Click for next page ( 18


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 17
Motor Vehicle Emissions and Regulation Motor vehicle emissions are a major source of air pollutants, and significant efforts over the past 3 ~ years have been directed at reducing these emissions. As required by the Clean Air Act Amen~nents of ~ 990 (CAAA90), regulatory agencies, particularly the U. S. Environmental Protection Agency (EPA), are pursuing multiple strategies to reduce emissions from mobile sources. These strategies include implementing vehicle-emissions inspection and maintenance (~/M) programs, setting tighter new vehicle tailpipe and evaporative emissions standards, and promoting alternative and reformulated fuels. This report fo- cuses on vehicle-emissions I/M programs, which are designed to identify vehi- cles that have higher than allowable emissions and to try to ensure that they are repaired or removed from the fleet. I/M programs attempt to control emis- sions throughout a vehicle's lifetime by ensuring that the emissions-control system is maintained and repaired when needed. As such, I/M programs can improve overall air quality. THE COMMITTEE'S CHARGE AND HOW IT ORIGINATED In its fiscal ~ 998 appropriations for EPA, the U. S. Congress called for the National Research Council (NRC) to assess the effectiveness of r/M pro- grams for reducing mobile-source emissions. The study is fended in two phases. The first phase, whichis the subjectofthis report, was charged to use available information to assess general relationships between motor vehicle 17

OCR for page 17
18 Evaluating Vehicle Emissions I/M Programs emissions characterized through I/M programs and emissions estimated by other methods. Also in phase I, the study is charged to develop criteria and methodologies for evaluating the design, implementation, and effectiveness of specific federal and state I/M programs. Phase 2 will evaluate several types of I/M programs. Specifically, phase ~ ofthe study, the subject ofthis report, is charged with the following: ~ . Describe the significance of emissions from motor vehicles subject to I/M programs relative to other emissions sources. The committee will assess the magnitude of emissions of carbon monoxide (CO), volatile organic com- pounds (VOC), oxides of nitrogen (NOx), and particulate matter (PM) from motor vehicles that exceed certified levels. 2. Compare motor vehicle emissions in areas with and without I/M pro- grams. The committee will compare motor vehicle emissions estimated from I/M programs with emissions estimated from other sources of data. (This will address the validity of using I/M program data to characterize fleet emissions levels.) The analysis will consider the merits and limitations of various ap- proaches used to estimate fleet emissions. 3. Identify criteria for the evaluation of I/M programs, including equip- ment needs, program costs, repair effectiveness, program effectiveness for vehicle categories, and effects of human behavior (such as vehicle tampering and I/M avoidance due to cost, inconvenience, program perceptions, and nonresponse to on-board diagnostic (OBD) indicator lights). 4. Develop methodologies for evaluating I/M programs based on the work described in items ~ -3. The methodologies should be applicable to exist- ing programs as well as programs to be implemented in the near future that will use OBD checks. Make recommendations for improving I/M programs. Identify research needs. COMMITTEE'S RESPONSE TO THE CHARGE REPORT CONTENTS This report documents the committee ' s response to the charge described above. The report consists of seven chapters and a summary. The committee found it useful to review passenger-car and light-truck emissions, the emissions

OCR for page 17
Motor Vehicle Emissions and Regulation 19 typically targeted by I/M programs, within the context of overall mobile-source and anthropogenic emissions. We also found it useful to review regulations covering motor vehicle emissions, especially/hose pertaining to I/M programs. These topics are discussed in Chapter ~ ofthe report. We considered it impor- tant to farther describe vehicle emissions control technologies and I/M testing techniques and to discuss how I/M has affected vehicle emissions. Chapters 2 and 3 examine these subjects. Chapter 4 discusses emerging testing tech- niques that may be incorporated into future I/M programs. Chapter 5 exam- ines the modeling used for simulating end predicting emissions reductions from I/M programs. Chapters 6 and 7 examine criteria and methods for evaluating I/M programs. Chapter 6 discusses evaluation methods for estimating emissions-reduction benefits due to I/M programs. Chapter 7 covers other criteria important for evaluating I/M programs, including costs and enforce- ment issues. The committee's findings and recommendations are presented in the Summary ofthe report. We begin with discussions of vehicle emissions and regulations pertaining to emissions standards and I/M programs. AIR POLLUTANTS EMITTED BY MOBILE SOURCES The air pollutants that are directly associated with mobile sources include CO, PM, VOCs or hydrocarbons (HC),~ and NOx~the sum of NO andiron. Mobile sources also contribute to ground-level ozone as HC, CO, and NOX emitted by vehicles react in the presence of sunlight to form ozone. Figures I-1 through 1-3 present estimates of the contribution of mobile sources to national annual emissions of CO, NOX, and VOCs based on model- ing results (EPA 2000a). Although these data give a general indication ofthe magnitude of mobile emissions compared with other major sources (fuel com- bustion in the figures refers to stationary sources), these emissions inventory estimates are lower than those estimated from on-road vehicle studies and The terms VOCs and HC are used in this report to denote organic compounds that are emitted as vapors under typical atmospheric conditions. Unless quoting an emissions inventory source or a regulation that uses another term, the report uses the term HC exclusively. Appendix B describes the differences among the terms used to refer to HC.

OCR for page 17
20 Evaluating Vehicle Emissions I/M Programs Note: Some fluctuations in the years before 1970 are the result of different methodologies. FIGURE 1-1 Trends in nationwide CO emissions, 1940-1998. Some fluctuations in the years before 1970 result Tom the use of different methods. Source: EPA 2000a. ambient measurements in urban areas, especially for HC and CO (IngalIs et al. 1989; Pierson et al. 1990; Fujita et al. 1992; Singer and Harley 1996; Gertler et al. 1997; Watson et al. 2001~. The mobile emissions reportedin Figures I-l through I -3 are categorized into on-road and non-road emissions; on-road emissions refer to both light-duty vehicles (LDVs) and heavy-duty vehicles (HDVs). The differentiation be- tween LDVs and HDVs historically has been 8,500 pounds gross vehicle weight (the weight of the vehicle plus the weight of the rated load-hauling capacity). LDVs are fueled primarily by gasoline, and HDVs use both diesel fuel and gasoline. The heavier HDVs, however (those with gross vehicle weights Heater than 26,000 pounds), are fueled almost exclusively with diesel fuel. Although these heavy HDVs make up about ~ % of the total number of vehicles, they represent about 5% of total vehicle miles traveled (VMT) and about ~ 4/O of total fuel consumption (Davis ~ 999~. Heavy HDVs are also an important source of PM and NOX emissions (EPA ~ 99Sa). Non-road emis- sions come from a wide variety of vehicles, including construction, logging,

OCR for page 17
Motor Vehicle Emissions and Regulation 21 .......... ... . ~iii~ i i i i ,,,,,,,,,,, . i ..i ............ Note: Some fluctuations in the years before 1970 are the result of different methodologies. FIGURE 1-2 Trends in nationwide NOxemissions, 1940-1998. Some fluctuations in the years before 1970 result from the use of different methods. Source: EPA 2000a. mining, and farm equipment; lawn and garden equipment; marine vessels; recreational vehicles; industrial, light commercial, and airport services; locomo- tives; and aircraft. In WAS, or-road end non-road mobile sources were est~matedto contribute over 75/0 of CO emissions nationwide (about 70,000 thousand short tons), about 50/0 of the NOx (about ~ 3,000 thousand short tons), and 40/0 of the VOCs (about 7,800 thousand short tons) (EPA 2000a).2 The remaining emis- sions came pnmanly from stationary source fuel combustion and industnal processes. The role of mobile-source emissions as a source of PM is not well understood. Emissions inventories indicate that mobile-source exhaust is a Mobile-source emissions are estimated with the MOBILE and NONROAD models, which are reviewed in NRC (2000~. To the extent that these emissions are projected using these models, they are subject to the types of uncertainties discussed in that report.

OCR for page 17
22 hi: Evaluating Vehicle Emissions I/M Programs 1.~0 4~5 ~~D i9~5 1960 1~S i~0 1~ Note: Some fluctuations in the years before 1970 are the result of different methodologies. FIGURE 1-3 Trends in nationwide VOC emissions, 1940-1998. Some fluctuations in the years before 1970 result from the use of different methods. Source: EPA 2000a. Source: EPA 2000a. relatively minor source of PM~o (atmospheric PM < ~ 0 Em in aerodynamic diameter) and sulfur dioxicle. Mobile-source contribution to fine particles (~2.5 Em in aerodynamic diameter and referred to as PM2 s) is an area of continuing study. Recent studies reported higher than expected PM2 5 emissions from LDVs at higher elevations (Fujita et al. 1998; Lawson and Smith 1998; Watson et al. ~ 998; Cadle et al. ~ 999a,b; Yanowitz et al. ~ 9994. Understanding these contributions is important as increased emphasis is now being placed on i ine PM (PM2 s) because of its potential effect on human health (Dockery et al. ~ 993~. A substantial fraction of PM2 s consists of secondary particles con- verted from gaseous pollutants by atmospheric processes (NRC 1998~. Within urban areas, mobile sources contribute an even greater fraction of air pollutant emissions than suggested by the national data shown in Figures ~ - l through ~ -3 . For example, emissions from mobile sources (both on-road and non-road) contribute 50% oftotal HC, 89/O of NOx, and94% of CO emissions in the South Coast Air Basin (which encompasses Los Angeles and Orange

OCR for page 17
Motor Vehicle Emissions and Regulation 23 Counties end the urbanized portions of Riverside and San Bernardino Counties in California) (TMRC 2000~. On the basis of its models, EPA suggests that vehicles typically contribute between 35% and 70% of HC end NOx emissions, and 90/O or more of CO emissions in cities with high levels of air pollution (EPA 1993a). However, as described previously andin NRC (2000), on-road vehicle emissions studies and comparisons of ambient data with emissions inventories have shown that mobile-source emissions are significantly higher than estimated by the models. VEHICLE TYPES AND STANDARDS Vehicle Types Mobile-source emissions can be categorized by the type of vehicle and engine system generating the emissions and by the type of fuel used. Table ~ - ~ (Sawyer et al. 2000) summarizes vehicle types, engines, and fuels that are commonly used. A single vehicle type may use any of several engines (e.g., a light-duty on-road vehicle might use a spark-ignition or compression-ignition engine, hybrid gasoline-electric, or electric engine) and any of several fuel types (e.g., gasoline, diesel, or liquefied petroleum gas). Most vehicle usage is associated with gasoline-powered, spark-ign~tion, light-duty on-road vehicles. Ofthe approx~mately200 million vehicles reg~steredin the United States, about two-thirds are light-duty, gasoline-powered passenger cars. There are also approximately 57 million light-duty trucks and 9 million heavy-duty trucks and buses in the current vehicle fleet (Sawyer et al. 2000~. As a rough guide to the significance of these vehicle categories, national emissions inventories for ~ 998 indicate that gasoline-powered vehicles ac- counted for 65% and 95% of on-road NOX and VOC emissions, respectively, with diesel trucks and buses contributing the remaining 35% of NOX and 5% of VOCs (EPA 2000a). Figure I-4 presents estimates of emissions by fuel type for on-road vehicles. Gasoline-powereU vehicles dominate VOC and CO emissions, and vehicles that operate on diesel fuel represent a significant frac- tion of NOX and direct PM emissions. As noteipreviously, however, mobile- source emissions have been underestimated and are subject to considerable uncertainty. I/M programs in the United States have been designed primarily for gasoline-powered, spark-ignition, light-duty on-road vehicles, although smaller

OCR for page 17
24 Evaluating Vehicle Emissions I/M Programs TABLE I-l Mobile-Source Vehicles, elm., ~~ ~~ Vehicles Light-duty on-road Heavy-duty on-road Heavy-duty off-road Light-duty off-road Aircraft Ships Locomotives Engines Fuels Spark ignition Compression ignition Gas turbine Electric Steam turbine (marine) Gasoline Diesel Jet fuel Residual fuel oil Liquefied petroleum gas Natural gas Electricity Alcohols Note: These sources are listed in approximate order of their use; boldface indicates the most important sources of HC and NOX. Source: Adapted from Sawyer et al. 2000. and newer programs have been introduced for HDVs in a few areas, such as California, Colorado, and several East Coast states. Therefore, most of this report focuses on LDV emissions, even though they are not the only important category of mobile-source emissions. I/M programs for HDVs represent an issue of increasing interest. Be- cause these vehicles are significant sources of PM and NOX emissions and they have not been subjected to extensive I/M programs, significant emissions reduction opportunities may exist. Currently, 1 ~ states have testing programs for assessing smoke emissions from heavy-duty diesel vehicles.

OCR for page 17
Motor Vehicle Emissions and Regulation 25 OXIDES OF NITROGEN 7~7R 5089 VOLATILE ORGANIC CARBON 5104 PARTICULATE MATTER (PM-~) CARBON ~ONOXIOE SULFUR DiOXIOE PARTICULATE MATER (PM-2.5) GAS Dl:ESEL ~3 FIGURE 1-4 Estimated emissions by fuel type from the on-road motor vehicle fleet in the United States. Emissions are shown in thousands of tons. Source: EPA 2000a. Vehicle Exhaust Emissions Standards Over the past 3 decades, efforts to reduce emissions from mobile sources have focused on reducing exhaust and non-tailpipe emissions from on-road

OCR for page 17
26 Evaluating Vehicle Em issior~s I/M Programs vehicles, specifically passenger cars and light-duty trucks. Passenger cars, as the name implies, refer to personal vehicles used primarily to transport people. For a given model year, a single set of emissions standards applies to all pas- senger cars (except for California, which has separate standards) regardless of size, passenger occupancy, oruse. Except for the firsts years of regulation 968- ~ 969), emissions have been regulated on a grams per mile basis. For 1968-1969, the standards were specified on a concentration basis, 275 parts per million (ppm) for total HC ant! ~ .5% for CO. This form of the standard allowed larger vehicles (with higher exhaust flow) to produce more emissions by mass than smaller models. The test procedure and the standards were revised for 1970-1972 to include an estimate for the exhaust volumetric flow based on the weight ofthe vehicle and a calculation ofthe mass of emissions produced per mile traveled. The test procedure and standards were modified again for model year ~ 972 and later vehicles to utilize a constant volume sam- pler (CVS) technique, which allowed a more accurate estimate ofthe vehicle mass emissions. New vehicles are certified by a chassis dynamometer test, such as the Federal Test Procedure (FTP).3 Table ~ -2 displays historical categories of standards (in grams per mile) for passenger cars for up to 50,000 miles and from 50,001 to 100,000 miles.4 Table ~ -3 summarizes some ofthe majormilestones forthese standards. The first federal emissions standards began with ~ 968 model-year vehicles and controlled engine ("crankcase") and tailpipe emissions of HC and CO. The ~ 970 Clean Air Act Amendments required that HC, CO, and NOX be Towered as soon as possible by at least 90%. Originally, the emissions standards were set at 0.4 I ,3.4, and 0.4 grams per mile (g/mi) for HC, CO, and NOX respec- tively, to be implemented starting with the ~ 975 model year. These standards were slightly modified and delayed, however, and were not fully implemented until the Clean Air Act Amendments of ~ 990 (CAAA90), which produced the Tier ~ emissions standards beginning with the ~ 994 model year. The final fed- 3In a chassis dynamometer test, the whole vehicle is mounted on a dynamometer for testing. In contrast, in an engine dynamometer test only the engine, not the whole vehicle, is mounted on a dynamometer. The FTP was designed as a standardized test for measuring the emissions from new vehicles. Manufacturers are allowed to certify compliance using low-mileage cars and an agreed-upon deterioration assumption. They are not required to recall and test in- service vehicles. (Vehicles may be recalled if emissions control systems are found to be faulty.)

OCR for page 17
Motor Vehicle Emissions arid Regulation 2 7 TABLE I-2 Passenger-Car Exhaust Gaseous Emissions Standards (all values in grams per mile except as noted) 50,000 miles 100,000 miles HC CO NOx NMHCa CO NOx . Model Year Precontrolb 10.6 84.0 4.1 1968-1969 275 ppm 1.5% 1970- 1971 4.1 34.0 1972 3.4 39.0 1973 3.4 39.0 3.0 1975-1976 1.5 15.0 3.1 1977-1979 1.5 15.0 2.0 1980 0.41 7.0 2.0 Category Tier 0 (1981-1993) 0.41 3.4 1.0 Tier 1 (beginning with 0.41 3.4 0.4 0.31 4.2 0.6 model year 1994) (0.25)a NLEV (beginning with 0.09 4.2 0.3 model year 1999) Tier 2 Default set in CAAA90 (beginning with model year 2004) Tier 2 Current proposed standards (beginning with model year 2004) Emissions standards were originally written for total HC and later for nonmethane HC (NMHC, shown in parentheses in the second column). Appendix B describes the differences among the terms used to refer to gaseous organic compounds. This report, unless otherwise noted, refers to the general class of gaseous organic compounds as HC. bStandards adjusted to current test procedure methods. CThe proposed Tier 2 standards are a corporate average standard with a focus on NOX emissions. This allows NMHC and CO emissions standards to "float," in that fleet emissions rates depend on the mix of vehicles used to meet the NOX standard. The emissions standards shown for NMHC and CO are those that would result given the mix assumed in the Notice of Final Rulemaking (EPA l 999a) to meet the NOX standard. Sources: Chrysler Corporation 1998; EPA 1998b, l 999a; Davis 2000. 0.125 1.7 0.2 >O.O9c >4.2c 0.07

OCR for page 17
Motor Vehicle Emissions and Regulation 35 Although a small population of vehicles is responsible for a large fraction of each category of exhaust emissions, the same vehicles are not necessanly super emitters in all emissions categories. Sawyer et al. (2000) conclude that although "there is significant overlap in the subset ofthe vehicle fleet that are high emitters of CO and HC...the NOX high emitters comprise a different, mostly disjoint set of vehicles from the CO and HC gross polluters." Figure -8 shows this relationship developed from roadside pullover tests in California. An important implication ofthe skewness of emissions distributions is that emissions targeted by I/M programs (those above the I/M cutpoints9) are concentrated in a small group of vehicles. For example, in the California I/M pilot study cited above, the data in Figure 1-9 have been used to demonstrate that about 75% of the excess aggregated CO, HC, and NOX emissions were produced by only 10% ofthe fleet. Excess emissions are defined here as the difference between a vehicle's current emissions rate and two times its certifi- cation standard. Aggregate excess emissions are determined by summing ~ ~ /7(CO) + NOX+ HC). This equation for aggregating excess emissions from multiple pollutants is described by the California Inspection and Maintenance Review Committee (IMRC ~ 993~. It should be noted that there is no standard definition of excess emissions or of how excess emissions relate to the amount of emissions that are "repairable." This concentration of excess emissions is also shown in the data from the ~ 999 California roadside survey, which purled over 8,443 vehicles. The rank- orderedexcess CO, HC, end NOx emissions are shownin Figure 1-10, which indicates that only 5% ofthe fleetproduces 76%, 83%, and 85% ofthe excess CO, HC, and NOX emissions, respectively. These results, however, might have been influenced by the methodology used to gather the test data. In an attempt to determine acceleration-simulation-mode fleet emissions rates with 95% confidence and a relatively small variance, a sampling method was used to ensure that a large proportion ofthe data collected are from older-model-year vehicles. Older vehicles have higher emissions and thus were selected more frequently for testing in this study. If new vehicles were adequately repre- sented in this data set, the degree of skewness would be even larger. Skewness is not limited to exhaust emissions. 1: iquid gasoline leaks are present in a small, but significant, fraction of current in-use vehicles. Recent investigations have indicated that although the frequency of liquid leaks is low, 9Cutpoints are emissions levels that are used in an I/M program to determine whether a vehicle passes or fails.

OCR for page 17
36 Evaluating Vehicle Emissions I/M Programs HI / 2.4% (I 6.1% \~/ Cam - 0.2%~ 78.0% 40.2% - \NO 8.3% / FIGURE 1-8 Degree of overlap among the highest 10% of emitters of CO, HC, and NOX in the LDV fleet. This figure shows the number of vehicles in each category. The figure is based on results of ASM 2525 emissions tests (controlled load) administered on 12,977 vehicles in California random roadside inspections tested June 9, 1998, until October 29, 1999. Sizes of the smaller overlapping areas are not drawn to scale. Of the vehicles tested, 78% did not fall in the top 10% for CO, HC, or NOx Source: Diagram prepared by Gregory S. Noblet, University of California, Berkeley. the gasoline lost becomes a significant contributor to the HC emissions inven- tory. A recent Coordinating Research Council/American Petroleum Institute leak study (McClement et al. ~ 997) examined ~ ,000 vehicles 500 in Arizona and 500 in Ohio curing the fall of 1997. Halfthe vehicles selected were 1991 end order; the other halfwere 1992 end newer. Vehicles were recruited from I/M lanes and given a physical inspection next to the I/M lane. Twenty-two significant leaks were found in the older sample (4.4%~. No significant leaks were found in the newer vehicles. Significant leaks were defined as expected to have "immediate, measurable reduction in whole gasoline emissions if re- paired." Another study recruited ~ 5 1 vehicles ~ ~ 971 -1 99 ~ model year) from an Arizona I/M lane during the summer of 1996 and measured the 24-hour diurnal emissions (Haskew and Liberty 1999~. This study included only vehi- cles from model-year 1991 and older. Liquid leaks were identified in 32 ofthe vehicles tested (21%~; 5 ofthem had significantleaks of greater then 50 g/day. The skewness of non-tailpipe emissions is also described by Pierson et al. (1999~.

OCR for page 17
Motor Vehicle Emissions and Regulation 3 7 100 - v, ._ 80 ~ v: cat .s v' can can o TO Ox 40 20 O 0 20 40 60 80 100 % of Fleet Ranked Dirty to Clean % of Total FIGURE 1-9 Aggregated excess FTP CO, HC NOX emissions from the California I/M pilot study rank-ordered from Lightest to lowest emitters. Excess emissions for each pollutant are aggregated using the equation (1/7(CO) + NOx + HC). OVERVIEW OF VEHICLE I/M PROGRAMS Over the past 30 years, a range of I/M programs have been created to reduce vehicle emissions. Vehicle I/M programs were identified as an option for improving air quality in the ~ 970 Clean Air Act, and the first T/M program was implemented in New Jersey in ~ 974. In that program, exhaust emissions at idle conditions were tested for light-duty, gasoline-powered vehicles manu- factured dunog or after 1968. The 1977 Clean Air Act Amendments man- dated the use of I/M for areas with Tong-ter~n air-quaTity problems, and, in law- ~ 80 . l can c' 40 . x ~ 20- ; lo _ ~ _ . . . : . 7 _ . . ~ ~ ~ _ 10 20 30 40 50 60 7C, % of neet Ranked Dirty to Clean ~ ~ of CO % of HC - % of NOx FIGURE 1-10 Excess emissions based on the California roadside pullover study.

OCR for page 17
38 Evaluating Vehicle Emissions I/M Programs ~ 978, EPA issued guidance for I/M programs, including minimum emissions reduction requirements, administrative requirements, and implementation schedules. This guidance was broad, and, consequently, a variety of state programs emerged. The CAAA90 was much more prescriptive about I/M, and EPA was required to develop enforceable guidance for "basic" and "enhanced" I/M programs. Basic programs were required for areas in moderate nonattainment of NAAQS, and enhanced programs were required for serious, severe, and extreme nonattainment regions. is The CAAA90 further mandated that en- hanced T/M programs be annual (unless biennial programs were proven to be equally effective), centralized (unless decentralized was shown to be equally effective, and enforced through registration denial (unIess a preexisting enforcement mechanism was shown to be more effective). The CAAA90 also required the use of on-road testing and the biennial assessment of I/M program effectiveness. However, the assessment requirement has not been enforced. In ~ 992, EPA published its rule (EPA ~ 992a) requiring enhanced I/M programs in response to the CAAA90. The rule required specific tailpipe, evaporative, and visual inspections of vehicles, including use of the IM240 emissions test for model-years ~ 986 and newer. The IM240 test, which is described in Chapter 3, is a 240-second test simulating actual driving with the engine in gear. The ~ 992 T/M rule also specified guidelines that required en- hanced EM programs to collect TM240 emissions tests on a random sample of 0. ~ % of the fleet. In addition, programs were supposed to perform on-road testingofan additional 0.5% ofthe fleet using either remote sensing orroad- side pull-overs. It should be noted that EPA never described how these data would be used, and the agency has not enforced these requirements. Implementation ofthe model I/M program was projected to achieve reduc- tions of 28% in HC, 3 ~ % in CO, and 9% in NOX for enhanced I/M regions, compared with emissions in the absence of an I/M program, by the year 2000 iNonattainment areas are areas violating federal air-quality standards for the criteriapollutants: sulfur dioxide, particulate matter, nitrogen dioxide, carbon monoxide, ozone, and lead. VIA centralized network consists of a relatively small number of stations that perform emissions tests only. A decentralized testing network consists of a larger number of low-volume stations that do both emissions testing and vehicle repairs. These networks are described in more detail in Chapter 3.

OCR for page 17
Motor Vehicle Emissions arid Regulation 39 (EPA 2000b). Basic [/M programs were projected to give emissions reduc- tions of 5% in HC and ~ 6% in CO (EPA 2000b). These estimates were made based on emissions simulations for average national conditions using EPA's Mobile Source Emissions Factor (MOBILE) mode. Chapter 5 contains a discussion of the use of MOBILE for estimating I/M emissions reductions. The enhanced I/M rule included a 50% credit discount for I/M programs with decentralized vehicle testing, based on EPA's observations ofthe degree of improper testing found in such programs (EPA ~ 993c). This discount was incorporated into the ~ 992 rule and addressed the implicit requirement in the CAAA90 that EPA distinguish between the relative effectiveness of central- ized versus decentralized programs. The discount for decentralized programs evolved into a major area of contention between EPA and states, including California (IMRC 1995a,b). The National Highway Systems Designation Act of ~ 995 included a provi- sion allowing decentralized I/M programs to claim ~ 00/O of the emissions- reduction credits afforded a similar centralized program. States were required to provide a good faith estimate of program effectiveness, which was to be substantiated with an evaluation using program data ~ ~ months after program approval. This ~ 8-month demonstration is based on criteria developed by the Environmental Council of States and is separate from the biennial evaluation requirement. Recent regulations have provided for more flexibility in I/M program de- sign and evaluation from those set out by the CAAA90 and the ~ 992 rule. EPA ~ ~ 99Sc) removed the requirement that evaluation be based on IM240 or other mass emissions transient test data. It called for evaluation to be based on "sound" methodologies; some of which have been discussed in further guidance memoranda (EPA ~ 998d,2000b). EPA ~ ~ 999c) proposed rule revi- sions to the motor vehicle T/M program requirements to incorporate recent policy decisions and statutory requirements. This proposed rule would provide states additional flexibility to tailor their I/M programs to better meet current and fixture needs. Among these is the need to maximize program efi iciency i2The MOBILE model is used to estimate vehicle emissions and the effectiveness of control strategies such as I/M, reformulated fuels, and emissions standards. The MOBILE estimates are critical as they quantify the emissions-reduction benefits that a state can claim for their I/M program. The benefit estimated in the 1992 rule were made using version 4.1 of MOBILE using 1992 national default assumptions for vehicle fleet characteristics and other factors.

OCR for page 17
40 Evaluating Vehicle Emissions I/M Programs . and customer convenience by capitalizing on newer vehicle testing options, such as OBD system testing and remote sensing. EPA (2000c) recently released a notice of proposed rule-making concerning the use of OBD in I/M. It proposes to provide states the flexibility to replace traditional I/M tests with OBD checks for cars equipped with OBDTI (1996 model year and newer). This rule was finalized April 5, 2001 (EPA 2001~. More than 30 states now operate I/M programs. Table I-4 summarizes some of these programs. Each program has distinct rules, test types, and frequencies of operation. Program effectiveness data, collected in response to the mandate for biennial evaluations, are beginning to emerge. Chapter 3 discusses a selection of state-sponsored and independent evaluations. As data on program effectiveness have become available, comparisons have been made between the emissions reductions initially projected for I/M programs and the emissions reductions suggested by the program evaluation data. These comparisons are critically important, given the significant role of I/M programs in developing state implementation plans (SIPs).~3 Moreover, evidence thus far suggests that actual emissions reductions attributable to I/M programs are considerably less, at least for exhaust emissions, than those credited to states on the basis of modeling using EPA's mobile-source emissions model MO- BlI-E. }4 . EVOLVING ISSUES AFFECTING I/M IN THE FUTURE Surrounding all these issues is yet another consideration involving the nature of current testing protocols and improvements in late-mode] vehicle emissions-controltechnolog~es. For thelate-modelvehiclefleet,current testing programs are not inspection and maintenance programs but rather inspection and repair programs. The distinction is substantive and points out a very ~ m p o r t a n t t e c h n o ~ o g ~ c a ~ ~ e v e ~ o p m e n t . W i t h r e g a r ~ t o e m i s s i o n s c h a r a c t e r i s t i c s , the emissions controls in current cars (including those dating back to the intro- duction of computer-controlled filet injection and emissions controlmost of subregions in nonattainment of NAAQS must develop a SIP detailing how they will come into compliance. Included in SIPs are estimates of the emissions benefits from an array of control programs, including I/M programs. ~4Emissions-reduction credits in SIPs are developed from modeling using MOBILE, not measurements. In California, the EMFAC model is used for SIP devel- opment.

OCR for page 17
41 . au O ~ ~ m ~ `= ~ ~ C;, ._ V, o ~4 a' C) V: o cn o . ~ Ct Ct For 1 Ct _ o CD Ct Ct U. o En' Em ~ ~ ~ ? 5 ~ X O t ~ ~ ~ 0 ~ O at O I, '_' ~ ~ ~ O ~ ~ ~ ~ of of ~ ~ ON ~ _ _1 ~ t_ _ Cal Ct - o C) ._ C' I. .= At Ct .> i_ Ct o Ct V] ~ En ct Y ;> V E~ C~= ~ ~ ~ ~ ~ =g ~ 00 ~ + + ~ + X + X + X X ~ c~~ ,= =~ a~ ~ GN ~ C7N ~ ~ ~ V ~ V ~ V ~ V ~ V V ~ ~ ~ ~ ~ ~ V C: _ ~ SC=~= o5= _ _ _ _ _ Ct _ Ct C~ Ct Ct _ .g ~ g g g g m <: m m m ~ m v ~O ~O ON ~v O~ ~v o a~ c~D ~ ~ ~ C) ct ~ ~ U~ C~:l <~s (t Ce Z ~ ~ ~ ~ C: C: _ h ~ ~U ~ =- ~ ~ ~ ~ ~ _ E C~ .= ~ ~ ~ ~ o o _ o ~ ~ ~ ~ _ oo ~ o ~ ~ o V CN CM ~ ~ CM ~q .= X ~ V ~ ~ V ~ ~ ~ ~ ~ V o o ~= ~ o =- U ~o ~ ~ ~ O 5 ~ E .m E o c

OCR for page 17
42 x q Fin En _ ~ .o ~ ~ ~ Ct ~ on m ~ C`: . ~ ~ ~ V) Cal Cal o ~ ~ ~ =.e Ct o Ce U] Cal Cal U. _ ~ ~ V Cal . o v o Z ~ O O ~ > 0 0 0 ~ ~ ~ _ = 00 01) O O ~ OXx + +X v ~ ~ v >~> >- >- vVv VV VV ~ Ct -3v ~ 'e cd .~ m m C) ~ V, C) cn ~ ~> ct C: ~ ~ V Q~ ~ eC ~ =~ ~ =- ce ~ _ _ O O O O ~ ~ ~ a., ~ V + ~ V' 0 C) Z ~ ~c-4 ~ O O ~V ~ ~ ;~ O C\S ~ ^= 0 =- ~ ^= O O =.~ (~ v V ~ ~ ~ a v Ct o ; - 00 ._ au Ct ._ OX Ct V V O C~ U. a ._ Ct ~q t~O C) au ._ o v ._ v ~0 Ct U U. ~ a., ._ O ~ V ~ Ct _4 o U. ct cn E~ o e ~s e o o - ~ u, c e u, s~ ~ o ct ~ c': X 3 Ct: ~ o o N .c O ~ C<: 3 3 :> o Q Ct ~ o U, . ~ <~, O ~ C~ _~ U . _ ~ ~V . - - Cd O O V: - Ct - .=

OCR for page 17
Motor Vehicle Emissions and Regulation 43 the fleet) are relatively maintenance-free. If a late-model car has excessive emissions, it is often the result of a system component failure. The component would have to be replaced, as opposed to undergoing maintenance in the tradi- tional sense of carburetor or other engine function adjustments. An increase in vehicle durability, including durability of emissions-control components, has accompanied these and other technological improvements. According to Davis (2000) the average age of in-use passenger cars has in- creased from a mean age of 5.6 years in ~ 970 to 8.8 years in ~ 998. Addition- ally, the average lifetime of a 1990 model-year passenger car is 2.7 years longer (14.0 years) than that of a ~ 970 model-year car. These trends have resulted in a large change in the percentage of older vehicles in the fleet. In ~ 970, the percentage of vehicles ~ 5 years and older was only 2.9%; in ~ 998, the percentage had risen to 13.2%. These changes in the nature of vehicle technology, durability, and vehicle lifetimes have implications for future I/M programs. The increased durability and lack of need for periodic maintenance in the sense of engine "tuning" should allow for reducing the testing burden through an increased use of clean screening of vehicles or increases in model-year exemptions. is Technological innovations in OBD systems might greatly speed up the inspection process, and eventually make the remote monitoring and reporting of vehicle emissions a reality. Indications are that the new technology vehicles are cleaner, and capa- ble of remaining cleaner for a longer period of time. The possible need for high-cost repairs towards the end of vehicle life remains. Older vehicles prob- ably will still tend to be owned by people in Tower-income groups who are least able to afford repairs. Thus, behavioral and economic issues might continue to play important roles in maintaining Tow emissions throughout vehicle life- times. Emerging air-quaTity issues also have implications for the future of emis- signs testing programs. For example, the South Coast Air Quality Manage- ment District (2000) recently reported that mobile-source emissions are the most significant contributor to human exposure to air toxic s. Increased under- standing ofthe effects of air tonics end PM, as well as the implementation of stricter standards for ozone and PM, might place new demands on future vehicle I/M programs. i5Clean screening is a method for exempting vehicles from regularly scheduled inspections through low-emitter profiles or remote sensing. Both are discussed in Chapter 4.

OCR for page 17
44 Evaluating Vehicle Emissions I/M Programs SUMMARY Chapter ~ ofthis reportbegan with a statement ofthe committee's charge, how the charge originated, and the committee's response to the charge. It then describes the air pollutants associated with motor vehicle use and characteris- tics ofthese pollutant emissions. This includes a categorization of emissions by vehicle type, exhaust standards, evaporative standards, emissions distribu- tions, a brief overview of existing T/M programs, and an assessment of future I/M issues. Motor vehicles represent a significant fraction of overall emissions, espe- cially in urban areas, and a relatively small fraction of on-road vehicles are responsible for a large fraction ofthe emissions. Typical numbers reported in the literature (usually obtained from measurements of in-use vehicles) suggest that for any given pollutant, 50-60% of LDV exhaust emissions are produced by about ~ 0% of the highest-polluting LDVs. The skewness of excess emis- sions is even greater, with 5% of vehicles producing 75% or greater of excess emissions. Identifying and repairing high-emitting vehicles clearly has the potential to reduce mobile-source emissions. Vehicle I/M programs provide the primary method for obtaining these reductions. Despite their widespread use in air-quality management, a number of concerns are associated with I/M programs. As data on program effective- ness have become available, comparisons have been made between the emis- sions reductions initially proj ected with MOB lLE and those estimated by pro- gram evaluation data. These comparisons are critically important given I/M programs' significant role in SIPs. Evidence thus far suggests that actual emissions reductions attributable to I/M programs are considerably less than those credited to states on the basis of simulations using MOBILE (emissions- reduction credits in SIPs are developed from simulations using MOBILE, not measurements). This evidence has raised questions about the effectiveness of I/M as a strategy for improving air quality. Additionally, the proposed rule for implementing the enhanced I/M program also created controversy by mandating the use ofthe TM240 test at centralized facilities and discounting by 50/O the emissions-reduction benefits for programs that relied on decentralized tests. These issues were at the forefront during a 1995 hearing ofthe House Subcommittee on Oversight and Investigation on the effectiveness of vehicle I/M programs and how the MOBILE model credits these programs (U.S. Congress ~ 995~. They also prompted Congress to request the NRC study reported here.

OCR for page 17
Motor Vehicle Emissions and Regulation 45 An important consideration for the late model vehicle fleet discussed in this chapter is that current testing programs are not I/M programs but rather in- spection end repair programs. With regard to emissions characteristics, cur- rent cars (including those dating back to the introduction of computer-con- trolled fuel injection and emissions control, as discussed in Chapter 2) are essentially maintenance-free. If a late-mode! car has excessive emissions, then a system component has failed and must be replaced. Maintenance in the traditional sense of carburetor or other engine injunction adjustment, for exam- ple, no longer applies. Changes in the nature of vehicle technology, durability, and lifetimes have serious implications for future I/M programs. These implications are discussed throughout this report and are a focus of the second phase of this study. In- creased durability and the lack of need for periodic maintenance should reduce the testing burden throughincreased use of clean screening of vehicles. Tech- nological innovations in OBD systems might speed up the inspection process and perhaps eventually make the remote monitoring or reporting of vehicle emissions characteristics a reality. New technology vehicles are cleaner and capable of remaining cleaner for a longer period oftime; however, the technol- ogy is so new that it remains to be seen what their emissions and repair re- quirements will be at the end of their useful lives.