The Effectiveness and Cost of Retrofit For Reducing Automobile Emissions

EPA has recently evaluated several automobile retrofit devices that are potentially applicable to pre-1975 vehicles. The results of this evaluation are described and used to estimate the effectiveness and cost of retrofit for reducing total automobile emissions in the period 1975-1985. It is estimated that retrofit combined with inspection/maintenance can potentially achieve reductions in automobile emissions of 33% to 60% in 1975, depending on pollutant, and 10% to 20% in 1985. The estimated present value discounted to 1972 of the average cost per retrofit vehicle is $30 to $152 over the period 1975-1985 depending on the retrofit system used. The corresponding annualized cost is $6 to $28 per vehicle.     

A Multipurpose Flare Stack for Control of Chemical Process Wastes

Abstract

To test the effectiveness of California’s vehicle inspection/ maintenance (I/M) program, exclusive of vehicle-owner intervention, a fleet of more than 1,100 vehicles that previously had failed California’s Smog Check test were sent to randomly selected Smog Check stations in the Los Angeles area for covert inspections and repairs. The two-speed idle test was used for repairs. For those vehicles that were repaired at the first inspection, their FTP emission reductions were 25%, 14%, and 11% for hydrocarbons (HC), carbon monoxide (CO), and nitrogen oxides (NO_x), respectively, although emissions testing for NO_x was not performed at the Smog Check stations. Idle HC and CO emissions increased for 35% and 43% of the vehicles, respectively, after repairs. This data set shows that most vehicles that fail the Smog Check inspection are only marginal emitters, with 61% and 44% of the total potential for HC and CO emission reductions, respectively, coming from only 10% of the vehicles that currently fail the inspection. When the vehicles were rank-ordered by idle emissions from dirtiest to cleanest, emission reduction costs for the highest-emitting 10% of the fleet averaged $l,100/ton and $250/ton for HC and CO, respectively, attributing all the costs to each pollutant exclusively. For the remaining vehicles, costs increased dramatically.     

Speciated hydrocarbon profiles and calculated reactivities of exhaust and evaporative emissions from 82 in-use light-duty Australian vehicles

Mass emissions of non-methane hydrocarbon (NMHC) from 26 pre-1986 and 56 post-1985 catalyst-equipped in-service vehicles were determined from measurements made on a chassis dynamometer using an urban drive cycle. Evaporative emissions were measured on a subset (4 pre-1986 and 8 post-1985) of these vehicles. Average ADR emissions (mg/km) of the individual HCs from the older pre-1986 vehicles were generally 4–7 times the emissions from newer catalyst-equipped vehicles. Evaporative emissions from the older vehicles are also much higher than those of newer vehicles. Exhaust from newer catalyst-equipped vehicles had lower proportions of substituted aromatics and alkenes and higher proportions of lower molecular weight alkanes. The effect of fuel type on the exhaust emissions was also investigated by refuelling 9 of the pre-1986 vehicles with both unleaded and leaded petrol. A 20–40% reduction in HC mass emissions was observed when unleaded petrol was used instead of leaded petrol. Reactivities of the emissions and the contributions from different classes of compounds are also reported. The specific reactivity of the exhaust emissions from newer vehicles was lower than that for older vehicles owing to the smaller proportions of highly reactive alkenes and substituted aromatic species. Moreover, as older vehicles have higher average mass emissions, when considered on a per-km basis, the pre-1986 vehicles have a greater ozone-forming potential than post-1985 vehicles. The specific reactivities of the NMHC (gO₃/gNMHC) of both the heat build and hot soak evaporative emissions were much lower than the exhaust emissions.     

Expected Decline in Carboxyhemogldbin Levels As Related to Automobile Carbon Monoxide Emission Standards

Carboxyhemoglobin (COHb) levels in blood are principally due to inhalation of carbon monoxide, although a low level (approximately 0.3%) of COHb is endogenous. A carboxyhemoglobin level above 1.5 % in non-smokers indicates exposure to CO in excess of the 10 mg/m³ air quality standard established under the Clean Air Act Amendments of 1970.

In most major U.S. cities, automobile emissions constitute the principal source of CO; in Chicago, according to EPA estimates,¹ light duty vehicles are responsible for 69.3% of all CO emissions. Thus as new automobiles incorporating emission controls enter the automotive fleet and older, emission-uncontrolled automobiles are phased out, ambient CO concentrations should decline and corresponding reductions in blood carboxyhemoglobin levels of nonsmokers can be expected.     

An Investigation of Idle Emissions Inspection and Maintenance at Altitude

Idle emissions inspection and maintenance was evaluated using a sample of 300 privately owned 1964 through 1973 model-year vehicles operating in the Denver metropolitan area. Ten privately owned stations, licensed by the State of Colorado to perform vehicle safety inspections, were utilized to conduct idle emissions inspection and subsequent maintenance of failed vehicles. Exhaust hydrocarbon (HC) and carbon monoxide (CO) reduction as measured by the 1975 Environmental Protection Agency (EPA) mass emission testing procedures was indicated to be 13% and 8% respectively at a 50 % rejection rate. The average maintenance cost to achieve the reduction was $11.32 per failed vehicle.

The adjustment and repair procedures provided to participating garages were sufficient to achieve significant emissions reduction and training provided to garage personnel was adequate. However, several problems were experienced with station personnel relative to data transmittal and inspection pass/fail limits. Problems were also experienced with respect to correlations between laboratory and garage-type analytical instrumentation.     

The Role of the Computer in Air Pollution Control

The search for ways of reducing vehicular emissions has led to numerous investigations of the relationships between fuel composition and the pollutants discharged from automobiles. The most obvious fuel effects result from evaporation of gasoline components from the fuel tanks and carburetors of vehicles which lack effective mechanical devices (such as those required on all 1971 model cars) to control evaporative losses. Thus, several laboratories and cooperative study groups (Coordinating Research Council and American Petroleum Institute) have investigated the ways in which fuel properties (especially the amounts and types of C₄-C₅ hydrocarbons) influence both the amount and the potential atmospheric reactivity of evaporative emissions.^1–6 But fuel evaporation accounts for only a small portion of the total hydrocarbons emitted by automobiles, and gasoline modifications (such as volatility reductions) that reduce evaporative losses can lead to higher levels of hydrocarbons in automobile exhaust.^4–6     

Modeling In-Use Vehicle Emissions and the Effects of Inspection and Maintenance Programs

Different ways for modeling the impact of vehicle emission inspection and maintenance programs on fleet hydrocarbon emissions are examined. A dynamic model is developed for forecasting fleet emissions in which individual vehicle performance is modeled as a stochastic process and vehicle emissions are tracked over time. Emissions inspection and repair are incorporated into the model, allowing for the stochastic aspects of both testing and repair. This model is compared to EPA’s model for evaluating the impact of vehicle emissions inspection and maintenance. We find that the way vehicle emission equipment deterioration overtime is modeled is important for forecasting emissions from the fleet and for assessing the success of inspection and maintenance programs. For inspection programs, we find that factors such as the proportion of vehicles tested, and repair effectiveness and duration have the greatest impact on emission reductions. The ability of different emission testing regimes to identify polluting vehicles has less impact on a program’s overall potential for emissions reduction. Policy recommendations for I&M testing and predictions of emission reduction credits from these tests will depend in important ways on the methods used in the underlying emissions models.     

A Realistic Vehicle Emission Inspection System

A one minute mass vehicle exhaust emission inspection system is described to aid those seeking to establish an appropriate effort in compliance with the Air Quality Act of 1967. This proposed inspection system is suitable for both emission controlled and pre-emission controlled automobiles. It is oriented to appropriate Auto Service Industry corrective measures since it provides guidelines as to probable cause of unnecessarily high emissions for each rejected vehicle. This capability is essential to avoid abnormal corrective expense due to excessive repairs or continued high emissions due to inadequate repairs. Low skill levels may be used during the inspection process, yet results are repeatable and consistent in a plurality of inspection stations. The net gain in lower emissions through inspection of in-use vehicles is large because the point of rejection is no longer dictated by the inherently high emission vehicles. Not more than two rejection standards are required for emission controlled vehicles, and only one for pre-emission controlled vehicles. This includes all domestic and foreign makes. These standards are flexible in that they may be initially liberal and gradually tightened as circumstances warrant. Such policy changes or future spark ignition engine designs will not significantly obsolete the hardware required to perform this inspection. The fundamental techniques employed permit the use of greatly simplified equipment and instrumentation with resultant lower initial cost and increased reliability than previously envisioned meaningful inspection methods. A study of nearly 1000 automobiles over a one year period evolved and documented this proposed inspection system. Qualified persons may study this work to any depth desired by contacting the authors.     

New Jersey Repair Project: Tune-Up at Idle

Idle hydrocarbon and carbon monoxide measurements have been made on over 2500 cars at a New Jersey Inspection Station. These studies have shown that the idle test can be integrated into the present periodic motor vehicle inspection system with a minimum cost, testing time, and ease of operation.

Instrumentation at a low cost has recently become available, test procedures have been developed and potential emission reductions have been demonstrated for idle testing. High emissions indicate a car malfunction and the need for a tune-up. Effective low cost tune-ups can be made with exhaust instrumentation and garage training.

In the New Jersey REPAIR Project, preliminary idle cut-off levels were selected at 6% carbon monoxide and 1000 ppm hydrocarbon for pre-68 cars, 4% and 500 ppm for 1968–69 cars, and 3% and 300 ppm for later years. Volunteered vehicles which exceeded these levels were further tested at the New Jersey laboratory. Federal hot cycles, ACID mass cycles, Key Mode, and Idle tests were conducted before and after maintenance.

At idle, uncontrolled pre-1968 vehicles had an average reduction from 8.2 to 3.3% carbon monoxide and 2153 to 459 ppm hydrocarbons as hexane. Average mass reductions from the ACID-cycle were 45 g/mi CO and 6.3 g/mi hydrocarbons. Carbon monoxide idle reductions obtained for emission controlled 1968, 1969, and 1970 cars were about equal to those obtained for the pre-emission controlled vehicles, but hydrocarbon reductions were lower. Reductions obtained in federal hot cycles were from 4.1 to 2.1% CO and 1418 to 580 ppm hydrocarbons for pre-1968 cars, and 2.6 to 0.7% and 502 to 308 ppm for 1968–1969 cars.

Idle adjustments lower emissions in the idle, deceleration, and cruise modes up to 30 mph, thus urban driving areas should show the greatest reduction. Total motor vehicle emission reduction in New Jersey would be about 920,000 ton/yr of CO and 101,000 ton/yr of hydrocarbon; a 20 and 32% reduction.     

Long-Term Efficiency of Catalytic Converters Operating in Mexico City

ABSTRACT

A 1999 ordinance by the Government of Mexico City bans 1993 model-year vehicles from on-road operation if their catalytic converters are not replaced with new ones. To validate the benefits of this action, we examined three issues related to exhaust emissions of vehicles equipped with catalytic converters. After selecting representative fleets of in-use vehicles, a comparison between emissions and catalyst efficiency in cars with two categories of exhaust emission limits was carried out. For that purpose, two fleets were selected, each made up of 10 vehicles run under similar conditions. A third, larger fleet with emissions control systems was used to evaluate and simulate real-world conditions of vehicles in a controlled laboratory. Finally, the aging effect on the catalytic converter was studied on vehicles run for 100,000 km, replacing their old emission control devices for new ones.

The 1991-1992 model-year vehicles showed a high percentage of compliance with the corresponding emissions standard (90%) in comparison with 1993 model-year and later vehicles (Tier 0). However, NO_x emissions were higher for the newer vehicles. Fifty percent of the 1991-1992 model-year vehicles evaluated under the official inspection/maintenance (I/M) procedure did not meet the regulated emissions standard when the results were compared with those of the U.S. Federal Test Procedure     

Costs, emissions reductions, and vehicle repair: evidence from Arizona

The Arizona inspection and maintenance (I/M) program provides one of the first opportunities to examine the costs and effectiveness of vehicle emission repair. This paper examines various aspects of emission reductions, fuel economy improvements, and repair costs, drawing data from over 80,000 vehicles that failed the I/M test in Arizona between 1995 and the first half of 1996. We summarize the wealth of data on repair from the Arizona program and highlight its limitations. Because missing or incomplete cost information has been a serious shortcoming for the evaluation of I/M programs, we develop a method for estimating repair costs when they are not reported. We find surprising evidence that almost one quarter of all vehicles that take the I/M test are never observed to pass the test. Using a statistical analysis, we provide some information about the differences between the vehicles that pass and those that do not. Older, more polluting vehicles are much more likely never to pass the I/M test, and their expected repair costs are much higher than those for newer cars. This paper summarizes the evidence on costs and emission reductions in the Arizona program, comparing costs and emissions reductions between cars and trucks. Finally, we examine the potential for more cost-effective repair, first through an analysis of tightening I/M cut points and then by calculating the cost savings of achieving different emission reduction goals when the most cost-effective repairs are made first.     

Long-term efficiency of catalytic converters operating in Mexico City

A 1999 ordinance by the Government of Mexico City bans 1993 model-year vehicles from on-road operation if their catalytic converters are not replaced with new ones. To validate the benefits of this action, we examined three issues related to exhaust emissions of vehicles equipped with catalytic converters. After selecting representative fleets of in-use vehicles, a comparison between emissions and catalyst efficiency in cars with two categories of exhaust emission limits was carried out. For that purpose, two fleets were selected, each made up of 10 vehicles run under similar conditions. A third, larger fleet with emissions control systems was used to evaluate and simulate real-world conditions of vehicles in a controlled laboratory. Finally, the aging effect on the catalytic converter was studied on vehicles run for 100,000 km, replacing their old emission control devices for new ones. The 1991-1992 model-year vehicles showed a high percentage of compliance with the corresponding emissions standard (90%) in comparison with 1993 model-year and later vehicles (Tier 0). However, NOx emissions were higher for the newer vehicles. Fifty percent of the 1991-1992 model-year vehicles evaluated under the official inspection/maintenance (I/M) procedure did not meet the regulated emissions standard when the results were compared with those of the U.S. Federal Test Procedure FTP-75. Our results suggest that the replacement of old catalytic converters with new ones will have little effect on decreasing polluting emissions because these vehicles were in very bad mechanical condition. Results of catalytic activity as a function of mileage indicated inefficient catalyst performance for the fleets tested. All pollutant conversions were below 90% efficiency, and they deteriorated by an average of 30% after the vehicles were run for 100,000 km.     

Costs,Emissions Reductions,and Vehicle Repair: Evidence from Arizona

ABSTRACT

The Arizona inspection and maintenance (I/M) program provides one of the first opportunities to examine the costs and effectiveness of vehicle emission repair. This paper examines various aspects of emission reductions, fuel economy improvements, and repair costs, drawing data from over 80,000 vehicles that failed the I/M test in Arizona between 1995 and the first half of 1996. We summarize the wealth of data on repair from the Arizona program and highlight its limitations. Because missing or incomplete cost information has been a serious shortcoming for the evaluation of I/M programs, we develop a method for estimating repair costs when they are not reported. We find surprising evidence that almost one quarter of all vehicles that take the I/M test are never observed to pass the test. Using a statistical analysis, we provide some information about the differences between the vehicles that pass and those that do not. Older, more polluting vehicles are much more likely never to pass the I/M test, and their expected repair costs are much higher than those for newer cars.

This paper summarizes the evidence on costs and emission reductions in the Arizona program, comparing costs and emissions reductions between cars and trucks. Finally, we examine the potential for more cost-effective repair, first through an analysis of tightening I/M cut points and then by calculating the cost savings of achieving different emission reduction goals when the most cost-effective repairs are made first.     

Emissions from lit-use Motor Vehicles in Los Angeles: A Pilot Study of Remote Sensing and the Inspection and Maintenance Program

As part of a major field study to understand the causes of persistent, elevated carbon monoxide pollution episodes in Los Angeles, we performed a project to understand the emissions of vehicles in use. In this experiment, we assessed the accuracy of a remote sensing instrument designed to measure CO concentrations from vehicles as they were driven on the road. The remote sensor was shown to be accurate within ten percent of the directly measured tailpipe value. We performed a roadside inspection on 60 vehicles and demonstrated that the remote sensor could be used as an effective surveillance tool to identify high CO-emitting vehicles. We also compared the roadside data set to the biennial Smog Check (I/M) tests for the same vehicles, and observed that carbon monoxide and exhaust hydrocarbons from high emitters were much higher than when the vehicles received their routine inspection. Furthermore, for the high-emitting vehicles in this data set, the length of time since the biennial Smog Check had little influence on the cars’ emissions in the roadside inspection.     

Influence of Oxygenated Fuels on the Emissions from Three Pre-1985 Light-Duty Passenger Vehicles

Tailpipe and evaporative emissions from three pre-1985 passenger motor vehicles operating on an oxygenated blend fuel and on a nonoxygenated base fuel were characterized. Emission data were collected for vehicles operating over the Federal Test Procedure at 40,75, and 90°F to simulate ambient driving conditions. The two fuels tested were a commercial summer grade regular gasoline (the nonoxygenated base fuel) and an oxygenated fuel containing 9.5 percent methyl tert-butyl ether (MTBE), more olefins, and fewer aromatics than the base fuel. The emissions measured were total hydrocarbons (THCs), speciated hydrocarbons, speciated aldehydes, carbon monoxide (CO), oxides of nitrogen (NO_x), benzene, and 1,3-butadiene.

This study showed no pattern of tailpipe regulated emission reduction when oxygenated fuel was used. Tailpipe emissions from the 1984 Buick Century without a catalyst and the 1977 Mustang with catalyst decreased with the MTBE fuel. However, emissions from the 1984 Buick Century and the 1980 Chevrolet Citation, both fitted with catalysts increased. The vehicles emitted more 1,3- butadiene and, in general, more NO_x when operated with the base fuel.

THC, CO, benzene, and 1,3-butadiene emissions from both fuels and all vehicles, in general, decreased with increasing test temperature, whereas NO_x emissions, in general, increased with increasing test temperature. Formaldehyde, acetaldehyde, and total aldehydes also showed a decrease in emissions as test temperature increased. More formaldehyde was emitted when the MTBE fuel was used.

Evaporative, diurnal, and hot soak emissions from the base fuel were greater than those from the MTBE fuel. The evaporated emissions from both fuels increased with increasing test temperatures. Diurnal data indicate that canister conditioning (bringing the evaporative charcoal canister to equilibrium) is required before testing.     

The effects of repairs on tailpipe emissions for on-board diagnostics II-equipped vehicles with the malfunction indicator light illuminated

A total of 77 On-Board Diagnostics II (OBDII)-equipped vehicles with illuminated malfunction indicator lights (MILs) and non-evaporative codes were tested before and after repair. The test cycles included the Federal Test Procedure (FTP), IM240, and steady-state cycles. A total of 17 vehicles were found with emissions greater than 1.5 times their respective FTP emissions standards. Repair of these vehicles resulted in dramatic reductions in overall emissions for all the cycles. A majority of the remaining vehicles were found to have emissions below the certification standard for the FTP both before and after repair. Repairs for the vehicles with emissions <1.5 times the standard resulted in some smaller but quantifiable emission reductions over the FTP and IM240 but larger reductions over the steady-state driving tests. Misfires, bad oxygen sensors, and exhaust gas recirculation (EGR) problems were the most common non-evaporative causes for triggering the MIL. The results show some fundamental differences between identifying malfunctioning vehicles using OBDII as opposed to more traditional dynamometer tests. In particular, for many systems, OBDII identifies components that are operating outside their design specification rather than for a specific emissions threshold.     

The Effects of Repairs on Tailpipe Emissions for On-Board Diagnostics II-Equipped Vehicles with the Malfunction Indicator Light Illuminated

Abstract

A total of 77 On-Board Diagnostics II (OBDII)-equipped vehicles with illuminated malfunction indicator lights (MILs) and non-evaporative codes were tested before and after repair. The test cycles included the Federal Test Procedure (FTP), IM240, and steady-state cycles. A total of 17 vehicles were found with emissions greater than 1.5 times their respective FTP emissions standards. Repair of these vehicles resulted in dramatic reductions in overall emissions for all the cycles. A majority of the remaining vehicles were found to have emissions below the certification standard for the FTP both before and after repair. Repairs for the vehicles with emissions <1.5 times the standard resulted in some smaller but quantifiable emission reductions over the FTP and IM240 but larger reductions over the steady-state driving tests. Misfires, bad oxygen sensors, and exhaust gas recirculation (EGR) problems were the most common non-evaporative causes for triggering the MIL. The results show some fundamental differences between identifying malfunctioning vehicles using OBDII as opposed to more traditional dynamometer tests. In particular, for many systems, OBDII identifies components that are operating outside their design specification rather than for a specific emissions threshold.     

Estimated Validity and Reliability of On-Board Diagnostics for Older Vehicles: Comparison with Remote Sensing Observations

ABSTRACT

Based on requirements under the Clean Air Act Amendments of 1990, most state vehicle inspection and maintenance (I/M) programs have, since 2002, replaced the tailpipe emission testing with the on-board diagnostic (OBD) II testing for 1996 model and newer vehicles. This test relies on the OBD II system to give the pass or fail result, depending on certain conditions that might cause the vehicle to emit pollution 1.5 times higher than the regulated standard. The OBD II system is a computer and sensors installed in the vehicle to monitor the emission control units and signal if there is any malfunction. As a vehicle ages, its engine, pollution control units, and OBD II system deteriorate. Because the OBD II system's durability directly influences the test outcome, it is important to examine the fleetwide trend in the OBD II test results in comparison with an alternative measure of identifying high emitting vehicles. This study investigates whether the validity and reliability of the OBD II test is related to the age of the OBD II system installed in the fleet. Using Atlanta's I/M testing records and remote sensing device (RSD) data collected during 2002–2005, this research establishes the convergent validity and interobserver reliability criteria for the OBD II test based on on-road emissions measured by RSDs. The study results show that older vehicles exhibit significantly lower RSD–OBD II outcome agreement than newer vehicles. This suggests that the validity and reliability of the OBD II test may decline in the older vehicle fleets. Explanations and possible confounding factors for these findings are discussed.

IMPLICATIONS This research demonstrates the potential worsening validity and reliability of the on-board diagnostic (OBD) II test in old vehicles. If the main source of low validity and reliability comes from the OBD II system malfunction, we expect this malfunctioning OBD II fleet will continue to grow in the future. If unchecked, the deterioration of OBD II system may impair the effort of the inspection and maintenance (I/M) program to identify high-emitting vehicles and the ultimate objective of reducing the air pollution from automobiles. This result is especially important in a regulatory context where technological and emissions standards dominate environmental policy and yet little attention is paid to the possible degradation of environmental monitors themselves.     

Review of light-duty diesel and heavy-duty diesel gasoline inspection programs

Emissions from diesel vehicles and gas-powered heavy-duty vehicles are becoming a new focus of many inspection and maintenance (I/M) programs. Diesel particulate matter (PM) is increasingly becoming more recognized as an important health concern, while at the same time, the public awareness of diesel PM emissions because of their visibility have combined to increase the focus on diesel emissions in the United States. This has resulted in an increased interest by some states in including heavy-duty vehicle testing in their I/M program. This paper provides an overview of existing I/M programs focused on testing light-duty diesel vehicles, heavy-duty diesel vehicles, and heavy-duty gasoline vehicles (HDGVs). Information on 39 I/M programs in 27 different states in the United States plus 9 international inspection programs is included. Information on the status of diesel emissions technology and current test procedures is also presented. The goal is to provide useful information for air quality managers as they work to decide whether such I/M programs would be worth pursuing in their respective areas and in evaluating the emissions measurement technology to be used in the program. Testing of HDGVs is generally limited to idle testing, because dynamometer testing of these vehicles is not practical, and most were not certified on a chassis basis. Testing of diesel vehicles has mostly been limited to SAE J1667 "snap-idle" opacity testing. Cost-effective technology for measuring diesel emissions currently does not exist, and, therefore, opacity-type measurements, although not effective at reducing the pollutants of most significant health concern, will continue to be used.     

High-mileage study of on-board diagnostic emissions

The 1990 Clean Air Act amendments require the U.S. Environmental Protection Agency (EPA) to set guidelines for states to follow in designing and running vehicle inspection and maintenance (I/M) programs. Included in this charge was a requirement to implement an on-board diagnostic (OBD) test for both basic and enhanced I/M programs. This paper provides the results to date of an ongoing EPA study undertaken to assess the durability of the OBD system as vehicles age and as mileage is accrued. The primary results of this effort indicate the points described below. First, the majority of high-mileage vehicles tested had emission levels within their certification limits, and their malfunction indicator light (MIL) was not illuminated, indicating that the systems are capable of working throughout the life of a vehicle. Second, OBD provides better air quality benefits than an IM240 test (using the federal test procedure [FTP] as the benchmark comparison). This statement is based on greater emissions reductions from OBD-directed repairs than reductions associated with IM240-identified repairs. In general, the benefits of repairing the OBD fails were smaller, but the aggregate benefits were greater, indicating that OBD tests find both the high-emitting and a number of marginally high-emitting vehicles without false failures that can occur with any tailpipe test. Third, vehicles that truly had high-tailpipe emissions as confirmed by laboratory IM240 and FTP testing also had illuminated MILs at a statistically significant level. Last, field data from state programs have demonstrated MIL illumination rates comparable with those seen in this work, suggesting that the vehicles sampled in this study were representative of the larger fleet. Nonetheless, it is important to continue the testing of high-mileage OBD vehicles into the foreseeable future to ensure that the systems are operating correctly as the fleet ages and as changes in emission certification levels take effect.