1

South Bronx Environmental Health Policy Study (SBEHPS)

FINAL REPORT OF NYU SCHOOL OF MEDICINE RESEARCH

February 2007

Professor George D. Thurston Ariel Spira-Cohen, Ph.D Candidate

Professor Lung Chi Chen

New York University School of MedicineDepartment of Environmental Medicine

Nelson Institute of Environmental Medicine57 Old Forge Rd., Tuxedo, NY 10987

2

Introduction

The overall objective of this South Bronx Environmental Health Policy Study (SBEHPS) was tocharacterize the nature and causal factors of the ambient air pollution in communities of the South Bronxthat experience high concentrations of diesel trucks and waste transfer facilities. Epidemiological, humanclinical, and animal toxicological studies have shown that exposure to ambient air pollution is likely to beone of the important outdoor environmental risk factors for asthma and other airway diseases (U.S. EPAPM Criteria Document, 2004). In the Bronx, the rate of asthma hospital admissions for all ages between1991 and 1996 was twice that of Manhattan and Brooklyn. In addition, most neighborhoods in the Bronxexperienced a 110 to 120 percent increase in asthma hospitalizations between 1987 and 1996, as comparedto 35 to 50 percent increases in most other neighborhoods in New York City. Although the origin ofasthma is unquestionably multi-factorial, with important genetic, immunologic, and environmentalcomponents, the marked geographic and temporal variations strongly suggest that environmental factorsmake a significant contribution to the pathogenesis and expression of asthma, underscoring the need for anevaluation of this factor in urban communities like the South Bronx.

To learn more about these conditions, New York University initiated a community-based feasibility andplanning study that would begin to examine the current environmental conditions, specifically as theyrelate to traffic, including that caused by the waste management facilities, and the regulatory and publicpolicy issues that affect environmental decision-making in the South Bronx. This feasibility and planningstudy was carried out by the Nelson Institute of Environmental Medicine (NIEM) at New YorkUniversity’s School of Medicine and by a multi-disciplinary team from NYU’s Robert F. Wagner GraduateSchool of Public Service in collaboration with a Community Advisory Committee consisting of severalcommunity groups in the South Bronx.

During the course of this EPA study, the New York University School of Medicine (NYUSOM)research team has developed a strong collaboration with the collaborating community groups in the SouthBronx, including: Nos Quedamos, the Point, the Youth Ministry for Justice and Peace, and the SportsFoundation. Nos Quedamos/We Stay represents a broad-based, grassroots coalition of residents andgroups who have a longstanding commitment to Melrose Commons (a 35-contiguous block area east ofYankee Stadium) in the South Bronx. The Point is an emergent neighborhood organization dedicated tothe cultural and economic revitalization of the Hunts Point section of the South Bronx. Youth Ministriesfor Peace and Justice works in the Bronx River, Bruckner and Soundview neighborhoods in the SouthBronx. The Sports Foundation, Inc. has also contributed immensely to the South Bronx communitythrough the development and implementation of various youth programs.

These local resident organizations have worked to change public policy and improve thecommunity's quality of life by participation in government/science advisory groups, education ofcommunity residents, and promotion of sustainable development. Concerns about the air pollution createdby waste transfer facilities, and by ever-increasing traffic volumes in the South Bronx, have caused them tocome together to actively seek scientific evidence in support of their confrontation with environmentalracism in their community. A three-year research grant was funded through EPA in 2000 to perform acommunity-based feasibility and planning study that would begin to examine the current environmentalconditions in the South Bronx, specifically as they relate to the waste management facilities, and theregulatory and public policy issues that affect environmental decision-making. The community groupsactively participated since the inception of the concept in the planning, preparation, and execution of thisresearch program. The purposes of that original planning and feasibility study were to 1) survey existingresearch and data; 2) to begin to collect baseline environmental quality data; and 3) to build a knowledgebase in the community.

3

As described in more detail below, we achieved the following goals in Phases 1, 2 and 3 of thisstudy:

1. Renovation of the NYU-EPA mobile air pollution monitoring laboratory;2. Installation and calibration of air sampling/measurement equipment in the mobile laboratory;3. Preliminary study to survey the black carbon concentrations in the South Bronx for site

selection;4. Selection of the sampling sites for more detailed air pollution monitoring;5. Arrangement for the deployment of the mobile air pollution monitoring laboratory at the

sampling sites.6. Monitored criteria pollutants at 6 selected sites in the South Bronx, focusing on ambient

particulate matter (PM) concentrations;7. Established protocols and standing operating procedure (SOP) of X-ray fluorescent

spectroscopy (XRF) to measure elemental concentrations of PM in the South Bronx;8. Established a web site (http://www.nyu.edu/projects/southbronxhealth) to provide project area

residents with meaningful, time-relevant, reliable data about the levels of selected local airpollutants that are easy to understand and directly accessible to community residents.

9. Analysis of filter samples for PM2.5 mass, trace elements, specific organic compounds, EC andOC, and gases concentrations; and,

10. Implementation of a transportable aerosol time-of-flight mass spectrometer in June, 2003 todetermine single particle speciation.

This air pollution field monitoring survey work was also accompanied by community educationprograms conducted in association with the Wagner School and the collaborating South Bronx communitygroups.

These monitoring and education programs were then followed by a multi-year health study ofpersonal air pollution exposures to children with asthma in South Bronx elementary schools, and thispollution’s associated adverse health effects. A separate assessment of air pollution experienced withinconventional vs. hybrid city buses was also conducted in December, 2004. Results of these efforts are alldetailed in the following sections.

FIELD MONITORING ASSESSMENT RESULTS

Diesel Particle Pilot Study

The Aethalometer continuous carbon sampler was used in a pilot study to proved data to help indetermining the final sampling sites to commence in the summer of 2001. The Aethalometer was installedin a car equipped with a video camera to obtain preliminary data and traffic analysis to be used in theselection of the NYU-EPA van’s comprehensive sampling sites. In this work, the black carbonconcentrations were measured during the morning (6 am to 9 am) and evening (4 pm to 7 pm) traffic rushhours, as well as between 10 am and 1 pm for a period of two weeks at a variety of prospective vanlocations. The ambient air was drawn through a cyclone with a 2.5-micron size cut located on thepassenger side rear view mirror.

Sampling was performed at each location for approximately 5 minutes while the traffic was beingrecorded by the video camera. Figure 1 shows the preliminary sampling sites suggested by the community.As shown in Figure 2, these sampling sites were chosen in part according to the traffic density in the SouthBronx area, as well as local knowledge regarding nearby affected populations (e.g., elementary schools).The thickness of the lines in Figure 2 corresponds to the traffic density and the small numbers adjacent to

4

the lines were percentage of truck traffic in each route. These data were supplied by Dr. Jose Holguin-Veras, who is collaborating with Dr. Zimmerman of the Wagner School at NYU.

Figure 1. Exploratory Elemental Carbon Sampling Sites.

A. Noble Field Park (E. 179th St, Noble and BronxRiver Ave.

B . Corner of Bronx River Ave. and Cross BronxExpressway

C. Bronx River Ave. and E.174th St.D. I.S. 123 Morrison Ave. and Bronx River ParkwayE. Hunts Point Riverside Park: Lafayette Ave. and

Bronx River ParkwayF. Bronx River Ave and Story Ave.G. ABC Carpet: 1055 Bronx River Ave.H. Westchester Ave. and Whitlock Ave.I. (163rd St. and Bruckner Expressway: Hunts Point)J. I.S 52: 681 Kelly St.K. P.S. 39: Southern Blvd and LongwoodL. Hunts Point Ave. and Lafayette Ave.M. P.S. 48: 1290 Spofford Ave. and Coster St.

N. M.S. 201 School of Law and Eng: 770 BryantAve.

O. Lafayette Ave. and Edgewater RoadP. Hunts Point Ave and Food Center DriveQ. East Bay Ave and Cassanova Ave.R. Tiffany Street Pier: Tiffany St. and Viele Ave.S. 138th St. and Bruckner Blvd.T. P.S. 43: 135th St. and Brown PlaceU. P.S. 154: 135th St. and Alexander Ave.V. 138th St. and Grand ConcourseW. Hostos Community College: 149th St. and Grand

ConcourseX. Lincoln Hospital: 149th StreetY. 161st St. and Sheridan Ave.Z. 161st St. and Third Ave.AA. Corona Park: Control Area

Figure 3 shows the median concentrations of the preliminary black carbon datafor each site. As shown in Figure 4, the black carbon concentrations in the morning werehigher than that in the evening at all sites, likely due to poorer mixing of the atmospherein the morning hours. In addition, as shown in Figure 5, it is apparent that black carbonconcentrations were highest at sampling sites closest to high traffic area as depicted inFigure 2.

Figure 2. Heavy Duty Truck Traffic Densities in the South Bronx (numbers noted onroadways are Annual Average Daily Traffic, AADT, for both directions).

0

2000

4000

6000

8000

10000

12000

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA

Site ID

Con

cent

rati

on (

ng/m

3)

Weekdays AM vs. PM Median Black Carbon Concentration Across All Samples for the Different Sites

0

5000

10000

15000

20000

25000

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA

Site

Con

cent

ratio

n (n

g/m

3)

AMPM

Figure 3: Median Concentrations of Black Carbon (Soot) for Figure 4: AM Vs. PM Median Concentrations of Blackthe 27 Sampling Sites Carbon for the 27 Sampling Sites

6

Figure 5. Black Carbon Concentrations at selected sites in the South Bronx

Since Dr. Thurston maintained another sampling site at lower Manhattan (25th St.and 1st Ave.), we were able to compare the concentrations we measured in the SouthBronx with those obtained in Manhattan for the same period of time. Both PM2.5

(measured by TEOM) and black carbon (measured by R&P Instrument) concentrationswere available at the Manhattan sampling site. As shown in Figures 6 and 7, blackcarbon is only a fraction of PM2.5 in both the South Bronx and Manhattan. The blackcarbon concentrations in the South Bronx were considerably higher than those measuredin Manhattan, especially during the morning rush hours. PM2.5 and black carbonconcentrations increased during the weekdays, usually peaking on Thursday. However, itis important to note that two different methods were used to measure carbon in these twolocales (i.e. aethelometer vs. R&P methods), so that may have contributed to theapparently higher concentrations in the Bronx. Concentrations of these particlesdecreased dramatically during the weekends.

7

NYU-EPA Mobile Air Monitoring Laboratory Sampling

The NYU mobile air monitoring laboratory was first deployed, in July of 2001, inthe South Bronx to measure a comprehensive suite of air pollutants and meteorologicalvariables at multiple locations in the South Bronx for one month each. The mobile labsampled air pollution and weather data continuously. In this planning research, NYUworked very closely with the community groups, who provided information of the localenvironment and available resources (laboratory assistants, security guards, electricity,water, toilet facilities, etc). From the commencement of that study, we held regularmeetings (at least once a month) with these community groups. Furthermore, thecommunity groups actively participated in the selection of the air monitoring sites. Theyalso provided continuous support for the monitoring programs by working with the BronxBorough President’s office, NYC Park Commissioner, and local police and trafficofficers to assure that the monitoring sites were efficiently secured.

The results of our monitoring efforts showed that (see attached manuscriptpublished in the scientific journal “Atmospheric Environment”), although the mediandaily PM2.5 concentrations agreed within 20%, the median hourly EC concentrations werehigher at all South Bronx sites ranging from 2.2 to 3.8 mg/m3, compared to 1.0 to 2.6mg/m3 at Hunter College. Continuous aethelometer measurements at additional 27sampling sites in the South Bronx were conducted along major highways. ECconcentrations varied within each site, depending on time-of-day, with a large spatialvariability from site-to-site. Median EC concentrations varied from 1.7 to 12 mg/m3 onthe weekdays, and were lower (0.50 to 2.9 mg/m3) on the weekends. Elementalconcentrations were higher at all South Bronx sites than those at Hunter College for allmeasured elements but Ni and V, and at the Hunts Point, an industrial location, wereapproximately 2.5-fold higher. The average sum of 35 PAHs was 225 ng/m3, which is 4.5times larger than representative urban concentrations in Jersey City, NJ. Example PAHsdata that we measured in the South Bronx are shown in Figure 8, indicating that thesecompounds are at very elevated and easily detectable levels in this community. Amongthe individual PAHs, 3,6-dimethylphenanthrene had the highest concentrations, and theoverall PAH fingerprint differed from urban signal for Jersey City. Our data indicates that

Con

cent

rati

on (µ

g/m

3 )

S. Bronx Black Carbon

Manhattan Black Carbon

PM2.5 (Manhattan)

Weekend Weekend

0

10

20

30

40

50

60

4/24 4/25 4/26 4/27 4/28 4/29 4/30 5/1 5/2 5/3 5/4 5/5 5/6 5/7

Con

cent

rati

on (µ

g/m

3 )

S. Bronx Black Carbon

Manhattan Black Carbon

PM2.5 (Manhattan)

Weekend Weekend

0

10

20

30

40

50

60

4/24 4/25 4/26 4/27 4/28 4/29 4/30 5/1 5/2 5/3 5/4 5/5 5/6 5/7

Figure 6: Preliminary Morning Black Carbon (Soot) Figure 7: Preliminary Afternoon Black Carbon (Soot)Concentrations in the South Bronx vs. Manhattan Concentrations in the South Bronx vs. Manhattan

8

highways encircling the South Bronx are having a measurable adverse influence onresidents’ exposure to pollutants compared to other NYC areas.

Figure 8. Comparison of mean concentrations (ng/m3) of individual PAHs between thosemeasured by NYU at Crotona Park in the South Bronx vs. three-year average Jersey City,NJ. (Maciejczyk et al., 2004)

Single Particle ATOF-MS InstrumentationWe then conducted an intense campaign at the 163rd St. and Bruckner Ave. to

further characterize the air pollutants of South Bronx in detail using all of the instrumentsin the van, as well as a co-located aerosol time-of-flight mass spectrometer. This noveltransportable aerosol time-of-flight mass spectrometer (model RSMS-III, custom-built byA.S. Wexler, University of California at Davis) was be employed at the South Bronxduring this sampling period for continuous single-particle measurements of size andcomposition. In this instrument, particles pass through a laser beam that blasts eachparticle into positively and negatively charged ions. These ions are then analyzed in atime-of-flight mass spectrometer to determine the atomic or molecular weight of eachion. This instrument samples ambient aerosol in nine different sizes ranging from 40 nmto 1.5 mm. A result is that chemical and size information is obtained from individualparticles in real time. The sampling procedure involves collecting mass spectra ofindividual particles at a selected size. A single scan through the entire size range lastsabout 30 minutes.

As a result of this sampling we identified 1,133 particle classes with the Aerosoltime-of Flight Mass Spectrometer (ATOF-MS) that were grouped into 16 types(Maciejczyk et al. 2004). Looking at specific organic molecular markers by winddirections provides insight into the traffic specific tracers. Figure 9 shows “wind rose”plots of the data collected, indicating that much of the EC and the total OC tends to come

9

primarily from the direction of the nearest interstate, although other directions alsoaffected the site.

Figure 9. ATOF-MS data collected by NYU at Hunts Point in the South Bronx during2003, plotted as a function of wind direction.

COMMUNITY EDUCATION PROGRAMS

The SBEHPS team also reached out to the South Bronx community to address theresidents’ important environmental health issues that adversely impact men, women andchildren in the South Bronx community, including air pollution and metals, and to informthe residents the progress of the planning study. A Town Hall/Community Forum washeld on June 16th, 2001 at Eugenio Maria de Hostos Community College in the Bronx tocommunicate with the public about environmental risks in the South Bronx, and topublicize this study to the press and the community. The speakers and topics included:Mr. Luis Torres (Special Counsel for Environmental Justice: Congressman Serrano’sOffice), Moderator, Mr. Robert Williams (Sports Foundation), Asthma, Ms. YolandaGarcia (Nos Quedamos), Air Pollution/Moderator, George Thurston, ScD, Lead in theUrban Environment, Max Costa, PhD. (Director of the Nelson Institute of EnvironmentalMedicine), Gas and Diesel Generators, Mr. Mathy V. Stanislaus, Esq. (Director ofEnvironmental Compliance for Enviro-Sciecences Inc.), and C o m m u n i t yEmpowerment/Involvement, Ms. Alexie Torres-Flemming (Youth Ministries for Peace &Justice Inc.). As part of the agenda, Dr. Chen introduced to the community to the SouthBronx Air Pollution Study. Dr. Chen also presented the EPA Monitoring Van, whichwas parked outside of the college for a tour of the van and to answer any questions afterthe forum. Everyone who attended, including the TV News broadcasters were very

10

interested and enthusiastic with our efforts to help the South Bronx community gain theknowledge and encouragement they need to make a change. Reports about that studywere carried on WINS1010-AM (reportedly the most listened to radio station in theU.S.), WOR-TV (Channel 9), Bronx Net Cable TV, and in New York’s Newsdaynewspaper. In addition, a web site (http://www.nyu.edu/projects/southbronxhealth/) wasset up to provide project area residents with meaningful, time-relevant, reliable data aboutthe levels of selected local air pollutants that were easy to understand and directlyaccessible to community residents.

Youth Participation & Leadership Program

The objective of the Youth Participation and Leadership Program was to involvehigh-school students from the South Bronx in a training program focused onenvironmental science, policy and health. The six-week program ran from October 7,2003 to November 13, 2003 and took place at Hostos Community College in the SouthBronx on Tuesday and Thursday evenings from 3:30 PM to 6:30 PM. Students workeddirectly with NYU researchers, namely Carlos Restrepo (ICIS) and Jessica Clemente(NIEM), and Community Liaison Anthony Winn to learn about the scientific findings ofthe project and to gain research and communication skills that will help them becomeactive participants in the public health and well-being of their neighborhood. Thestructure of the Youth Participation & Leadership Program also allowed students toexplore career opportunities within the fields of public health and environmental science.

Youth Participation & Leadership Program Team Leaders:

Carlos Restrepo (NYU-ICIS), Curriculum Development, Workshop FacilitatorJessica Clemente (NYU-NIEM), Administration / Scheduling, Curriculum Development,Workshop FacilitatorAnthony Winn (Community Liaison / South Bronx Environmental Health Center),Student Recruitment, Curriculum Development, Workshop Facilitator

Curriculum Overview

In addition to drawing from the workshop facilitators’ own scientific research and workexperiences, instruction materials for the Youth Participation & Leadership Programwere compiled from a variety of sources, including:

• EPA’s Project A.I.R.E.• EPA’s “Air Pollution: What’s the Solution?” Program• American Lung Association’s “Open Airways” program• Various web pages and reading materials

Lesson plans addressed the issues of air quality, the sources of air pollution, theenvironmental health effects of air pollution, and environmental justice. A number ofspeakers from the U.S. EPA, the New York Environmental Justice Alliance, The PointCDC, Youth Ministries for Peace and Justice, NYU - Nelson Institute of Environmental

11

Medicine (NIEM), and NYU – Office of Federal Policy among others supported theefforts of the facilitators and presented a practical context for the field of environmentalhealth and policy. Students were also taught basic computer skills, with Microsoft Exceland Power Point, and were directed to use the Internet as a research and informationresource.

Mock HearingU.S. Senate Committee on Environment & Public WorksImpacts of Ambient Concentrations of Fine Particulate Matter on Air Quality

The culminating event for the Youth Participation & Leadership Program was a mockhearing, which required students to research, prepare, and present remarks before theU.S. Senate Committee on the Environment & Public Works. The topic of the mockSenate hearing was the current standard for ambient concentrations of fine particulatematter (PM2.5). Students were instructed that the U.S. EPA is required to revise allavailable information on the health outcomes and effects of each major pollutant everyfive years. If new evidence suggests the current standard is not adequate to protect humanhealth, the EPA revises the standard and makes it more stringent (i.e. lowers the currentlevel of allowed ambient concentrations of PM2.5).

Students, playing the role of scientists, presented information from the following fields ofexpertise, namely: atmospheric science, epidemiology, environmental engineering andenvironmental economics. Their presentations to the Senate committee focused on PM 2.5

trends, health effects of PM 2.5, technologies to reduce PM 2.5 emissions, and the costs andbenefits of reducing PM 2.5 emissions. These students made their 3-5 minutepresentations using Power Point and had prepared and practiced their remarks overseveral weeks with Youth Program facilitators. Supported by the latest research on airquality and drawing upon their own personal experiences, the scientists’ goal was toenlist the support of the Senate panel in making the current PM2.5 standard more stringent.

Other students played the role of legislative aides to Senators on the Committee onEnvironment & Public Works. Their preparation involved researching the environmentalvoting record of their Senator, figuring out how their Senator should vote given theirconstituency, and briefing the panel on their recommendations. Community leaders andstaff from the NYU – Office of Federal Policy played the roles of U.S. Senators and wereasked to make a decision, based on the cogency of the information presented to them.

The mock hearing was held at the Bronx Zoo on Tuesday, November 11, 2003. CherylSimmons-Oliver from Congressman José E. Serrano’s Office opened up the hearing witha few remarks and offered the support of the Congressman to the students’ achievementsas well as the overall work of the South Bronx Environmental Health and Policy Study.Local media coverage of the event was printed by the Bronx Times, Washington SquareNews, and NYU Today. Due to the success of the mock hearing, a similar format wasalso implemented for the Spring 2004 Youth Program session.

12

NYUSOM BACKPACK STUDY

Asthma Study Background

NYC has one of the highest rates of asthma mortality and morbidity in the U.S.(Weiss et al., 1990; Carr et al., 1992). Within this city, the South Bronx has among thehighest incidences of asthma hospital admissions. Indeed, a recent city survey of asthmain the South Bronx’s Hunts Point district found an asthma prevalence rate in elementaryschools of 21-23% (Leighton et al., 1999). Furthermore, the South Bronx has largevolumes of heavy vehicle traffic passing through it along several major highways (i.e.,Interstates 95, 87, 278, and 895), as well as over two dozen waste transfer facilities (thatconcentrate diesel trash hauling trucks from around the city), and at the Bronx Terminaland Hunts Point Wholesale Produce Markets. All of this traffic results in highconcentrations of truck activity and diesel engine emissions in the proximity of schoolsand residences in the South Bronx. At Hunts Point Market alone, some 12,000 trucksmove in and out daily. Indeed, it is very common to see schools, playgrounds, single-family housing, waste transfer stations, and large apartment buildings, all within a fewcity blocks of each other. This type of zoning is generally less common in other parts ofNYC, and the South Bronx represents one of NYC’s few remaining mixed-use zoneswhere residential neighborhoods come into such close contact with heavy truck traffic(Claudio et al., 1999).

The community’s concerns that living in this polluted environment has adversehealth impacts is a valid one, given the increasing body of evidence, mostly fromEuropean studies, indicating that traffic-related exposures and residential proximity tovehicular traffic are associated with various adverse health outcomes. These includeincreased respiratory conditions and symptoms in children, including increasedprevalence of asthma, wheezing, recurrent respiratory illnesses, and hospital admissionsfor asthma. However, a major limitation in these studies is that there has been a lack ofinformation on actual exposures of the subjects to traffic-related airborne particulatematter (PM), instead they have relied on surrogates of PM exposure, such as proximity toroadways. Through NYU SOM field campaigns in the South Bronx, this knowledge gapis reduced by measuring actual exposure levels for each individual via personalmonitoring.

PM, especially in the form of environmental tobacco smoke (ETS) and DPM, hasbeen suggested as an environmental trigger of asthma. Published epidemiological studiescollectively indicate that short-term exposure to PM air pollution can be associated withacute adverse human health effects (e.g. see U.S. EPA, 2004). In addition, the literaturealso suggests that asthmatics represent a subpopulation that can be especially affected byacute exposures to air pollution (e.g., see Koren and Utell, 1997). In particular,prospective epidemiologic studies of panels of individuals confirm the asthmaexacerbation-air pollution association. In a study in adult asthmatics in France over a six-month period, Diakite et al. (1994) found that asthma symptoms in mild and moderateasthmatics increased with increasing concentrations of carbonaceous PM (measured asblack smoke) and SO2. Forsberg et al. (1993) followed a panel of 31 asthmatic patients

13

residing in the town of Piteá in northern Sweden. Severe symptoms of shortness ofbreath, wheeze, cough, and phlegm were recorded in an asthma diary together withsuspected causes. Daily variations in the PM had significant effects on the risk ofdeveloping severe symptoms of shortness of breath. Ten subjects had at least fiveincident days with severe shortness of breath. In a wintertime study in Denver by Ostroet al. (1991) of some 207 adult asthmatics, it was found that cough was associated withPM2.5, and shortness of breath with sulfates. Pope et al. (1991) found, in a diary study ofa group of persons with asthma (8-72 yrs. of age), in an industrialized valley in Utah, thatthe probability of the use of asthma medication on the day with the highest dailyconcentration of thoracic PM (PM10) day, was six times that on the lowest pollution day.Roemer et al. (1993) found consistent positive associations between PM10 and SO2 airpollution and both wheeze symptoms and bronchodilator use by schoolchildren withchronic respiratory symptoms. Margolis et al. (1994) collected data on lung functionpeak expiratory flow rate (PEFR) for 95 non-smokers with asthma in Anaheim, CA,

finding that particulate sulfate (SO4=) was consistently related to decreased PEFR. Ostroet al. (1995) found that shortness of breath symptoms experienced by a panel of 83African-American children were associated with both PM and O3 experienced during thesummer of 1992 in Los Angeles, CA. Subsequently, Thurston et al. (1997), in a study ofsome 166 children attending an asthma summer camp over three summers, found that notonly are both increased respiratory symptoms and decreased lung functions significantlyassociated with summertime haze air pollution, but that the relative risk of an asthmaexacerbation requiring doctor-prescribed rescue medication administration rose byroughly 40 percent on the highest pollution day, relative to the average pollution day inthis group of children with asthma. Thus, past epidemiologic studies have indicated thatexposure to ambient air pollution, and often PM in particular, can be associated withacute exacerbations of asthma.

While the PM associations with adverse health effects among asthmatics andothers are well documented, the type/source(s) of PM most associated with adversehealth effects are not known at this time. One suspected causal PM agent, especially forthose with asthma, is PM of diesel engine combustion origins. Indeed, initial in vitroresearch results suggest that short-term exposures to DPM can act as adjuvants in theimmune response, and may lead to the enhancement of allergic inflammation (e.g., Diaz-Sanchez et al., 1997, Kobayashi, 2000). Also, several epidemiological studies in Europehave found relationships between proximity to traffic flow and increased risks ofchildhood hospital admissions (Edwards et al., 1994), respiratory symptoms (Wjst et al.,1993; Duhme et al., 1996), and decreased lung function (Brunekreef et al., 1997: seeFigure 10). In such studies, data on traffic flow were obtained by self-report, using thehighest traffic volume in the school district, by assigning traffic density derived frommeasured distances from residences to streets on maps, or by linking traffic flow data toresidential postal zip codes. Similarly, an analysis of traffic and asthma incidence andprevalence in San Diego, CA used the Geographic Information Systems (GIS) andexplored whether childhood residence near busy roads was associated with asthma. Noassociation was found between asthma prevalence and traffic proximity (English, 1999).But those with asthma residing near high traffic flows (measured at the nearest street)were more likely to have acute medical visits than those residing near lower traffic flows,

14

suggesting that exposure to traffic exhaust may aggravate asthmatic symptoms inindividuals. In a recent California study, McConnell et al. (2006) examined therelationship of local traffic-related exposure and asthma and wheeze in southernCalifornia school children (5-7 years of age), finding a higher risk of asthma near a majorroad that decreased to background rates at 150-200 m from the road. An analysis ofbronchitis prevalence in schools near San Francisco by Kim et al. (2004), found arelationship with Black Carbon, but not PM2.5 in general. Black Carbon is likelydominated by traffic emissions in California, but has various sources, and is notnecessarily specific to diesel pollution there. However, such traffic studies have not yettested whether personal exposures to diesel PM are themselves directly associated withthe adverse asthma effects being found to be associated with traffic proximity.

1,860

1,920

1,980

2,040

2,100

7500 9500 11500 13500 15500 17500 19500

Truck Traffic Density

Lun

g F

unct

ion

in li

ter

FEV1

Figure 10. Lung function has been found to be associated with truck traffic density (Brunekreef, 1997).

The presence of elemental carbon (EC) in the lung has now been directlyassociated with adverse effects, supporting the hypothesis that it is the DPM that iscausing the associations of traffic with asthma exacerbations. Kulkarni et al. (2006) justreported in the New England Journal of Medicine (NEJM) that they were able to assessthe carbon content of airway macrophages in 64 of 114 healthy children (56 percent).Each increase in primary PM10 of 1.0 µg per cubic meter was associated with an increaseof 0.10 µm2 (95 percent confidence interval, 0.01 to 0.18) in the carbon content of airwaymacrophages, and each increase of 1.0 µm in carbonic content was associated with areduction of 17 percent (95 percent confidence interval, 5.6 to 28.4 percent) in forcedexpiratory volume in one second, of 12.9 percent (95 percent confidence interval, 0.9 to24.8 percent) in forced vital capacity (FVC), and of 34.7 percent (95 percent confidenceinterval, 11.3 to 58.1 percent) in the forced expiratory flow between 25 and 75 percent ofthe FVC. The authors concluded that: “There is a dose-dependent inverse associationbetween the carbon content of airway macrophages and lung function in children.” Theyalso conclude that they “found no evidence that reduced lung function itself causes anincrease in carbon content.” Thus, it appears that respiratory doses of EC have been

15

directly related to lung function changes, but the source of these EC exposures is not yetdetermined.

In NYC, EC variation is an extremely reliable indicator of traffic-generatedpollution. Published NYU data for the South Bronx included EC concentrations rangingfrom 1.7 to 12 ug/m3 on weekdays and from 0.50 to 2.9 ug/m3 on the weekends(Maciejczyk et al. 2004). Ground level concentrations and rooftop monitors at PS154 inthe S. Bronx were compared to those at Hunter College on E. 25th street in Manhattan,and were found to be higher for both PM2.5 and EC. The median hourly BCconcentrations were higher at all South Bronx sites ranging from 2.2 to 3.8 ug/m3,compared to only 1.0 to 2.6 ug/m3 at Hunter College. In another study in Harlem(another NYC community with high rates of asthma and traffic volumes), the authorsfound that PM2.5 concentrations exhibited only modest site to site variation, while ECconcentrations varied 4-fold across sites and were associated with bus and truck countson adjacent streets (Kinney et al. 2000). Variations in EC concentrations on sidewalks inthe South Bronx were also shown to be related to local truck traffic density, and wereparticularly elevated in Hunts Point (Lena 2002), indicating that the South Bronx is ahotspot for both PM2.5 and EC, with EC more directly correlated with local trafficpatterns.

Backpack Study Design and MethodsNYU SOM investigated the documented link between acute adverse health effects

and exposure to surrogates for traffic-generated pollution by directly measuring personalpollution exposure. In order to test our hypothesis that exposure to EC, a component ofPM closely associated with diesel exhaust in the area, is a causal agent in the adversehealth effects seen in the studies cited above we conducted four field sampling campaignsduring Spring 2002, Spring 2004, Fall 2004 and Spring 2005 at each of four South Bronxschools (PS154, MS302, CS152, MS201) located at varying distances from the Bronx’shighways.

Ten children with asthma from each school were followed for approximately onemonth each. The students moved about with a rolling backpack containing air pollutionmonitoring equipment attached. Two of the schools that participated were locatedadjacent to Interstate highways, and two schools were located several city blocks distantfrom the major Interstates. 24-hr filters with a PM2.5 inlet were collected for eachparticipant. Personal exposure measurements also included continuous measurements ofPM by the MIE Personal Data Ram (pDR), which uses optical scattering to measureparticles in the sub-micron size range. A parent or guardian completed backgroundquestionnaires on home environment and child’s asthma history, and each subjectcompleted an asthma baseline evaluation. NYU personnel met with the students twice aday during weekdays to check activity and asthma symptom diaries, and to change filtersand download personal pollution data.

At each school site, sampling by the NYU-EPA PM Center mobile laboratory vanprovided a central monitoring site for ambient monitoring outside the school of PM2.5,O3, NOx, CO, and SO2. The van collected PM2.5 daily filter data as well as continuoushourly measurements of the gases and a semi-continuous index of PM2.5 mass (R&PTEOM). An aethelometer tape sampler provided additional central site measures of ECand OC ambient concentrations. Substantial state air monitoring data and meteorologicaldata was also collected, since two of the four schools have NY DEC monitoring stations

16

located on their roofs. Data were collected during spring and fall seasons, avoidingpossible effects of summer haze air pollutants such as ozone and acidic aerosols. Aftersampling, the 24-hr filters collected were analyzed for PM2.5 mass and Elemental Carbonconcentration (EC). In urban New York City, EC is primarily from diesel soot, and is agood indicator of exposure to pollution from diesel trucks.

We confirmed and calibrated the equivalence of measuring Black Smoke (BS)reflectance as an index of EC in this South Bronx locale. The samples were collectedsimultaneously with the personal exposure samples at the NYU van and, therefore, thePM2.5 collected represented local PM2.5 concentrations and composition during theexposure study. A calibration curve of 24 hour Harvard Impactor PM2.5 samples wasconstructed for EC determination, according to the methods of Kinney et al. (2000).Samples were collected with three co-located PM2.5 Harvard Impactors (Air Diagnostics,Harrison, ME, USA) on 37 mm Teflon filters (Andersen) or on 37mm quartz fiber filters(SKC Inc., Eighty Four, PA, USA). The impactor samplers were placed on the roof of theNYU van located at the corner of Alexander Ave. and 135th Street in the South Bronx,adjacent to the Major Deegan Expressway (I-87). “Sets” of samples were thereforecomposed of 2 Teflon filters and one quartz fiber filter. The reflectance of each filter wasmeasured three times with a Smokestain Reflectometer M434 (Diffusion Systems LTD,London, UK) just after they had been weighed. The reflectance of 10 lab blanks, 6 fieldblanks, and 20 sampled filters from the South Bronx was measured ten times for qualityassurance purposes. For direct EC determination, the quartz fiber filters were sent toSunset Laboratory (Hillsborough, NC) for EC analysis. The samples were analyzed bothwith the NIOSH and the IMPROVE method. EC concentrations were measured by anAethalometer (Andersen, RTA9 Series, with a PM2.5 inlet; 10 minute averaging time),adjacent to the PM2.5 Harvard Impactors.

The Black Smoke reflectance-EC calibration curve is shown in Figure 11. Thereflectance in the EC calibration curve is expressed in a standardized absorptioncoefficient (ISO9835, 1993):Absorption Coefficient = (0.5 x A x ln (R0/ Rf))/V

where: A = soiled filter area [m2]

R0 = reflectance value of clean filter [%]

Rf = reflectance value of soiled filter [%]

V = volume of air sampled [m3]

17

y = 1.25x

R2 = 0.99

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50

NIOSH EC Concentration [ug/m 3]

Ab

sorp

tio

n C

oef

fici

ent

*10-5

[m-1]

Figure 11. Black smoke reflectance absorption readings versus NIOSH ECmeasurements of simultaneous South Bronx filters.

Characterization of Backpack Study Pollutant Exposure Levels

Particle air pollution levels at each of the four South Bronx school sampling sitesshow high variation both temporally and spatially. At all school sites, personal EC levelsmeasured in the study were lower than ambient levels, indicating that exposure to EC isfrom outdoor sources, of which diesel trucks are the predominant source. The oppositetrend was found for overall PM2.5 levels, which were lower outdoors than measured bythe personal monitors. Personal levels were weakly correlated with, and usually higherthan, central site PM2.5, suggesting indoor sources contribute significantly to PM2.5

exposures. In the pilot school, PS154, personal PM2.5 concentrations were higher for themost ETS (Environmental Tobacco Smoke) exposed children. In the following schools,we targeted children from non-smoking households to avoid heavy ETS exposure.Personal EC levels were also variable in the pilot school, but more strongly correlatedwith (r2=0.72), and usually lower than, central EC, suggesting outdoor sources dominatedpersonal EC exposures. Unlike PM2.5, EC exposures were similar for ETS vs. non-ETSaffected subjects, providing further evidence that outdoor sources, predominantly dieseltrucks, dominate personal EC levels. At PS154, located adjacent to the Major Deegan,stationary EC concentrations were lower by almost a factor of two during the weekend ascompared to the weekdays.

We found high inter and intra-subject variability in the personal pollutionmeasurements. Daily personal PM2.5 concentrations ranged from 1.2 – 167 ug/m3 acrossall of the subjects from all of the schools, with the highest PM exposures from ETSexposed subjects from the pilot school. Daily personal exposure levels of EC rangedfrom 0.11 – 14.3 ug/m3 during the study. Of the individual daily measurements, 1– 35%

18

of the total PM2.5 mass was composed of EC, though overall, on a daily basis, ECcomposed only about 6% of total PM exposure. These trends reflect the high temporaland spatial variability of PM pollution in the area, particularly the levels of EC. The veryhighest personal EC levels were likely from cooking, since this was indicated in the time-activity diaries. Outdoor levels of EC were higher than personal levels at all fourschools, with mean daily concentrations ranging from 2.0-3.2 ug/m3 across the schools,and mean daily personal EC levels ranging from 1.6-2.4 ug/m3 (Table 1). Mean ambientconcentrations of total PM2.5 mass at ground-level ranged from 11-19 ug/m3 across thefour schools, while mean personal PM2.5 mass was higher than outdoor levels, rangingfrom 26-32 ug/m3 (Table 1).

Table 1: Pollutant levels across all four South Bronx schoolsPS154* CS152* MS302 MS201

Season Spring Fall Spring SpringPersonal† PM2.5

(ug/m3)31 26 32 32

Personal EC†

(ug/m3)1.8 1.7 2.4 1.6

Van PM2.5

(ug/m3)14 11 19 11

Van BC (ug/m3) 3.2 2.8 3.1 2.0 *Schools closest to highways † Weekdays only

Outdoor ground level concentrations as measured on an hourly basis by the NYUvan were highly variable and did not differ significantly based on proximity of the schoolto the highway (Figure 12). All schools had median hourly Black Carbon concentrationsabove 2 ug/m3, with the exception of MS201, a school set farther from the highways thathad the lowest EC levels of the four schools sampled. These concentrations are abovepreviously measured carbon levels at Hunter College in Manhattan. Minimum hourlycarbon concentrations across the four schools ranged from 0-0.49 ug/m3 and maximumconcentrations ranged from 7.2-26.5 ug/m3. Up to 75% of the PM was composed of ECat a resolution of hourly measurements.

Figure 12: Hourly concentrations of EC and total PM2.5 varied within school samplingsite and across schools.

19

Although the highest ground-level PM2.5 levels, (19 ug/m3 mean concentration),were found at one of the schools farther from the highways, MS302, this school also hadthe highest background PM2.5 levels, and ground-level PM was actually lower than at thePS154 DEC rooftop monitoring station. Comparing pollution levels measured by theNYU van at the school field sites with the levels measured by the DEC monitor on theroof of PS154 and the DEC monitor at Kew Gardens (a site in Queens that serves as aNYC background site), shows that all school sites, and the South Bronx area in general,had only slightly higher overall PM levels than the Kew Gardens site during the onemonth sampling period (see Table 2). Much of the PM pollution in the NYC area isfrom transported (non-local) aerosols, accounting for this trend.

Table 2: PM levels during school sampling periodsPS154* CS152* MS302 MS201

NYU Van 14 11 19 11PS154 DECmonitor

13 10 21 11

Kew GardensDEC monitor(reference site)

11 9.4 18 10

*Schools closest to highways

Although each school’s proximity to the highway was not a reliable indicator ofoverall personal exposure to either PM or EC over time, proximity to the highways showsa differential impact on the overall fraction of EC present in the PM2.5 total mass (Figure13). We expected to see high concentrations of traffic-related air pollution downwind ofthe major highways, however, high EC levels were found at all of the schools measured,since this area is so beset by Interstates and truck traffic. In the schools monitored thatwere closer to the major highways, the overall fraction of EC in PM2.5 was greater.Although absolute levels of pollution varied across schools and time periods, for the twoschools that were closest to high levels of truck traffic, PS154 and CS152, the fraction ofEC mass was about 30-50% higher than at the schools farther away from traffic

Figure 13: Traffic impact is the highest fraction of measured PM2.5 at schoolsbeside the Bronx’s highways (30-50% higher). Grey bars beside the pie chartsindicate mean PM2.5 levels in ug/m3 from the NYU van over the study period.

20

Traffic contribution to pollution exposure

Particle pollution exposure from traffic in the Bronx was assessed via directtraffic counting at the pilot school. PS154 was selected for the pilot study as it wasadjacent to the Major Deegan Expressway (Interstate 87), one of the major trafficthoroughfares in the south Bronx and because there was a New York State Department ofEnvironmental Conservation (NYSDEC) air quality monitoring station on it’s roof. NYUwas able to collect one week’s worth of traffic data for this location to determine therelationship of traffic counts to pollutant levels. It was clear that variations in EC levels,together with variations in NOx, an indicator of fresh traffic exhaust, matched closely thevariation in hourly truck counts (Figure 14).

Figure 14: Elemental Carbon (soot) pollution was closely related to hourlyvariations in highway truck traffic. NOx levels are an indicator of freshtraffic exhaust.

Using a modeled traffic exposure variable, we determined the contribution oftraffic-generated PM pollution to overall PM2.5 exposure at the pilot school, PS154. Thedispersion model used for this analysis is a recent formulation developed by Dr. Holguin-Veras. This approach assumes that of an isotropic pattern of diffusion of the air pollutionis generated by the traffic. The model determines the effect that a given number ofvehicles going through a traffic link would have on a child on a given location i, byconsidering the coordinates of two ends of the traffic link, the coordinates of the child,wind direction and speed (For model details, see Spira-Cohen et al. (2004)). Usingfifteen-minute intervals to match the traffic data totals, a time-evolving relationship ofexposure to pollution from traffic throughout the day was determined. The resultsprovide a dispersion-adjusted index of 15-minute traffic counts.

21

Regression analysis was used to test the association of PM2.5 with traffic densityat PS154 by using Dr. Holguin-Veras’ dispersion-adjusted index of traffic counts. Forthis analysis, the dispersion-adjusted index by vehicle class was collapsed from theoriginal 13 class categories into two categories: cars and trucks. Simple and multipleregressions were performed of the dispersion adjusted traffic exposure index on thevarious PM2.5 concentrations from the different sources of measurement, one personal:the pDR; and the others stationary: the van and NY DEC TEOMs, which serve to verifythe source. Only measurements during school hours (9:00 AM DST to 3:00 PM DST)from May 13th to May 17th were included. An average of PM2.5 concentrations from therooftop and the van TEOMs on an hourly basis was used for this regression, and hourlyaverages from the pDR data were also regressed with the model output. In the multipleregressions, the truck traffic index remained significant, while the car traffic index didnot.

The results of the regression analysis indicate that almost half of the children’sPM2.5 exposure was attributed to truck traffic. Using personal exposure measurements(pDR data), 49±32% (95% CI, p<.05) of PM2.5 was attributed to traffic on the MajorDeegan Expressway and other streets near the school used in the traffic-density exposuremodel. Using stationary monitoring site (TEOM) data, the percent mass was similar tothat calculated using the pDR data: 44±13% (95% CI, p<.05) of the total PM2.5

concentration (Figure 15). Finding similar associations using PM2.5 mass data from theoutdoor TEOM sources at the school validates the influence of traffic emissions on thechildrens’ exposure. These figures probably encompass some background particlepollution from other traffic emissions on other Bronx roadways that are also correlatedwith traffic flow closest to the school. Still, the major findings of our analysis supportevidence that PM2.5 exposure is heavily influenced by truck traffic in the vicinity of thiselementary school.

Figure 15: Association of Personal PM2.5 Exposure and Outdoor PM2.5 at PS154 withTruck Traffic Index

Health effects analysis

An analysis across the four schools sampled indicates that exposure to PMpollution, particularly the EC fraction, exacerbates asthma. Exposure to EC was linked todecreased lung function and an increase in asthma symptoms. Mean daily personal PM2.5

22

mass and mean daily personal EC concentrations were regressed on mean daily lungfunction metrics, and mean scores of daily cough and wheeze. Mean pollution valuesshow a significant association with both lung function and asthma symptoms. Anincrease in mean personal pollution exposure yielded significant decreases in lungfunction measures, and an increase in asthma symptoms. Afternoon “peak flow” (PEF)and FEV1 values, two lung function metrics, were significantly associated (p<0.05) withexposure to EC (Figure 16). EC was also significantly associated (p<0.05) with cough,wheeze and shortness of breath (Figure 17). The findings represent an increase infrequency and/or severity of cough, wheeze and shortness of breath with increasing ECexposure. In fact, this severity-related index of asthma symptoms nearly doubles on hightraffic (EC) pollution days.

Figure 16: Afternoon lung function metrics, PEF and FEV1, were more negativelyassociated with EC than with PM2.5 total mass.

23

Figure 17: Daily symptoms nearly doubled in severity with increasing exposure toEC than with PM2.5 total mass

Our results show significant associations between personal EC exposure and threekey asthma symptoms, and between EC exposure and lung function. Although the slopesgenerally followed the same direction when PM2.5 mass instead of EC was regressed onthe health outcomes, the associations were not significant for total PM2.5 mass.Therefore, daily symptoms and lung function analyses results indicate that traffic sootparticles, a large contributor to EC concentration in the South Bronx, is more related toasthma symptoms than PM2.5 overall (Figures 16 & 17). We also conducted regressionanalyses of the asthma effects after controlling for the possibly confounding effects ofboth ozone and temperature on the PM-health effects relationships. Even afterconsidering the influences of both Temperature and Ozone concentration, the adverse

24

effects of EC on the children’s asthma symptoms and lung function remain significant.Sampling was conducted during spring and fall periods to avoid significant exposure toozone.

The primary conclusions from this study indicate that traffic is a major source ofparticle pollution exposure for children in the South Bronx. This exposure is of concernto the South Bronx community’s health, as we found significant associations ofincreasing personal pollution exposure with adverse health outcomes in this group ofchildren with asthma.

Hybrid vs. Conventional Bus Exposures of Passengers to Diesel Soot

Background

Diesel combustion pollution, such as that from buses and trucks, is one suspectedcausal particle pollution source component in the overall PM-health effects associations.This is especially true in densely populated urban areas, where the almost exclusive useof diesel buses for school and transit buses can be a major contributor to local PMpollution. Published in-vitro research results suggest that short-term exposure to dieselexhaust particle air pollution can act as adjuvants in the immune response and may leadto the enhancement of allergic inflammation (e.g., Diaz-Sanchez et al., 1997, Kobayashi,2000). Also, several epidemiological studies in Europe have found a relationshipbetween proximity to traffic flow and increased risks of childhood hospital admissions(Edwards et al, 1994), respiratory symptoms (Wjst et al, 1993; Duhme et al., 1996), anddecreased lung function (Brunekreef et al, 1997). Moreover, the U. S. EnvironmentalProtection Agency and the International Agency for Research on Cancer have alsoconcluded that there is considerable evidence of an association between exposure todiesel exhaust and an increased risk of lung cancer (IARC, 1989; US EPA, 2002). Thus,exposure to diesel particles has been indicated by available research to date to be asignificant health risk.

Aside from their inherent toxicity, diesel buses and trucks are also a majorcontributor to the total mass of PM emitted in general, both in the ambient environmentand when traveling in vehicles such as buses. Diesel buses and trucks account for over100,000 tons per year of PM2.5 in the U.S., or some 70% of all fine particles directlyemitted by on-road vehicles in the U.S. (U.S. EPA, 2003). Diesel fuel use has increasedrapidly in recent years, rising some 45% during the 1990’s (EHHI, 2002). As shown inFigure 18, most of the diesel particle mass is as black elemental carbon. However, theworst impacts of this vehicular pollution are not necessarily to the people outdoors, but tothe passengers in vehicles. Most of the nation’s school bus fleets run on diesel fuel. Arecent study of diesel pollution exposures in school buses found that children ridinginside of a diesel-powered school bus may be exposed to as much as 4 times the level ofdiesel exhaust as someone riding in a car just ahead of the bus (NRDC, 2001). An evenmore recent study by Sabin et al. (2004) compared PM concentrations in a conventionaldiesel school bus vs. a compressed natural gas (CNG) fueled bus. Pollutants weresignificantly higher inside conventional diesel buses compared to the CNG bus. Meanblack carbon, PB-PAH, benzene and formaldehyde concentrations were higher when thewindows were closed, compared with partially open, in part, due to intrusion of the bus’sown exhaust into the bus cabin, as demonstrated through the use of a tracer gas added to

25

each bus’s exhaust. These same pollutants tended to be higher on urban routes comparedto the rural/suburban route, and substantially higher inside the bus cabins compared toambient outdoor measurements. These studies show that riding in diesel buses results inelevated exposures to diesel PM by passengers. While no such comprehensive study oftransit buses has yet been reported in the literature, the available studies, combined withour own recent bus sampling conducted in conjunction with the NYC Transit Authority,indicate that conventional diesel transit buses are a significant source of particulate matterexposures, especially for urban dwellers, who more often travel by bus than non-urbandwellers.

Figure 18. Diesel Vehicle Exhaust Particle Mass Composition (U.S. EPA, 2002)

The Bronx, one of the NYC boroughs, has over 70% non-white population (vs.25% for the USA as a whole), and a high poverty rate: the per capita income of Bronx isthe lowest in NYC, and 31% of the families in the Bronx live in poverty (vs. 12 percentfor the USA as a whole) (U.S. Bureau of the Census, 2000).

DIESEL BUS PM2.5 SAMPLING METHODS

With the cooperation of the NYC Transit Authority, we have conductedpreliminary measurements of the fine particle mass and number concentration in aconventional diesel transit bus and a hybrid-electric transit bus. New York City is in theprocess of transitioning from old conventional bus to hybrid buses that reduce fuel use,and also have PM traps, which reduce particle emissions. While many are aware of theenergy and outdoor pollution benefits associated with the hybrid buses, most are unawareof the potential PM exposure benefits to inner-city riders from reductions in pollutionexposures inside the bus. This has been documented in the literature for school buses,but not for transit buses, so we conducted a pilot study of two buses in the NYC Transitsystem to see if reductions in emissions outside the bus translated into lower exposures toriders inside the bus. We were given access to dedicated buses by the NYC TransitAuthority for these tests, which in itself demonstrates the cooperation that we have fromthe NYC Transit for our research efforts. Each bus rode the same route on the same day(December 9, 2004), although not at the same time of day.

26

Two portable MIE DataRAMs and pTrak samplers were used to measure fineparticle mass and number concentrations, respectively, in both the front and back of theconventional and hybrid buses, and both were also were carried outdoors at bus stops.

DIESEL BUS PM2.5 SAMPLING RESULTS

As seen in Figure 19, the PM mass concentrations measured in the conventionaldiesel bus were noticeably higher than for the hybrid bus. Also, concentrations at thefront of the bus (light line) were generally slightly higher than in the back of the bus(dark line). The hybrid bus cut the PM2.5 exposures to riders by roughly half, ascompared to the conventional diesel bus. The overall front/back mean PM2.5

concentration (excluding outdoor measurements) in the conventional diesel bus averaged40.1 ug/m3 (SE=4.2 ug/m3), while PM2.5 concentrations in the hybrid bus averaged 18.1ug/m3 (SE=0.4 ug/m3), or less than half the conventional bus levels (See Table 3). It isnotable that the more than 50% reduction in exposure in the hybrid bus was measured inspite of the fact that the outdoor PM2.5 concentrations measured during the hybrid bus run(in the afternoon) averaged some 7 ug/m3 higher than during the conventional diesel busmeasurements (in the morning), which would tend to offset some of the hybrid PM2.5

benefits measured, and cause an underestimation of the exposure reduction measured inour pilot comparison. Comparison between the inside bus with the outdoorconcentrations at the time (noted as “stars” in Figure 19) are also presented.

Table 3. Comparison of Mean (and Standard Error) of PM Concentrations MeasuredInside a Conventional Diesel Bus vs. a Hybrid-Electric Bus on December 9, 2004

Bus Type PM2.5 MassConcentration

(ug/m3)

Black CarbonConcentration

(ug/m3)

Ultra-Fine ParticleNumber Conc.

(#/cm3)

Conventional Diesel 40.1 ug/m3

(SE=4.2)11.6 ug/m3 (SE =0.4)

337,000/cm3 (SE =49,000)

Hybrid-ElectricDiesel

18.1 ug/m3

(SE=0.4) 4.2 ug/m3 (SE =0.2)

71,000/cm3 (SE =2,000)

27

The diesel pollution exposure reductions achieved by the hybrid buses are evenmore clearly shown by the black carbon (BC) measurements we also made in the frontand back of each bus, as displayed in Figure 5. The mean BC concentration for theconventional diesel was 11.6 ug/m3 BC (SE = 0.4 ug/m3 BC), versus only 4.2ug/m3 BC(SE = 0.2 ug/m3 BC). These results indicate that the Black Carbon concentration (abetter indicator of diesel particles than PM2.5 mass) inside a conventional diesel bus isapproximately three times the levels found inside a hybrid-electric diesel bus.

Figure 19. Comparison of PM2.5 mass concentrations measured inside a conventional diesel transit busversus a hybrid-electric diesel transit bus.

Figure 20. Comparison of Black Carbon concentrations measured inside a conventional diesel transit busversus a hybrid-electric diesel bus

28

The number concentration data indicate the greatest improvement in pollutioninside the hybrid-electric diesel bus versus a conventional diesel bus, as shown in Figure6. The mean indoor number concentration recorded inside the conventional bus was337,000 per cm3 (SE = 49,000), while the mean number concentration recorded inside thehybrid diesel bus was 71,000 per cm3 (SE = 2,000). In most cases, the particle numberconcentration was actually higher outdoors than inside the hybrid bus, while the oppositewas true for the conventional bus. Overall, these results indicate that ultrafine particleconcentrations inside of conventional diesel buses are roughly five times theconcentrations seen in a hybrid diesel bus.

Figure 21. Particle concentrations measured in a conventional diesel bus versus a hybrid-electric bus.

29

REFERENCESBrunekreef B, Janssen N, de Hartog J, Harssema H, Knape M, Van Vliet. P. (1997) Airpollution from truck traffic and lung function in children living near motorways.Epidemiology 8:298-303.

Carr W, Zeitel L, Weiss K. (1992) Variations in asthma hospitalizations and deaths inNew York City. Am J Pub. Health 82:59-65.

Claudio L, Tulton L, Doucette J, Landrigan PJ. (1999). Socioeconomic factors andasthma hospitalization rates in New York City. J Asthma. Jun;36(4):343-50.

Diakite B, Aubier M, Korobaeff M, Quenel P, LeMoullec Y, Neukirch F. (1994)Relationships of air pollution levels to asthma symptoms and peak flow variabilityrecorded daily during six months. Preliminary results. Am J Respir Crit Care Med;149(4):A662.

Diaz-Sanchez, D; Tsien, A.; Fleming, J; Saxon, A. (1997). Combined diesel exhaustparticulate and ragweed allergen challenge markedly enhances human in vivo nasalragweed-specific IgE and skews cytokine production to a T helper 2-type pattern. J.Immunol. 158 (5): 2406-2413.

Duhme H, Weiland SK, Keil U, Kraemer B, Schmid M, Stender M, Chambless L. (1996)The association between self-reported symptoms of asthma and allergic rhinitis and self-reported traffic density on streets of residence in adolescents. Epidemiology 7:578-582..

Edwards J, Walters S, Griffits RK. (1994). Hospital admissions for asthma in preschoolchildren: relationship to major roads in Birmingham, UK. Arch Environ Health 49:223-227.

Environmental and Human Health Inc. (EHHI) (2002). Children’s exposure to dieselexhaust on school buses. Available at: http://www.ehhi.org/reports/diesel/summary.htm

English P, Neutra R, Scalf R, Sullivan M, Waller L, and Zhu, L (1999). ExaminingAssociations between Childhood Asthma and Traffic Flow Using a GeographicInformation System. Environmental Health Perspectives Volume 107, Number 9,September 1999

Forsberg B, Stjernberg N, Wall S.(1997) People Can Detect Poor Quality well belowguideline concentrations: a prevalence study of annoyance reactions and air pollutionfrom traffic. Occupational Environ Med 54(1): -48.

International Agency for Research on Cancer.(IARC) (1989):Monographs on theEvaluation of Carcinogenic Risks to Humans: Diesel and Gasoline Engine Exhausts andSome Nitroarenes. Vol 46. Lyon, France: International Agency for Research on Cancer;458 pp

30

Kim JJ, Smorodinsky S, Lipsett M, Singer BC, Hodgson AT, Ostro B. (2004) Traffic-related air pollution near busy roads: the East Bay Children’s Respiratory Health Study.Am J Respir Crit Care Med. 2004 Sep 1;170(5):520-6.

Kinney PL, Aggarwal M, Northridge ME, Janssen NAH, Shepard P. (2000) Sidewalkexposures to PM2.5 and diesel exhaust particles in Harlem. Environ Health Perspect108:213–218.

Kobayashi T. (2000) Exposure to diesel exhaust aggravates nasal allergic reaction inguinea pigs. Am J Respir Crit Care Med 162 (2 Pt 1): 352-356.

Koren HS, Utell MJ. (1997) Asthma and the environment. Environ Health Perspect..105:534-537.

Kulkarni N., Pierse N, Rushton L, and Grigg J. (2006) Carbon in Airway Macrophagesand Lung Function in Children. N Eng J Med. Volume 355: 21-30. July 6, 2006

Leighton J, Matte T, Findley S et al. (1999) Asthma prevalence among school children ina Bronx community. Asthma Conference, February 9, 1999. Atlanta, GA: Centers forDisease Control and Prevention. p. 26.

Lena TS, Ochieng V, Carter M, Holguin-Veral J, Kinney PL. (2002) Elemental carbonand PM2.5 levels in an urban community heavily impacted by truck traffic.Environmental Health Perspectives. 100(10):10009-15

Maciejczyk, P.B., Offenberg, J.H., Clemente, J., Blaustein, M., Thurston, G.D., andChen, L.C. (2004) Ambient pollutant concentrations measured by a mobile laboratory inSouth Bronx, NY. Atmos. Environ. 38:5283-5294.

Margolis HG, Colome SD, Westerdahl FD, Lebowitz MD. Pulmonary function andsymptomatic responses of asthmatics to acidic ambient air. Paper #36. Presented atISEE-ISEA Joint Conference, Research Triangle Park, NC, Sept. 1994

McConnell R, Berhane K, Yao L, Jerrett M, Lurmann F, Gilliland F, Kunzli N,Gauderman J, Avol E, Thomas D, Peters J. (2006) Traffic, susceptibility, and childhoodasthma. Environ Health Perspect. May;114(5):766-72.

Natural Resources Defense Council (NRDC) (2001). No breathing in the aisles: Dieselexhaust inside school buses. January 2001, New York, NY. Available at:http://www.nrdc.org/air/transportation/schoolbus/sbusinx.asp

Ostro BD, Lipsett MJ, Wiener MB, Selner JC. (1991) Asthmatic response to airborneacid aerosols. Am J Pub Health. 81:694-702.

31

Ostro, B.D., M.J. Lipsett, J.K. Mann, H. Braxton-Owens, M.C. White. (1995). Airpollution and asthma exacerbations among African-American children in Los Angeles.Inhalat. Toxicol. 7:711-722.

Pope CA III. (1991) Respiratory hospital admissions associated with PM10 pollution inUtah, Salt Lake, and Cache Valleys. Arch Environ Health. 46:90-97.

Roemer W, Hoek G, Brunekreef B. (1993) Effect of ambient winter air pollution onrespiratory health of children with chronic respiratory symptoms. Am. Rev. Respir. Dis.147:118-124.

Sabin L., Behrentz E, Winer AM, Jeong S, Fitz DR, Pankratz DV, Colome SD, and FruinSA. (2004) Characterizing the range of children's air pollutant exposure during schoolbus commutes. JEAEE (in press, online at:http://www.nature.com/jea/journal/vaop/ncurrent/).

Spira-Cohen A, Holguin-Veras J,Zorrilla JC, Kendall M, Gorczynski J, Clemente J,Blaustein M, Thurston GD. (2004) The Role of Traffic in Fine Particle PollutionExposure Among Children at an Elementary School in the South Bronx. Proceedings ofthe XII Panamerican Conference of Traffic and Transportation Engineering, Sept 26-29,Albany, NY.

Thurston, G. D., Lippmann, M., Scott, M.B., and Fine, J.M. (1997) Summertime haze airpollution and children with asthma. Am. J. Respir. and Crit. Care Med. 155: 654-660.US Census Bureau (2000) 2000 Census of Population and Housing. State & CountyQuickfacts (http://quickfacts.census.gov/qfd/states/36/36005.html). U.S. Government,Washington, DC.

US Environmental Protection Agency (U.S.EPA). (2002) Health Assessment Documentfor Diesel Engine Exhaust. Washington, DC: Office of Research and DevelopmentNCfEA; 2002. EPA/600/8–909/057F

U.S. EPA, (2003) National Emissions Inventory Average Annual Emissions, All CriteriaPollutants Including 1980, 1985, 1989-2001. Research Triangle Park, NC.

U.S. EPA (2004) Air Quality Criteria for Particulate Matter. EPA/600/P-99/002aF.National Center for Environmental Assessment-RTP Office. Office of Research andDevelopment. Research Triangle Park, NC.

Weiss KB, and Wagener, DK. (1990) Changing patterns of asthma mortality. Identifyingtarget populations at high risk. J Am Med Assoc. 264:1683-1687.

Wjst M, Reitmeir P, Dold S, Wulff A, Nicolai T, von Loeffelholz-Colberg EF, vonMutius E. (1993) Road traffic and adverse effects on respiratory health in children. BrMed J 307:596-600.

32

Appendix

ARTICLE IN PRESS

1352-2310/$ - se

doi:10.1016/j.at

�Correspond845-351-5472.

E-mail addr

Atmospheric Environment 38 (2004) 5283–5294

www.elsevier.com/locate/atmosenv

Ambient pollutant concentrations measured by a mobilelaboratory in South Bronx, NY

Polina B. Maciejczyka, John H. Offenbergb, Jessica Clementea, Martin Blausteina,George D. Thurstona, Lung Chi Chena,�

aNelson Institute of Environmental Medicine, New York University School of Medicine, Tuxedo, NY 10987, USAbDepartment of Environmental Science, The State University of New Jersey, Rutgers, New Brunswick, NJ 08901, USA

Received 8 August 2003; received in revised form 30 January 2004; accepted 9 February 2004

Abstract

The objective of this study is to characterize the ambient air quality of the South Bronx, New York City (NYC),

having high concentrations of diesel trucks and waste transfer facilities. We employed a mobile laboratory for

continuous measurements of concentrations of fine particulate matter (PM2.5), black carbon (BC), and gaseous

pollutants at 6 locations for three–four weeks each during the period of April 2001–February 2003. Integrated 24-hr

PM2.5 samples were also collected for elemental and PAHs analyses. South Bronx PM2.5 and BC levels were compared

to those at Bronx PS 154 (NYSDEC site) and at Hunter College in the Lower Manhattan. Although the median daily

PM2.5 concentrations agreed within 20%, the median hourly BC concentrations were higher at all South Bronx sites

ranging from 2.2 to 3.8mgm�3, compared to 1.0–2.6 mgm�3 at Hunter College. Continuous Aethelometer

measurements at additional 27 sampling sites in the South Bronx were conducted along major highways. BC

concentrations varied within each site, depending on time-of-day, with a large spatial variability from site-to-site. Daily

median BC concentrations varied from 1.7 to 12 mgm�3 on the weekdays, and were lower (0.50–2.9 mgm�3) on the

weekends. Elemental concentrations were higher at all South Bronx sites than those at Hunter College for all measured

elements but Ni and V, and at the Hunts Point, an industrial location, were approximately 2.5-fold higher. The average

sum of 35 PAHs was 225 ngm�3, which is 4.5 times larger than representative regional concentrations in Jersey City,

NJ. Among the individual PAHs, 3,6-dimethylphenanthrene had the highest concentrations, and the overall PAH

fingerprint differed from signal for Jersey City. Our data indicates that highways encircling the South Bronx are having

a measurable adverse influence on residents’ exposure to pollutants compared to other NYC areas.

r 2004 Elsevier Ltd. All rights reserved.

Keywords: Urban air pollution; Particulate matter; Organic compounds; Diesel traffic; Black carbon

1. Introduction

In the Bronx, the most northern of the five boroughs

of New York City (NYC), the rate of asthma

e front matter r 2004 Elsevier Ltd. All rights reserve

mosenv.2004.02.062

ing author. Tel.: +1-845-731- 3560; fax: +1-

ess: chenl@env.med.nyu.edu (L. Chi Chen).

hospitalizations increased 110–120% between 1987 and

1996, as compared to 35–50% increases in most other

neighborhoods in NYC (Leighton et al., 1999). Ad-

ditionally, these rates between 1991 and 1996 were

higher for all ages of the Latino residents of the South

Bronx than those of Manhattan and Brooklyn (Ray

et al., 1998; Claudio et al., 1999). Although the origin

of asthma is unquestionably multi-factorial, with

d.

ARTICLE IN PRESSP.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–52945284

important genetic, immunologic, and environmental

components, the marked geographic and temporal

variations strongly suggest that local environmental

factors make a significant contribution to the pathogen-

esis and exacerbation of asthma, underscoring the need

for an evaluation of this factor in urban neighborhoods

within the South Bronx.

The South Bronx has large volumes of heavy vehicle

traffic passing through it along several major highways

(i.e., Interstates 87, 95, 278, and 895) that encircle the

South Bronx, creating pollution that can affect local

residents under any wind direction. In addition, there

are multiple local industries and facilities that generate

truck traffic, including Hunts Point Wholesale Markets

(world’s largest wholesale market), a municipal sewage

sludge processing plant, a privately owned sludge drying

plant, and 19 public and private waste transfer stations.

The area also hosts a municipal waste water treatment

plant and a large number of manufacturing facilities. All

of this traffic results in high concentrations of truck

activity and diesel emissions in the proximity of schools

and residences in the South Bronx. At Hunts Point

Market alone, some 12,000 trucks move in and out daily

(New York State Department of Transportation (NYS

DOT), 2001). Indeed, it is very common to see schools,

playgrounds, single family housing, waste transfer

stations, and large apartment buildings, all within a

Fig. 1. Location of sampling sites in South Bronx. Preliminary sites ar

for mobile laboratory measurements are marked by flags.

few city blocks of each other. This type of zoning is

generally less common in other parts of NYC, and the

South Bronx represents one of the city’s few remaining

mixed-use zones where residential neighborhoods come

into such close contact with heavy truck traffic (Claudio

et al., 1998). Local community groups have raised the

question whether this high concentration of truck

activity and diesel emissions in the proximity of schools

and residences are contributing to their very high

prevalence of asthma. The New York State Department

of Environmental Conservation (NYS DEC) measures

neighborhood air pollution in the South Bronx at three

centralized monitoring sites, currently located at public

elementary and intermediate grade schools PS 154, IS

52, and IS 74 (sites U, S, and N in Fig. 1, respectively).

Currently, these sites report the following air pollution

hourly parameters: PS 154–PM2.5, IS 52–PM2.5, PM10,

SO2, O3, and NOx, IS 74–PM2.5. Given the elevation of

these sites tens of feet above the street level, questions

have arisen on whether these measurements represent

the pollution at the ground level of the residents.

Additionally, no speciation data is provided by NYS

DEC sites.

To learn more about local South Bronx conditions,

New York University (NYU) initiated a community-

based South Bronx Air Pollution Study to examine the

current environmental conditions, specifically as they

e marked by shaded circles and letters. Final comprehensive sites

ARTICLE IN PRESSP.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–5294 5285

relate to the siting of waste management facilities and

traffic, and the regulatory and public policy issues that

affect environmental decision-making. This study is

carried out by the Nelson Institute of Environmental

Medicine (NIEM) at NYU School of Medicine in

collaboration with NYU’s Robert F. Wagner Graduate

School of Public Service and four community groups in

the South Bronx (The Point, Youth Ministries, Nos

Quedamos, Inc., and the Sports Foundation). In this

paper, we report the elemental and organic composition

of ambient PM2.5 samples collected during this study.

2. Experimental methods

The van for NYU’s Mobile Laboratory was provided

by the US EPA for use in an EPA Cooperative

Agreement, and remains on permanent loan, with

support provided by the NYU Particulate Matter

Health Effects Research Center. It is equipped with

new environmental monitoring equipment, and heavy-

duty electrical wiring, which powers the air and

meteorological monitoring equipment. It also houses

on-board computers to store and process the exposure

measurement data collected in the field. This mobile unit

is a well-established facility that has maintained a good

rapport with South Bronx residents, police and city

government. The NYU Mobile Laboratory was em-

ployed periodically in 2000–2003 to collect the real-time

measurement of pollutant levels needed for this research.

Measurements included continuous PM10 mass (sequen-

tial sampler, NYU), PM2.5 mass (TEOM with ACCU

Sampler, R&P, inlet temperature 50 1C, inlet design was

PM2.5 sharp-cut cyclone), BC (Aethelometer, Andersen

Instruments, Magee Scientific, AE-14 dual channel,

wavelength for the BC channel 880 nm, absorption

cross-section for channel 1 (BC) was 12.6, channel 2 (UV

carbon) was 30), CO (Model 48C, Thermo Environ-

mental Instrument Inc.), O3 (Model 103-PC, Thermo

Environmental Instrument Inc.), nitrogen oxides (Model

8840, Monitor Labs.), and SO2 (Model 8850, Monitor

Labs.). Additionally, PM2.5 samples for gravimetric and

elemental elements (via X-ray fluorescence (XRF))

analyses were collected daily at all sites for 24 h on

Teflon filters (Gelman ‘‘Teflo’’, 37mm, 0.2 mm pore). In

this paper, we discuss only the measurements of BC,

PM2.5 mass by TEOM, and PM2.5 filter composition

via XRF. Filter samples were stored at constant

temperature and humidity (2170.5 1C, 4075%RH)

until analyzed.

The integrated PAH samples were obtained only at

the Crotona site during the period 14 August–25

September 2002, using the method of Naumova et al.

(2002). Sampled air was first passed through a PM2.5

cyclone and then through two 47mm quartz fiber filters

(QFFs) to collect particulate-phase PAHs, then into a

stainless-steel cylinder containing two polyurethane

foam plugs (PUF) (diameter 25mm, height 100mm) to

retain the gas-phase PAHs. The sampler was operated at

a flow rate of 16.5 Lmin�1; leading to sample volumes

ranging from 73 to 188m3 depending upon the number

of days over which the integrated sample was collected.

After sampling, PUF plugs were placed in a pre-baked

glass jar with aluminum foil-lined lids, while QFFs

were placed in aluminum foil pouches. All samples

were transported in a cooler and stored at 4 1C until

analyzed.

Filters collected for gravimetric and elemental analy-

sis were equilibrated at NYU’s weigh laboratory at

constant temperature and humidity (2170.5 1C,

4075%RH) for 24 h. Filter mass was measured on a

microbalance (Model MT5, Mettler-Toledo). Analysis

for 34 elements followed by non-destructive ED-XRF

(Model EX-6600 –AF, Jordan Valley), and spectral

software XRF2000v3.1 (US EPA and ManTech Envir-

onmental Technology, Inc.).

Analysis of QFFs and PUFs was performed under

clean trace organic sample handling and techniques

developed previously (Offenberg and Baker, 1997,

1999a, b). Briefly, the PUF and QFF samples were

spiked with a surrogate standard consisting of d10-

acenaphthene, d10-anthracene, d10-fluoranthene, and

d12-benzo[e]pyrene to determine the analytical recov-

eries of the PAHs. The QFFs and PUFs were then

extracted for 24 h in Soxhlet extractors with a mixture of

petroleum ether:dichloromethane or acetone:hexane.

The filter samples were solvent exchanged to hexane

during the rotary evaporation step. An internal standard

solution consisting of deuterated d8-naphthalene, d10-

phenanthrene, d10-pyrene, and d12-benzo[a]pyrene was

added to concentrated samples after the final nitrogen

blow-down step. The final extracts were concentrated

and analyzed on a high resolution GC/MS capillary

detector (Hewlett-Packard 6890/5973) equipped with a

0.25mm� 30m DB-5 capillary column (0.25mm film

thickness; J&W Scientific). The MS was operating in

Electron Impact ionization mode (70 eV), with data

collected for target ions only (selected ion monitoring).

No significant breakthrough (i.e.,o5% on the back-

up PUF) was observed for the PAH species reported

here, with the exception of naphthalene, which is not

included in the following results. Nearly 30% of the 3,6-

dimethylphenanthrene was found on the second QFF,

indicating that there may have been some adsorption to

the filter surface or volatilization losses from the

particles loaded onto the front filter during the multi-

day sampling times, and resulting in less than quanti-

tative collection. All other analytes exhibited no

significant mass on the backup filter. 3,6-dimethylphe-

nanthrene was kept for the results presented here,

however the relatively high fraction on the backup filter

must be noted.

ARTICLE IN PRESSP.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–52945286

Site selection for comprehensive sampling was based

on preliminary sampling for BC (Aethelometer) at 27

locations (as shown in Fig. 1) chosen according to the

traffic density as well as local knowledge regarding the

nearby affected population (e.g., elementary school).

This preliminary sampling was conducted during April

and May 2001. The BC concentrations were measured at

each location (but not concurrently) for approximately

5min during the morning (6 a.m.–9 a.m. local time) and

evening (4 p.m.–7 p.m. local time) traffic rush hours.

Each site was sampled at least four different weekdays

and four different weekend days to get a representative

mean concentration (Table 1). Based on the results, we

selected six intensive sampling sites (shown in Fig. 1)

that represent both high and low concentrations of BC,

and were widely distributed within the South Bronx.

Each site characteristics are listed in Table 2 along with

Annual Average Daily Traffic (AADT) as reported by

NYS DOT for 2001. Additional sampling site was

located at lower Manhattan (on the roof of a two-

storied building of Hunter College School of Health

Science, 25th St. and 1st Ave.) and operated by another

NYU research project. While the South Bronx is

surrounded by three major interstates laden with semi-

Table 1

Arithmetic mean BC concentration (mgm�3) measured at 27 prelimin

Site All Day of the week

Weekday

A 5.274.8 7.475.2

B 3.973.3 5.473.3

C 5.478.0 6.174.9

D 3.373.1 4.973.5

E 3.875.6 6.376.4

F 5.275.0 7.775.3

G 5.877.1 9.978.2

H 9.4712.4 15.1714.7

I 5.476.0 8.976.7

J 2.472.0 3.572.2

K 3.073.5 4.574.2

L 3.273.4 4.874.0

M 2.672.4 3.972.7

N 2.272.0 3.272.4

O 4.573.4 5.373.1

P 5.376.2 6.777.6

Q 2.973.0 4.373.5

R 3.173.1 4.673.5

S 5.976.3 8.676.6

T 6.878.8 9.9710.1

U 4.274.0 5.873.6

V 3.672.8 5.072.8

W 6.279.2 9.4711.0

X 3.672.9 5.072.8

Y 4.373.2 4.672.7

Z 3.273.3 4.573.7

AA 3.673.3 4.973.6

tractor trailers, Hunter College diesel traffic emissions

are limited to delivery trucks on First and Second

Avenues and local cross streets only, since no trucks are

allowed on the nearby FDR Drive (reported AADT for

FDR Drive between 23rd and 34th St. 125,637). The

Hunter College site was used in this work for

comparison of ambient pollutant levels.

3. Ambient results and discussion

The BC concentrations measured at 27 different sites

(shown in Table 1) indicate variability within each site

depending on day of the week, with weekdays being

highest, as well as a large spatial variability from site-to-

site. Mean BC concentration varied from 1.7 to almost

12mgm�3 on the weekdays, and was significantly less

(0.5–2.9 mgm�3) on the weekends (both a.m. and p.m.

rush hours were averaged in weekday and weekend

values). The BC levels in the morning were higher that

those in the evening at all sites, likely due to the poorer

mixing of the atmosphere in the morning hours. A

separate investigation of the relationship between traffic

counts and air pollutants conducted at PS 154 revealed

ary sites

Weekdays

Weekend a.m. p.m.

2.672.5 8.676.3 6.173.5

2.272.3 8.272.4 2.971.5

4.6710.4 8.775.1 3.072.2

1.670.9 6.374.3 3.471.4

1.271.0 9.177.5 3.373.2

2.272.3 10.575.5 4.873.2

1.771.5 11.679.1 8.076.8

3.072.7 23.3716.6 7.476.5

1.871.2 11.578.5 6.272.6

1.170.5 4.472.4 2.671.5

1.370.6 6.475.2 2.671.5

1.571.5 6.474.0 2.873.0

1.270.6 5.273.0 2.671.6

1.170.6 3.772.5 2.872.2

3.773.6 4.772.4 5.973.7

3.873.9 10.078.9 3.273.3

1.571.0 4.872.2 3.674.6

1.471.0 6.473.6 2.572.1

1.571.1 10.177.8 7.375.3

1.971.0 13.2712.2 6.174.9

1.773.3 6.573.9 5.173.1

1.570.8 6.272.9 3.972.0

1.972.0 10.8710.4 8.2711.6

1.771.3 6.072.8 4.072.4

3.973.7 6.672.3 2.771.3

1.571.5 6.373.9 2.271.7

1.671.3 6.873.9 2.971.9

ARTICLE IN PRESS

Table 2

Site information for the intensive measurements with the mobile laboratory

site ID in

Fig. 1

Site name Sampling dates Site characteristics

U PS 154 27 July–29 August 2001, and

23 April–20 May 2002

On a corner of 135th St. and Alexander Ave. next to Bronx public

school 154. PS 154 is also a location of NYS DEC monitoring

station. This site was in immediate proximity of I-97 (Major

Deegan Expressway) that had 110, 000 AADT for 2001.

I Hunts

Point

November 8–29 November,

2001

Above the underground Hunts Point Station of the NYC subway

system, in immediate proximity to an elevated section of I-278

(Bruckner Expressway) and the exit ramp to Hunts Point Avenue

that leads to the Hunts Point Wholesale Market district. The

AADT count for 2001 was over 117, 000.

A Noble

Field

4, December 2001–2 January,

2002

In residential area, on the grounds of the Noble Field Park, a small

recreational park that borders I-95 (Cross Bronx Expressway) and

the Bronx River Parkway, a 6-lane limited access road that

excludes the truck traffic. For this location the 2001 AADT was

over 124, 000.

J IS 52 27 March–21 April, 2002 4 blocks northwest of I-278. This area is more residential, and also

has an established NYS DEC monitoring station.

AA Crotona 13 August–11 October, 2002 On the grounds of Crotona Park, a large public recreational park

located 2 blocks south of I-95. Despite the park-like setting, this

site is influenced by traffic on the Crotona Avenue that runs

through the park, in addition to some 148, 200 AADT from I-95

under northerly winds.

S 138th

Street

6 January–11 February, 2003 On a corner of 138th street and Bruckner Boulevard, in immediate

proximity of I-278. This site was on the same section of route for

AADT as the Hunts Point.

P.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–5294 5287

that both peak BC and oxides of nitrogen were related

to hourly variations in nearby I-87 truck traffic

(Thurston et al., 2003). We did not have the traffic

count data for each individual site, and use of NYS

DOT annual average daily traffic densities sited in the

Methods section did not provide a good correlation,

most likely due to our limited dataset. Additionally, the

true daily volumes on highways may vary widely from

the AADT. Considerably higher or lower values often

result in areas of seasonal activities (e.g., holidays,

weather, pollution episodes) when counting weekend

versus weekday traffic.

We compared ambient TEOM PM2.5 and BC

concentrations measured by the van at various parts of

the South Bronx to the ones simultaneously measured at

Hunter College and by NYS DEC at PS 154. The

comparison of these three sites indicated that although

the median daily TEOM PM2.5 concentrations (shown

in Fig. 2) at all three concurrent sites agreed within 20%,

the median hourly BC concentrations (shown in Fig. 3)

were higher at all south Bronx sites. Notably, the

intervals within which lies 50% of the data (each box in

Fig. 2 and 3) were wider for the samples collected at the

South Bronx intensive sampling sites but the April-May

2002 campaign. A paired Student’s t-test (equal or close

to equal variance) of the PM2.5 distributions revealed

that for sampling in July–August 2001 the difference in

the means are all statistically significant. However, for

November 2001 and April–May 2002 only van and DEC

sites were statistically different; and the mean of the

distributions sampled in December 2001 were not

different.

The average daily PM2.5 mass concentration for all six

intensive sampling sites was 16.275.1 mgm�3. However,

as shown in Fig. 4, these concentrations were on average

30% higher on weekdays than those on weekends except

for Crotona, where weekend PM2.5 levels were slightly

higher, likely due to the active use of the park by the

local residents. The reason why a weekend decrease in

PM2.5 was not observed at 138th St. is not clear.

Probability associated with an unpaired Student’s t-test

(unequal variance) with two- tailed distribution yielded

p40:05 for cross comparison of weekend and weekdayPM2.5 data for Crotona and 138th St. Arithmetic mean

of daily BC concentration for all six sites was

2.771.1 mgm�3. Here, however, the daily averages were

about 1.8 times higher on weekdays vs. weekends. Since

one of the major sources of black carbon in this urban

environment is diesel exhaust, we concluded that week-

end decrease in BC could be explained by reduced diesel

truck traffic on the surrounding highways. Although we

might have expected to observe higher BC concentra-

tions due to residential woodburning during winter

months ((Noble Field and 138th St. sampling periods),

ARTICLE IN PRESS

Fig. 3. Concurrent hourly arithmetic mean BC concentrations (mgm�3) measured at three South Bronx sites and Hunter College (HC).

Line in the box is a median, box size 25th–75th percentiles. Circles are outliers (measurements outside the 10th and 90th percentile).

July-August 2001

van (26) HC (28) DEC (31) van (16) HC (17) DEC (21) van (24) HC (24) DEC (26) van (23) HC (23) DEC (22)

PM

2.5

(µg

m-3

)

0

10

20

30

40

50

60

November 2001 December 2001 April-May 2002

Fig. 2. Daily mean PM2.5 concentrations (mgm�3) measured concurrently at the NYS DEC (DEC), Hunter College (HC), and at 3

sites in the South Bronx (van). Line in the box is a median, box size 25th–75th percentiles. Circles are outliers (measurements outside

the 10th and 90th percentile). Number of samples (N) for each box plot is in parenthesis.

P.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–52945288

seasonal trends in BC concentrations were not obvious:

two second highest BC concentration ratios were

observed at PS154 during spring-summer months. The

similarity of the BC weekend to weekday ratio at PS154

during two different time periods, however was not

reflected in overall BC concentrations at this site: mean

BC were 4.34 and 2.77mgm�3 for July–August 2001,

and April–May 2002, respectively.

ARTICLE IN PRESS

PM2.5

10

20

30

40weekdaysweekends

BC

n= 24 10 15 6 21 8 16 6 25 10 43 16 25 10

PS 154-1 Hunts Point Noble Field IS 52 PS 154-2 Crotona 138th St.0

2

4

6

1.4

1.4

1.4

1.4

1.5

1.51.6

0.99 1.0

1.3

2.3

1.62.0

2.2

Fig. 4. The arithmetic mean and standard deviation of daily mean concentrations of PM2.5 and BC (mgm�3) measured at the South

Bronx sites. Error bars are the standard deviation of the arithmetic mean. Number of samples (n) applies to both PM2.5 and BC.

Numbers immediately above the bars indicate the weekdays to weekend ratios of the measurements for each site.

P.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–5294 5289

Table 3 presents average mass fraction concentrations

of elements observed in the PM2.5 samples at the six

sites. Although the XRF analysis was done for 34

elements, here we report only elements detected in more

than 25% samples per site. In our treatment of XRF

data, the element was considered to be detected if the

value was larger than three times the uncertainty of the

measurement. In calculating the arithmetic averages, the

values of the non-detected elements were replaced with

the three times uncertainty of the measurement. S, K,

Ca, Fe, Ni, and Zn were detected in 475% of samples

at all sites. Elements that were detected in less than 50%

of samples include Cr, Sb, and As. The sum of all

analyzed elements contributed 10.5–17.3% of the PM2.5

mass. As shown in Fig. 5, elemental concentrations

measured in this work agree well with previously

reported data for samples collected in 3 NYC neighbor-

hoods of Harlem, Bronx, and Queens (Kinney et al.,

2002). On average, S was higher in samples collected

during summer months due to increased photochemical

gas-to-particle conversion processes. In winter, we

observed higher concentrations of Ni and V, two

elements known to be enriched in residual fuels and

their combustion products (Cooper and Watson, 1980).

Notably, these elements were higher at the Hunter

College site than at the concurrent Bronx site, pre-

sumably due to the influences of local power plants

located in Manhattan and across the East River in

Queens.

An inter-comparison within all sites (Table 3) showed

that mean fraction concentrations of all elements but Cl,

Ni, and V were somewhat higher at the South Bronx

than those concurrently collected at Hunter College.

Comparison of concurrent South Bronx and Hunter

ARTIC

LEIN

PRES

S

Table 3

Summary of elemental composition of PM2.5 samples, mean (standard deviation), ppm

27 July–29 August 2001 8–29 November 2001 4 December 2001–2 January

2002

27 March–21

April 2002

23 April –20

May 2002

13

August–11

October 2002

6 January–11

February

2003

PS154 Hunter

College

Hunts Point Hunter

College

Noble Field Hunter

College

IS52 PS154 Crotona 138th St.

n ¼ 34 n ¼ 34 n ¼ 21 n ¼ 21 N ¼ 29 n ¼ 23 n ¼ 22 n ¼ 37 n ¼ 59 n ¼ 35

Na 3919 (5779) 5482 (6775) 6924 (6698) 6990 (9754) 6320 (4634) 4092 (2190) 15,927

(13,180)

10,161 (8283) 9877 (11,338) 11,325 (9231)

Mg 1221 (714) 1374 (1009) 1940 (923) 1673 (1296) 2294 (908) 1733 (603) 3210 (1615) 5145 (5240) 2419 (1659) 2287 (1724)

Al 2457 (1253) 2790 (1280) 3733 (1508) 3602 (1516) 4299 (2103) 3700 (1477) 5226 (2714) 3620 (1914) 3193 (1382)

Si 4530 (1790) 5526 (2775) 7693 (2039) 6197 (2916) 6995 (2311) 7097 (2684) 12,081 (8044) 13,011 (6833) 5596 (2176) 5383 (1907)

S 91,663

(38,084)

110,317

(29,025)

74,688

(18,052)

63,284

(15,832)

84,271

(17,445)

68,657

(13,005)

114,496

(17,424)

95,674 (23,064) 102,872

(33,796)

106,061

(19,238)

Cl 2510 (2128) 3640 (7734) 1271 (690) 1670 (1280) 1565 (1661) 1354 (2135) 4272 (4725)

K 1563 (670) 1890 (646) 3513 (818) 3094 (886) 3742 (1132) 3013 (705) 4112 (1703) 4394 (2366) 3441 (1885) 4101 (944)

Ca 2285 (1121) 2756 (1111) 5741 (1936) 3762 (1555) 5770 (1586) 4165 (1661) 5049 (1452) 5549 (1880) 3903 (2187) 3727 (1159)

Ti 288 (223) 333 (156) 497 (274) 419 (175) 213 (112)

V 303 (147) 466 (239) 496 (240) 711 (306) 533 (258) 666 (281) 488 (181) 704 (607) 375 (198) 441 (179)

Cr 227 (128) 177 (95)

Mn 184 (87) 361 (652) 468 (223) 334 (179) 340 (139) 337 (158) 319 (131) 713 (729) 225 (102) 197 (108)

Fe 9654 (4549) 9771 (7511) 34,359

(24,368)

10,940 (4009) 18,305 (5595) 11,104 (3978) 11,384 (3366) 20,388 (6575) 11,642 (4818) 397 (282)

Co 98 255 (100) 189 (125) 275 (126) 214 (119) 174 (72) 394 (384) 126 (55) 25,763

(13,277)

Ni 384 (155) 755 (352) 1980 (1026) 2415 (1331) 2464 (1215) 2968 (1662) 1482 (810) 1002 (692) 804 (394) 184 (76)

Cu 299 (171) 291 (189) 748 (637) 333 (184) 530 (168) 366 (274) 396 (182) 808 (604) 310 (134) 1637 (469)

Zn 1020 (429) 1233 (568) 5699 (2651) 2544 (1567) 4769 (1638) 3163 (1132) 2870 (1068) 2868 (2124) 1727 (826) 344 (123)

As 355 (182)

Se 154 (75) 262 (123) 193 (72) 340 (164) 298 (159) 315 (106)

Br 238 (97) 233 (105) 472 (284) 366 (146) 529 (200) 352 (92) 603 (221) 914 (821) 234 (146) 226 (83)

Sr 150 (102) 224 (126) 162 (80) 274 (141) 282 (138) 898 (1312) 232 (166) 491 (171)

Sb 556 (332) 1114 (478)

Ba 792 (417) 539 (320) 2320 (1725) 615 (280) 1815 (667) 677 (272) 843 (355) 2447 (2331) 1161 (570) 548 (278)

Pb 776 (559) 640 (387) 410 (230) 1936 (1124)

%mass 12.2 (4) 14.4 (2.9) 15.8 (2.6) 11.1 (3.5) 14.7 (3.2) 11.4 (2.6) 17.9 (3.8) 19.1 (5.7) 15.5 (2.6) 17.5 (3.8)

Values in bold—elements detected 476% of samples, in underlined—51–75%, regular font—26–50%, not reported—o25%.

P.B

.M

aciejczy

ket

al.

/A

tmo

sph

ericE

nviro

nm

ent

38

(2

00

4)

52

83

–5

29

45290

ARTICLE IN PRESS

Na Mg Al S K Ca Ti V Mn Fe Co Ni Cu Zn Pb PM2.5

mea

n co

ncen

trat

ion,

ng

m-3

0.1

1

10

100

1000

10000

South BronxHunter Collegewinter, Kinney 2002summer, Kinney 2002

Fig. 5. Arithmetic mean and standard deviation of daily mean elemental concentrations (ngm�3) in samples measured concurrently at

the South Bronx sites and Hunter College. Winter and summer means are adapted from Kinney et al. (2002). PM2.5 concentrations are

in mgm�3.

P.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–5294 5291

College samples showed consistently higher Fe, Cu, Zn,

and Ba at the former, but little spatial variation of

elements of crustal origin such as Al and Si. Average Cr,

Mn, Fe, Co, Cu, Zn, Ba, and Pb in Hunts Point site

samples were up to 2.5-fold higher for these elements at

the other sites. We suspect that this location was affected

by the emissions from the nearby subway entrances,

since high concentrations of Fe, Ba, and Cu were

similarly observed in a subway study by Furuya et al.

(2001). At other Bronx locations, high Ba levels are

attributed to the wear of automobile break pads, and,

thus, are expected to be elevated in higher traffic density

area. The subway could also be the source of elevated

Na and highest median Cl concentration (median data

not shown) observed in Hunts Point samples as was

reported for the Washington, DC, subway (Birenzvige et

al., 2003). Concentrations of most elements at PS 154

were almost twice as large in 2002 than in 2001. Since

the prevalent wind directions for both sampling periods

were comparable, i.e., the frequency of southerly winds

were 50.1% in 2001 and 58.1% in 2002, and northerly

winds 35.9% and 23.3%, respectively, we speculate that

the observed concentrations could partially be attributed

to local activities: during 2001 sampling period the

school was not in session.

Total (gas phase+particle-bound) S35-PAH concen-

trations (the sum of 35 individual PAHs listed below)

collected at the Crotona site ranged from

106–374 ngm�3 (geometric mean of all S35-PAH of

225 ngm�3). These observed concentrations approach

the highest ever reported in the US. For comparison,

S-PAH concentrations ranged 27–430 ngm�3 in Chica-

go (Simcik et al., 1997), 92 ngm�3 in Denver, CO

(Foreman and Bidleman, 1990), and 19.5–114 ngm�3 in

urban Baltimore, MD (Offenberg and Baker, 1999a).

Fig. 6 shows the comparison of Crotona S-PAHs tothe closest representative regional urban site at Jersey

City, NJ, where total S-PAHs measured once every

twelve days over the period 5 July 1998–3 October 2000,

and averaged 52.8710.5 ngm�3 (arithmetic mean and

standard deviation) (Gigliotti et al., 2000). This trans-

lates into the total S-PAH concentrations at Crotona

being 4.5-fold higher than those measured in Jersey City.

More than 97.5% of the total S-PAH measured at

Crotona was found to be in the gas phase. For

comparison, concentrations of PAHs in during summer

1996 in urban Baltimore averaged 43.7732.5 ngm�3,

with 92% found in the gas phase (Offenberg and Baker,

1999a). While the gas-phase concentrations reported

here were higher than those measured in Jersey City by a

factor of 4.9, the particle bound concentrations were

only higher by a factor of 1.3. Additionally, the Crotona

concentrations shown in Fig. 6 were measured as

integrated samples over periods of three–eight days

and as such represent time weighted average concentra-

tions over several day periods. This several day

integration dampens extremes that otherwise would

have been seen in shorter integrated samples, such as

the 24 h sampling strategy utilized by Gigliotti et al.,

(2000).

Average concentrations of individual gas-phase and

particle-bound PAHs are shown in Fig. 7. Gas-phase

concentrations were dominated by phenanthrene, 3,6-

dimethylphenanthrene, and mono-methylphenanthre-

nes+mono-methylanthracenes. Particle-phase concen-

trations were dominated by 3,6-dimethylphenanthrene,

with significant contributions of benzo[g,h,I]perylene,

indeno[1,2,3-c,d]pyrene and coronene. For all samples,

3,6-dimethylphenanthrene was greatly enhanced, repre-

senting approximately 35% of the total S-PAH, and

ARTICLE IN PRESS

0

100

200

300

400

8/14-8/21 8/21-8/28 8/28-9/05 9/05 - 9/9 9/09 - 9/12 9/12- 9/18 9/18-9/25

2002

Sum particle-bound PAHs

Sum gas phase PAHs

Con

cent

ratio

n Σ-

PAH

, ng

m-3

Jersey City average July 5, 1998 – October 3,2000

Fig. 6. Total (gas phase+particle-bound) S-PAH concentrations (ngm�3) measured at the Crotona site. Jersey City, NJ average from

Gigliotti et al. (2000).

0.001

0.01

0.1

1

10

conc

entr

atio

n, n

g/m

3 co

ncen

trat

ion,

ng/

m3

0.0001

0.001

0.01

0.1

1

10

100

Flu

oren

e

1Met

hylfl

uore

ne

Dib

enzo

thio

phen

e

Phe

nant

hren

e

Ant

hrac

ene

Met

hylp

hena

nthr

enes

4,5-

Met

hyle

neph

enan

thre

ne

9,10

-Dim

ethy

lant

hrac

ene

Met

hyld

iben

zoth

ioph

enes

3,6-

Dim

ethy

lphe

nant

hren

e

Flu

oran

then

e

Pyr

ene

Ben

zo[a

]fluo

rene

Ret

ene

Ben

zo[b

]fluo

rene

Cyc

lope

nta[

cd]p

yren

e

Ben

z[a]

anth

race

ne

Chr

ysen

e/T

riphe

nyle

ne

Nap

htha

cene

Ben

zo[b

]nap

htho

[2,1

-d]

thio

phen

eB

enzo

[b+k

]fluo

rant

hene

Ben

zo[e

]pyr

ene

Ben

zo[a

]pyr

ene

Per

ylen

e

Inde

no[1

,2,3

-cd]

pyre

ne

Ben

zo[g

,h,i]

pery

lene

Dib

enzo

[a,h

+a,c

]ant

hrac

ene

Cor

onen

e

Jersey City, NJCrotona, South Bronxgas phase

particle-bound

Fig. 7. Comparison of mean concentrations (ngm�3) of individual PAHs between the Crotona site (n ¼ 7) and three-year average

Jersey City, NJ (from Gigliotti et al., 2000).

P.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–52945292

suggesting a strong local source of this PAH. Overall,

the relative contributions to total PAHs (the PAH

‘‘fingerprint’’) differed from that in Jersey City, as

shown in Fig. 7. A majority of the difference in total

PAH concentrations relate to large enrichment in 3,6-

dimethlyphenanthrene, phenanthrene, mono-methyl-

phenanthrenes+mono-methylanthracenes at the Croto-

na location.

Jersey City is a highly urbanized location on the west

bank of Hudson River across the Lower Manhattan.

ARTICLE IN PRESSP.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–5294 5293

It is just east of I-95 (NJ Turnpike), and is a major

commute route to Holland Tunnel via I-78. This area

experiences daily rush hour traffic and slow downs for

toll booths similar to the South Bronx. While Jersey City

is located upwind of Manhattan, South Bronx is

downwind of Manhattan and New Jersey, possibly

experiencing the cumulative effect of pollution sources

as well as local. Lee et al. (2004) describe the

apportionment of ambient concentrations to sources of

PAHs in the airshed over the lower Hudson River

Estuary. The large differences in total PAH concentra-

tion and relative contributions may be suggestive of

completely different sources rather then just traffic

contributes to the level of organic pollutants in the

South Bronx.

4. Summary

An intensive PM2.5 speciation sampling study was

carried out at six sites in South Bronx, NY during the

period of April 2001–February 2003. The results for BC

and elemental composition were compared to those at

Lower Manhattan (Hunter College). The BC concentra-

tions varied within each site depending on time of day,

with a large spatial variability from site-to-site. The

average daily PM2.5 mass concentration for all six sites

was 16.275.1 mgm�3. The sum of 34 elements analyzed

by XRF contributed 10.5–17.3% of the PM2.5 mass.

Elemental concentrations were higher at all South Bronx

sites than those at Hunter College for all measured

elements but Ni and V, with highest concentrations

for most transition group elements reported at Hunts

Point site.

Low volume integrated sampling for gas -and particle-

phase PAHs exhibited high concentrations in the

Bronx. While the sum concentrations were similar to

those observed in other heavily populated urban centers,

such as Chicago, Baltimore and Denver, the relative

composition was distinctive. The relative contri-

bution of 3,6-dimethylphenanthrene was high, and

suggests a strong source of this compound, and

possibly other PAHs proximate to the sampling

location.

Overall, our sampling data indicate that the major

highways encircling the South Bronx, as well as other

local sources, are having a measurable adverse impact

on residents’ exposures to air pollution, even relative to

other NYC area locales.

Acknowledgements

Supported by EPA (R827351, Agreement X-982152)

and NIEHS (ES00260).

References

Birenzvige, A., Eversole, J., Seaver, M., Francesconi, S.,

Valdes, E., Kulaga, H., 2003. Aerosol characteristics in a

subway environment. Aerosol Science and Technology 37,

210–220.

Claudio, L., Torres, T., Sanjurjo, E., Sherman, L.R., Land-

rigan, P.J., 1998. Environmental health sciences educa-

tion—a tool for achieving environmental equity and

protecting children. Environmental Health Perspectives

106 (Suppl. 3), 849–855.

Claudio, L., Tulton, L., Doucette, J., Landrigan, P.J., 1999.

Socioeconomic factors and asthma hospitalization rates in

New York City. Journal of Asthma 36 (4), 343–350.

Cooper, J.A., Watson, J.G., 1980. Receptor oriented methods

of air particulate source apportionment. Journal of Air

Pollution Control Assosiation 30, 1116–1125.

Foreman, W.T., Bidleman, T.F., 1990. Semi-volatile organic

compounds in the ambient air of Denver, Colorado.

Atmospheric Environment 24A, 2405–2416.

Furuya, K., Kudo, Y., Okinaga, K., Yamuki, M., Takahashi,

S., Araki, Y., Hisamatsu, Y., 2001. Seasonal variation and

their characterization of suspended particulate matter in the

air of subway stations. Journal of Trace and Microprobe

Techniques 19 (4), 469–485.

Gigliotti, C.L., Dachs, J., Nelson, E.D., Brunciak, P.A.,

Eisenreich, S.J., 2000. Polycyclic aromatic hydrocarbons in

the New Jersey coastal atmosphere. Environmental Science

and Technology 34, 3547–3554.

Kinney, P.L., Chillrud, S.N., Ramstrom, S., Ross, J., Spengler,

J.D., 2002. Exposures to multiple air toxics in New York

City. Environmental Health Prospectives 110 (Suppl. 4),

539–546.

Lee, J.H., Gigliotti, C.L., Offenberg, J.H., Eisenreich, S.J.,

Turpin, B.J., 2004. Sources of polycyclic aromatic hydro-

carbons to the Hudson River Airshed. Atmospheric

Environment, in review.

Leighton, J., Matte, T., Findley, S., 1999. O1_MRKO1_MR-

KAsthma prevalence among school children in a Bronx

community. Asthma Conference, February 9, 1999. Centers

for Disease Control and Prevention, Atlanta GA 26pp.

Naumova, Y.Y., Eisenreich, S.J., Turpin, B.J., Weisel, C.P.,

Morandi, M.T., Colome, S.D., Totten, L.A., Stock, T.H.,

Winer, A.M., Alimokhtari, S., Kwon, J., Shendell, D.,

Jones, J., Maberti, S., Wall, S.J., 2002. Polycyclic aromatic

hydrocarbons in the indoor and outdoor air of three cities in

the US. Environmental Science and Technology 36 (12),

2552–2559.

New York State Department of Transportation (NYS DOT),

2001. Traffic volume report for Bronx County. www.dot.

state.ny.us/tech_serv/high/countfiles/bronx.pdf.

Offenberg, J.H., Baker, J.E., 1997. Polychlorinated biphenyls in

Chicago precipitation: Enhanced wet deposition to near-

shore Lake Michigan. Environmental Science and Technol-

ogy 31, 1534–1538.

Offenberg, J.H., Baker, J.E., 1999a. Influence of Baltimore’s

urban atmosphere on organic contaminants over the

Northern Chesapeake Bay. Journal of the Air and Waste

Management Association 49, 959–965.

Offenberg, J.H., Baker, J.E., 1999b. Aerosol size distributions

of polycyclic aromatic hydrocarbons in urban and over-

ARTICLE IN PRESSP.B. Maciejczyk et al. / Atmospheric Environment 38 (2004) 5283–52945294

water atmospheres. Environmental Science and Technology

33, 3324–3331.

Ray, N.F., Thamer, M., Fadillioglu, B., Gergen, P.J., 1998.

Race, income, urbanicity, and asthma hospitalization in

California—a small area analysis. Chest 113 (5), 1277–1284.

Simcik, M.F., Zhang, H.X., Eisenreich, S.J., Franz, T.P., 1997.

Urban contamination of the Chicago/coastal Lake Michi-

gan atmosphere by PCBs and PAHs during AEOLOS.

Environmental Science and Technology 31, 2141–2147.

Thurston, G., Kendall, M., Maciejczyk, P., Lall, R., Gorczyns-

ki, J., Blaustein, M., Clemente, J., 2003. A pilot study

of traffic exposures and their health effects among

South Bronx children with asthma. Presented at 2003

American Association for Aerosol Research Particulate

Matter Meeting. Particulate Matter: Atmospheric

Sciences, Exposure and the Fourth Colloquium

on PM and Human Health, Pittsburgh, PA, 31 March–

4 April.