Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Office of Planning, Environment, & Realty (HEP)
PlanningEnvironmentReal Estate

HEP Events Guidance Publications Awards Contacts

Strategic Workplan for Air Toxics Research

1: Characterization

A. Introduction

1. Background

This workshop session focused on ways to improve vehicle emission measurements. The primary methods for measuring criteria air pollutants from motor vehicles have been laboratory/dynamometer tests, onboard measurements, and remote sensing. Some of the same techniques may be viable for air toxics emission measurement, but there are additional steps, or research efforts, needed to extend current measurement methods in order to measure air toxic compound emissions.

2. Current Information Summary

Current tailpipe and evaporative emissions testing for organics involves collecting samples in canisters and sending these canisters to a laboratory for analysis of toxic species. Because most of the organics measurements available to date measure only total hydrocarbons (HCs), emission estimates for specific HC compounds typically rely on applying species profiles to the total HC values. With species profiles being based on a limited number of samples, estimates derived using this technique are extremely uncertain. For metals and particulate emissions, samples are collected on particulate filters and analyzed. Metals analysis provides estimates of the total amount of metal emitted.

Existing references for developing motor vehicle emission values for air toxics include the Auto/Oil Air Quality Improvement Research Program (light-duty gasoline), the European Programmes on Emissions, Fuels and Engine Technologies (light-duty gasoline and diesel), Heavy-Duty Vehicle Chassis Dynamometer Testing for Emissions Inventory - Coordinating Research Council (CRC) Projects E55-E59 (heavy-duty diesel), recent studies by the Center for Environmental Research and Technology completing detailed chemical speciation of diesel exhaust (primarily heavy-duty), and the Clean Fleet Study - which mostly consists of tests of Federal Express vehicles (medium- and heavy-duty gasoline and alternative fuels).

The EPA, CRC, National Renewable Energy Laboratory (NREL), FHWA, State and Territorial Air Pollution Program Administrators/Association of Local Air Pollution Control Officials (STAPPA/ALAPCO) study is an active effort to characterize exhaust emissions from light-duty gasoline-powered vehicles (LDGVs). It is planned that 480 randomly selected LDGVs in the Kansas City area will undergo emissions testing. Study goals include determining PM emission distributions, identifying high emitter percentages, and characterizing gaseous and PM toxics exhaust from a portion of these vehicles.

Emissions measurements for the Kansas City study include dynamometer testing with gaseous air toxics analyses planned. Samples are being collected by summa canisters [for volatile organic compounds (VOCs)] and DNPH cartridges (for aldehydes or ketones). Aldehydes and ketones are two classes of oxygenate HCs. (Acetone is an example of a ketone.) HC, carbon monoxide (CO), oxides of nitrogen (NOx), and PM continuous measurements are also being obtained, in addition to the HAP samples.

In addition to dynamometer testing, researchers have been using portable emissions measurement systems (PEMS) that make onboard emission (and activity) measurements during actual driving. Current PEMS devices record measurements second-by-second for environmental conditions (ambient temperature, humidity, barometric pressure, etc.), vehicle parameters (engine revolutions per minute, vehicle speed, air conditioning on, onboard diagnostic codes, etc.), date/time, and emissions [HC, NOx, and carbon dioxide (CO2)]. PEMS primary advantages are that they capture real world driving, and they are less expensive than dynamometer testing. Disadvantages include being less accurate than dynamometer tests, and an inability to measure emissions at low concentrations.

Remote sensing of vehicle emissions works by measuring the absorption of beams of infrared or ultraviolet (UV) light by the CO, CO2, HCs, and NOx in vehicle exhaust plumes. Based on the absorption, a computer is able to calculate the ratios of CO, HC, and NOx to CO2 in the exhaust. From this information, it is possible to compute the concentrations of CO, HC, and NOx in the exhaust at the instant the vehicle passes the remote sensing device. Videos of license plates are made at the time the vehicle passes the remote sensing device, so that emission concentrations can be correlated with makes and model years of vehicles. Remote sensing technology currently does not measure air toxic compound emissions although some recent advances show promise for a subset of the air toxic compounds. Evaluations performed to date have been limited to single lane situations, and some 2-lane interstates. An advantage of remote sensing is its ability to provide a large number of emission measurements for vehicles in on-road conditions. Its main disadvantages are that it does not provide accurate measurements at very low emission levels and since the operating mode of the vehicle is generally controlled by the site selected for testing, vehicle operating modes are restricted.

B. Workshop Discussion Summary

1. Introduction

The workshop participants recommended that researchers continue to consider examining compounds that do not appear on EPA's MSAT list. This is especially true for PM. Other potential sources and compounds, besides diesel PM plus diesel exhaust organic gases, that may prove to be important in toxic PM include the following:

In designing a research plan for vehicle emissions characterization, it was specifically recommended that two works in progress be consulted before implementing any plans to collect additional emissions measurements. These include the Department of Energy (DOE)-sponsored study of diesel/gasoline vehicle contributions to ambient PM in the South Coast Air Basin and the Health Effects Institute (HEI) study of PM, which may identify a health effects marker other than EC. Researchers want to know whether EC is an adequate surrogate for diesel-emitted PM in the atmosphere. The current method is to begin by estimating how much diesel exhaust is comprised of EC. Then, how much ambient EC is contributed by diesels is estimated. Then, these two estimates are used to estimate how much diesel exhaust is in the atmosphere. Research by the California Air Resources Board (ARB) has shown that EC is not always a good marker of diesel or other mobile source exhaust. Depending on the area and season, there may be other significant or overwhelming sources such as residential wood burning.

The South Coast research study mentioned above is known as DOE's Gasoline/Diesel PM Split Study, because DOE's Office of Heavy Vehicle Technologies is the primary sponsor. The study is designed to assess the relative contribution of PM from spark ignition (SI) and compression ignition (CI) engines in California's South Coast Air Basin. In this study, 59 light-duty vehicles were emissions tested over a modified Unified Cycle, and 34 heavy-duty vehicles were tested over a variety of test cycles. In addition, ambient sampling was conducted at a variety of locations in the Los Angeles area in the summer of 2001. Data analyses are still ongoing, with results expected to appear in the peer-reviewed literature. This study is a follow-up to the 1996/97 Northern Front Range Air Quality Study (NFRAQS), which concluded that emissions from SI engines were three times those from CI engines in the Denver area in wintertime (Norton, et al., 1998). These NFRAQS study findings were counter to the findings in previous South Coast analyses, which showed diesels being much more significant PM contributors than SI engines.

The HEI study mentioned above is a diesel epidemiology project that was initiated in 1998. This is a multi-faceted research and assessment effort to (1) develop new research initiatives, including feasibility studies to identify potential new cohorts or to improve exposure assessment methods, and (2) evaluate the strengths and limitations of the published epidemiologic studies available for quantitative risk assessment. As part of this project, the expert panel reported its findings and recommendations in the 1999 HEI Special report "Diesel Emissions and Lung Cancer: Epidemiology and Quantitative Risk Assessment." Other reports and publications are listed on the reference page of this report.

In 2000, the Diesel Epidemiology Working Group was formed to continue the work of the Diesel Epidemiology Expert Panel. It was charged with reviewing reports from six feasibility studies funded by HEI, and developing an agenda for research that would provide better information for assessing quantitative risk from diesel exposure and adverse health effects, including lung cancer.

In its report, the Diesel Epidemiology Working Group (2002) concluded that full studies of cohorts that had been characterized in the feasibility studies would not generate substantially more accurate exposure-response information. The Working Group also concluded from the feasibility studies that the available methods for assessing exposure to diesel exhaust were not sufficiently specific. Working Group members also found that EC may be a useful marker for exposure to diesel exhaust if diesel engines are the dominant source of particles. The Working Group noted, however, that EC by itself lacks the necessary sensitivity and specificity to serve as a signature of diesel exhaust in ambient exposure settings, where particles typically include EC from other combustion sources. Therefore, they recommended identifying more specific markers, or a set of markers (a signature), for diesel exhaust that could be used to enhance exposure assessment for past studies, strengthen future epidemiologic studies, and assess population exposures.

2. Emission Measurements

New emission measurement programs should include size segregated PM measurements wherever possible. This means that data collection should include measurements/laboratory analysis of particle size, mass, number, shape, as well as chemical and biological characterization. (Note that methods first need to be developed to allow laboratory collection of size segregated emissions. Besides the need to measure PM size distributions for motor vehicle emissions, the capability to measure PM continuously should be utilized in some programs. Differentiating operating mode contributions to PM emissions is also important.)

There are some emission measurement and sampling issues that need to be solved in order for emission characterizations to properly support analyses. These emission measurement issues are likely to become more challenging with lower-emitting new technology vehicles that meet 2005-2007 emission standards. One issue raised during the workshop was that there can be occasions when the exhaust concentrations of a pollutant are lower than the concentration measured in ambient air. (This is an issue for organics, and it may also be for metals, as well.) Hence, there is a general need for an up-to-date look at measurement technology.

Measuring ultrafine particles from vehicles is relatively new and may prove to be more significant from a health perspective than the mass concentrations now used for regulatory purposes. There are many significant trace species whose emissions will be affected by controls - with increases and decreases being possible. Examples include the potential for increased ultrafine particle emissions when total mass is lowered, the possible increase in the ratio of ammonia and nitrous oxide to NOx when NOx emissions are lowered, and potentially increased ratios of nitrogen dioxide/NOx and concentrations of nitro- polycyclic aromatic (e.g., 1-nitronaphthalene) hydrocarbon formation with diesel PM controls. Emission control systems need to be evaluated with respect to all pollutants before they are applied on a widespread basis.

EPA has promulgated Tier 2 motor vehicle emission standards and gasoline sulfur control requirements (Tier 2) and heavy-duty engine and vehicle standards and on-highway diesel fuel sulfur control requirements. Compliance with these emission standards and changes to fuel characteristics are likely to result in significant changes to the combustion chemistry and emission characteristics of these vehicles. Emission measurements need to be performed for these vehicles/fuel combinations. Vehicle types/technology types that need to be included in new emission testing programs include hybrid-electric vehicles, heavy-duty gasoline-powered vehicles, and diesels equipped with oxidation catalysts.

It is also important to study the emissions performance of current technology vehicles on these low sulfur fuels because low sulfur fuels will be in widespread use by the end of this decade. Knowing how these (diesel) engines operate with low sulfur fuels will provide evidence to help us determine whether these current technology engines are candidates for certain retrofit technologies (PM traps, for instance).

For any measurement studies, it was strongly suggested that they be designed to obtain repeat measurements. This is important because there is often more variability in individual vehicle emission performance than there is in the measuring devices. Replicate testing should be performed until the variability (or lack thereof) is shown not to be an issue. High emitter variability is what we need to know the most about because of skewness, with the mean, median, and mode being located at very different points of the cumulative frequency distribution.

Conditions/control measures for which emissions data will be needed include: before and after diesel emission inspections and repairs, with pollution control retrofits, emulsion fuels, and to capture fuel variability. It was noted that the renewable fuels mandate may change fuel characteristics. Renewable fuel standards are important because they will affect the magnitude of HC/PM emissions and the proportion of toxics emitted. Ethanol, which contains oxygen, in fuel increases the proportion of the carbonyl compounds (acetaldehyde, formaldehyde, and acrolein) in exhaust HC. It will also increase permeation (evaporative) HC emissions in excess of what is predicted by Reid vapor pressure changes alone. The use of gasoline oxygenate blends in areas which also have non-oxygenated gasoline results in commingling of the two types of fuels in a vehicle's fuel tank. The combination of the oxygenate and non-oxygenate fuel results in increased vapor pressure, which will be higher than that of either fuel individually. Because virtually all of the current testing has been performed on renewables-free fuel, the effect of renewable fuel standards on air toxic emissions is poorly understood.

For heavy-duty diesel trucks, a lack of speciated HC emission measurements was noted as a data gap. At the time of the workshop, there were about 50 runs (measurements) of speciated HCs on chassis dynamometers available for use in establishing gaseous toxic emission rates (Lev-On et al., 2002). Since May 2003, there are an additional 60 to 80 samples from studies sponsored by the National Renewable Energy Laboratory and the Coordinating Research Council. This can be contrasted with there being 8,000 measurements from about 2,000 separate heavy-duty diesel vehicles (HDDVs) for particulates. Note that most of these PM measurements are not speciated. It is estimated that a few hundred of these measurements have had elemental carbon and organic carbon determined from the filters. Very few PM samples have been speciated beyond that.

For all vehicle types, it is important to identify the relative importance of cold start emissions. Then, it can be determined how much effort should be devoted to sampling during other modes. It is also important to test a certain fraction of gross emitters in any sampling scheme. Numerous studies have shown the relative importance of gross emitters to total motor vehicle emissions in urban areas.

New HDDV emissions testing should include tests that isolate idle emissions. Test samples should include trucks equipped with exhaust gas recirculation and after treatment.

3. Real World Conditions

There is also a data gap between the lab measurements that are available now, and real world in-use diesel activity. We need to fill this hole, especially for HDDVs. Suggestions for additional data collection techniques to fill this data gap included PEMS and remote sensing. One of the strengths of PEMS is its ability to include measurements of activity patterns. Remote sensing helps by collecting data from many vehicles, and provides estimates of emissions variability, although there are associated site limitations. It was noted that the current remote sensing status for estimating PM emissions is based on UV-fuel specific (CO2 specific) opacity. Testing programs are needed to assess how well this technology can estimate PM emissions. Two firms currently have the capability to perform such measurements.

For PEMS, it was suggested that there are instrument and data quality issues that need to be resolved before researchers can gain confidence in the data being collected. Inexpensive units have proved to produce inaccurate emission measurements. Therefore, protocols need to be developed that define the required repeatability of measurements that must occur, plus other data quality objectives to be met before an instrument can be used in a research study. PEMS do not currently have the ability to measure toxic compounds. It was suggested that more emphasis be placed on developing the technology to measure some toxic compound emissions using PEMS.

What can be done to inform the remote sensing industry about the needed measurement capabilities for air toxics from this technology? A remote sensing industry representative suggested that they need to know what some appropriate surrogates are that they might have the ability to measure (1 or 2 surrogates for HC species, for example). This would allow them to focus on developing the instrumentation to measure those surrogates. It was mentioned that benzene emissions would be very difficult to measure via remote sensing because of interference from CO2 and water vapor. Recent technology advances indicate that measurement of 1,3-butadiene, acrolein, acetaldehyde, and formaldehyde may be feasible using special remote sensing techniques. If such techniques prove useful, a key question is "Can these (or a subset of these) be useful surrogates to predict total MSAT content?"

Users of emissions data (and those who generate it) need to understand the conditions under which the data was generated. Lab-based vehicle and engine emissions measurement and test methods were developed and are intended for use in supporting regulatory activities (i.e., emissions certification). The methods mandated for use in certification testing are not necessarily suited to generating data for ambient air modeling or exposure assessment. An example is the choice of driving cycle. Most light-duty vehicle emissions measurements make use of the regulatory test cycle (FTP-75) which is notoriously unrepresentative of real-world driving cycles - so much so that a small industry has developed that produces "adjustment factors" for use in emissions modeling to attempt to predict emissions from modes of operation that are outside the FTP-75 test procedure. Another example is the new 2007 heavy-duty diesel engine test procedure that requires PM samples to be collected at 47+5oC. The reason for this requirement is to have a test procedure that is repeatable day to day and lab to lab. Using this sampling procedure to collect samples for toxic compound analysis (e.g., PAH) will produce a sample that is quite different from what a person would be exposed to in the real world. Finally, the chemical composition of emissions will be changing significantly over the next few years as new emissions standards come into effect. This requires a comprehensive screening of emissions to identify new, exotic, and previously unidentified pollutants that may be generated as a result of the new technologies and fuels.

4. Activity Data Collection

As more sophisticated emission measurements are made, it may be necessary to examine associated travel data needs in order to perform more complex analyses of transportation networks, whether it is at the micro-scale (intersections or roadway segments) or urban scale. For example, if operating modes are important, then traffic data collection will need to be redesigned to capture this information. So, part of this issue is to be able to describe to the transportation practitioner what is needed from them to support the use of more sophisticated emissions data (modal, or the equivalent). Thus, pursuit of this research will require interagency cooperation (environmental and transportation agencies). One specific travel data improvement that was identified was the need to be able to estimate truck travel percentages by roadway type. In most situations, analysts are using a single fraction [truck vehicle miles traveled (VMT) as a percentage of total VMT] to estimate truck emissions. It is likely that truck travel is not distributed evenly across roadway types (freeway, arterial, etc.).

For trucks, there are separate needs to investigate the differences in fleet populations and travel for important sub-categories of trucks. These sub-categories can be defined as: long-range fleet, regional fleet (300 to 500 miles), and local fleet (primarily delivery trucks). For each of these categories, it is expected that there will be spatial operating characteristic differences and time-of-day operating differences that need to be enumerated. Evaluations of the long-range fleet need to provide data about potential out-of-country trucks, especially in areas near the northern and southern US borders. An important operating characteristic to study for these truck sub-categories is idling time. Some studies have estimated that heavy trucks in urban areas spend as much as 45 percent of their operating time in idle mode. The result of any research on truck operating characteristics by sub-category should include the ability to link spatial operating characteristics to geographic information system (GIS) and truck travel detail by facility. Truck activity characteristics should include estimates of volumes, speeds, and time-of-day of travel. Especially useful would be good activity data for on-road and non-road sources at ports, intermodal terminals, large distribution warehouses, and other similar locations. Information on idling and other activity at these facilities is sparse, while public interest in air quality analysis of these facilities is increasing.

5. Off-Road Engines/Vehicles

For off-road engines/vehicles, diesel-powered construction equipment was identified as a major emissions source (of PM) with large uncertainties in emission estimates. There are many long-term construction projects that involve diesel-powered construction equipment. The error bounds in estimating these engine populations alone were estimated to be plus or minus 50 percent. Activity patterns and emission rates are added sources of uncertainty. It was noted that there is more instrumentation now (PEMS) that can be used to measure activity patterns. Then, emissions can be measured in a laboratory setting replicating the activity patterns that are measured in the field. An important issue for this source type is identifying surrogates for toxic air pollutants. In other words, which measured pollutants can serve as surrogates for the specific toxic compounds that are of interest, because it is unlikely that all toxic pollutants emissions can be measured directly? Also useful would be identification of thresholds; i.e., what level of activity at a construction site would be significant enough that mitigation is warranted to protect public health?

There is already a need in California to have a tool that can be used to estimate the emissions from highway construction projects. Projects are being criticized on this basis. Projects in Boston and New Haven have required mitigation measures (though not directly on air toxic grounds) for diesel construction equipment, including retrofit of diesel oxidation catalysts, and limitations on certain types of diesel activity and idling in sensitive areas.

6. Near Roadway Pollutant Behavior

It was also suggested that experiments be carried out to examine near roadway concentration patterns (especially for highways). There seems to be evidence that concentrations of air toxics near roadways may be appreciably higher than those 300 to 500 meters or more, downwind. Experiments to examine these gradients are likely to assist in steering future research toward examining pollution emissions, fate and transport in the immediate vicinity of highways compared with examining regional scale chemistry and transport.

One of the expected outcomes of any such near-roadway situation would be specification of the needed data precision for different data uses. This could be thought of as identifying different tiers of needs, such as more precision being needed for intersection scale analyses (modal information, that is), and less for regional scale work.

7. Coordination with Planned Research

At one of the sessions, the upcoming joint agency funded light-duty gasoline powered vehicle emission measurement effort scheduled in Kansas City was discussed. It was suggested that the FHWA piggyback some additional data collection efforts on this project, where possible. This might include some PEMS data collection, although it was noted that PEMS data collection was already planned as part of the work scope. There was also additional discussion of the use of the resulting emissions data set to develop facility-based versus trip-based emission estimates. Note that the current plan in the Kansas City study is to conduct dynamometer testing using an aggressive test cycle [the Los Angeles 1992 (LA92) Unified Driving Cycle]. While this is an aggressive test cycle, it also encompasses a number of operating conditions that are representative of typical urban driving. The LA92 cycle consists of a cold start Phase 1 (first 310 seconds), a stabilized Phase 2 (311-1,427 seconds), a 600 second engine off soak, and a warm start Phase 3 (repeat of Phase 1 of LA92). It was mentioned that data from this cycle will not provide information about any operation on specific facilities, and concerns were expressed about having these test results in a form that could provide inputs to MOBILE6, or the planned EPA Motor Vehicle Emission Simulator (MOVES) model.

The above project is one example of the need for performing research that is coordinated with other agencies and organizations. There are many others provided in this report.

8. Quantifying Uncertainty

For all new vehicle emissions characterization studies, the importance of characterizing variability and uncertainty in the emission estimates was noted. Variability refers to real differences in emissions among multiple emission sources at any given time, or over time, for any individual emission source. Variability in emissions can be attributed to variation in fuels, ambient temperature, technology, maintenance, or operation. Uncertainty refers to the lack of knowledge regarding the true value of emissions. Sources of uncertainty include small sample sizes, bias or imprecision in measurements, non representativeness, or lack of data. Quantitative methods for characterizing both variability and uncertainty are available.

Uncertainty in emission factors, and in emission inventories, is typically not quantified (Frey and Li, 2003). Therefore, it is not known, in many cases, how robust regulatory or management decisions are with respect to uncertainty. If management decisions are based upon point estimates of emissions that are biased, or if the range of uncertainty in emissions is much larger than any predicted change in emissions resulting from an air quality management strategy, then the decision-making process for developing management strategies could be ineffective.

Participants mentioned that there is significant uncertainty in estimating PM emissions from re-entrained road dust on paved road surfaces. Many researchers believe that EPA's models overestimate these emissions, and that new measurement studies are needed to provide estimates of whether such emissions are important components of regional transport. Because the conventional wisdom is that re-entrained road dust emissions are unlikely to be important contributors to regional haze and fine particle nonattainment, it has been difficult to attract funding for efforts to provide new measurement studies for this source type.

Meanwhile, PM emission estimates made using EPA's current AP-42 emission factors will continue to identify re-entrained road dust as a major PM source in many attainment plans until such time that research provides enough new data to revise the emission factor algorithm.

C. Priority Research Recommendations

Three major research areas emerged as being consistently important to researchers and transportation practitioners in meeting needs for air toxics emissions characterization. These include funding work to improve emission measurement technology so that lower emission levels can be measured, establishing experiments to examine gradients in near roadway concentrations of air toxics, and expanding the available set of information about air toxic emissions and activity patterns for on-road and off-road vehicles.

Proposed Programmatic Initiatives

Programmatic Initiative P1: Fund research for improvements in emission measurement technology that are needed to measure the lower emission levels of air toxic compounds expected with improved emission control technologies and lower sulfur fuels expected in the 2005 to 2007 time frame.

EPA has promulgated Tier 2 motor vehicle emission standards and gasoline sulfur control requirements and heavy-duty engine and vehicle standards and on-highway diesel fuel sulfur control requirements. Compliance with these emission standards and changes to fuel characteristics are likely to result in significant changes to the emission characteristics of these vehicles. Emission measurements need to be performed for these vehicles/fuel combinations in order to provide appropriate data sets.

There are some emission measurement and sampling issues that need to be solved in order for researchers to have confidence in these emission estimates. These emission measurement issues will be more challenging with lower-emitting new technology vehicles that meet 2005 to 2007 emission standards. The conference summary paper from the Coordinating Research Council Workshop on Vehicle Exhaust Particulate Emission Measurement Methodology, October 21, 2002 (www.crcao.com) provides an excellent summary of relevant issues here. The purposes of the Workshop were to: discuss recent progress in understanding the formation and fate of vehicle exhaust particulate emissions, discuss new measurement techniques (particle size, number, surface area, and composition) that supplement current mass measurements, discuss types of PM measurements and identify knowledge gaps that hinder their development, and to discuss what needs to be done to progress toward agreed upon methods for regulatory and research needs.

For PEMS, onboard instrumentation of vehicles during on-road operation enables data collection under real-world conditions. However, PEMS do not currently have the ability to measure toxic compounds. More emphasis should be placed on developing the technology to measure some toxic compound emissions using PEMS.

For remote sensing, studies should be performed to assess ability to measure and to predict PM mass, and surrogates for various classes of MSATs should be identified that the remote sensing instruments might have the ability to measure. This would allow the remote sensing technology development to proceed with developing the instrumentation to measure those surrogates.

Estimated Cost: $2 million
Duration: 2 years

Programmatic Initiative P2: Design and initiate experiments to examine near roadway concentration patterns (especially for highways). There is evidence that air toxic concentrations near roadways are appreciably higher than those 300 to 500 meters or more downwind. Experiments to examine these gradients are likely to assist in steering future research toward examining pollutant emissions, fate, and transport in the immediate vicinity of highways compared with examining regional scale chemistry and transport.

This program area would generate field study data collection efforts similar to those of the Los Angeles Catalyst Study during the 1970s. The Los Angeles Catalyst Study occurred because of public concern over the possible adverse impact of emissions from catalyst-equipped automobiles in ambient air. The study was to be a comprehensive long-term investigation of the ambient levels of sulfuric acid aerosol, sulfates, and other potential catalyst emission products in areas adjacent to a heavily traveled freeway in Los Angeles. EPA ultimately designed a roadway study to monitor pollutant levels on both sides of a freeway for a 3-year period (EPA, 1977).

A more recent measurement effort involving ultrafine particulate measurement near a major highway is described in the May 2002 issue of JAWMA (Zhu et al., 2002). Similar research applied to MSAT emissions or an established surrogate will aid in the future development of practical tools for roadway assessments. Several studies under various conditions will be necessary to establish fate and transport of MSAT pollutants for a micro-scale assessment.

Estimated Cost: $3-5 million
Duration: 2 - 3 years

Programmatic Initiative P3:

Conduct research to expand the available set of information about air toxic emissions and activity patterns for the on-road and off-road vehicle types with the most significant contributions to ambient air toxic concentrations of concern and the greatest uncertainty in their emission estimates. Given the current state-of-knowledge and planned research projects in this area, two sub areas within this programmatic initiative have been identified as likely first priorities for study.

Sub-Area 1: There are needs to investigate the differences in fleet populations and travel characteristics for important sub-categories of trucks. These sub-categories can be defined as: long-range fleet, regional fleet (300 to 500 miles), and local fleet (primarily delivery trucks).

For each of the three truck categories, it is expected that there will be spatial operating characteristic differences and time-of-day operating differences that need to be enumerated. Evaluations of the long-range fleet need to provide data about potential out-of-country trucks, especially in areas near the northern and southern U.S. borders. An important operating characteristic to study for these truck sub-categories is operating time.

Estimated Cost: $500,000
Duration: 18 months

Sub-Area 2: For off-road engines/vehicles, diesel-powered construction equipment is a major emissions source (of PM) with large uncertainties in emission estimates.

An important issue for this (and other) source types is identifying surrogates for toxic air pollutants. In other words, which measured pollutants can serve as surrogates for the specific toxic compounds that are of interest, because it is unlikely that emission measurement programs would have sufficient funding for all toxic pollutant's emissions to be measured directly.

Estimated Cost: $750,000
Duration: 24 months

Updated: 07/06/2011
HEP Home Planning Environment Real Estate
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000