Page 67

3
Criteria for Designing and Evaluating Ozone Reduction Strategies

Introduction

This chapter provides background information related to State Implementation Plans (SIPs), which the nation's states are required to use to attain the National Ambient Air Quality Standard (NAAQS) for ozone. The Clean Air Act, as amended in 1990, provides the legal foundation for this process. This chapter also discusses weaknesses in the existing SIPs' use of ambient air-quality data and emissions inventory data, the connections between air quality and emissions, new strategies for control, and the effectiveness of existing controls.

The Clean Air Act

The Clean Air Act was the first modern environmental law enacted by Congress. The original act was signed into law in 1963, and major amendments were made in 1970, 1977, and 1990. The act establishes the federal-state relationship that requires the Environmental Protection Agency (EPA) to develop uniform air-quality standards (NAAQS) and empowers the states to implement and enforce regulations to attain them. The act also requires EPA to set NAAQS for common and widespread pollutants after preparing criteria documents summarizing scientific knowledge of their detrimental effects. EPA established NAAQS for each of six criteria pollutants: sulfur dioxide, particulate matter, nitrogen dioxide, carbon monoxide, ozone, and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 67
Page 67 3 Criteria for Designing and Evaluating Ozone Reduction Strategies Introduction This chapter provides background information related to State Implementation Plans (SIPs), which the nation's states are required to use to attain the National Ambient Air Quality Standard (NAAQS) for ozone. The Clean Air Act, as amended in 1990, provides the legal foundation for this process. This chapter also discusses weaknesses in the existing SIPs' use of ambient air-quality data and emissions inventory data, the connections between air quality and emissions, new strategies for control, and the effectiveness of existing controls. The Clean Air Act The Clean Air Act was the first modern environmental law enacted by Congress. The original act was signed into law in 1963, and major amendments were made in 1970, 1977, and 1990. The act establishes the federal-state relationship that requires the Environmental Protection Agency (EPA) to develop uniform air-quality standards (NAAQS) and empowers the states to implement and enforce regulations to attain them. The act also requires EPA to set NAAQS for common and widespread pollutants after preparing criteria documents summarizing scientific knowledge of their detrimental effects. EPA established NAAQS for each of six criteria pollutants: sulfur dioxide, particulate matter, nitrogen dioxide, carbon monoxide, ozone, and

OCR for page 67
Page 68 lead. These pollutants are emitted from numerous and diverse sources, and at certain concentrations and length of exposure they are anticipated to endanger public health or welfare. The NAAQS are threshold concentrations based on a detailed review of the scientific information contained in criteria documents prepared by EPA and peer reviewed. Pollution concentrations below the NAAQS are intended to expected have no adverse effects for humans and the environment. For each criteria pollutant, NAAQs comprise: a primary standard, which is intended to protect the public health with a margin of safety, and a secondary standard, which is intended to protect the public welfare as measured by the effects of the pollutant on vegetation, materials, and visibility. Primary and secondary NAAQS for ozone, which were originally called NAAQS for oxidants, were established by EPA in 1971. Photochemical oxidants, a group of chemically related pollutants, are defined as those compounds giving a positive response using the iodide oxidation technique. In 1979 EPA revised the NAAQS for oxidants to the current NAAQS for ozone only. Both the primary and secondary NAAQS for ozone are now defined as a daily maximum 1-hour average concentration of 0.12 parts per million (ppm), or 120 parts per billion (ppb), not to be exceeded on average more than once each year. The average number of days exceeding the standard is calculated for a 3-year period. Based on recent health effects studies, the ozone NAAQS requires the shortest averaging time of any of the criteria pollutants' NAAQS. An area, whose boundaries are designated by EPA, is considered to exceed this threshold and is classified as being in ''nonattainment'' if a violation occurs anywhere within the area. For ozone, areas have consisted of Metropolitan Statistical Areas (MSAs), Consolidated Metropolitan Statistical Areas (CMSAs), and counties. The Clean Air Act requires that a SIP be developed for areas in nonattainment to reduce precursor emissions enough to bring air quality into compliance with the NAAQS. SIPs must be adopted by local and state governments and then approved by EPA. Once a SIP is fully approved, it is legally binding under both state and federal law. In the Clean Air Act amendments of 1970, Congress set 1975 as the deadline for meeting the NAAQS. By 1977, 2 years after this deadline, many areas were still in violation of the ozone NAAQS. The 1977 amendments to the Clean Air Act delayed compliance with the ozone and carbon monoxide NAAQS until 1982, and areas that demonstrated they could not meet the 1982 deadline were given extensions until 1987. In 1990, 3 years after the final deadline, more than 133 million Americans were living in the 96 areas that were not in attainment of the ozone NAAQS the year before (EPA, 1990b). The 1990 amendments classify nonattainment areas according to degree of

OCR for page 67
Page 69 noncompliance with the NAAQS. The classifications are extreme, severe, serious, moderate, or marginal, depending on the area's ozone design value and the percentage by which the value is greater than the NAAQS. Ozone design values are ozone concentrations that are statistically determined from air-quality measurements for each nonattainment area. If monitoring data for an area are complete, the design value is the fourth highest monitor reading over the past 3 years. The U.S. Code of Federal Regulations, Appendix H) provides a method to account for incomplete monitoring data (40 CFR 50.9(a). Design values are used to determine the extent of control needed for an area to reach attainment. These values and the target attainment years are shown in Table 3-1. TABLE 3-1 Classification of Nonattainment Areas Designation % above 0.12 ppm ozone Ozone design value rangea, ppm Years allowed to attain ozone NAAQSb Number of areas, 1989 Extreme > 133 > 0.280 20 1 Severe 50-133 0.180-0.280 15 c 8 Serious 33-50 0.160-0.180 9 16 Moderate 15-33 0.138-0.160 6 35 Marginal 0-15 0.121-0.138 3 36 aDetermined as the fourth highest value over 3 consecutive years. bNumber of years from November 15, 1990, allowed in the 1990 amendments to the Clean Air Act. cSevere areas with design values between 0.19 and 0.28 ppm are allowed 17 years to attain the ozone NAAQS. The 96 areas out of compliance in 1989, and their design values for 1983-1985, 1985-1987, and 1987-1989 are listed in Table 3-2. The areas classified as extreme or severe are in four major regions of the nation: the South Coast basin (Los Angeles) and San Diego, California; the greater Houston area; the Northeast Corridor (which extends from the Washington, D.C. area to Boston

OCR for page 67
Page 70 TABLE 3-2 Classification of Nonattainment Areas for Ozonea   Design value, ppmb Area 1983-1985 1985-1987 1987-1989 Extreme, design value 0.2.8 ppm or higher     Los Angeles/Long Beach CMSAc 0.36 0.35 0.33 Severe, design value 0.18 to 0.28 ppm     Baltimore 0.17 0.17 0.19   Chicago 0.20 0.17 0.19   Houston CMSA 0.25 0.20 0.22   Milwaukee 0.17 0.17 0.18   Muskegon, MI 0.14 0.17 0.18   New York City, NY/NJ/CT CMSA 0.22 0.19 0.20   Philadelphia, PA/NJ CMSA 0.18 0.16 0.19   San Diego 0.21 0.18 0.19 Serious, design value 0.16 to 0.18 ppm     Atlanta 0.16 0.17 0.16   Bakersfield 0.16 0.16 0.17   Baton Rouge 0.16 0.14 0.16   Beaumont/Port Arthur, TX 0.16 0.13 0.16   Boston 0.16 0.14 0.17   E1 Paso 0.16 0.16 0.17   Fresno 0.17 0.17 0.17   Hartford 0.23 0.17 0.17   Huntington/Ashland, WV/KY/OH 0.14 0.14 0.16   Parkersburg/Marietta, WV/OH — 0.13 0.17   Portsmouth/Dover/Rochester NH/MA 0.13 0.13 0.17   Providence CMSA 0.18 0.16 0.16 (Table continued on next page)

OCR for page 67
Page 71 (Table continued from previous page)   Design value, ppmb Area 1983-1985 1985-1987 1987-1989 Serious, design value 0.16 to 0.18 ppm (continued)     Sacramento 0.18 0.17 0.16   Sheboygan, WI — — 0.17   Springfield, MA — — 0.17   Washington, DC/MD/VA 0.16 0.15 0.17 Moderate, design value 0.138 to 0.16 ppm     Atlantic City 0.19 0.14 0.15   Charleston, WV 0.13 — 0.14   Charlotte/Gastonia/Rock Hill, NC/SC 0.13 0.13 0.16   Cincinnati, OH/KY/IN 0.17 0.14 0.16   Cleveland 0.14 0.13 0.16   Dallas/Fort Worth 0.16 0.16 0.14   Dayton/Springfield 0.13 0.13 0.14   Detroit 0.13 0.13 0.14   Edmonson County, KY — — 0.14   Grand Rapids 0.13 0.13 0.14   Greensboro/Winston-Salem/High Point, NC — — 0.15   Hancock County, ME 0.13 0.13 0.13   Jefferson County, NY — 0.13 0.14   Knox County, ME — 0.15 0.16   Louisville, KY/IN 0.15 0.16 0.15   Kewaunee County, WI — 0.13 0.15   Knoxville — — 0.14   Memphis, TN/AR/MS 0.15 0.13 0.14 (Table continued on next page)

OCR for page 67
Page 72 (Table continued from previous page)   Design value, ppmb Area 1983-1985 1985-1987 1987-1989 Moderate, design value 0.138 to 0.16 ppm (continued)     Miami/Hialeah 0.13 0.15 0.14   Modesto 0.15 0.15 0.14   Nashville 0.14 0.14 0.14   Pittsburgh 0.13 0.13 0.15   Portland, ME 0.16 0.14 0.16   Poughkeepsie — — 0.13   Raleigh/Durham — 0.13 0.14   Reading, PA 0.13 — 0.14   Richmond/Petersburg, VA 0.13 0.13 0.14   Salt Lake City/Ogden 0.15 0.15 0.14   San Francisco CMSA 0.17 0.14 0.14   Santa Barbara/Santa Maria/Lompco, CA 0.16 0.14 0.14   St. Louis, MO/IL 0.16 0.16 0.16   Smyth County, VA — — 0.14   Visalie/Tulare/Porterville, CA 0.13 0.15 0.15   Worcester, MA 0.13 0.13 0.15 Marginal, design value 0.121 to 0.138 ppm     Albany/Schenectady/Troy — — 0.13   Allentown/Bethelehem, PA 0.14 0.13 0.14   Altoona, PA — — 0.13   Buffalo — — 0.13   Birmingham 0.13 0.15 0.13   Canton, OH — — 0.14   Columbus — — 0.13   Erie, PA 0.13 — 0.13   Essex County, NY — — 0.13   Evansville, IN/KY — — 0.13   Fayetteville — — 0.13 (Table continued on next page)

OCR for page 67
Page 73 (Table continued from previous page)   Design value, ppmb Area 1983-1985 1985-1987 1987-1989 Marginal, design value 0.121 to 0.138 ppm (continued)     Greenbrier County, WV — — 0.13   Harrisburg/Lebanon/Carlisle 0.13 — 0.14   Indianapolis 0.13 — 0.13   Johnson City/Kingsport/Bristol, PA — — 0.13   Johnstown, PA. — — 0.13   Kansas City, MO/KS 0.14 — 0.13   Lake Charles, LA 0.14 — 0.13   Lancaster, PA 0.13 — 0.13   Lexington, KY — 0.13 0.13   Lewiston/Auburn, ME — — 0.14   Lincoln County, ME — 0.13 0.13   Livingston County, KY — — 0.13   Manchester, NH — — 0.14   Montgomery, AL — 0.14 0.14   Norfolk — 0.13 0.13   Owensburg, KY — — 0.14   Scranton/Wilkes Barre — — 0.13   South Bend/Mishawaka, IN — — 0.12   Stockton 0.15 0.14 0.13   Sussex County, DE — — 0.13   Tampa/St. Petersburg/Clearwater 0.13 0.13 0.13   Waldo County, ME — — 0.13   York, PA — — 0.13   Youngstown, Warren, OH — — 0.13 aBased on data from 1987-1989;bRounded to the nearest hundredth;cConsolidated Metropolitan Statistical Area. Source: OTA, 1989 and EPA, 1990b.

OCR for page 67
Page 74 and beyond); and the Chicago area, including downwind areas in Wisconsin and Michigan. The 1990 amendments to the Clean Air Act require EPA to designate the boundaries and classifications of the nonattainment areas. These designations are important in that they determine the SIP's geographic extent as well as the severity of the control program for these areas. From a scientific perspective they are no less important. Ozone formation is complex and time dependent. Given the time scales for the transport of pollutants from other areas, ozone maximum concentrations can be triggered by ozone precursor sources far upwind from the affected area. These atmospheric processes and the regional transport of ozone are described in Chapters 5 and 4, respectively. Criteria for determining the size of the areas within a state have been established using jurisdictional boundaries (Metropolitan Statistical Areas or Consolidated Metropolitan Statistical Areas). In general, these areas should be as large as possible, and they should include sources and ozone-monitoring sites within the same area while allowing for expansion in accordance with population growth. Section 184 of the 1990 amendments of the Clean Air Act establishes an interstate ozone transport region extending from the Washington, D.C. metropolitan area to Maine. In this densely populated region, ozone violations in one area might be caused, at least in part, by emissions in upwind areas. A transport commission is authorized to coordinate control measures within the interstate transport region and to recommend to EPA when additional control measures should be applied in all or part of the region in order to bring any area in the region into attainment. Hence areas within the transport region that are in attainment of the ozone NAAQS might become subject to the controls required for nonattainment areas in that region. EPA will likely establish other interstate ozone transport regions and transport commissions. How well these commissions carry out their responsibilities will be an early test of the effectiveness of the 1990 Clean Air Act amendments. The State Implementation Plan The State Implementation Plan is the technical and regulatory process for demonstrating attainment and maintenance of the requirements of the NAAQS. Once approved by EPA, the plan is legally enforceable under federal law and thus is a powerful tool for achieving the NAAQS. The current SIP mechanism is represented in Figure 3-1. Each side of the triangle represents a fundamental aspect of the SIP. The horizontal base (a) represents the time an area is allowed to achieve the NAAQS; this schedule is established by the

OCR for page 67
Page 75 Figure 3-1 Conceptual diagram of SIP mechanism. (a) Time allowed to attain the ozone  NAAQS; (b) required reduction in VOC and/or NOx emissions (c) reasonable  further progress line. Clean Air Act. The vertical ordinate (b) is the reduction in precursor emissions needed to reach attainment. The slope that completes the triangle is the rate of emission control progress needed to attain the NAAQS; it is called the reasonable further progress (RFP) line (c). The new RFP provision of the act will require a reduction in VOC emissions below the base year inventory by 15% over the first 6 years and 3% per year thereafter for all but "marginal" nonattainment areas. Areas classified as moderate, serious, or severe may choose an alternative to the 15% and 3% requirements, which includes, for example, the installation of all feasible controls on existing sources of emissions. The 1990 amendments of the Clean Air Act also address the use of NOx controls in addition to, or instead of, VOC controls. Section 182(c) of the 1990 amendments allows states, with EPA guidance and approval, to supplement or replace VOC controls with NOx controls to an extent "that would result in a reduction in ozone concentrations at least equivalent to that which would result from the [required] mount of VOC emission reductions." In addition, Section 182(f) mandates that the control provisions required for major stationary sources of VOCs also apply to major stationary sources of NOx unless EPA determines that net air-quality benefits in an area are greater in the absence of NOx reductions, or that NOx reductions would not con-

OCR for page 67
Page 76 tribute to attainment of the ozone NAAQS in that area. Except in California, NOx emission reductions have not previously been a major component of most SIPs. SIPs will rely on enhanced monitoring of ozone, NOx, and VOCs for the demonstration of attainment and maintenance of the NAAQS in serious, severe, and extreme areas. A less stringent process is allowed for moderate and marginal areas. Figure 3-2 diagrams the different phases of a SIP. The demonstration phase of the plan will take from 1 to 4 years after the plan begins, depending on an area's classification. The implementation phase ends when attainment is reached in accordance with the act's deadline or when EPA determines that the SIP is deficient and issues a recall or, as a last resort, develops a Federal Implementation Plan (FIP) for the area. The FIP could be a complete replacement of the SIP or a supplement to the SIP to correct its deficiencies. Figure 3-2 Three components of state implementation planning process. In the demonstration phase, base-year modeling is performed to verify the air quality-emissions relationship for a specified meteorological episode. The base year is the year of the latest available emissions inventory. Target-year modeling, in which emissions are reduced according to the control measures

OCR for page 67
Page 77 selected by the area, is performed next. In general the meteorological episode is a typical episode of high ozone concentrations for which adequate meteorological data are available. Figure 3-2 illustrates the three SIP data bases: air quality, meteorology, and emissions. Because of the importance of these data bases to attainment, each is discussed in detail in subsequent chapters. Once established by the base-year analysis, the same meteorological input data are applied to future years, and modeled VOC or NOx emissions are incrementally reduced by applying estimates of reductions to be brought about by control measures proposed for mobile, stationary and area sources. If attainment is not demonstrated by the target year, the inventory is further reduced by introducing additional control measures for total and reactive VOC emissions and possibly for emissions of NOx. Only after the target-year modeling demonstrates attainment with the NAAQS and after a period of public comment will EPA approve the SIP. In the demonstration phase, considerable reliance is placed on the air-quality model and its required inputs to determine the emission reductions necessary to achieve the NAAQS. There are always uncertainties in the base-year inventory and in the projected control measure estimates. These uncertainties are further compounded by the projection of modeling inputs to a future year for attainment demonstration. The tracking mechanism used in the implementation phase is also unsatisfactory. In past SIPs, states were required only to estimate emissions reductions periodically and track them in relation to the RFP line shown in Figure 3-1. Projected air-quality improvements were not checked with models. It is anticipated that at least one mid-course modeling demonstration will be required for the upcoming SIPs. Limited federal guidance was incorporated in the tracking process, as the federal government's major emphasis shifted from oversight to rule development and implementation strategies. Limited use of audits was employed by EPA, and those that were conducted were compromised by their lack of independence: the agency conducting the audit was the same one held responsible for the success of the process. The committee concludes that the lack of an adequate mechanism for tracking progress and for taking corrective action, if needed, significantly limits the ability of areas to effectively monitor and maintain progress toward achievement of the NAAQS. Other processes—less demanding of resources than the process based on air quality and emissions—could be used. For example, the accommodative SIP would rely on emission reductions from federally mandated control measures and existing controls and thus allow for expected emissions growth in some areas. This approach would be applied only in the most marginal non-attainment areas. These areas would not need to develop more detailed

OCR for page 67
Page 82 Control Strategies The primary purpose of the SIP is to set forth a control program of NAAQS attainment strategies that are legally enforceable at both the state and the federal level. As case examples for this report, a generic process for developing controls is illustrated by an OTA report (OTA, 1989; Rapoport, 1990), and an extreme example is demonstrated in the 1988 air-quality management plan (AMP) for California's South Coast Air Quality Management District (SCAQMD, 1989). The OTA study characterized the emissions reduction potential of various control strategies and then applied these reductions to many of the nonattainment areas listed in Table 3-2. Each area was ranked by ozone design value, in ppb (130-140, 150-170, 180-260, > 260), and estimates were made of each area's potential to achieve the NAAQS by 1994 and by 2004. Although major assumptions were made with the emissions data, and the EKMA model was used in the comparison, the OTA study provides a common data base to compare control strategies nationally and examine their ability to attain the NAAQS. Near-term VOC strategies were analyzed first. These included application of all reasonable available control technology (RACT) controls now required by any state to all large (>25 tons/yr) sources in nonattainment areas. OTA also estimated the cumulative benefits of additional strategies including emission controls on hazardous waste treatment, storage, and disposal facilities (TSDFs); federally regulated controls on architectural surface coatings; on-board motor vehicle controls; modification of gasoline station pumps to trap escaping vapors; enhanced inspection and maintenance (I/M) programs; more stringent motor vehicle exhaust standards and gasoline volatility standards; and the use of methanol in centrally owned motor vehicle fleets. The estimated results of these strategies were projected to target years 1994 and 2004, taking into account increased vehicle mileage and economic activity. Figure 3-3 shows the estimated reductions for these years. The OTA report concluded that given the most optimistic estimates of these potential VOC emission reductions, nonattainment areas with design values of less than 160 ppb have some prospect of achieving the NAAQS. It was concluded that additional reductions are needed in areas with design values of 160 ppb or greater, even in the most optimistically modeled cases. OTA then expanded the control options using categories that might not yet be available. These included NOx controls on major existing stationary sources; I/M for NOx; more stringent NOx standards for motor vehicle exhaust, control of organic solvent evaporation; alternative fuels for passenger vehicles; and transportation control measures (TCMs). (TCMs include modified work

OCR for page 67
Page 83 schedules, highway lanes for carpools, bicycle lanes, road use tolls, and improved public transit.) Emission reductions from these future controls, however, were not estimated or modeled. OTA estimated that emissions from solvents, highway motor vehicles, and gasoline refueling will account for 70% of the remaining VOC inventory, suggesting the need for longer-term strategies that include lowering or restricting organic solvent emissions, implementing long-term TCMs, and using alternative motor vehicle fuels such as methanol and compressed natural gas. OTA also concluded that in many nonattainment areas, exhaustive exploration of potential control measures will be required to identify reductions to attain "or come as dose as possible to" the NAAQS. In areas where TCMs are needed, close cooperation with state and local transportation agencies and land-use planners is essential. The South Coast Air Quality Management Plan (SCAQMP) is an example of a plan to identify potential control measures that could be available in the near term (Tier I) and longer term (Tiers II and III). Tier I controls are those measures that can be reasonably adopted in the next 5 years and implemented over the next 20 years with currently available technology. Tier II would require significant advancement of current technology and further regulatory controls through technology-forcing standards or emission fees. Tier III would require technologic advances in industrial, commercial, and residential solvent and coating applications and the use of essentially emission-free vehicles. Major transportation and land use planning efforts also would be needed. All tiers together were calculated to result in an 84% and an 80% reduction for VOCs and NOx, respectively. Recently, EPA proposed for the South Coast area a FIP that is designed to augment the SCAQMD's 20-year plan (CFR, July 21, 1990). It was issued in compliance with a court order that resulted from a suit filed by several environmental organizations for the SCAQMD's failure to attain the NAAQS. The plan is believed to represent EPA's view of what future SIPs should contain in severe and extreme areas. The EPA plan is designed to achieve the NAAQS by 2010 by promoting market-based incentives and by conforming with the 1990 amendments to the Clean Air Act. Similar to the tier concept, the FIP calls for the promulgation of "core" measures and compliance with a schedule for reasonable further progress. The program will implement backstop measures, such as innovative technology standards and economic incentive, programs if the area does not achieve emission reductions on schedule. The core measures, to be implemented during the first 5 years of the FIP, include further limitations on the seasonal changes in gasoline volatility, use of oxygenated fuels and reformulated gasoline (as described in Chapter 12),

OCR for page 67
Page 84 Figure 3-3 Predicted percent reductions of VOC emissions in 1994 and  2004 compared with 1985 emissions, by control method. *Indicates that emissions reductions would also be achieved in areas in attainment of the ozone NAAQS. Percent reductions from methanol fuels would occur only in areas where use is required. TSDF = hazardous waste

OCR for page 67
Page 85 Figure 3-3 (continued). treatment, storage, and disposal facilities. RACT = reasonable available control technology required for existing stationary sources. Enhanced I/M = inspection and maintenance of motor vehicles. Stage II = control devices on gas pumps to capture gasoline vapor during motor vehicle refueling. CTG = new control technique guidelines for RACT on existing stationary sources. Onboard controls = emission controls on motor vehicles. Source: OTA, 1989.

OCR for page 67
Page 86 controls on marine vessel tanks, and further control of evaporative emissions from gasoline-fueled motor vehicles. EPA has proposed to implement a regulatory-based FIP to obtain the remaining reductions needed after the first 5 years. Under this approach, highly restrictive performance standards would be developed, and as new technologies become available the regulations would be adjusted as needed. The regulatory FIP also calls for an ultra-dean-vehicles program that will require new vehicle standards that can be met through any combination of fuel mixes and new vehicle design. Additionally, the program calls for a composite in-use standard instead of separate measures for exhaust, evaporative, and running loss and for loss during refueling. The motor vehicle program would allow marked-based practices such as banking, averaging, and trading of emissions. The stationary-source program would encourage pollution prevention and market-based approaches. Public education would encourage the use of fewer consumer product solvents. The EPA plan also would allow sources to decide how to obtain VOC reductions, including the use of such measures as reformulation or substitution of compounds contributing to ozone formation, use of control equipment, or purchase of emission reduction credits from more effectively controlled facilities. In addition, industrial and commercial sources would be required to reduce emissions at a rate of 6% per year. Even with these measures, EPA estimates that because of growth, there will need to be further reductions. EPA will monitor growth projections at least every 3 years and propose such regulations as needed. These future controls are likely to be far less cost effective than the controls now available (Wilson et al., 1990), and regional control flexibility could be a necessary ingredient to maximize cost effectiveness in attaining the ozone NAAQS. Likewise, compliance by an emission source with air-quality permits would not be fully effective, especially when the maximum emissions allowed by the permit are exceeded due to unusual operating conditions or high-temperature days. It is likely that with further controls in severe and extreme nonattainment areas becoming increasingly limited, these areas will follow the SCAQMP and EPA approach. Rule Effectiveness A major fallacy of existing SIPs is the presumption that control measures achieve 100% effectiveness. The number and diversity of VOC sources, the uncertain reliability of existing controls, and limitations in state and local resources make 100% compliance extremely unlikely. When determining the

OCR for page 67
Page 87 extent of emissions controls required to meet the NAAQS. States need to account for the fact that rules are less than fully effective. Specific data or studies that compare the effectiveness of rules have been limited. The SCAQMD conducts a periodic audit that includes engineering assessments, field inspections, and source testing. In 1988 it found that more than 70% of the 180 facilities visited by the audit teams had underestimated emissions by an average of 15% (Guensler, 1990). An American Petroleum Institute (API) evaluation of SIPs in seven cites reported the most common SIP deficiencies and inadequacies in implementation (API, 1989). The report concluded: Actual effectiveness of implemented control measures was less than that estimated in control strategies. This was observed for both motor vehicle I/M programs (which in some cases fell far short of targets) and controls on stationary and area VOC sources. Technological and enforcement weaknesses are both likely to contribute to this problem. The magnitude of shortfalls in effectiveness is uncertain, due to the general lack of data in all areas regarding in-use controls. Moreover, the seven-city report found that the RFP reports were not a reliable means of identifying weaknesses during the SIP implementation phase. Even when deficiencies in the 1982 SIP estimates had been found and corrected, there was little feedback into the regulatory process, so there was not enough effort to upgrade the measures or implement further controls. EPA has released a guidance document for new SIPs that provides the basis for preparing base-year inventories for stationary sources (EPA, 1990e). In addition, the Clean Air Act amendments of 1990 require EPA to review emission factors at least every 3 years. Emission factors must be established for sources that have not had them in the past. Also, the 1990 amendments require stationary sources to submit annual statements of VOC and NOx emissions. EPA's guidance procedure presumes that each rule is 80% effective. However, the real effectiveness is highly variable, and one option would allow tracking and establishment of an appropriate effectiveness level for each area. As an example, during the EPA study of regional ozone modeling for northeast transport (ROMNET), each of the 12 participating states took part in a survey to evaluate its own rule effectiveness for existing measures (EPA, 1988a). Three aspects of source categories were investigated: technology control efficiency, enforcement compliance, and projected losses due to rule cutoff levels, waivers, and exemptions. An example of technology control effectiveness is shown in Table 3-3. The states reported that outright prohibi-

OCR for page 67
Page 88 TABLE 3-3 Maximum Technology Control Levels for VOC Area Sources Description Control level, % efficiency Gasoline marketing     Stage I 95   Stage II 86 Prescribed forest burninga 100 Agriculturala 100 Degreasing 83 Drycleaning 70 Graphic arts (printing) 85 Rubber and plastics manufacturing 83 Architectural coating 52 Auto body repair 88 Motor vehicle manufacturing 88 Paper coating 90 Fabricated metals coating 57 Machinery manufacturing 90 Furniture manufacturing 90 Flat wood products coating 90 Other transportation equipment 88 Electrical equipment 90 Ship building and repairinga 47 Miscellaneous industrial manufacturinga 85 Miscellaneous industrial solvent usea 85 Miscellaneous nonindustrial solvents 20 Publicly owned treatment works 90 Cutback asphalt paving 100 Fugitive emissions, SOCMIb 56 Bulk terminals/bulk plants 91 Refinery fugitive emissions 93 Process emissions, bakeries 90 Pharmaceutical emissionsa 90 Synthetic fibersa 85 Crude oil- and gas-products fields 93 Hazardous-waste TSDFsc 90 aValues provided by states based on in-use experience;bsynthetic organic chemical manufacturing industry;ctransportation, storage, and disposal facility.

OCR for page 67
Page 89 tions that require minimal enforcement provide the most effective control, and small-source controls, which require the most enforcement, provide the least effective control. The states reported that the product of the three aspects reduced the overall control effectiveness for many source categories to below 50%. During the initial phase of the upcoming SIPs, an effort will be made to upgrade existing rules. EPA has indicated that efforts will be made to require consistency among the states and regions with regard to the effectiveness of similar rules. The more stringent operating permit and enforcement programs mandated in the 1990 amendments of the Clean Air Act also should improve effectiveness. In addition, the 1990 amendments contain an improved RFP tracking requirement that imposes a fixed annual percentage reduction averaged over a set number of years. Failure to meet this requirement would subject an area to the sanctions provided for in the act. Possible sanctions include more stringent offset requirements for new construction, withholding of federal highway funds, and, in extreme and severe areas, fines on major stationary sources. Major emphasis will be placed on effectiveness improvement during the implementation phase of the SIP. Summary The heart of the Clean Air Act for ozone attainment is the State Implementation Plan (SIP), a plan for emission reductions designed to reduce ambient concentrations of ozone to concentrations that do not exceed the National Ambient Air Quality Standard (NAAQS). Despite considerable efforts over the past 2 decades in developing and implementing SIPs, NAAQS violations are still widespread. The essential components of SIP development include • monitoring of ambient pollutant concentrations to determine whether the NAAQS has been exceeded and, if so, by how much and where; • collection and analysis of meteorological data and air-quality data needed to develop the appropriate emissions-air quality relationship; • inventorying of emissions from point, area, and mobile sources to determine the emission reductions necessary to attain the NAAQS; • projecting the emissions inventory to future years; • identifying and selecting specific emissions control measures and demonstrating that the control strategy will be adequate to achieve the air-quality goal.

OCR for page 67
Page 90 However, the SIP must be effectively implemented. The necessary steps in implementation include adoption and enforcement of regulations, tracking of progress in achieving emission reductions, and adjustment of the plan as necessary to achieve emission reduction targets. If the system works properly, implemented emission reductions should provide attainment. Failure to attain the ozone NAAQS can result from ineffective implementation of the SIP or from defects in the SIP itself. The SIP will not succeed if it does not properly estimate emissions, if it does not adequately relate emissions to ambient concentrations, or if it fails to identify sufficient emission reductions. Although the SIP appears to be a fundamentally correct approach to air-quality management, major weaknesses in both its methodology and its application are apparent. Base-year emissions inventories have underestimated actual emissions, in some cases by substantial amounts (see Chapter 9). As a result, future-year emissions inventories have been consistently underpredicted, and emission reductions brought about by controls often have a proportionately smaller effect on total emissions than the SIPs originally estimated. In addition, the reductions required to attain the ozone NAAQS were in most cases developed using EKMA (empirical kinetic modeling approach), an emissions-air quality model that did not account sufficiently for regional characteristics (see Chapter 6). Finally, individual control measures for motor vehicles and stationary sources were falsely assumed to meet their emission reduction targets, and provisions were not made for evaluating actual reductions. The magnitude of shortfalls in effectiveness is uncertain because of the general lack of data in all areas regarding in-use controls. Thus, SIPs have overstated the effectiveness of controls in reducing ozone concentrations and have understated the emission reductions needed to attain the NAAQS. Improvements in implementation must be directed toward the use, where possible, of realistic descriptions of the relationships between emissions and air quality and toward control programs that account for uncertainties and potential shortfalls in the effectiveness of controls. Procedures are needed for tracking actual emissions and reductions. Finally, and most important, feedback must be provided from the implementation phase to the SIP development phase. The following actions are recommended to improve the effectiveness of State Implementation Plans: • Establish air pollutant transport commissions in appropriate areas and determine their effectiveness. • Concentrate SIP modeling resources in nonattainment areas requiring relatively large efforts to achieve compliance with the NAAQS, including multistate areas. Accommodative or technology-based SIPs should generally be used elsewhere.

OCR for page 67
Page 91 • Conduct a review of the appropriate numbers and siting of ozone monitors for urban, suburban, and rural areas. • Establish a reference method for a continuous total VOC (volatile organic compound) monitoring system that can speciate major VOC classes, and establish VOC monitoring in the most severe nonattainment areas (see Chapter 7). • Develop independent validation techniques for the SIP components. For example, statistically based surveys could compare stationary source inventories, and roadside screening surveys should be compared with mobile source inventories. • Establish an effective audit program to track SIP progress. • Establish feedback between the SIP development and implementation phases. If control measures are not being implemented effectively for technical or other reasons, adjustments to the plan must be made.

OCR for page 67