air days. In analyzing the GCVTC definition of reasonable progress there are several observations: the principles, the criteria they used, and the definition are inextricably tied to HOW they chose to implement control strategies; in the GCVTC region it was possible for the existing control plans to meet the health standards to eliminate most of the existing visibility impairment (This may not be adequate in the East.); the commitment to emission targets and/or declining caps will require much more detailed emissions inventories for the region; and the issues raised and discussed by GCVTC (such as costeffectiveness, equity, and areas for control) are the same ones being raised in the Ozone Transport Assessment Group (OTAG), and equity was a key item for agreement. The GCVTCs definition of reasonable progress combines an emissions budget with monitored reductions in visibility impairment. If meteorology changes in the short term, they would still be able to assess the emissions budget to ensure progress is being made. (The Commission also set a firm budget for SO2.)Reducing visibility impairment can be thought of as rearranging the distribution of clear and hazy days. In other words, the hazy days could be converted to not-so-hazy days, and the almost-clear days could be transformed into clear air days. The GCVTC developed a dual strategy which concentrates on improving hazy days and preserving clear days. The controls for the health standards would help on the hazy days. The Commission examined the entire spectrum of days because of equity issues. The Commission identified at least three types of equity considerations, including sectorbysector concerns, existing versus future impairment, and hazy versus clear days. F. How Long Should the SIP/TIP Planning Take?   At some point the regional haze measures, like the PM and Ozone measures, must be included in a SIP, TIP, or SPIP. Should EPA give direction in the regional haze rule on how long the planning should take? SIPs for PM and Ozone standards are due by a specific date. Should similar dates be set for regional haze? The planning for PM and Ozone begins with the designation of AOVs of the standards. These are known for regional haze - all the Class I areas. The time between now and when PM and Ozone designations are made may be the time to develop the quantitative objectives for the Class I areas. Once the planning and SPIP or SIP development starts for all three, it makes sense that they all proceed on as similar a time frame as possible. There might be cases where the whole AOI is in attainment for PM and Ozone and regional haze is the only issue. Such an area could proceed immediately or be tied to attainment area SIP dates. At a practical level, the planning process for Ozone, PM, and regional haze may need to be longer. While a lot can be done simultaneously, it still may take longer because of the broader regional coordination needed and the additional complexity of the analysis. It was recognized that the control issues involve a political process, and it would be preferable to only go through that once. Also, it is not generally understood that the outputs of concentrations from either the Ozone models or the PM models can be converted to visibility impacts without a separate analysis (it is more of a post processor converting concentration and humidity to scattering or extinction and summing over all the constituents to get the total). For an integrated approach, the SIP dates should be similar. If there is a need, there should be incentives for areas that are addressing all three. G. How Often Should Implementation Followup Progress be Reported? The PM and Ozone NAAQS programs require progress reports (Reasonable Further Progress [RFP]) on a specified timeframe. EPA is required to report to Congress on progress to improve regional haze every five years. The current regulations require an update of the regional haze longterm strategies every three years. There is flexibility in the regional haze program in reporting. RECOMMENDATION: Use the flexibility of the regional haze program to keep the frequency of the same as the progress reporting for PM and Ozone (RFPs). The current rule requires the update of the longterm strategies every three years. Recommend that the longterm strategies be required to be updated less frequently. IV. METHODS FOR MEASURING REASONABLE PROGRESS IMPROVE is the visibility monitoring network that resulted from the cooperative effort between EPA, several federal land management agencies, and several States. IMPROVE was established in response to the requirements of section 169A of the CAA to protect the visibility of 156 ClassI federal areas. IMPROVE was not the first effort at monitoring visibility in the Class I areas. Subsequent to the 1977 CAAA, the FLMs used cameras to establish whether visibility was an important Air Quality Related Value (AQRV). Cameras, telephotometers, and nephelometers have been used to obtain an inventory of standard visual range, or scattering and extinction and many more of the Class I areas than are currently monitored with IMPROVE today. It became clear that speciated aerosol data (PM10 and PM2.5) would be needed to do detailed source attribution, so the threemodule filter system was added. With the National Research Council review of the IMPROVE monitoring protocol, many of the other types of monitors have been replaced with IMPROVE sites. Some IMPROVE sites have collected eight years of data, and some sites in the East have three years of data.  The main objectives of this program are: (1) to establish the background visibility levels necessary to assess impacts of potential new sources; (2) to determine the sources and levels of reasonably attributable visibility impairment; and (3) to collect data useful for assessing progress toward the national visibility goal. A. Aerosol and Visibility Monitoring The EPA proposal to use section 169 of the CAA for addressing the issues that a section 109 secondary (or welfare) standard would do, still raises an issue of how to measure progress. Experience with cameras and visual equipment alone has strongly indicated the need for speciated aerosol data in combination with meteorological data, extinction measurements - in short, the IMPROVE type of protocol. The proposed FRM for PMfine will not provide the speciated data, nor the continuous extinction measurements needed to address regional haze issues. The IMPROVE network with its limited number of sites would also not be capable of addressing regional haze on the scale that a pseudo-secondary standard would. There may be ways, however, to combine the elements of the Ozone precursor and PMfine monitoring, the IMPROVE, and any transport and background networks, such as the Clean Air Act Status and Trends Monitoring Network (CASTNET).   OPTIONS Option 1: In the regional haze rule, specify that Federal visibility monitoring guidelines be developed for the IMPROVE protocol. In the short term, use any reliable data (e.g., IMPROVE, cameras, telephotometers, airport data, speciated PM2.5, stacked filters, etc.) that are available for assessing visibility impairment. Pros: The data exists and should be used. The protocol is documented. Time and money can be spent on extending network and analyzing the data. Cons: The data analysis for the aerosol samples in the IMPROVE protocol is a very special process and not available from many contractors or labs. There are not enough sites in urban and rural areas. Option 2: Specify no method, but speed up the process to have IMPROVE aerosol monitoring approved as an equivalent method for monitoring PM2.5. Deploy PM2.5 monitors with a nephelometer (to measure extinction) at sites important for the modeling of transport or background. Pros: This leaves the monitoring methods more flexible. Cons: The value of the existing data may be diminished just at the time that it is needed for evaluating control strategies. Option 3: Devise a protocol different than the IMPROVE protocol for aerosols that is more widely available. Accept in the rule any compatible visibility data taken for five years or more in the past (e.g., IMPROVE, cameras, telephotometers, airport data, speciated PM2.5, stacked filters, etc.), but recommend speciated aerosol data, relative humidity, and extinction monitoring as a minimum. Pros: Might get a new technology that could be used with PM2.5 samplers for speciation. This would give longer period of record for data. Cons: Development time would delay the program, when some data is available now. Use of the other data might involve using a different metric for improvement in regional haze (e.g., airport data would qualify, but the metric would be visibility in miles with no speciation).    RECOMMENDATION: Option 1. The IMPROVE protocol is available now, and the types of data that it collects are needed in PM2.5 speciation. It is an internationally approved protocol. Federal visibility monitoring guidance that builds upon the current IMPROVE system should be updated as soon as possible to improve the capture of ammonium nitrate and volatile organic particulates. Available compatible data should be used in the short term where it will help provide an understanding of conditions in a region, even though some places do not have IMPROVE sites. B. Other Scientific and Technical Data Issues The Tools and Resources Timelines paper will address some of these issues in greater detail. V. CONTROLS In section 169A of the CAA, States are required to revise their SIPs to make reasonable progress toward meeting the national visibility goal. Revised SIPs must include:  " "Best Available Retrofit Technology (BART)  " "long-term strategies (10-15 years) for making reasonable progress,  " "estimates of future population growth and associated changes in emissions,  " "a plan to ensure reasonable progress under these anticipated conditions, and  " "strategies for mitigating all types of visibility impairment, including regional haze. If regional haze can be integrated with the considerations of controls necessary for fine particulates and Ozone, then there may be some opportunities to address the specific requirements of CAA section 169A in some more innovative ways. This part of the paper explores some of the requirements of the regional haze program and the issues as they relate to an integrated approach. A. Longterm Strategies  Longterm strategies are a particular feature of the visibility part of the CAA and the associated rules. Longterm strategies for regional haze resulted in some important components of the GCVTC. Following are recommendations that could add flexibility and longerterm effectiveness if integrated with the PMfine and Ozone plans:  " "pollution prevention,  " "studies and information needed by the next check point, " "estimates of growth through 2040 to explore plant retirements,  " "availability of better control technologies in the future, and  " "tradeoffs where shortterm controls result in longterm pollution problems. Longterm strategies may have an added importance in the eastern U.S., where the initial progress comes from Title IV capandtrade programs. A longterm integrated strategy would analyze any or all of the following basic elements:  " "environmental benefits and disbenefits for PMfine, Ozone, and regional haze,  " "credit for existing related local, State, and national pollution control programs (especially Title IV),  " "prioritization schemes for analyzing benefits and disbenefits,  " "economies achieved by an integrated strategy, common reporting, and pollution control requirements,  " "pollution prevention opportunities,  " "growth,  " "tradeoffs where shortterm controls result in longterm pollution problems, and  " "retirement of pollution sources. B. Best Available Retrofit Technology and/or Alternatives Section 169A(b)(2) requires that rules for regional haze address the issue of BART for any major source (250 tons per year) placed in operation between 1962 and 1977 that emits any air pollutant which may reasonably be anticipated to cause or contribute to any impairment of visibility in a mandatory Federal Class I area. Current visibility rules limit BART to major stationary sources whose contribution is reasonably attributable to impairment in a Class I area. Under the existing visibility program, the BART process has involved extensive technical assessments to demonstrate that emissions from a specific major source contribute a specific amount of impairment at a specific Class I area. BART is an expensive attribution and analysis process that has only been considered in a few cases.    OPTIONS Option 1: Continue to pursue existing sources on a sourcebysource basis where impairment at a Class I area is reasonably attributable. Option 2: Pursue existing sources on a sourcecategorybysourcecategory basis where impairment at a Class I area is reasonably attributable. Option 3: Eliminate the reasonably attributable test and provide regulations specifying BART as a floor for retrofits, but continue to allow States and Tribes to accommodate better solutions. Option 4: Allow the SPIP process to consider a broader range of sources contributing to regional haze and to determine the appropriateness of BART in the context of other requirements to address the new NAAQS and regional haze. These other measures may include marketbased strategies, technological advances, and criteria for the impact of source categories, such as proximity, amounts of emissions, and types of particle formation.    RECOMMENDATION: Option 4. EPAs mandated BART provisions should not preclude the development of additional or more innovative strategies.