cw» Li.(l; gt} . ?3?X;§ L ~ '4 87-436 RCO 5:.’ v- 1,: a, . _ LHW£tE{ ‘ ywr~« 2 .7'~ - ? - «K 4' ‘H1 i __ 5 4 En; K-2;, - cRs REPORT FOR CONGRESS r 4_ lcaflons G°V9’"ment U H ‘W6 7 7 1994 Wash1ngt0,, Univ St. Louis’ ersity Libraries Mo 63130 EMISSIONS IMPACT OF OXYGENATED (ALCOHOL/GASOLINE) FUELS Emissions of carbon monoxide from auto engines fueled with alcohol/gasoline blends are significantly lower than CO emissions from engines fueled by gasoline itself. Emissions of hydrocarbons from blends are sometimes lower and sometimes higher, with the overall impact apparently a slight reduction. However, blends have higher volatilities than gasoline itself, so total hydrocarbon entry into the atmosphere might rise. Emissions of nitrogen oxides appear to be somewhat higher with blends. Thus, reductions in CO might come at the expense of increases in ozone precursors. ulu 0 C by David E. Gushee Senior Specialist in Environmental Policy May 20, 1987 mini M iii ‘iii O10-103940443 lillltiiml “ a Ill! The Congressional Research Service works exclusively for the Congress, conducting research, analyzing legislation, and providing information at the request of committees, Mem- bers, and their staffs. The Service makes such research available, without parti- san bias, in many forms including studies, reports, compila- tions, digests, and background briefings. Upon request, CRS assists committees in analyzing legislative proposals and issues, and in assessing the possible effects of these proposals and their alternatives. The Service’s senior specialists and subject analysts are also available for personal consultations in their respective fields of expertise. CONTENTS INTRODUCTION THE AIR POLLUTION IMPERATIVE THE AGRICULTURAL IMPERATIVE OXYGENATED (ALCOHOL/GASOLINE) FUELS Blend Compositions ....................................... Estimated Effects on Auto Emissions Estimated Effects on Carbon Monoxide Levels in Air Increased Fuel Volatility Estimated Effects on NOX and HC Levels in Air Aldehyde Emissions Conclusions 000000000000 CD®\l\lO\-D-L‘ EMISSIONS IMPACT OF OXYGENATED (ALCOHOL/GASOLINE) FUELS Congressional interest in alcohol as a fuel or fuel component, at a fever pitch during the energy crises of the 1970's, has resurfaced in the late 1980's. This time around, the two basic driving forces behind this interest are environmental and agricultural--environmental because of the perception that alcohol in vehicle engines reduces air pollution, and agricultural because one of the alcohols (ethanol) can be made from surplus crops such as corn and other grains. In the 1970's, the driving force was to substitute domestic raw materials (agricultural crops or coal) for imported inputs (crude oil); this interest remains but is not as intense at this time. Two compounds--methanol, sometimes known as wood alcohol, and ethanol, also known as grain alcohol-are under consideration. Two types of fuel-- mostly gasoline with some alcohol, or mostly alcohol, perhaps with some gasoline--are involved. Many other alcohols exist but are not prime candi- dates for a variety of physical and chemical reasons. And many other types of fuel mixtures are possible, but are not frontrunners, again because of a variety of physical and chemical characteristics. As with everything else'in life, the situation, when treated in depth, is really more complex than two alcohols, two fuel forms, and two driving forces. But these six factors permit one to develop a reasonable picture of the forces at work. This report will focus on the environmental dimension of the alcohol-in- gasoline issue. It will first describe briefly the air pollution situation with respect to ozone and carbon monoxide, two of the six "criteria pollu- tants" for which the Clean Air Act requires ambient standards and deadlines for their attainment. It will then highlight the agricultural pressures which underlie current farm belt interest in fuel as an outlet for crop excesses. It will then discuss what is assuredly known and what is not-so-assuredly known about emissions effects from alcohol/gasoline fuels. THE AIR POLLUTION IMPERATIVE The Clean Air Act as amended1 requires, among other things, that the concentrations of specified pollutants in the air not exceed specified levels (National Ambient Air Quality Standards, or NAAQS) set by the Environmental Protection Agency (EPA) to protect public health. The Act specifies six such 1 42 u.s.c. 7401-7626 CRS-2 pollutants, and gives EPA the responsibility of identifying others over time. Two of the original six are ozone and carbon monoxide (CO). CO is emitted directly as a pollutant, mostly as a product of imperfect combustion in vehicle engines and other combustion sources in industry. Ozone is not emitted directly but forms in the air by reaction in the presence of sunlight of photoreactive, volatile, organic compounds (called VOC's, for short) and nitrogen oxides (NOX), both of which are emitted from vehicles, industry, and a number of other types of sources. The Act also specifies that, in areas of the country where concentra- tions of these pollutants exceed the standards, the States are to develop plans (State Implementation Plans, or SIP's) to bring under control emission sources so that emissions and hence ambient air concentrations are reduced so that the standards will be met. The Act specifies further that these steps are to be taken and the standards achieved by no later than December 31, 1987, for these two pollutants.2 A The EPA has taken a number of steps at the national level to reduce the relevant emissions. Perhaps the best known is the emission control program for new vehicles; other steps are New Source Performance Standards (NSPS) for new industrial sources and identification of cost effective control techno- logies for a variety of types of existing sources. The States with nonattainment areas have already taken a number of steps to reduce the offending emissions. They have required retrofit for pollution control of the categories of sources for which EPA has issued guidance. They have installed inspection/maintenance systems for cars in use. They have established a number of other controls such as improved public transit, improved traffic flow, and staggered work hours. There are some 76 areas of the country (Table 1) currently not meeting the ozone standard and about 40 areas (Table 2) not meeting the CO standard. Most of these will not meet the deadline. For most of these areas, most of the easy steps have been taken. The next increments of emission reduction will be achieved by--or at least will be seen by many as-imposing signifi- cant cost or inconvenience. If alcohol fuels were to offer significant reductions in CO and/or VOC emissions, then perhaps many of the additional future control actions currently being contemplated in nonattainment areas might not be necessary. In addition, many areas now in attainment would find it possible to maintain attainment without--or with fewer--of the more costly or more inconvenient control actions seen now as necessary in the future. That hope of relatively painless emission reduction is the source of the air pollution imperative propelling alcohol fuels. 2 For further details on ozone nonattainment, see Issue Brief IB 87066, "Clean Air Act: Ozone Nonattainment" by David E. Gushee, regularly updated. CRS-3 THE AGRICULTURAL IMPERATIVE3 The U.S. agricultural sector is currently plagued by a multitude of complex problems, not the least of which is the high level of government spending on farm programs. Yet a number of farmers will continue to go bankrupt over the next few years. Land values have declined markedly since 1981 and are still falling, placing highly leveraged producers and financial institutions in a precarious position. Exports of U.S. agricultural commodi- ties have dropped steadily since 1981. The underlying theory behind the 1985 Farm Bill is that crop exports hold the key to the economic future of American agriculture. It is widely recognized that the domestic market for agricultural products is relatively mature and that further growth in consumption will be slow to develop. In fact, the rate of growth in domestic consumption is not large enough to keep pace with productivity gains. Now the United States is pursuing policies to try to regain markets through lower prices. Market response to these reduced price levels has been slow. Consequently, many individuals are doubting whether the 1985 Farm Bill can achieve the desired objectives of reducing U.S. prices to be competitive, expanding U.S. export markets, maintaining farm income, and keeping Federal expenditures for agriculture to a minimum. Large corn crops and lagging exports over the past two years have caused inventories to build and have pushed market prices below price support program levels. In the 1986/7 crop year, farm level prices of corn are estimated to average only $1.60 to $1.65 per bushel, 30-35 cents below the $1.92 price support level. Even with a stronger acreage control program in 1987, corn supplies should hold the corn price to the $1.65 to $1.75 per bushel range. The U.S. Department of Agriculture projects spending for agricultural price and income support programs at $25.3 billion for the current fiscal year, down only slightly from last fiscal year's level of $25.8 billion. The corn and sorghum programs alone are estimated to exceed $13 billion. If the nation were to embark on a full scale ethanol fuel program, from two billion to four billion bushels of grain would be converted each year. Such an increase in domestic demand for grain would eliminate our grain surpluses and reduce, or possibly eliminate, the need for crop production controls and price supports. Grain prices would rise considerably above current levels; rising with them would be the incomes of grain producers, the cost of producing meat and other animal products, and consumers‘ food costs. The desire to free U.S. agriculture from the vagaries of export markets and government farm programs is the source of the agricultural imperative propelling alcohol fuels. 3 This section was prepared by A. Barry Carr, CRS Specialist in Agricultural Policy CRS-4 OXYGENATED (ALCOHOL/GASOLINE) FUELS Blend Compositions The most clearly demonstrable emissions impact of oxygenated gasoline is reduced emissions of carbon monoxide. The major variable affecting emission performance of these fuels is the oxygen content. Currently, the favored range is 2-3.5% oxygen, limited on the low end by the emission reduction impact (oxygen content of less than 2% fails to lead to meaningfully reduced carbon monoxide (CO) emissions) and on the high end by the impact on fuel characteristics (storage stability, for example). Formulating blends to meet the varied requirements for vehicle fuels is a complex task and can lead to complex fuel mixtures. Volatility over a range of weather conditions, tendency to pick up water, behavior in storage tanks, tendency to affect gaskets, hoses, and other materials in the fuel systems, energy content per unit of volume, and octane rating (which affects the tendency of the engine to knock under load) are several of the factors to be considered. The three basic blend fuels currently under active consideration are gasoline/ethanol, gasoline/methanol, and gasoline/MTBE (MTBE is methyl tertiary butyl alcohol and is made from methanol). Gasoline/ethanol blends (commonly called gasohol) are usually about 90% gasoline and 10% ethanol. Gasoline/methanol blends usually contain no more than 5% methanol and are formulated with additional additives to cope with negative aspects of the basic blend such as vapor lock, corrosion, damage to hoses and gaskets, and tendency to pick up water. Gasoline/MTBE blends are usually between 5 and 10% MTBE. Estimated Effects on Auto Emissions Enough studies have been carried out to provide an idea of the emissions performance of oxygenated fuels. EPA has a bibliography4 of 43 journal arti- cles, 72 reports and meeting presentations, a number of personal communica- tions, and the information contained in documents submitted in support of requests for fuel waivers under the Clean Air Act. The documents cited in this bibliography identify about 500 cars as having undergone tests. Several of the studies involved fleets of up to 50 cars. — The State of Colorado has for several years been looking at oxygenated fuels as a control technique for its carbon monoxide nonattainment problem. 4 "List of References on Fuels Containing Alcohols and Other Additives" available from Emission Control Technology Division, Office of Mobile Sources, Environmental Protection Agency, Ann Arbor, MI. CRS-5 In its deliberations, it has used5 a data base of seven studies involving about 110 vehicles to make its estimates of the benefits of a gasohol program. Automobile manufacturers, fleet owners, fuel additives makers, and others have also carried out emissions tests on various fuel mixes in various cars over various test cycles. In all, there are a least several hundred careful experiments and probably several thousand indicative tests reported both in the refereed literature and in the "gray literature." The emissions effects measured in these tests derive from the effect of oxygen in the fuel on the "air/fuel ratio" in the engine (air/fuel ratio is the basis on which changes in many engine operating characteristics are described). The oxygen in the fuel "leans out" the mixture of vapor and air in the engine combustion chamber (the cylinder). From a host of carefully controlled studies of engine models over the years9, it has become well documented that as the air/fuel ratio increases (leans out), carbon monoxide emission levels continuously decline, hydrocarbon emission levels first decline and then rise (the rise occuring at a rather slow rate compared to the earlier decline), and nitrogen oxide emission levels rise sharply and then decline sharply. The NOX peak and the HC valley are very close to each other. As a car is started, warms up, and is driven through a test cycle, the changing conditions continually change the air/fuel ratio in the engine. Changes in the test cycle for any given car will cause changes in the measured emission results. Add changes in the car as it ages, and then add in the very wide range of makes, models, states of mechanical repair, and test cycles used, and it becomes clear why there is a range of results and a certain uncertainty in reaching overall conclusions--except for CO, where oxygenated fuels indisputably reduce emissions. These studies show beyond a doubt that carbon monoxide (CO) emissions are reduced. Nitrogen oxide (NOX) emissions are apparently increased (there are some reports showing no effect and a few showing decreases). Hydrocarbon (HC) emissions sometimes increase and sometimes decrease, with the prepon- derance of evidence pointing to a net decrease. These effects are most noteworthy on pre-1981 model year cars, rather small on 1981-4 model year cars (which have three-way catalysts and oxygen sensors in the exhaust systems controlling the air-fuel ratio in the engine), and even smaller in 1985 model year and later cars (which in addition have fuel injection and/or "adaptive learning” electronic systems controlling air-fuel ratios). For all model years, the effects increase with increasing altitudes. 5 "Ethanol-Blended Fuel as a CO Reduction Strategy at High Altitude" prepared by Mobile Sources Program, Air Pollution Control Division, Colorado Department of Health, August 1985. 6 For a summary of many of these studies, see "Performance of Alco- hol/Gasoline Blends in Automobile Engines" by Robert E. Trumbule, Oct. 26, 1978. CRS-6 Test data indicate that, for pre-1981 models, C0 reductions of 20% or more can often be expected, while for 1981 and later models, C0 emissions might go down by somewhere between 5 and 10%. At higher altitudes, greater reductions can be expected; Denver specialists have calculated reductions as high as 30-40% for the earlier model year cars, and a more modest 10-20% or so for the later model year cars. For nitrogen oxides, most studies show that emissions rise 6-9%, apparently across all model years. For hydrocarbons, there is not yet a good estimate of the aggregate change in emissions. Estimated Effects on Carbon Monoxide Levels In Air Estimating aggregate emissions effects from the limited data base available on vehicles using alcohol fuel blends is fraught with uncertainty. Although some thousands of cars have been tested, the data are not plentiful in terms of the multiple models of cars in use, the wide range of circum- stances in which they are used, and the great variability in state of repair and type of driving. Given that caveat, order of magnitude estimates can be made. The fact that newer model cars show less reduction in C0 than older ones has major implications. The change in car control systems since 1981 means that whatever air quality benefits are achieved from widespread adoption of oxygenated fuels would be highest in the first year of adoption and would decline over time as the older vehicles are phased out of the fleet and are replaced with cleaner new vehicles. By the mid-1990's, with essentially all vehicles operating with adaptive learning controls, there would be only a modest carbon monoxide benefit (estimated by EPA7 to be 5-10% maximum, based on test results from late model new cars with adaptive learning or fuel injection). ’ Since motor vehicles contribute two-thirds to three fourths of all C0 in most urban areas,8 and cars contribute two-thirds to three-fourths of vehicle emissions, car C0 reductions do not result in one-to-one reductions in C0 ambient levels. An order of magnitude estimate would be two to one--that is, each 2% reduction in car C0 emissions would lead to a 1% reduction in C0 levels. At most, a 1.5 to one ratio might be appropriate in a few cities with little industry and minimal truck traffic. Thus, urban C0 concentrations might initially decline between 3% and 10% at sea level and up to perhaps 20% at high altitudes. Such a decline might bring several areas (see table 2 for CO noncompliance areas) into C0 com- 7 Personal communication from Jeff Alson, Emission Control Technology Division, EPA, April 22, 1987. 8 "National Air Quality and Emissions Trends Report, 1984" Environ- mental Protection Agency EPA-450/4-86-001, April 1986 CRS-7 pliance for a brief period, but the decay in CO reductions caused by fleet turnover coupled with historic growth in population and vehicle use would mean that gasoline/alcohol blends would not solve the CO nonattainment problem in very many places and even where they did, not for very long. Increased Fuel Volatility Adding methanol or ethanol to gasoline increases fuel volatility.9 At the levels currently under consideration (10% or less), fuel Volatilities are about 1.5 to two times as high as the volatility of gasoline alone. That translates into an increase in Reid Vapor Pressure (RVP, the standard measure of volatility used for fuels) of 1-2 psi. Gasoline RVP's for many years averaged about 9 psi, but over the past several years, they have been increasing and are now often above 11 psi. Volatilities of the blends would be commensurately higher. The blends currently being sold today are sold under waiverlo from EPA, despite their higher volatility. MTBE does not have the same magnitude of effect on blend volatility. This may turn out to mean more in the future than it has heretofore. Higher volatility means that more fuel evaporates during and after vehicle operation, and more fuel escapes into the air during vehicle refu- eling. Even though methanol and ethanol are themselves less photochemically reactive than gasoline components, the increased volatility is made up mostly of gasoline components. EPA is currently considering a new vapor pressure regulation which reportedly would put an upper limit on RVP of 9 psi; EPA might propose such a regulation later this year. Gasoline evaporation matters to ozone attainment, because gasoline is a "volatile organic compound" (VOC) which contributes to ozone formation. Widespread adoption of alcohol fuels would, in the absence of compensating changes in fuel formulation, exacerbate the evaporation problem, which would in turn tend to increase local ozone levels. Regulating RVP of gasoline will introduce some controversy, as refiners would have to reformulate their fuel composition by removing some of the lighter components. This would cause problems at the refineries, where these light ends are currently in oversupply and have no other use of comparable value. To regulate the RVP of blends would not only exacerbate this problem but would also introduce additional complications and costs. Estimated Effects on NOX and HC Levels in Air Emissions data for HC from blend-fueled cars are equivocal, in that, as indicated earlier, they are lower for some cars and higher for others. From 9 Letter from Charles E. Gray, Jr., Director, Emission Control Tech- nology Division, EPA, to Charles P. Anders, Arizona Department of Health Services, Oct. 16, 1986. 10 Under Section 21l(f) of the Clean Air Act. CRS-8 the limited available evidence, the net effect is more likely to be a lowering than an increase for the vehicle fleet as a whole. EPA is studying this issue but has not yet concluded whether the potential net reduction in engine emissions would or would not outweigh the effects of increased volatility. In anticipation of the results of the EPA study, one could legitimately assume that whatever emissions benefits would initially occur would decay over time via fleet turnover in the same way that CO reductions would, while the volatility effects would continue on. Thus, the net effect on the combination of emissions and evaporation over time is more likely to be a net increase in VOCs than a net reduction. As indicated earlier, NOX emissions would increase 6-9%. Aldehyde Emissions When alcohols are oxidized (combustion is one way to oxidize chemicals), among the resulting products will be aldehydes. Methanol partially oxidizes to formaldehyde; ethanol partially oxidizes to acetaldehyde. Aldehydes are more toxic, more reactive, and are basically less desirable contaminants than either the alcohols or products of more complete oxidation such as organic acids and ultimately carbon monoxide and carbon dioxide. Aldehydes are of considerable concern in Brazil, for example, where most cars use either "neat" (100%) ethanol or a blend including 20% ethanol and most cars do not have catalytic afterburners to clean up exhausts.11 Experi- ments on U.S. cars show that aldehyde emissions do increase when alcohol blends are used as fuels, but the increases are modest in older cars and miniscule in newer ones.12 The EPA considers that aldehydes are not of concern so long as alcohol/gasoline blends make up a small fraction of the fuel mix, but the agency has not reached a final position on aldehydes under a 100% blend fuel mix. Conclusions Whatever benefits from reduced CO emissions are achieved will be associated with disbenefits in terms of increased NOX emissions and, at best, after a few years when older cars leave the fleet, in increased HC emissions as well. This effect may not matter in the colder months, when ozone for- mation from HC and NOX is not a significant factor, but it would matter in the warmer months (the so-called smog season) from Spring through Fall in most parts of the country and almost all year round in Southern California and most of the year in parts of Arizona and the Gulf Coast. Even if HC emissions do not increase as a result of the higher volatilities, the higher 11 Automotive Use of Alcohol in Brazil and Air Pollution Related Aspects. Alfred Szwarc and Gabriel M. Branco. SAE Technical Paper Series 850390, February 1985. 12 Ethanol-Blended Fuel as a C0 Reduction Strategy at High Altitude. op. cit. CRS-9 NOX emissions will in themselves exacerbate ozone formation. And ozone nonattainment is a more ubiquitous and more intransigent air quality problem than CO nonattainment is. The fact that alcohol/gasoline blends would not be a CO compliance panacea does not mean, however, that such a step should not be taken in at least some places or for at least some specified periods during the year. Experience has shown that air pollution control, and particularly ozone control, is the result of a large number of different actions, each of which makes its modest contribution to the total effort. That is the reason why Denver, for example, is close to a decision on mandating alcohol/gasoline blends for wintertime use as the latest in its long series of control actions. Not only does it not have an ozone problem in the winter but its altitude (the "mile-high city") magnifies the CO reductions potentially available, particularly in the first years of adoption. Despite the inevitable trade-off in air pollution effects, strong support from the farm sector for alcohol/gasoline blends can be expected to continue. Carry-over corn stocks this year will be close to 6 billion bushels; were 50% of the gasoline market converted to an ethanol/gasoline blend, as is proposed in H.R. 2052 (Rep. Durbin, D., Ill.), a market for some two billion bushels of corn per year would develop. Until such time as foreign markets are found for excess U.S. agricul- tural production capacity, farmers will seek new markets in domestic indus- trial uses such as ethanol. And agricultural policymakers, staggering under the $26 billion farm subsidy burden, will continue to search for ways to reduce subsidy costs and crop surpluses. 4-‘J-\-I-\-l>-|>J>L»UJLaJUJLAUJUJbdL»JLnhJI\JhJl\Jl\Jl\Jf\J[\Jl\Jl\>r-9-4»--9-r-up-at-‘v-0-do-— L11J>UJl\Jr-'O\Dm\lO\U'|-I-\UJt\JP-'O\DCD\lO'\U'|-l>UJl\>9-‘C>\OCn\lO‘U1-l>UJl\>r—-O\OCX)\lO\U1-l>u.>t\Jo—- Table 1 Air Quality Areas Exceeding Ozone Standard (0.12 ppm) 1983-5 Name of Area Design Expected Value* Exceedances Los Angeles Ca 0.36 123.2 Houston/Galveston/Brazoria Tx 0.25 20.2 Greater Connecticut Ct 0.23 35.6 New York, NY-NJ-Ct NY 0.22 31.3 San Diego Ca 0.21 11.2 Chicago/Gary/Lake Co. I1 0.20 8.5 Atlantic City NJ 0.19 10.3 Providence/Pawtucket/Fall River RI 0.18 10.3 Philadelphia Pa 0.18 9.9 Sacramento Ca 0.18 8.6 Baltimore Md 0.17 10.1 Cincinnati/Hamilton Oh 0.17 7.1 Fresno Ca 0.17 6.6 Milwaukee/Racine Wi 0.17 6.2 San Francisco Ca 0.17 3.7 Atlanta Ga 0.16 7.5 Bakersfield Ca 0.16 27.2 Baton Rouge 1a 0.16 3.4 Beaumont/Port Arthur Tx 0.16 8.3 Boston/Lawrence/Salem Ma 0.16 6.6 Dallas/Ft Worth Tx 0.16 7.3 El Paso Tx 0.16 13.6 New Bedford Ma 0.16 10.8 Phoenix Az 0.16 4.8 Portland Me 0.16 6.8 Santa Barbara/San Maria/Lompoc Ca ‘ 0.16 3.0 St. Louis Mo 0.16 9.4 Washington, DC-Md-Va DC 0.16 8.5 Longview/Marshall Tx 0.15 2.3 Louisville Ky 0.15 8.0 Memphis Tn 0.15 2.9 Modesto Ca 0.15 13.0 Salt Lake City/Ogden Ut 0.15 4.4 Seaford De 0.15 4.6 Stockton Ca 0.15 9.0 Worcester Ma 0.15 3.0 York Co. Me 0.15 9.5 Allentown/Bethlehem Pa 0.14 3.8 Cleveland/Akron/Lorain 0h 0.14 3.3 Dover De 0.14 3.7 Gardner Me 0.14 2.2 Huntington/Ashland, WV-Ky WV 0.14 5.2 Jacksonville Fl 0.14 1.7 Kansas City, Mo-Ks Mo 0.14 2.4 Lake Charles La 0.14 1.7 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 Table 1 Air Quality Areas Exceeding Ozone Standard (0.12 ppm) 1983-5 Name of Area Design Expected Value Exceedances Muskegon Mi 0.14 3.6 Nashville Tn 0.14 2.8 Northampton Co. Va 0.14 2.6 Acadia National Park Me 0.13 1.4 Birmingham Al 0.13 2.4 Charleston WV 0.13 1.3 Charlotte/Gastonia/Rock Hill NC 0.13 1.8 Dayton/Springfield Oh 0.13 1.5 Denver/Boulder Co 0.13 1.7 Detroit/Ann Arbor Mi 0.13 2.2 Erie Pa 0.13 1.4 Grand Rapids Mi 0.13 1.3 Hancock Co. Me 0.13 1.5 Harrisburg/Lebanon/Carlisle Pa 0.13 1.7 Iberville Parish La 0.13 2.4 Indianapolis In 0.13 2.5 Janesville/Beloit Wi 0.13 1.9 Lancaster Pa 0.13 3.3 Miami/Ft Lauderdale Fl 0.13 1.3 Pittsburgh/Beaver Valley Pa 0.13 2.6 Point Coupee Parish La 0.13 3.1 Portland/Vancouver, Or-Wa Or 0.13 1.4 Portsmouth/Dover/Rochester, NH-Me NH 0.13 1.3 Reading Pa 0.13 2.1 Richmond/Petersburg Va 0.13 2.6 St. James Parish La 0.13 3.9 Tampa/St. Petersburg/Clearwater Fl 0.13 1.7 Tulsa- Ok 0.13 1.9 Visalia/Tulare/Porterville Ca 0.13 8.8 York Pa 0.13 2.5 Yuba City Ca 0.13 3.0 * Fourth highest reading in 3-year period (three readings over 0.12 ppm allowed in three years) Source: Environmental Protection Agency Table 2 Metropolitan Statistical Areas with Carbon Monoxide Levels Greater Than Standard (9 ppm) 1983-5 No. of Design Exceed- Name of Area Value* ances 1 Los Angeles/Long Beach Ca 27.4 52 2 Denver 2 Co 24.0 59 3 Phoenix Az 20.3 85 4 Provo/Orem Ut 19.1 73 5 Fort Collins Co 17.8 7 6 Fairbanks Ak 17.7 40 7 Albuquerque NM 17.2 57 8 Medford Or 16.3 43 9 Sacramento Ca 16.3 31 10 Las Vegas Nv 16.3 21 11 Reno Nv 16.2 26 12 Greeley Co 16.2 12 13 Anchorage Ak 16.0 43 14 Boise Id 15.5 15 15 Spokane Wa 15.4 15 16 San Jose Ca 14.3 19 17 Yakima Wa 13.9 6 18 El Paso Tx 13.3 8 19 Colorado Springs Co 13.2 4 20 Oklahoma City 0k 12.8 10 21 Kansas City Mo 12.8 6 22 Anaheim/Santa Ana Ca 12.7 7 23 Missoula Mt 12.6 8 24 San Francisco Ca 12.3 4 25 Tacoma Wa 12.2 15 26 Portland Or 12.2 3 10 27 Salt Lake City/Ogden Ut 12.0 9 28 Fresno Ca 12.0 9 29 Tucson A2 11.8 14 30 Seattle Wa 11.5 10 31 Grants Pass NM 11.3 12 32 Santa Barbara/Santa Maria/Lompoc Ca 10.5 5 33 Des Moines Ia 10.4 6 34 Dubuque Ia 10.3 2 35 Dallas Tx 10.3 2 36 San Diego Ca 10.0 4 37 Boulder/Longmont Co 9.9 2 38 Salem 0r 9.8 4 39 Vallejo/Fairfield/Napa Ca 9.7 4 40 Chico Ca 9.6 2 * Fourth highest reading in 3-year period (three readings over 9 ppm allowed in three years) Source: Environmental Protection Agency LIBRARY OF WASHINGTON UNIVERSITY ST. LOUIS - MO.