Surface Transportation Environmental Research: A Long-Term Strategy -- Special Report 268 (2002)

Through the use, manufacture, maintenance, and scrappage of vehicles (including passenger cars, trucks, buses and railcars) and the maintenance of the infrastructure they use, the transportation system creates significant harmful effects on human health. Surface transportation is linked to human health (both mortality and morbidity) through vehicle emissions; road dust; degradation of drinking and stream water by road runoff from substances in vehicles, deicing chemicals, or other substances; exposure to toxic contaminants through vehicle and battery scrappage (either directly or through contamination of ground or surface water); and engine and tire noise.¹ The effects of these emissions, pollutants, and wastes on health occur at different spatial scales. Carbon monoxide (CO) emissions, for instance, are generally of concern at the local level; the same is true for larger particulate emissions, which tend to settle out quickly. Finer particles and ambient ozone, on the other

hand, can travel for hundreds of miles or more, so that pollution in one state can affect the health of people in many other states and even other countries. While the ground-level emissions from mobile transportation sources tend to be less of a problem with regard to long-range transport relative to stationary sources with tall stacks, they are still significant. And some transport emissions, such as carbon dioxide (CO₂), have global effects (EPA 2001a).

Responsibility for research on transportation, pollution control, and human health is divided among several agencies, and while the surface transportation community may have an interest in every linkage between transportation emissions and health, some are more directly within its purview than others. As a result, gaps in research sometimes develop. Research on transportation and air quality offers a case in point. Under the Clean Air Act, the Environmental Protection Agency (EPA) is responsible primarily for estimating emissions from mobile, area, and stationary sources, linking emissions to concentrations of air pollutants, and linking the resulting air quality to human health. The agency places particular emphasis on documenting the emissions from transportation sources and seeking transportation policies and measures to reduce pollution. Little of EPA’s work, however, has addressed the broader question of how current transportation programs affect human health, or how they could be redesigned to avoid or reduce pollution. The Federal Highway Administration (FHWA) takes an active interest in EPA’s research and sponsors its own studies on how to meet the requirements of the Clean Air Act. In particular, FHWA is responsible for ensuring that local transportation plans conform to the Clean Air Act. However, neither FHWA nor any other transportation agency conducts research on the health effects of its programs on a regular basis. Thus although both agencies have focused attention on transportation and air quality for several decades, significant research gaps exist.

In this chapter, the Advisory Board proposes a research agenda to address these gaps. This agenda encompasses epidemiological topics of particular significance to surface transportation, as well as research issues related to estimation of the economic health benefits of reducing transportation-related emissions. Also, because health effects and the monetary value of those effects figure prominently in cost/benefit and cost-effectiveness analyses of environmental and transportation regulations, these issues are included in the recommended agenda.

The consequences of transportation for human health are not straightforward, and extensive research is needed, in some cases to substantiate the linkages and in other cases to establish the direct and indirect links. Much of the problem is due to a lack of sufficient data. Determining which transportation activities

pose the greatest threat to human health as a result of environmental impacts and which should be modified is critically important. There is also a need to achieve the greatest possible reduction in risk for the costs incurred; thus the relationship between risk reduction and cost is an important research topic as well.

The specific linkages between surface transportation and human health are described in some detail in the next section. The discussion focuses primarily on air pollution, from which most health effects of surface transportation are believed to occur. Some attention is also paid to water pollution, solid wastes, and noise. Water pollution and noise are both addressed further in the discussion of ecology and natural systems in Chapter 3.

Fuel burning due to transportation activities, whether under congested or uncongested conditions, has the potential to increase the risk of death from respiratory and cardiovascular disease; raise the risk of developing certain chronic diseases [including cancer (Pope et al. 2002); chronic bronchitis; and, according to very recent evidence, asthma (McConnell et al. 2002)]; aggravate various existing chronic conditions; and lead to acute cardiopulmonary symptoms, such as cough, a runny nose, and other signs of a cold (Key 1998; Liao et al. 1999; Tolbert et al. 2000). Burning gasoline still leads to significant emissions of volatile organic compounds (VOCs) (some of which are carcinogenic), CO, nitrogen oxide (NO_x), and particulate matter, although each of these pollutants has been heavily regulated under the Clean Air Act by means of emission standards, mandated inspection and maintenance programs, and fuel quality standards. Moreover, because of the large amount of gasoline burned in the United States, even the very small amounts of sulfur present in gasoline can lead to significant emissions of sulfur dioxide (SO₂). (Lead, however, has been virtually eliminated from gasoline.) Burning of diesel fuel (primarily by trucks in the United States) emits NO_x, SO₂, and fine particulates (under 1 micron in diameter), some of which are carcinogenic (EPA 2000). In addition, mobile source air toxics from the combustion of gas and diesel fuel pose potential human health risks.² Air toxics are

not regulated under the National Ambient Air Quality Standards, but are still subject to regulations specified in the Clean Air Act Amendments of 1990. Moreover, in 2001 EPA set new performance standards for gasoline, requiring that refiners maintain their average 1998–2000 performance levels for air toxics. It is estimated that mobile source air toxics may account for 21 to 42 percent of urban air toxics (EPA 1998a; EPA 1998b; HEI 1999).

Efforts to reduce emissions from fuels have consisted primarily of the use of fuel additives, which generate a whole new set of potential health concerns. For example, methyl tert-butyl ether (MTBE), an oxygenate added to gasoline to reduce CO emissions, may be associated with headache, nausea, and sensory irritation in sensitive individuals, and has also been linked to neurotoxic effects and cancer in animals at high exposure levels (HEI 1996). Similarly, the health effects of the cerium-based diesel fuel additive Eolys, which has been used to reduce particulate emissions, have yet to be evaluated quantitatively, though the additive is thought potentially to target the lungs or lymph nodes at high exposure levels (HEI 2001). Alternative fuels (such as ethanol) emit their own signature suite of pollutants, including formaldehyde.³

The following discussion of the effects of mobile sources of air pollutants on health begins with a review of the emissions and concentrations of the various air pollutants. The health effects of these pollutants are then examined, followed by a brief discussion of approaches for determining the monetary benefits of mitigating those effects. Each of these linkages is explored in great detail in EPA’s regulatory impact analyses of its ambient air quality standards and in both the prospective and retrospective analyses of the costs and benefits of the Clean Air Act and its 1990 amendments (EPA 1997a; EPA 1997b; EPA 1999a).

Emissions of automotive air pollutants vary according to many factors, such as the following:

• VOC emissions vary with the Reid vapor pressure of the fuel and the air and engine temperature. VOC emissions also result from certain high-speed accelerations, as well as from fuel leaks and spills and from evaporation.

• CO emissions tend to move with VOC tailpipe emissions over many operating conditions; that is, both VOC and CO emissions are high when engines are cold and speeds are low.

• NO_x emissions vary with the heat of the engine. Overall NO_x emissions have a U-shaped relationship with speed: they are high at low speeds as the vehicle accelerates, fall as the vehicle gains speed, and then rise again at higher speeds.

• SO₂ emissions are directly related to the sulfur content of the fuel.

• Road dust varies with the dryness of the pavement and the number and speed of vehicles on the roadway, as well as with certain pavement designs.

In addition to these so-called “conventional pollutants,” motor vehicles emit CO₂, a greenhouse gas, proportional to the amount of fuel burned; indeed, motor vehicles are estimated to be responsible for approximately 30 percent of all U.S. emissions of CO₂ (EIA 2001). Global warming attributed to greenhouse gases has been linked to a variety of health effects, some positive (due to warmer temperatures in colder climates), many more negative (due to warmer temperatures in warmer climates), and some exotic (due to the increased range of mosquito populations). In addition, vehicle air conditioners leak chlorofluorocarbons (CFCs), a critical element in the destruction of the ozone layer, which is linked in turn to higher rates of melanoma. This problem is being mitigated over time by the phase-out of production of the most potent CFCs both in the United States and worldwide in accordance with multilateral agreements under the Montreal Protocol.

While policy discussion has focused historically on the long-term health effects of global warming, research has indicated that reductions in mobile source greenhouse gas emissions—provided these reductions are based on lowered fossil-fuel combustion—may provide immediate health benefits. In a study of four major cities, Cifuentes et al. (2001) found that adoption of greenhouse gas mitigation technologies can significantly reduce ambient particulate matter (PM) and ozone concentrations, thereby leading to large reductions in premature mortality, chronic bronchitis, and restricted activity days.

The polluting role of on-road mobile sources is captured by EPA’s emission estimates for such sources. As an example, during the period 1989–1998, emissions of NO_x from these sources increased by 1 percent, whereas emissions

of the other key pollutants fell⁴ [SO₂ by 50 percent, PM₁₀ (particulates of 10 microns or less in diameter) and VOCs by 25 percent]—and this in spite of large increases in vehicle-miles traveled. Change in the composition of the vehicle fleet appears to have played a major role in the latter improvements, as newer vehicles meeting tighter emission standards gradually replaced older vehicles. In their study of PM emissions from light-duty gas and diesel vehicles, Cadle et al. (1999) estimate that 26.8 percent of total PM₁₀ mass recorded in the study was from cars dated 1986–1990, and that only 3.8 percent of the total observed PM₁₀ was from cars and trucks dated 1991–1996.⁵

In the atmosphere, the conventional pollutants disperse according to atmospheric conditions and their own properties (reactivity, weight, particle size). Some of the pollutants transform, often in highly nonlinear processes. For instance, SO₂ converts to sulfates (a fine particulate) in the presence of ammonia. NO₂ also converts to nitrates (a fine particulate as well) in the presence of ammonia, but the ammonia reacts preferentially with the SO₂; thus, reductions in SO₂ can lead to increases in nitrates. Similarly, VOCs and NO_x react to form ozone in the presence of sunlight. Depending on atmospheric and emission conditions, however, an area may be NO_x- or VOC-limited, meaning that reductions in the nonlimiting pollutant will not affect ozone concentrations. Also, NO_x emissions into an existing ozone-rich atmosphere will “scavenge” (i.e., reduce the concentration of) ozone. An important point to note with respect to transportation is that ozone concentrations evidence a strong association with daytime hours because of their dependence on sunlight; emissions of ozone precursors at night will not create much ozone, but will be available for downwind ozone formation in the morning. Also important for transportation, ozone and fine particulates travel long distances, so emissions in one location can affect human health in downwind locations even 1,000 miles away (TRB 1995).

With regard to trends in air quality and emissions, EPA estimates that as of 2000, 121 million people nationwide lived in nonattainment areas with pollution levels above at least one of the National Ambient Air Quality Standards (set to be protective of health). Yet EPA analysis reveals much improvement.

About 85 percent of areas had downward trends in concentrations of one or more of the criteria pollutants (ozone, particulates, NO_x, SO₂, CO, and lead), while only about 10 percent had upward trends. Most of the areas with upward trends had air quality that significantly exceeded the air quality standards, while the rest reflected a lack of compliance with the new (not yet implemented) 8-hour ozone standard. Still, NO_x and ozone levels had fallen only 11 and 10 percent, respectively, overall. At the same time, some cities had experienced remarkable gains. From 1991 to 2000 in Los Angeles, the area with the most severe air pollution in the nation, the number of days for which the air quality index indicated unhealthy air fell from 168 to 48; days on which there were unhealthy levels of ozone declined from 126 to 45.

While monitoring for air quality changes is a principal means of demonstrating how well an area is doing in meeting the National Ambient Air Quality Standards, linking monitored changes in concentrations to mobile source emissions is not a simple matter. Complex air quality models are used to make “source attributions,” relating emissions from various source types and source locations to concentration changes in particular receptor locations (EPA 2001b). Key issues with these models include data quality and representativeness, both of which affect the statistical validity and associated error of the models; for photochemical models, representation of the underlying chemistry of pollutant formation and dissipation is also an issue.⁶

In addressing the linkage between motor vehicle emissions and human health, research and discussion have focused primarily on PM and ozone. PM exposure is considered an important risk factor for cardiopulmonary disease and mortality and is associated with other health endpoints, such as chronic bronchitis. Pope and Dockery (1999) review the epidemiology of PM effects. Other studies have addressed the effects of both acute exposure and chronic exposure to PM, though the former studies are more prevalent because of the availability of daily time-series datasets. These studies, such as Dockery et al. (1993) and Samet et al. (2000b), typically have suggested an increase in the daily mortality rate ranging from less than 0.5 percent to as high as 1.5 percent per 10 µg/m³ of PM₁₀. Chronic mortality studies, such as Pope et al.

(1995), have suggested a mortality response of 3 percent or greater per 10 µg/m³. Pope and Dockery (1999) note that these studies identify complementary effects, and suggest that the implications of chronic exposure may be more important from a public health standpoint. Chronic exposure increases the risk of chronic respiratory disease and of cardiorespiratory mortality, while acute exposure is believed to cause mortality by exacerbating existing cardiovascular and pulmonary disease, thus increasing the number of individuals in the population who die from complications related to such diseases.

While the relationship between PM and health has generally been accepted, there are large uncertainties about the relative potency of particulates by size and composition (NRC 2001). Health-based standards originally were set for total suspended particulates, but this was changed to PM₁₀; in 1997 EPA promulgated standards for PM_2.5 as well.⁷ It has been suggested that particles of smaller size may be more potent because they can be breathed more deeply into the lungs (Pope et al. 1995). With regard to composition, epidemiological studies have been most robust in linking PM₁₀ to health outcomes (Pope and Dockery 1992), but some very influential studies have shown a linkage to PM_2.5, and within this size category to sulfates (Wilson and Suh 1997). A number of studies have evaluated the effects of more than one size category. Krewski et al. (2000), in a reanalysis of work by Pope et al. (1995), finds that PM_2.5 has a significant association with chronic mortality, while PM_2.5–15 does not. Some acute mortality studies (Fairley 1999; Ostro et al. 2000) have reported similar findings. On the other hand, Lippmann et al. (2000) suggest comparable effects of PM₁₀, PM_2.5, and PM_2.5–10. While few studies have addressed the influence of PM₁₀ on health from an epidemiological perspective, a recent study by Wichmann et al. (2000) finds that PM_2.5 and PM₁₀ are similarly yet independently associated with daily mortality levels.

Within these different size classifications of PM, there is additional uncertainty about the relative toxicity of different particle types. One difficulty with pinpointing the potency of pollutants is that statistical results are often sensitive to the types of pollutants included in the analyses. In particular, some recent research (Spix et al. 1998) shows that SO₂concentrations can overshadow

particulate effects, potentially leading to an underestimation of the potency of particulate effects in certain cases. Recent research by Laden et al. (2000) attempts to link source-specific particle types within PM_2.5 to mortality. The results suggest that particles from mobile and coal combustion sources are associated with increased mortality, while crustal particles are not. Given the finding of Cadle et al. (1999) that the majority of PM mass from vehicle emissions is PM_2.5, the finding of Laden et al. (2000) appears to be consistent with the notion of PM_2.5 as more potent than larger particulate matter.⁸ Also of particular interest in Laden et al. is the finding that particles from mobile sources have a mortality effect about three times that of particles from coal combustion sources.

An additional source of uncertainty is the actual level of exposure to PM and other pollutants from vehicle emissions. Epidemiological studies typically measure ambient concentrations at monitoring stations within a locality and use these measurements as a proxy for actual exposure. The potential measurement error introduced by such an assumption has been cited as a major limitation of these studies (Samet et al. 2000a). Samet et al. note that individuals spend a small percentage of their time outside, making personal exposure levels highly dependent on the ability of pollutants to penetrate indoor spaces and on the quality of indoor air. It is perhaps even more difficult to estimate the percentage of an individual’s total level of exposure to air toxics that is due to vehicle emissions, which surely varies with distance from roads and traffic density, among other factors.

However, some estimates of actual exposure exist. Cohen and Nikula (1999) cite estimates from Cass and Gray (1995) and EPA (1993) that human populations in the general urban environment are likely to be exposed to average annual concentrations of diesel exhaust of 1 to 10 µg/m³. EPA (1999b) provides annual average exposure projections for benzene and diesel PM for a variety of locations and policy scenarios, ranging from approximately 0.25 to 0.83 µg/m³ and 0.24 to 0.40 µg/m³, respectively, for 2007. In-vehicle exposures, however, may be substantially higher. In their study of pollution concentrations inside vehicles on Sacramento and Los Angeles roadways, Rodes et al. (1998) find in-vehicle concentrations of benzene and PM₁₀ ranging from 3 to 15 µg/m³ and 20 to 40 µg/m³, respectively, in Sacramento, and 10 to 22 µg/m³and 35 to 105 µg/m³, respectively, in Los Angeles. Interestingly, pollutant levels in vehicles traveling in a carpool

lane were generally significantly lower than those in the righthand, slower lanes. In addition, roadway type, freeway congestion level, and time of day were shown to have some influence on in-vehicle pollutant levels.

Ozone has been the other pollutant of interest in research and policy discussion, though there is less of a consensus in the literature regarding the mortality effects of ozone than is the case for particulate matter. In their review of ozone mortality studies, Thurston and Ito (1999) suggest about a 0.3 percent increase in the risk of premature mortality for a 10-µg/m³ increase in daily 1-hour maximum ozone. Touloumi et al. (1997) also find an association between 1-hour maximum ozone levels and daily deaths. However, Samet et al. (2000b) report an association between ozone concentrations and mortality in the summer, when ozone concentrations are at their highest, but not over the full year. In his review of the existing literature, Schwartz (1997) finds no compelling evidence of a mortality effect.

A primary reason for disagreement in the literature is the complications that arise in modeling ozone effects. Thurston and Ito (1999) note that because ozone concentrations are highly correlated with temperature, studies that fail to control adequately for weather are likely to misrepresent the relationship between ozone and mortality. Research has also linked ozone exposure to reduced lung function, aggravation of preexisting respiratory ailments, and hospital admissions and emergency room visits linked to respiratory causes (Thurston and Ito 1999).

Because of the difference in molecular weight between ozone and fine particulates, it is not straightforward to compare the potency of these two substances on the basis of such measures as the percentage change in mortality risk per µg/m³ change in pollution concentration. Based on elasticities computed from various epidemiological studies (where the elasticity is the percentage change in mortality risk per average percentage change in pollution concentration across the study sample), acute PM₁₀-related mortality appears to be about four to five times more potent than that related to ozone.⁹

Mobile source air toxics pose the third major health concern associated with vehicular emissions. EPA has estimated that mobile sources of air toxics account for as much as half of all cancers attributed to outdoor sources of air toxics (EPA 1994). A particular concern is benzene, a known carcinogen that is a component of gasoline. Acetaldehyde, 1,3-butadiene, and formal-dehyde are also “reasonably anticipated” carcinogens under the National Toxicology Program’s classification system (HEI 1999). Uncertainties in estimating both the actual human exposure to these pollutants and the magnitude of the carcinogenic risk to humans are currently too great to allow for a meaningful assessment of the potential impact of reducing emissions of air toxics (HEI 1996).

Sensitive populations were identified in the process of setting the National Ambient Air Quality Standards under the Clean Air Act. These subgroups, identified through clinical, field, and epidemiological studies of the health effects of the six criteria pollutants, include children, the elderly, those with cardiopulmonary diseases, and pregnant women. Air pollution aggravates existing pathologies among these groups (Key 1998) while also affecting those with underdeveloped or vulnerable immune systems. Woodruff et al. (1997) found a relative increase in mortality risk for infants from PM₁₀ exposure similar to that estimated for adults by Pope et al. (1995), though over a shorter time frame of exposure, suggesting that infants face a greater risk from exposure than adults (Pope and Dockery 1999). Cohen and Nikula (1999) describe three studies that examined the association between exposure to vehicular air pollution and respiratory health in children. Brunekreef et al. (1997) and de Hartog et al. (1997) both found evidence of an inverse relationship between indices of pulmonary function and truck-traffic density and/or black smoke concentration in schools. Similarly, van Vliet et al. (1997) found that both truck-traffic density and in-school concentrations of black smoke were positively associated with prevalence of reported respiratory symptoms. Finally, a role for socioeconomic factors in determining risk from air pollution has been suggested. For example, in their reanalysis of the work of Pope et al. (1995), Krewski et al. (2000) found evidence of an inverse relationship between mortality risk from PM exposure and education.

Private and public preferences for reducing negative health outcomes are important to the extent that certain pollutants are more likely than others to

affect particular groups or types of health outcomes (e.g., mortality, hospital admissions, colds). Individuals make trade-offs in everyday living that reveal how much they care about reducing health risks versus meeting other needs or wants, such as spending their incomes elsewhere or arriving on time at a meeting (e.g., by driving faster than they would otherwise).

These preferences have been captured with two types of indexes—quality-adjusted life-years (QALYs) and disability-adjusted life-year (DALYs)—and with monetary values. The life-years approach involves totaling life-years saved as a result of various policies and adding this figure to years with quality-of-life improvements, thereby capturing reduced morbidity. The monetary valuation approach entails converting preferences for reducing mortality and morbidity risks to monetary measures. Both approaches rest on a variety of strong assumptions (with the QALY/DALY assumptions being more restrictive about preferences) and have significant limitations. Research is ongoing to improve these approaches and determine the advantages and disadvantages of each. The life-years approach is widely used in the health care system and has been employed in rulemakings, particularly by the Food and Drug Administration. The monetary approach is used in cost/benefit analysis and by many federal and state agencies, most notably EPA and the U.S. Department of Transportation (DOT).

In the three decades since passage of the Clean Water Act, vast improvements have been made in the United States and worldwide in cleaning up water resources and preventing future pollution. Cleanup of point (or stationary) sources accounts for most of these advances; little progress has been made on reducing non–point source pollution, including the road and parking lot runoff, airport runway runoff, and railroad runoff associated with transportation.

Studies have shown that increased traffic volume in watersheds increases pollutant loads (Benfield et al. 1999). Routine vehicle operations involve leakage of brake fluid, antifreeze compounds, lubricating oil, engine oil, grease, and gasoline onto the highway surface, which can lead to the contamination of surface water and groundwater, as can the washing of fine particles from tires and road surfaces into water supplies. Routine vehicle maintenance also can result in water contamination, despite regulatory efforts to limit such contamination. Moreover, routine activities such as road repair, deicing, right-of-

way maintenance,¹⁰ and the transport of hazardous materials pose a significant risk of polluting water resources.

People become subject to negative health effects of water pollution by drinking contaminated water (either surface or groundwater), swimming in or otherwise contacting contaminated water, or eating fish or other organisms that have come in contact with contamination or have bioaccumulated contamination in their tissues. Of particular concern is acid deposition to soil and water (as well as to forests; see Chapter 3) from emissions of SO₂ and NO_x due to burning of gasoline and diesel fuel. Such deposition leaches out aluminum, mercury, and other substances in soils, which, through runoff, can bioaccumulate in fish. In addition, there is speculation that high eutrophication levels (caused by excessive nitrification of estuaries, lakes, and streams through nitrate deposition) can trigger Pfiesteria outbreaks.

Upon discovering fish contamination, authorities may issue fish consumption advisories (FCAs) to limit the exposure of recreational fisherman and fish consumers. Direct health effects may occur until the elevated levels of pollution in fish are discovered and as a result of violations (knowingly or not) of the FCA. Economic losses to recreational and commercial fisheries (and consumers) may also result from an FCA. Such losses, while not themselves health effects, are the result of trying to limit human exposure to toxicants. None of the above linkages are well understood.

Transportation is a major generator of solid wastes through the disposal of vehicles and parts as well as construction materials. Human health effects from solid waste can occur through the release of toxic substances from scrapped vehicles or parts of vehicles. Although an active battery recycling market exists, exposure to acid or lead from batteries is a possible concern, as is exposure to heavy metals that exist in small but growing amounts in vehicle electronics. Two additional concerns are with oil from scrapped parts and vehicles and with used tires. Oil is regularly recycled through market mechanisms; however, some oil may leach into groundwater or end up in surface water. Used tires are increasingly being recycled in the United States for use in a variety of products,

as well as to fuel industrial processes and generate electricity, yet they are still frequently discarded. There are three potential health problems associated with these tires: pollution from tire fires, pollution from the use of tires as fuel (in boilers not environmentally suited to their burning), and the contribution of tire piles to the propagation of mosquito-borne diseases. The concern with tire fires is primarily with the particulates they emit, which, as noted above, are related to excess mortality; little is known about the frequency and toxicity of such fires, however. Inappropriate use of tires as fuel is not thought to be a problem in the United States, but is a problem in Mexico near the U.S. border (Blackman et al. 2001). Finally, tires provide an excellent habitat for mosquitoes. The extent of this problem in the United States is unknown, although Blackman et al. (2001) found it to be negligible in Juarez, Mexico, where large tire piles exist.

Noise pollution—defined as any form of unwanted or unhealthy sound (FHWA 1995)—has been studied extensively with respect to air transportation, and research also has addressed the noise effects of highways and transit (TRB 1987; TRB 1990; Balachandran 1992; Wyse 1994). Studies have shown that some of the most pervasive sources of noise, in both rural and urban environments, are those associated with transportation (FHWA 1995). Noise pollution is understood to have a variety of effects on humans, in addition to the obvious direct effects of annoyance and interference with hearing (Gottlob 1985; Green 1987). For example, noise has been shown to have both auditory and nonauditory adverse physiological effects (Kryter 1994), and some studies indicate linkages between noise and hypertension (Fay 1991; Kryter 1994).

In the absence of interventions, growth in surface transportation activities is likely to increase both overall noise levels and population exposure to noise (TRB 1990; Fleming et al. 2000). Strategies to address the problem include redesign of vehicles to reduce noise from engines and tires; redesign of pavements to reduce noise production; location and design of transportation infrastructure to avoid or minimize noise transmission; installation of barriers or other devices to deflect noise away from populations; and design and retrofitting of buildings and outdoor areas to reduce indoor noise exposure, mask noise, or cancel out noise (TRB 1987; TRB 1990). The relative effectiveness of these strategies, their public acceptability, and their efficacy (in terms of noise reduction and overall effects on human health) remain topics for additional

research (OECD 1991; Job 1991; Balachandran 1992; Lansdell and Cameron 1998; TRB 1998).

FHWA, EPA, and other government agencies have taken on the task of developing research agendas related to surface transportation and human health. In the following recommendations, the Advisory Board proposes research that cuts across the concerns of transportation, environmental, and health agencies and is aimed at the development of a stronger understanding of fundamental processes as well as policy linkages.

While knowledge about the linkage between emissions and human health impacts is increasing, a better understanding is still needed of how the release of emissions is altered as a result of congestion and travel behavior. Until quite recently, the standard data on emissions were developed for an average vehicle speed and then adjusted by a function for higher or lower average speed. Emissions of the major gasoline-vehicle pollutants (VOCs, CO, and NO_x) are highly sensitive to acceleration, deceleration, and other driving modes. Given this sensitivity, information on “modal” emissions is needed to understand how emissions respond to changes in congestion or in travel behavior that affect driving modes. While research has been conducted on this topic, much more needs to be known before the effects of policy on emissions can be estimated reliably. This need extends to alternative-fuel vehicles (including those with ethanol additives); hybrid and electric vehicles (for which attention also should be given to emissions from power plants for battery charging); and, in the future, vehicles running on fuel cells (with consideration of upstream emissions). This need also extends to increasing the numbers and types of vehicles tested to generate emission estimates.

Understanding of overall PM and VOC emissions has been the objective of considerable research, with the caveat about modal emissions noted earlier. However, better, more detailed characterizations of these vehicle emissions are needed in light of the increasing attention being paid to the VOCs in gasoline and the various fine particulates in diesel fuel, as well as the growing awareness of mobile source air toxics in general as possible carcinogens and contributors to a variety of respiratory ailments, immune and nervous system disorders, and birth defects.

The development of improved air quality models is critically important for appropriate assessment of the effects of mobile sources of pollutant emissions on human health, as it is through these models that one can trace changes in concentrations at given receptors back to changes in emission source locations and types. Until recently, the effects on ozone of VOCs and NO_x emissions and the effects on particulates of NO_x and SO_x emissions were modeled separately. More recent efforts have focused on the development of models that integrate the air chemistry for producing ozone and particulate species (such as sulfates and nitrates). These models are just beginning to be implemented, and significant validation, testing, and application will be needed before policymakers will trust their output. An additional concern is the data inputs to the models concerning background VOC and ammonia concentrations. Until such models are in the mainstream and the data underlying them are viewed as credible, the contribution of mobile sources to air quality cannot be assessed reliably.

Mobile source emissions comprise several pollutants whose health effects are not well understood. Little research has been done, for example, on the actual human cancer risk from mobile source air toxics at current exposure levels. Likewise, understanding of the potential health risks of fuel additives such as

MTBE at low exposure levels must be augmented, particularly as the use of additives becomes more prevalent and the additives themselves more varied. While the use of additives reduces emissions and thus the health effects of more conventional pollutants, a better understanding of any potential risk trade-offs is necessary.

Finally, road dust entrained by mobile sources has been found to contribute up to 60 to 70 percent of the inventory of total suspended particulates from mobile sources. Little is known, however, about the effect of exposure to road dust on mortality or morbidity risks. A better understanding of these risks is needed so that strategies for reducing road dust can be devised and prioritized.

Although EPA recently promulgated the world’s first standard for fine particulates (particles with a diameter of less than 2.5 microns), uncertainties exist about the toxicity of various particle types and sizes. Until this uncertainty is reduced, the risk remains that resources will be wasted on controlling pollutants that have little or no effect on public health, while the effects of other pollutants may be underestimated. A good example of this point is nitrates, which are formed from NO_x emissions, water vapor, and ammonia. None of the key epidemiological studies relied upon by EPA in its rulemakings provides information about the effect of nitrate exposure on health. Yet nitrates are routinely assumed to be as potent an agent of mortality risk as the average PM_2.5 particle. If this assumption were to prove incorrect and nitrates turned out to be far less potent, the rationale for controlling NO_x would be significantly weakened; if, on the other hand, nitrates were shown to be more potent, different strategies might have to be introduced. Research on this topic is sorely needed. At the same time, NO_x emissions have been shown to scavenge ozone, meaning that reductions in NO_x would, at times and in certain locations, lead to higher, not lower ozone concentrations. Also, few studies have found an effect of ozone or NO_x on mortality risk, and there is still no consensus on the effect of ozone on mortality risk. Generally, effects found are of an order of magnitude lower than those found for sulfates and PM_2.5. Research is needed to establish and rank the potency of various pollutants.

In addition, mobile sources emit a complex mixture of pollutants, making it extremely difficult to establish a direct causal linkage between emission controls (particularly driving reductions or fuel switching) and health benefits.

A better understanding is needed of the synergistic and antagonistic relationships between pollutants and their effects on human health.

Special concern: Particulate matter measured by mass. EPA’s newly promulgated standard for fine particulates is stated in terms of the weight of the particles in the air, i.e., µg/m³. However, an emerging school of thought supports measuring particles by the number of particles per volume of air, as well as by mass. Specifically, it has been asserted that the particle surface area to which the lung is exposed probably is more important for determining health effects than the mass of the particles (Key 1998). Additional research is clearly needed in this area.

Further investigation is needed of the environmental and genetic risk factors for respiratory and pulmonary disease and the interaction between these factors. Current data and research findings suggest that an exposure–response relationship exists between high levels of air pollution and aggravated conditions among vulnerable populations. More research is needed to determine the disease-specific conditions involved, the conditions of exposure that are most important, and the long-term implications of these findings. Research on the adverse effects of transportation-related exposures for vehicle occupants, travelers on congested highways, residents and workers in buildings near heavily traveled streets, and other vulnerable groups is particularly important for planners and decision makers, who will need to devise appropriate ways to avoid, minimize, or mitigate those effects.

Not all health effects resulting from air pollution are of equal importance. One criterion for allocating scarce research dollars should be to determine those health effects deemed most important by the public. Research has focused on estimating these preferences across health outcomes in terms of people’s willingness to pay for health improvements. However, the context and populations involved in these economic studies may have been inappropriate for generalizing such preferences, particularly preferences for reductions in mortality risk associated with air pollution. Addressing the preferences of at-risk

populations—including the elderly and the infirm—should be a major social science research priority.¹² Research should also address parental preferences concerning children’s health, including avoidance of problems that arise from fetal exposure, and how such preferences are affected by latency, dread, involuntary exposure, and other features of the air pollution context. Research should be directed as well at describing preferences through the life-years (QALY) approach and at contrasting this approach with the monetary valuation approach.

A comprehensive accounting is needed of the many pathways by which surface transportation activities affect human health through contact with polluted water or contaminated fish or through linkages with solid waste from vehicle or parts scrappage. Greater understanding is also needed of the social and economic consequences of attempts to limit such contact, for example, through fish consumption advisories or the creation of more comprehensive recycling markets.

Information on emissions, air quality, health effects, and valuation is most valuable when used to choose approaches and policies for controlling pollution. To help make these choices, policymakers need information relating the costs of various approaches and policies to the anticipated benefits—either in physical or QALY/DALY terms (which implies a cost-effectiveness analysis) or in monetary terms (which implies cost/benefit analysis). Research is needed on methods for assessing the economic costs of alternative approaches and policies and for combining them appropriately with physical or monetary measures of benefits. This need is particularly acute in light of the new regulatory concern for fine particulates and the need to assess as effectiveness measures not just emission reductions, but also (because they have common precursor

emissions) reductions in concentrations of both ozone and fine particulates combined. As regulations directed at transportation emissions may affect differentially the various phases of the transportation system, including the production of vehicles and their scrappage, as well as the mix of vehicles in use and the amount they are driven, research methods are needed for capturing these effects across the vehicle life cycle.

Special concern: Technological improvements. Technological improvements in transportation must be evaluated and their costs and benefits fully characterized in terms of their impacts on human health. For example, the fuel additive MTBE was introduced as a technological advance for improving air quality, but also adversely impacts water quality and thus poses potential risks to human health through both inhalation and ingestion. In the past, technological improvements designed to improve air quality have also exacerbated noise pollution. The tightening of standards for vehicle emissions of VOCs, NO_x, and CO and for the sulfur content of diesel fuel and gasoline—as well as the penetration of reduced emissions vehicles required by federal law and regulation as well as alternative-fuel vehicles—will diminish the air quality problems associated with these pollutants. But as new fuels and forms of transport emerge, so, too, will new types of pollution problems. Research efforts should outpace these major changes in transportation technology. More research is needed to understand the potential human health impacts of transportation technologies prior to their full implementation.

Special concern: Regulatory constraints. Conformity requirements in the Clean Air Act were intended to incorporate air quality considerations into transportation planning. In the process, significant requirements were placed on the transportation community not only to meet demands for reduced congestion, but also to achieve and maintain clean air goals. Some of these burdens are tied to the procedural aspects of the conformity process, others are associated with the state implementation plan process, and others are the inevitable result of tensions between the congestion and air quality goals. Research is needed to identify mechanisms for reducing these burdens and for developing constructive and cooperative ties between the agencies responsible for transportation planning and for maintaining air quality. Conduct of this research may necessitate closer linkages between air quality and transportation models.

Another regulatory issue arises because of growing understanding of the long-range transport of important pollutants, such as ozone and fine particulates. Both of these pollutants may travel across half the continent, and long-range transport is an issue for mobile as well as stationary sources. The Clean Air Act

was designed primarily under the premise that localities make their own messes and thus are responsible for cleaning them up. Understanding of long-range transport has challenged that assumption and called into question the appropriateness of making the locality (the air basin) the centerpiece of the regulatory strategy. This issue affects the locus of control of transportation emissions through transportation control measures and other, nonfederal approaches.

Further studies are needed to provide data that can be used to measure the singular and cumulative impacts of transportation-related noise separately— focusing on noise annoyance, noise exposure, and the interaction between the two.

EIA Energy Information Administration

EPA Environmental Protection Agency

FHWA Federal Highway Administration

HEI Health Effects Institute

NRC National Research Council

OECD Organization for Economic Cooperation and Development

TRB Transportation Research Board

Balachandran, C. G. 1992. Traffic and Transit Noise Challenges in the Nineties. In International Conference on Noise Control Engineering (Internoise 92), Vol. 2, Noise Control Foundation, New York, pp. 845–848.

Benfield, F. K., M. D. Raimi, and D. D. T. Chen. 1999. Once There Were Greenfields: How Urban Sprawl Is Undermining America’s Environment, Economy, and Social Fabric. Natural Resources Defense Council, Washington, D.C.

Blackman, A., A. Krupnick, and A. Palma. 2001. Tire Piles as an Environmental and Health Hazard on the U.S.–Mexico Boarder: A Preliminary Assessment Focusing on Paso Del Norte. Prepared for Applied Sciences Laboratory, Inc.

Brunekreef, B., N. Janssen, J. de Hartog, H. Harssema, M. Knape, and P. van Vliet. 1997. Air Pollution from Truck Traffic and Lung Function in Children Living Near Motorways. Epidemiology, Vol. 8, pp. 298–303.

Cadle, S. H., P. Mulawa, E. C. Hunsanger, K. Nelson, R. A. Ragazzi, R. Barrett, G. L. Gallagher, D. R. Lawson, K. T. Knapp, and R. Snow. 1999. Light-Duty Motor Vehicle Exhaust Particulate Matter Measurement in the Denver, Colorado, Area. Journal of the Air and Waste Management Association, Vol. 49, PM164–PM174.

Cass, G. R., and H. A. Gray. 1995. Regional Emissions and Atmospheric Concentrations of Diesel Engine Particulate Matter: Los Angeles as a Case Study. In Health Effects of Diesel Engine Emissions: Characterization and Critical Analysis, Health Effects Institute, Cambridge, Mass.

Cifuentes, L., V. H. Borja-Aburto, N. Gouveia, G. Thurston, and D. L. Davis. 2001. Hidden Health Benefits of Greenhouse Gas Mitigation. Science, Vol. 293, pp. 1257–1259.

Cohen, A. J., and K. Nikula. 1999. The Health Effects of Diesel Exhaust: Laboratory and Epidemiologic Studies. In Air Pollution and Health (S. T. Holgate, J. M. Samet, H. S. Koren, and R. L. Maynard, eds.), Academic Press, San Diego, Calif.

de Hartog, J., P. H. van Vliet, B. Brunekreef, M. C. Knape, N. A. Janssen, and H. Harssema. 1997. Relationship Between Air Pollution due to Traffic, Decreased Lung Function, and Airways Symptoms in Children (in Dutch). Ned Tijdschr Geneeskd, Vol. 141, pp. 1814–1818.

Dockery, D. W., C. A. Pope III, X. Xu, J. D. Spengler, J. H. Ware, M. E. Fay, B. G. Ferris, Jr., and F. E. Speizer. 1993. An Association Between Air Pollution and Mortality in Six U.S. Cities. New England Journal of Medicine, Vol. 329, pp. 1753–1759.

EIA. 2001. Emissions of Greenhouse Gases in the U.S. 2000. DOE/EIA-0573 (2000). Department of Energy, Washington, D.C.

EPA. 1993. Motor Vehicle–Related Air Toxics Study. EPA 420-R-93-005. Office of Mobile Sources, Ann Arbor, Mich.

EPA. 1994. Air Toxics from Motor Vehicles. EPA 400-F-92-004. Fact Sheet OMS-2.

EPA. 1997a. The Benefits and Costs of the Clean Air Act: 1970 to 1990. Office of Air and Radiation and Office of Policy, Washington, D.C.

EPA. 1997b. Regulatory Impact Analysis for Ozone and Particulate National Ambient Air Quality Standards. Washington, D.C.

EPA. 1998a. Draft Integrated Urban Air Toxics Strategy to Comply with Section 112(d), 112(c)(3), and Section 202(l) of the Clean Air Act; Notice. Federal Register, Vol. 63, No. 177, pp. 49,239–49,258.

EPA. 1998b. National Air Quality and Emissions Trends Report, 1997. http://www. epa.gov/oar/aqtrnd97.

EPA. 1999a. The Benefits and Costs of the Clean Air Act, 1990 to 2010. Office of Air and Radiation and Office of Policy, Washington, D.C.

EPA. 1999b. Estimation of Motor Vehicle Toxic Emissions and Exposure in Selected Urban Areas, Volume 1. Draft EPA420-D-99-002a. Office of Air and Radiation, March.

EPA. 2000. Health Assessment Document for Diesel Exhaust. EPA/600/8-90/057E. Washington, D.C.

EPA. 2001a. National Air Quality and Emissions Trends Report 1999. EPA454/R-01-004. Research Triangle Park, N.C.

EPA. 2001b. Models 3 Documentation. http://www.epa.gov/asmdnerl/models3/doc/.

Fairley, D. 1999. Daily Mortality and Air Pollution in Santa Clara County, California: 1989–1996. Environmental Health Perspectives, Vol. 107, pp. 637–641.

Fay, T. (ed.). 1991. Noise and Health. New York Academy of Medicine.

FHWA. 1995. Highway Traffic Noise Analysis and Abatement Policy and Guidance. Office of Environment and Planning, Washington, D.C.

Fleming, G. G., R. E. Armstrong, E. Stusnick, K. D. Polcak, and W. Lindeman. 2000. Transportation-Related Noise in the United States. TR News, No. 208, May–June, pp. 7–11.

Gottlob, D. 1985. Effects of Transportation Noise on Man. In International Conference on Noise Control Engineering (Internoise 85), Bundesanstalt fur Arbeitsschutz, Dortmund, Germany.

Green, C. (ed.). 1987. Traffic Noise: A Bibliography on Surface Transportation Noise, 1979–86. Technical Communications, Letchworth, Herts, England.

HEI. 1996. The Potential Health Effects of Oxygenates Added to Gasoline. A Review of the Current Literature. Charlestown, Mass., April.

HEI. 1999. Research on Air Toxics. Program Summary. Cambridge, Mass.

HEI. 2001. Evaluation of Human Health Risk from Cerium Added to Diesel Fuel. HEI Communication No. 9. Charlestown, Mass.

Hillman, M. 1997. Health Promotion: The Potential of Non-Motorized Transport. In Health at the Crossroads—Transport Policy and Urban Health (T. Fletcher and A. J. McMichael, eds.), John Wiley and Sons, New York.

Job, R. F. S. 1991. Internal Consistency and Stability of Measurements of Community Reaction to Noise. In Transportation Research Record 1312, TRB, National Research Council, Washington, D.C., pp. 101–108.

Key, S. 1998. Studies Hint That Pollution Leads to Cancer. Cancer Weekly Plus, Oct. 5.

Krewski, D., R. T. Burnett, M. S. Goldberg, K. Hoover, J. Siemiatycki, M. Jerrett, M. Abrahamowicz, and W. H. White. 2000. Reanalysis of the Harvard Six Cities Study and American Cancer Society Study of Particulate Air Pollution and Mortality: A Special Report of the Institute’s Particle Epidemiology Reanalysis Project. Health Effects Institute, Cambridge, Mass.

Kryter, K. D. 1994. The Handbook of Hearing and the Effects of Noise: Physiology, Psychology, and Public Health. Academic Press, San Diego, Calif.

Laden, F., L. M. Neas, D. W. Dockery, and J. Schwartz. 2000. Association of Fine Particulate Matter from Different Sources with Daily Mortality in Six U.S. Cities. Environmental Health Perspectives, Vol. 108, No. 10, Oct., pp. 941–947.

Lansdell, H., and C. Cameron. 1998. Night-Time Noise Levels: A State-of-the-Art Review. Research Report 323. ARRB Transport Research Ltd., Vermont South, Victoria, Australia.

Liao, D., J. Creason, C. Shy, R. Williams, R. Watts, and R. Zweidinger. 1999. Daily Variation of Particulate Air Pollution and Poor Cardiac Autonomic Control in the Elderly. Environmental Health Perspectives, Vol. 107, No. 7, pp. 521–525.

Lippmann, M., I. Kazuhiko, N. Nádas, and R. T. Burnett. 2000. Association of Particulate Matter Components with Daily Mortality and Morbidity in Urban Populations. Research Report 95. Health Effects Institute, Cambridge, Mass.

McConnell, R., K. Berhane, F. Gilliland, S. J. London, T. Islam, W. J. Gauderman, E. Avol, H. G. Margolis, and J. M. Peters. 2002. Asthma in Exercising Children Exposed to Ozone: A Cohort Study. The Lancet, Vol. 359, pp. 386–391.

McMichael, A. J., A. Haines, R. Slooff, and S. Kovats (eds.). 1996. Climate Change and Human Health. World Health Organization, Geneva, Switzerland.

NRC. 2001. Research Priorities for Airborne Particulate Matter: III. Early Research Progress. National Academy Press, Washington, D.C.

OECD. 1991. Fighting Noise in the 1990s. Paris.

Ostro, B. D., R. Broadwin, and M. J. Lipsett. 2000. Coarse and Fine Particles and Daily Mortality in the Coachella Valley, California: A Follow-Up Study. Journal of Exposure Analysis and Environmental Epidemiology, Vol. 10, pp. 412–419.

Pope, C. A. III, and D. W. Dockery. 1992. Acute Health Effects of PM10 Pollution on Symptomatic and Asymptomatic Children. American Review of Respiratory Disease, Vol. 145, No. 5, pp. 1123–1128.

Pope, C. A. III, and D. Dockery. 1999. Epidemiology of Particle Effects. In Air Pollution and Health (S. T. Holgate, J. M. Samet, H. S. Koren, and R. L. Maynard, eds.), Academic Press, San Diego, Calif.

Pope, C. A. III, M. J. Thun, M. M. Namboodiri, D. W. Dockery, J. S. Evans, F. E. Speizer, and C. W. Heath. 1995. Particulate Air Pollution as a Predictor of Mor-

tality in a Prospective Study of U.S. Adults. American Journal of Respiratory and Critical Care Medicine, Vol. 151, pp. 669–674.

Pope, C. A. III, R. T. Burnett, M. J. Thun, E. E. Calle, D. Krewski, K. Ito, and G. D. Thurston. 2002. Lung Cancer, Cardiopulmonary Mortality, and Long-Term Exposure to Fine Particulate Air Pollution. Journal of the American Medical Association, Vol. 287, No. 9, pp. 1132–1141.

Rodes, C., L. Sheldon, D. Whitaker, A. Clayton, K. Fitzgerald, J. Flanagan, F. DiGenova, S. Hering, and C. Frazier. 1998. Measuring Concentrations of Selected Air Pollutants Inside California Vehicles. California Air Resources Board, Sacramento.

Samet, J. M., F. Dominici, S. L. Zeger, J. Schwartz, and D. W. Dockery. 2000a. The National Morbidity, Mortality, and Air Pollution Study. Part I: Methods and Methodologic Issues. Research Report 94. Health Effects Institute, Cambridge, Mass.

Samet, J. M., S. L. Zeger, F. Dominici, F. Curriero, I. Coursac, D. W. Dockery, J. Schwartz, and A. Zanobetti. 2000b. The National Morbidity, Mortality, and Air Pollution Study. Part II: Morbidity, Mortality, and Air Pollution in the United States. Research Report 94. Health Effects Institute, Cambridge, Mass.

Schwartz, J. 1997. Health Effects of Air Pollution from Traffic: Ozone and Particulate Matter. In Health at the Crossroads—Transport Policy and Urban Health (T. Fletcher and A. J. McMichael, eds.), John Wiley and Sons, New York

Spix, C., H. R. Anderson, J. Schwartz, M. A. Vigotti, A. Letertre, J. M. Vonk, G. Touloumi, F. Balducci, T. Piekarski, L. Bacharova, A. Tobias, A. Poenkae, and K. Katsouyanni. 1998. Short-Term Effects of Air Pollution on Hospital Admissions of Respiratory Diseases in Europe: A Quantitative Summary of APHEA Study Results. Archives of Environmental Health, Jan.–Feb.

Thurston, G. D., and K. Ito. 1999. Epidemiological Studies of Ozone Exposure Effects. In Air Pollution and Health (S. T. Holgate, J. M. Samet, H. S. Koren, and R. L. Maynard, eds.), Academic Press, San Diego, Calif.

Tolbert, P. E., J. A. Mulholland, D. L. MacIntosh, F. Xu, D. Daniels, O. J. Devine, B. P. Carlin, M. Klein, J. Dorley, A. J. Butler, D. F. Nordenberg, H. Frumkin, P. B. Ryan, and M. C. White. 2000. Air Quality and Pediatric Emergency Room Visits for Asthma in Atlanta. American Journal of Epidemiology, Vol. 151, pp. 798–810.

Touloumi, G., K. Katsouyanni, D. Zmirou, J. Schwartz, C. Spix, A. P. De Leon, A. Tobias, P. Quennel, D. Rabczenko, L. Bacharova, L. Bisanti, J. M. Vonk, and A. Ponka. 1997. Short Term Effects of Ambient Oxidants Exposure on Mortality: A Combined Analysis Within the APHEA Project. American Journal of Epidemiology, Vol. 146, pp. 177–185.

TRB. 1987. Transportation Research Record 1143: Environmental Issues: Noise, Rail Noise, and High-Speed Rail. National Research Council, Washington, D.C.

TRB. 1990. Transportation Research Record 1255: Energy and Environment 1990: Transportation-Induced Noise and Air Pollution. National Research Council, Washington, D.C.

TRB. 1995. Special Report 245: Expanding Metropolitan Highways: Implications for Air Quality and Energy Use. National Research Council, Washington, D.C.

TRB. 1998. Transportation Research Record 1626: Environmental and Social Effects of Transportation. National Research Council, Washington, D.C.

van Vliet, P., M. Knape, J. de Hartog, N. Janssen, H. Harssema, and B. Brunekreef. 1997. Motor Vehicle Exhaust and Respiratory Symptoms in Children Living Near Freeways. Environmental Research, Vol. 74, pp. 122–132.

Wichmann, H.-E., C. Spix, T. Tuch, G. Wölke, A. Peters, J. Heinrich, W. G. Kreyling, and J. Heyder. 2000. Daily Mortality and Ultrafine Particles in Erfurt, Germany. Part I: Role of Particle Number and Particle Mass. Research Report 98. Health Effects Institute, Cambridge, Mass.

Wilson, W. E., and H. H. Suh. 1997. Fine Particles and Coarse Particles: Concentration Relationships Relevant to Epidemiologic Studies. Journal of the Air and Waste Management Association, Vol. 47, pp. 1238–1249.

Woodruff, T. J., J. Grillo, and K. C. Schoendorf. 1997. The Relationship Between Selected Causes of Postneonatal Infant Mortality and Particulate Air Pollution in the United States. Environmental Health Perspectives, Vol. 105, pp. 608–612.

Wyse, W. J. 1994. Transport Noise Pollution. Light Rail and Modern Tramway, Vol. 57, No. 675, March, pp. 65–69.