MEASUREMENTS OF OUTDOOR AIR POLLUTIONFROM SECONDHAND SMOKE ON THE UMBC CAMPUSJames Repace, MSc.Repace Associates, Inc.101 Felicia LaneBowie, MD 20720www.repace.comJune 1, 2005Introduction.Individual cigarettes are point sources of air pollution; smoking in groupsbecomes an area source. Outdoor air pollutants from individual point sources are subjectto plume rise if the temperature of the smoke plume is hotter than the surrounding air;however if the plume has a small cross-section, as for a cigarette, it will rapidly cool andlose its upward momentum, and then will subside as the combustion particles and gasesare heavier than air. Thus, in the case of no wind, the cigarette plume will rise to acertain height and then descend, and for a group of smokers, for example sitting in anoutdoor cafe, on a hospital patio, or in stadium seats, their smoke will tend to saturate thelocal area with secondhand smoke (SHS). In the case where there is wind, the amount ofthermally-induced plume rise is inversely proportional to the wind velocity -doubling the wind velocity will halve the plume rise. In this case, the cigarette plumewill resemble a cone tilted at an angle to the vertical. The width of the cone and its anglewith the ground will depend upon the wind velocity: a higher wind will create amore horizontal but wider cone (due to increased turbulence), with uncertain impact onexposure to SHS for downwind nonsmokers. If there are multiple cigarette sources,the downwind concentrations will consist of multiple intersecting cones, i.e., overlappingplumes. As the wind direction changes, SHS pollution will be spread in variousdirections, fumigating downwind nonsmokers.SHS contains a large quantity of respirable particles, which can cause breathingdifficulty for those with chronic respiratory diseases or trigger an asthmatic attack inthose with disabling asthma. For the remainder of nonsmokers, Junker et al. report eye,nasal and throat irritation thresholds for 24 healthy young adult females for repeatedexposures over the course of 2 hours, corresponding to an SHS-PM2.5 concentration ofabout 4.4 micrograms per cubic meter (µg/m3) (Junker, 2001).Very few published data are available on outdoor levels of SHS. A California AirResources Board study (CARB, 2003), measured 1 and 8 hour time-weighted averagenicotine concentrations outside an airport, college, government center, office complex,and amusement park, found that at these typical outdoor locations, Californians may beexposed to SHS levels previously associated only with indoor SHS concentrations.Concentrations were strongly affected by counts of the number of smokers andmoderately affected by the size of the smoking area and the measured wind speed. TheCARB study indicated that outdoor SHS concentrations are detectable and sometimescomparable to indoor concentrations, and demonstrates that the number of cigarettesbeing smoked (i.e., total source strength), the position of smokers relative to the receptor,

and atmospheric conditions can lead to substantial variation in average exposures.A more recent pilot study by Klepeis, et al. (2004) reported that mean outdoorSHS concentrations determined from field surveys of particle concentrations measured inbuildings, at outdoor patios, on airport sidewalks, and in parks and public sidewalksduring time periods spent in locations where smokers were intermittently active thatmean SHS particle concentrations in outdoor settings in some cases can be comparable tothose in indoor settings. However, mean outdoor SHS concentrations appear morevariable than indoors, because outdoor SHS does not accumulate and outdoor transientpeaks are more sensitive to source-receptor proximity and wind conditions.Long-term means for outdoor SHS concentrations are averaged over a largenumber of transient peaks, which only occur when smokers are active, whereas indoorconcentrations remain high long after cigarettes have ended, and the total dose to aperson indoors from each cigarette will be greater than for a cigarette smoked outdoors.Klepeis, et al. (2004) found from controlled experiments that, during periods of smokingactivity outdoor SHS levels can reach mean concentrations measured indoors, usingeither burning cigarettes or CO tracer gas release, and reported a decrease in meanpollutant concentrations as a function of distance such that a doubling of distance couldresult in a concentration reduction of up to 50% or more. At distances of 1-2 m from thesource, mean outdoor SHS particle concentrations declined by about 75%. Klepeis et al.found that changing wind directions can have a large impact on outdoor SHS exposure asdemonstrated by the differences between concentrations monitored on opposite sides ofan active point source.The plume is driven in the longitudinal direction by the wind, and in thetransverse directions by diffusion. A highly simplified expression which illustrates thephysics for the downwind concentration C on the plume line for a point source pollutantemitted at ground level is given by: C Q/k ykzx, where Q is the pollutant mass emissionrate, x is the longitudinal distance from the source to the receptor, and where the productk ykz represents the diffusion constants in the transverse vertical and horizontal planeswhich describe the increasing lateral spread of the pollutant concentration as it proceedsdownwind in the longitudinal direction. There are four key features of most modelswhich describe the dispersal of emissions from a point source at ground level:1. The downwind concentration at any location is directly proportional to the massemission rate of the source.2. The more turbulent the atmosphere, the more rapid the spread of the plume in thedirection transverse to the direction of propagation of the plume.3. The maximum concentration at ground level is directly downwind on the plumeline, and is inversely proportional to the downwind distance from the source.4. The maximum concentration decreases for higher wind speeds, even though onthe plume line there is no explicit dependence on wind speed, because thediffusion constants k ykz are inversely proportional to wind speed, due tomechanical turbulence. These empirically-determined constants also depend onthe vertical temperature gradient of the atmosphere, which determines the-2-

temperature difference between a rising parcel of plume air and the surroundingair. (Williamson, 1973)Thus, for each point source, the plume concentration will increase with source strength,and decrease with increasing distance from the source and with increasing wind speed.However, for a very large area source, while the pollutant concentration downwind willstill decrease inversely as the wind speed, it will increase with downwind distance fromthe source as the square root of distance, or if there is an atmospheric inversion, withincrease linearly with distance.With these considerations in mind, a field study and two controlled experimentswere designed and implemented on the campus of the University of Maryland atBaltimore’s (UMBC) Catonsville, MD campus, at the request of UMBC’s UniversityHealth Services, to perform experiments designed to quantify secondhand smoke levelsoutdoors in the vicinity of building entrances, in order to provide scientific data relatingto whether limitations on smoking in proximity to campus building entrances werejustified.Biographical Sketch of the Principal Investigator. I am a biophysicist and aninternational secondhand smoke consultant with more than 60 scientific papers publishedon the hazard, exposure, dose, risk, and control of secondhand smoke. I have received theFlight Attendant Medical Research Institute Distinguished Professor Award, the RobertWood Johnson Foundation Innovator Award, the Surgeon General’s Medallion, and aLifetime Achievement Award from the American Public Health Association. I am aVisiting Assistant Clinical Professor at the Tufts University School of Medicine. I was asenior policy analyst and scientist with the U.S. Environmental Protection Agency. Iserved as a consultant to the Occupational Safety and Health Administration, U.S.Department of Labor, on its proposed rule to regulate secondhand smoke and indoor airquality. I was also a research physicist at the Naval Research Laboratory in the OceanSciences and Electronics Divisions. My full CV may be viewed at UMBC Outdoor Secondhand Smoke Studies.Equipment and Methodology.I deployed continuous real-time monitors for respirable particles (RSP), i.e.,airborne particulate matter in the combustion size range below 3.5 microns in diameter(PM3.5), and carcinogenic particulate polycyclic aromatic hydrocarbons (PPAH), whichare appropriate markers for secondhand smoke and its toxicity. In addition I monitoredcarbon dioxide (CO2), carbon monoxide, temperature, and relative humidity. For SHStracer monitoring, I used real-time battery-powered instruments, including the activemode MIE personalDataRAM (pDR-1200) and the EcoChem PAS 2000CE, a real-timerespirable PPAH monitor. Outdoors, the major sources of PPAH particles are dieselexhaust and cars with defective catalytic converters. PPAH particles are submicron insize, or “nanoparticles.” The calibration and deployment of these instruments isdescribed in Repace (2004). The monitoring instruments were synchronized to eachother and to a wrist watch. A time-activity diary was used to record location and clock-3-

time from the watch at that location for comparison to the RSP and PPAH data measuredat various locations on the UMBC campus in the studies described below.Results.On April 5th and 14th, I performed one field study and two sets of controlledexperiments, as summarized in the figures below. Figure 1 illustrates the effect of a lightand heavy breeze on a cigarette smoke plume. Figure 2 illustrates the effect of no breezeon the cigarette plume, which rises and disperses until it cools and subsides (Repace,2000).Effect of increased wind is to blowthe plume to a more horizontalposition, and narrow the coneangle. Increased wind also increasesturbulence which widensthe conemepluetteracigLighter breezeHeavierbreezecigarette plumeEffect of wind on a smoke plumeJL Repace, 2001Figure 1. Plume cones in light & heavy breezes.Figure 2. Plume rises & falls with no breeze.Figure 3 shows a plot of the real-time data measured on the UMBC campus for RSP(PM3.5) in units of micrograms per cubic meter (µg/ m3) on the left axis, and PPAHconcentrations in nanograms per cubic meter (ng/m3) on the right axis, as a function ofelapsed time in minutes (lower horizontal axis) and clock time (upper horizontal axis).The PPAH monitor was housed in a camera bag mounted on top of a small wheeledsuitcase which housed the RSP monitor. The intakes and exhausts of the concealedmonitors were connected to the outdoor environment.The monitors were deployed about the UMBC campus in a variety of locations onTuesday, April 5th, 2005, including indoors in the Health Services conference room,outdoors where smokers were briefly encountered between the Mathematics andPsychology Buildings between 12:45 and 1:00 PM, on the Commons Building Plaza nearthe cafeteria entrance, and at various distances in the Plaza. A controlled experimentusing 5 smoldered cigarettes was conducted between 2:20 and 2:40, to simulate the effectof smokers outside the cafeteria entrance to the Commons building. The smolderedcigarettes each emit about 90% of the smoke a smoked cigarette. In all cases, the pointsources of smoking were subject to breezes blowing in various directions from WestSouthwest to North-Northwest from 3 to 7 mph. The study ended at about 3:10 PM. It isseen that in the proximity of smokers, both RSP and PPAH peaks are elevated well abovebackground concentrations.-4-

UMBC RESPIRABLE PARTICLE (RSP) AND CARCINOGEN (PPAH) TIME SERIES: OUTDOOR SMOKING12:150 3:001 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5 7 0 7 5 8 0 8 5 9 0 9 5 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 175 180100TUESDAY,APRIL 5, 200590RH% 20% (SD 8%)To C 24.7o C (SD 4o C)804 SMOKERSWITHIN 4-6 ft.,OUTSIDECAFETERIAENTRANCE90RSPPPAH80COMMONS BUILDING PLAZAOUTDOORSCONTROLLEDEXPERIMENTS:5 CAMELSSMOLDEREDFOR 17 MIN.WITHIN 6 IN10-30 ft4030INDOORS20HEALTHSERVICESCONF. ROOM1040PPAH (ng/m3)RSP µg/m37030CLOSESTSMOKERS50 TO 75 ft1-2 SMOKERSWITHIN15 TO 25 ft0-2SMOKERS9 ftONCAMPUSRETURNTOHEALTHBLDG.201000051 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5 7 0 7 5 8 0 8 5 9 0 9 5 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 175 180ELAPSED TIME, minutesRepace Associates, Inc. 2005Figure 3. April 5th field study. Winds were light 3-7 mph, blowing WSW-NNW. One indoor locationand several outdoor locations were sampled with smokers in close and distant proximity. Acontrolled experiment with cigarettes located at a point source was conducted for comparison.April 15th Controlled Experiments.A series of experiments were conducted on Thursday, April 14th to measure theconcentration of SHS as a function of distance from the source. Based on the results ofthe controlled experiment of April 5th, to eliminate variation in concentration due tochanges in wind direction during the time it takes to smoke a cigarette, the source wasarrayed in a ring at 8 -10 points around the compass, so that no matter which way thewind blew, the monitors would pick up the smoke-plume. Up to 10 smokers wererecruited by UMBC Health Services, and they smoked at 3 distances as shown inExperiments I (1-2 smokers only), III (9-10 smokers), and IV (10 smokers). ExperimentsII, V, and VI were conducted with smoldered Marlboro Medium Cigarettes only forcomparison. Initially (Experiment I) 2 smokers were set up upwind of the monitors at 2compass points. The levels are little different from 8 smoldered cigarettes at the samedistance (Experiment II). Similarly, there is little difference between 8 smolderedcigarettes at 1.5 meters and 9.4 smokers at 2 meters. Figure 4 shows the experimentaldesign overlaid on the smokers sitting in chairs around the centrally-located monitor.-5-

MONITORS WILL PICK UP SIGNALFROM DIRECTION OF WINDNNWNERESPIRABLEPARTICLE &CARCINOGENMONITORSradius:1.5, 2, 3,and 5 meters(5 to 16 ft)WWINDSWESESApril 14th Controlled ExperimentthFigure 4. Controlled experiment of April 14 involved simulating an area source, by locatingsmokers or smoldered cigarettes on chairs in a ring around the PAH and RSP monitors. The ringradius was started at 1.5 meters, and increased in steps to 2, 3, and 5 meters. A meter represents 3.28feet. No matter which direction the wind blows from, the receptor will always be downwind.UMBC OUTDOOR SMOKING EXPERIMENT: APRIL 14, 2005 - COMMONS BUILDING PLAZA12:00 PM1:00 PM12:30 PM1:30 PM1:55 PM200200150Expt. [email protected] 1.5meters8 [email protected] 1.5metersExpt. IIIRSPUMBC Physics Dept. Website:62 o F; 30% RH; Wind 11 mph (NE);Gusts ENE @ 1:04 PM 23 mph9.4 [email protected] 2 meters150Expt. IV10 [email protected] 3 meters100Expt. V8 [email protected] 3meters100Expt. VI8 [email protected] [email protected] CARCINOGENS (PPAH) ng/m3RESPIRABLE PARTICLES (RSP), µg/m3PPAHExpt. I0050100150200250300350400450500ELAPSED TIME, 10 second intervals550600650700Repace Associates, Inc. 2005Figure 5. April 14th field study. The diamonds represent the PPAH data in ng/m3, and the circlesrepresent the RSP data in µg/m3. One indoor location and several outdoor locations were sampledwith smokers in close and distant proximity. A controlled experiment with cigarettes located at apoint source was conducted for comparison.-6-

Figure 5 shows the data for RSP and PPAH for each of the experiments as the ringdiameter is increased. Figure 5 shows the data for each of the experiments as a functionof time, numbers of smokers or cigarettes, and ring diameter. RSP is shown on the righthand vertical axis, PPAH on the left-hand vertical axis, and the ring-radius (i.e., thesmoker-to-monitor distance) is shown on the horizontal axis. Figure 6 shows a plot of the3 smoldered cigarette experiments (II, V, and VI); an approximately inverse dependenceof SHS-RSP concentration with source-receptor distance is displayed, while the PPAHconcentration decays approximately as the square of the distance. In controlledexperiments indoors, Repace (2004) observed that PPAH concentrations decreasedapproximately twice as fast as SHS-RSP. Figure 7 plots all of the experiments (I-VI)together, adding the smokers to the smoldered cigarettes. There is considerably morescatter in the data, likely due to the more erratic pattern of smoking by real smokers thanfor smoldered cigarettes. Nevertheless the same dependence with distance emerges fromthe curve fits. Neither concentration appears to get close to background until a distance ofgreater than 7 meters is reached.35y 86.001 * x (-2.4225) R2 0.999773030y 13.127 * x (-1.1159) R2 0.88175PPAH smolder2525RSP smolder202015151010550335Est. SHS Respirable Particulate Concentration, µg/mParticle-bound Polycyclic Aromatic Hydrocarbons, ng/m3UMBC2 8-SMOLDERED CIGARETTE CONTROLLED EXPERIMENT(background-subtracted data)01234567Monitor-to-Cigarette Radius, metersFigure 6. April 15 Experiment. Smoldered cigarettes (Marlboro Medium 100s, filtered) located at8 equally spaced compass positions at ring radii 1.5, 3, and 5 meters. Curve fits to the PPAH andRSP curves are shown, and extrapolated to 7 meters (23 feet). PPAH declines as the inverse squareof the source-receptor distance x, whereas RSP declines inversely as the distance, as

35y 101.67 * x (-2.3883) R2 0.760073030y 23.394 * x (-1.1624) R2 0.1012425252020PPAHRSP15151010550335Est. SHS Respirable Particulate Concentration,µg/mParticle-bound Polycyclic Aromatic Hydrocarbons, ng/m3UMBC2 SMOKED & SMOLDERED CIGARETTE CONTROLLED EXPERIMENT(background-subtracted data)01234567Monitor-to-Cigarette Radius, metersFigure 7. Smoked cigarettes at 1.5, 2, and 3 meters overlayed on the smoldered cigarette plot offigure 5 with curve fits to the combined PPAH and RSP data, extrapolated to 7 meters (23 feet).Although there is more scatter in the data when the smokers are added, approximately the samedependence of PPAH and RSP with distance is seen.Discussion. What levels of SHS constitute clean air? PM2.5 is the RSP size rangethat encompasses combustion-related fine particulate by-products such as tobacco smoke,chimney smoke, and diesel exhaust. PM2.5 is legally regulated in the outdoor air. In1997, the EPA promulgated a 24-hour U.S. Annual National Ambient Air QualityStandard (NAAQS) for RSP for particulate matter PM2.5. The NAAQS for PM2.5 of 65µg/m3, also limited by an annually averaged NAAQS for PM2.5 of 15 µg/m3, based onprotecting human health. The NAAQS for PM2.5 is designed to protect against suchrespirable particle health effects as premature death, increased hospital admissions, andemergency room visits (primarily the elderly and individuals with cardiopulmonarydisease); increased respiratory symptoms and disease (children and individuals withcardiopulmonary disease); decreased lung function (particularly in children andindividuals with asthma); and against alterations in lung tissue and structure and inrespiratory tract defense mechanisms in all persons. PM2.5 and PM3.5 (measured in thisstudy) are closely-related RSP fractions, especially for the submicron SHS aerosol. TableI shows the federal Air Quality Index and the associated color-coded advisories.-8-

While these have averaging times associated with them, the levels may be used toinfer whether a given peak in figures 2 and 4 represent high or low levels of pollution.Each of these figures shows levels as high as 100 to 150 µg/m3 outdoors in proximity tosmokers, indicating that the air is in the unhealthy or Code Red range. Moreover,secondhand smoke causes a number of acute symptoms (eye, nose, and throat irritation,headaches, dizziness, and nausea) and chronic diseases (lung and nasal sinus cancer andheart disease) (CARB, 2003). Levels of irritation begin as low as 4 µg/m3 SHS-RSP andlevels of odor detection are as low as 1 µg/m3 (Junker et al. 2001). Thus SHS odorwould be detectable in our experiments as far as 7 meters from the source, and levels ofirritation would begin at 4 meters from the source.As for the PPAH carcinogens, Figures 2 through 6 show clearly that for thispollutant, levels close to smokers are elevated above background by up to 2 orders ofmagnitude (a factor of 100), relative to distances beyond 7 meters. Thus, it is clear thattobacco smoke pollution outdoors at significant distances from smokers must beconsidered as significantly unhealthy. Thus, while students or faculty asthmatics passthrough a cloud of smoke, levels might be sufficient to trigger an attack, and certainly arehigh enough to pose a nuisance to all. Moreover, smoking in proximity to doorways orair intakes might easily be inducted into the building through posing both acute andchronic threats to building occupants.Table 1. Levels of fine particulate (PM2.5) air pollution and corresponding federalhealth advisory descriptors with accompanying simplified color code (US EPA,1999).Color CodePM2.5 (µg/m3) AQI Air Quality Index CategoryBreak-points0.0 - 15.40 - 50GoodGreen15.5- 40.451 - 100ModerateYellow40.5 - 65.4101 -150Unhealthy SG*Orange65.5 - 150.4151 - 200UnhealthyRed150.5 - 250.4201 - 300Very unhealthyViolet250.5 - 350.4301 - 400HazardousMaroon350.5 - 500.4401- 500Very HazardousMaroon(Significant Harm)** 505500*SG sensitive groups; **exists, but is not a part of the AQI. Source U.S. EPA, 1999.[GUIDELINE FOR REPORTING OF DAILY AIR QUALITY - AIR QUALITY INDEX (AQI) UnitedStates Office of Air Quality EPA-454/R-99-010 Environmental Protection Planning and Standards July1999 Agency Research Triangle Park, NC 27711].Conclusions.These experiments dispel the common misconception that smoking outdoors canbe ignored because smoke plumes immediately dissipate into the environment. Thesecontrolled experiments with and without smokers show similar results: if a receptor suchas a doorway, air intake, or an individual is surrounded by an area source – and thiswould include an entranceway with a group of smokers standing nearby – then regardlessof which way the wind blows, the receptor is always downwind from the source.Cigarette smoke RSP concentrations decline approximately inversely with distancedownwind from the point source, as expected, whereas cigarette smoke PPAHconcentrations decline faster, at approximately inversely as the square of this distance.-9-

Based on these measurements, which involve a single ring of cigarettes orsmokers, the smoke levels do not approach background levels for fine particles orcarcinogens until about 7 meters or 23 feet from the source, which is likely to be thesmoke from no more than 1 or 2 smokers. Greater numbers of smokers in the area couldlead to higher concentrations. because a crowd of smokers constitute an area source,whose plumes may overlap downwind, potentially causing smoke concentrations toincrease locally before dissipating at greater distances. Secondhand smoke causes anumber of acute symptoms (eye, nose, and throat irritation, headaches, dizziness, andnausea) and chronic diseases (lung and nasal sinus cancer and heart disease). Students orfaculty passing through the cloud of smoke would encounter detectable levels at about 7meters (23 feet) from a smoker, and irritating levels at 4 meters (13 feet). Moreover,smokers in proximity to a doorway as persons enter or depart, may result in smoke beinginducted into the building, posing a chronic threat as well as an acute one, to buildingoccupants. Therefore it makes sense to post signs warning smokers not to smoke closerthan about 20 feet from building entrances, and to place ashtrays at that distance and nocloser. Moreover, because some persons suffer from severe asthma, and secondhandsmoke is a known asthmatic trigger, this is another good reason to keep smokers fromcongregating closer to building entrances than 20 feet.References.CARB (2003) "Technical Support Document for the Proposed Identification ofEnvironmental Tobacco Smoke as a Toxic Air Contaminant: Part A," Technical Report.California Environmental Protection Agency, California Air Resources Board, Office ofEnvironmental Health Hazard Assessment, Chapter 5, pp. V6-V19.Junker MH, Danuser B, Monn C, Koller T. Acute sensory responses of nonsmokers atvery low environmental tobacco smoke concentrations in controlled laboratory settings.Environ Health Perspect 2001 Oct;109(10):1045-52.Klepeis NE, Ott WR, Switzer P. Real-Time Monitoring of Outdoor EnvironmentalTobacco Smoke Concentrations: A Pilot Study. Stanford University Department ofStatistics, Sequoia Hall, Stanford, California 94305-4065. University of California, SanFrancisco Contract Number 3317SC, March 1, 2004Repace JL. Banning outdoor smoking is scientifically justifiable. (Invited review)Tobacco Control 9:98 (2000).Repace JL. Respirable Particles and Carcinogens in the Air of Delaware HospitalityVenues Before and After a Smoking Ban. Journal of Occupational and EnvironmentalMedicine, 46:887-905 (2004).Williamson SJ. Fundamentals of Air Pollution. Addison-Wesley, Reading MA, 1973.-10-

The monitors were deployed about the UMBC campus in a variety of locations on Tuesday, April 5. th, 2005, including indoors in the Health Services conference room, outdoors where smokers were briefly encountered between the Mathematics and Psychology Buildings between 12:45 and 1:00 PM, on the Commons Building Plaza near