Atmospheric Environment (v.38, #20)

Special issue of Atmospheric Environment on findings from EPA's Particulate Matter Supersites Program by Constantinos Sioutas; Spyros N Pandis; David T Allen; Paul A Solomon (3101-3106).

Pittsburgh air quality study overview by Ann E. Wittig; Natalie Anderson; Andrei Y. Khlystov; Spyros N. Pandis; Cliff Davidson; Allen L. Robinson (3107-3125).
Ambient sampling for the Pittsburgh Air Quality Study (PAQS) was conducted from July 2001 to September 2002. The study was designed (1) to characterize particulate matter (PM) by examination of size, surface area, and volume distribution, chemical composition as a function of size and on a single particle basis, morphology, and temporal and spatial variability in the Pittsburgh region; (2) to quantify the impact of the various sources (transportation, power plants, biogenic sources, etc.) on the aerosol concentrations in the area; and (3) to develop and evaluate the next generation of atmospheric aerosol monitoring and modeling techniques. The PAQS objectives, study design, site descriptions and routine and intensive measurements are presented. Special study days are highlighted, including those associated with elevated concentrations of daily average PM2.5 mass. Monthly average and diurnal patterns in aerosol number concentration, and aerosol nitrate, sulfate, elemental carbon, and organic carbon concentrations, light scattering as well as gas-phase ozone, nitrogen oxides, and carbon monoxide are discussed with emphasis on the processes affecting them. Preliminary findings reveal day-to-day variability in aerosol mass and composition, but consistencies in seasonal average diurnal profiles and concentrations. For example, the seasonal average variations in the diurnal PM2.5 mass were predominately driven by the sulfate component.
Keywords: Atmospheric aerosol; Supersite; PAQS;

Mass size distributions and size resolved chemical composition of fine particulate matter at the Pittsburgh supersite by Juan C. Cabada; Sarah Rees; Satoshi Takahama; Andrey Khlystov; Spyros N. Pandis; Cliff I. Davidson; Allen L. Robinson (3127-3141).
Size-resolved aerosol mass and chemical composition were measured during the Pittsburgh Air Quality Study. Daily samples were collected for 12 months from July 2001 to June 2002. Micro-orifice uniform deposit impactors (MOUDIs) were used to collect aerosol samples of fine particulate matter smaller than 10 μm. Measurements of PM0.056, PM0.10, PM0.18, PM0.32, PM0.56, PM1.0, PM1.8 and PM2.5 with the MOUDI are available for the full study period. Seasonal variations in the concentrations are observed for all size cuts. Higher concentrations are observed during the summer and lower during the winter.Comparison between the PM2.5 measurements by the MOUDI and other integrated PM samplers reveals good agreement. Good correlation is observed for PM10 between the MOUDI and an integrated sampler but the MOUDI underestimates PM10 by 20%. Bouncing of particles from higher stages of the MOUDI (>PM2.5) is not a major problem because of the low concentrations of coarse particles in the area. The main cause of coarse particle losses appears to be losses to the wall of the MOUDI.Samples were collected on aluminum foils for analysis of carbonaceous material and on Teflon filters for analysis of particle mass and inorganic anions and cations. Daily samples were analyzed during the summer (July 2001) and the winter intensives (January 2002). During the summer around 50% of the organic material is lost from the aluminum foils as compared to a filter-based sampler. These losses are due to volatilization and bounce-off from the MOUDI stages. High nitrate losses from the MOUDI are also observed during the summer (above 70%). Good agreement between the gravimetrically determined mass and the sum of the masses of the individual compounds is obtained, if the lost mass from organics and the aerosol water content are included for the summer. For the winter no significant losses of material are detected and there exists reasonable agreement between the gravimetrical mass and the sum of the concentrations of the individual compounds.Ultrafine particles (below 100 nm) account on average, for < 5% of the PM2.5 mass, and show different composition for the summer and the winter. During the summer the ultrafine mass is 50% carbonaceous material (organic material and elemental carbon) and 50% inorganic (mainly sulfate and ammonium); during the winter these percentages are 70% and 30%, respectively.
Keywords: MOUDI; PM x ; PM size/composition; Condensation mode; Droplet mode; Ultrafine PM; PAQS;

Regional composition of PM2.5 aerosols measured at urban, rural and “background” sites in the Tennessee valley by Roger L. Tanner; William J. Parkhurst; Myra L. Valente; W. David Phillips (3143-3153).
Aerosol composition data, simultaneously collected from collocated samplers at an urban, a rural and a background site in the Tennessee Valley, have been compared for all seasons in 2001. Consistent with previous data, organic aerosols and sulfates are the two largest contributors to fine mass throughout the year at all three sites. Levels of major constituents were not significantly different at the rural and background sites during any season, but levels of organic and elemental carbon were higher at the urban site during fall and winter periods. Seasonal trends at all sites showed maxima in summer for sulfate and significant nitrate levels only in winter, but no significant seasonal trend was observed for organic aerosol mass at any site. Year-to-year variability in aerosol composition at the background site is small compared to day-to-day variability within seasons. The appropriate factor for converting organic carbon to organic mass (conventionally *1.4), used in calculating aerosol chemical composition, may vary with season—larger in warm seasons, smaller in cool seasons.
Keywords: Aerosol chemical composition; Organic aerosols; Seasonal trends; Collocated FRM samplers;

PM data analysis—a comparison of two urban areas: by Shao-Hang Chu; Joseph W. Paisie; Ben W.-L. Jang (3155-3164).
Urban speciated fine particulate data from the Speciation Trends Network from January 2001 to February 2002 were studied in both eastern and western locations of the United States. The seasonal variability of PM2.5 mass, organic carbon, elemental carbon, sulfate ion, nitrate ion, and ammonium cation concentrations were analyzed. Their relationships with ozone and meteorology were also examined. The results reveal that differences in meteorology and emissions have a significant impact on the observed seasonality in species concentrations in Fresno and Atlanta. Based on a parallel analysis of regional PM2.5 episodic events, this influence appears to be general and may typify the difference between eastern and western cities in the United States.In Atlanta, ozone, sulfate, and ammonium were high in the summer when temperatures and humidities were high, whereas organic carbon concentrations were relatively flat year-round. In Fresno, however, ozone concentrations were high but sulfate concentrations were very low even in the summer, whereas PM2.5 concentrations were much higher in the winter and dominated by organics. Meteorologically, in Fresno, it was hot and dry in the summer but cool and humid in the winter. Organic carbon, nitrate, and ammonium ion concentrations were observed to be the highest in late fall and winter when the average relative humidity was the highest (above ∼60%). Much lower mixing heights and frequent stagnations in the winter in Fresno were the major factors influencing the observed high concentrations of various species. The wintertime organic aerosols in Fresno were predominately primary in origin. However, on some very high organic concentration days, up to 30% of the observed organic aerosols could be attributed to secondary organic aerosols (SOAs). These very high organic aerosol concentration days in the winter typically had mild temperatures, high humidities, low dilution rates, and an abundance of nitrate particles. These conditions were favorable for additional SOA formation through the acid catalyzed heterogeneous reactions at night on top of the already high primary organic emissions.
Keywords: PM2.5; Organics; Sulfate; Nitrate; Acid catalyzed; Heterogeneous; SOA; Meteorology;

Ambient fine particulate concentrations and chemical composition at two sampling sites in metropolitan Pittsburgh: a 2001 intensive summer study by William K. Modey; Delbert J. Eatough; Richard R. Anderson; Donald V. Martello; Satoshi Takahama; Leonard J. Lucas; Cliff I. Davidson (3165-3178).
The concentration and chemical composition of ambient fine particulate material (PM2.5) is reported for two sampling sites in the Pittsburgh, Pennsylvania metropolitan area: the Department of Energy, National Energy Technology Laboratory (NETL) PM study site south of the city center, and the Carnegie Mellon Pittsburgh Air Quality Study (PAQS) site 5 km east of central Pittsburgh established with funding by the EPA Supersites Program and by DOE-NETL. Data from these sampling sites were characterized by one to three-day episodes with PM2.5 concentrations (constructed from the sum of the chemical components) exceeding 40.0 μg m−3. The episodes were dominated by high concentrations of ammonium sulfate. The fine particle concentrations were compared with meteorological data from surface weather maps and a Hybrid Single Particle Lagrangian Integrated Trajectory model (HYSPLIT model), with back-trajectories estimated over 24 h. High PM2.5 concentrations were associated with transition from a high pressure to a low pressure regime in advance of an approaching frontal system indicating long-range transport of pollutants. In contrast, fine particulate organic material appeared to be dominated by nearby sources. Distinct differences were observed in the diurnal variations in concentration between the two sites. The NETL site showed clear maximum concentrations of semi-volatile organic material (SVOM) during midday, and minimum concentrations of nonvolatile organic compounds in the afternoon. In contrast, the Carnegie Mellon PAQS site showed an absence of diurnal variation in SVOM, but still with minimum concentrations of nonvolatile organic compounds in the afternoon and evening. Neither site showed significant diurnal variation in ammonium sulfate.
Keywords: PM2.5 episode; PM2.5 composition; Pittsburgh; Local emissions; Long-range transport;

Sources of fine particulate sulfate in New York by Vincent A. Dutkiewicz; Sumizah Qureshi; Adil R. Khan; Vincent Ferraro; James Schwab; Kenneth Demerjian; Liaquat Husain (3179-3189).
Daily PM2.5 sulfate measurements are reported for July 2001 through June 2002 from three New York State sites that were a part of the New York Supersite Program. The sampling sites were located at Queens, Pinnacle State Park, and Whiteface Mountain. Quarterly mean sulfate concentrations at these sites were respectively; 5.05, 5.08, and 3.14 μg/m3 during July–September (QIII); 2.84, 2.26, and 1.40 μg/m3 during October–December (QIV); 2.71, 2.39, and 1.55 μg/m3 during January–March (QI); and 4.17, 4.52, and 2.15 μg/m3 during April–June (QII). Although the sites are separated by several hundred kilometers and surrounded by highly varied population densities, their sulfate concentrations have very similar patterns. These data show that sulfate concentrations over broad regions of the Northeast are correlated. Backward air trajectories were used to evaluate the regional sources of sulfate impacting these sites. The highest sulfate concentrations at all three sites were associated with air masses that pass through the Ohio River Valley and the area around the Great Lakes Basin. Moderately high concentrations at Pinnacle and Queens were also associated with air flow through the Mid-Atlantic states. In addition, we used air trajectories to estimate local versus transported components. On an annual basis, 44–55% of the sulfate at Queens and 60% at Whiteface and Pinnacle was transported.
Keywords: Sulfate; Particulate matter; Aerosols; PM2.5; Source–receptor relationship;

Natural radionuclides in fine aerosols in the Pittsburgh area by Jeffrey S. Gaffney; Nancy A. Marley; Mary M Cunningham (3191-3200).
Natural radionuclides have been proposed for use as tracers in assessing the transport of ozone and aerosols in the troposphere. Beryllium-7 is produced in the upper troposphere and lower stratosphere and after formation rapidly attaches itself to fine aerosol particles. Measurements of 7Be at ground level can therefore be used as a tracer of stratospheric/tropospheric folding events, leading to injection of ozone and upper atmospheric aerosols into the lower atmosphere. Since its concentration varies little on regional scales it can also be used to detect and correct for sampling problems in particulate monitoring networks. Lead-210 and its progeny, 210Bi, and 210Po can also be used to determine the apparent tropospheric residence times for fine aerosols by looking at the 210Bi/210Pb and 210Po/210Pb activity ratios. Reported here are measurements of the natural radionuclides 7Be and 210Pb, taken at two sites near Pittsburgh, PA in the summer of 2001 and at a site near Centerton, New Jersey in 1999 during the NEOPS field campaign. Beryllium-7 results show no evidence of upper atmospheric input during the sampling period. Apparent residence times as calculated from 210Po/210Pb ratios are shorter for Pittsburgh than for Centerton, and shorter for both sites than those obtained previously in other areas, indicating a local aerosol source as well as a higher loading of water-soluble species such as sulfate and nitrate. A comparison of fine and course aerosol lifetimes shows no contribution of excess 210Po from wind-blown soil or from coal-fired power plants.
Keywords: Aerosols; Fine particulates; Aerosol residence times; 210Pb; 210Po; 210Bi; Natural radioactivity in aerosols; Lead-210 in aerosols;

Semi-continuous PM2.5 inorganic composition measurements during the Pittsburgh Air Quality Study by Ann E. Wittig; Satoshi Takahama; Andrei Y. Khlystov; Spyros N. Pandis; Susanne Hering; Brent Kirby; Cliff Davidson (3201-3213).
A method for semi-continuous (10 min time resolution) PM2.5 nitrate and sulfate measurements, based on the humidified impaction with flash volatilization design of Stolzenburg and Hering (Environ. Sci. Technol. 34 (2000) 907), was evaluated during the Pittsburgh Air Quality Study (PAQS) from July 2001 to August 2002. The semi-continuous measurements were corrected for several operating parameters. The overall corrections were less than 10% on average, but could be quite large for individual 10 min measurements. These corrections resulted in an improvement in the agreement of the measurements with the filter-based measurements, with a major axis regression relationship of y=0.83x+0.20 μg m−3 and R 2 of 0.84 for nitrate and y=0.71x+0.42 μg m−3 and R 2 of 0.83 for sulfate. The corrected semi-continuous measurements were calibrated over the entire year using collocated denuder/filter-pack-based measurements. These calibrated semi-continuous measurements are used in conjunction with temporally resolved gas-phase measurements of total (gas- and aerosol-phase) nitrate and meteorological measurements to investigate short-term phenomena at the Pittsburgh Supersite. The gas-to-particle partitioning of nitrate varied daily and seasonally, with a majority of the nitrate in the particle phase at night and during the winter months.
Keywords: Atmospheric aerosols; PM2.5 nitrate; PM2.5 sulfate; Continuous monitors;

Ultrafine nitrate particle events in Baltimore observed by real-time single particle mass spectrometry by Michael P. Tolocka; Derek A. Lake; Murray V. Johnston; Anthony S. Wexler (3215-3223).
Ambient particles in Baltimore, Maryland were characterized from April through November 2002 using the real-time single particle mass spectrometer, RSMS III. When particles containing nitrate were examined, two types of ultrafine particle events were revealed: a large burst of nominally “pure” nitrate particles in the 50–90 nm size range, and a smaller (and less frequent) burst of “pure” particles in the 50–90 nm size range that grew to 110–220 nm with time. Coincident with both of these events was an increase in the number of mixed composition particles containing nitrate, suggesting that they were formed by condensation of ammonium nitrate onto pre-existing particles. Meteorological variables, particle number concentrations and continuous nitrate mass measurements were compared to the single particle data. Number and mass concentrations estimated from RSMS III correlated well with similar measurements with other techniques. Ultrafine nitrate particle events were observed during periods of low temperature and high relative humidity as expected from ammonium nitrate equilibrium considerations. During these events, the partitioning of ammonium nitrate to the particle phase strongly influenced the particle number concentration as well as the chemical composition.
Keywords: Ambient aerosol nitrate mass spectrometry;

Seasonal and spatial trends in organic carbon (OC) concentrations in fine particulate matter (PM) were examined using data collected through the regulatory fine PM monitoring network in southeast Texas, and data collected during the Gulf Coast Aerosol Research and Characterization Study (GC-ARCH or Houston Supersite). Primary OC concentrations in the aerosol were estimated by establishing a linear relationship between primary OC and elemental carbon (EC) from 24-hour integrated samples. The relationship between primary OC and EC varied by site and season. Secondary OC (SOC) was estimated as the difference between total OC and primary OC and was found to be between 0.65±1.11 and 1.15±1.52 μg m−3 among sites (mass of carbon only), averaged over 2 years. The mean fraction of SOC in PM2.5 was found to be between 0.05±0.06 and 0.10±0.33. Primary OC concentrations were on average higher than SOC concentrations at all locations, and were between 1.50±0.69 and 2.89±1.03 μg m−3 (mass of carbon only). Both primary and SOC have highest monthly mean concentrations in early fall through late winter throughout the region. Spatially, sites that are closer to urban emissions have, on average, higher primary and SOC concentrations. One exception to this trend is a remote site located in a forested area north of Houston, where primary and SOC tend to be as high as at urban/industrial locations.
Keywords: Particulate matter; PM2.5; Secondary organic aerosol;

Seasonal variation of the particle size distribution of polycyclic aromatic hydrocarbons and of major aerosol species in Claremont, California by Antonio H. Miguel; Arantzazu Eiguren-Fernandez; Peter A. Jaques; John R. Froines; Bill L. Grant; Paul R. Mayo; Constantinos Sioutas (3241-3251).
As part of the Southern California Particle Center and Supersite (SCPCS) activities, we measured, during all seasons, particle size distributions of 12 priority pollutant polycyclic aromatic hydrocarbons (PAHs), concurrently with elemental carbon (EC), organic carbon (OC), sulfate (SO4 2−), and nitrate (NO3 ) size distributions, from October 2001 to July 2002 in Claremont, CA, a receptor site located about 40 km downwind of central Los Angeles. Samples were collected approximately once every week, for 24-h periods, from midnight to midnight. MOUDI impactors collected samples at 30 LPM which were composited for analysis into monthly periods in three aerodynamic diameter size intervals, defined for the purpose of this work, as: 0–0.18 μm (ultrafine mode), 0.18–2.5 μm (accumulation mode), and 2.5–10 μm (coarse mode). For the monthly composites from October to February, the size distributions of the target PAHs are similar. However, from March to July, notable differences are observed: a significant fraction of the PAH mass is found in the coarse mode, as compared with the previous period. During the entire 1-year period, the form and shape of the EC size distributions did not vary much and are distinguished by prominent mass in the ultrafine and accumulation size mode. For the individual modes of the major species, the highest Pearson's correlation coefficients for the variation of temperature with species concentration were found in the ultrafine mode for both SO4 2− (0.92) and EC (0.90), and in the coarse mode for both OC (0.85) and NO3 (0.54). High SO4 2− correlations are consistent with increased gas-to-particle formation during the warmer months from (precursor) SO2 emissions in the Los Angeles and Lon Beach seaport areas and, similarly for EC, increased atmospheric transport to Claremont as the season progresses from winter to summer. Although not statistically significant, the correlations were negative for the less volatile or particle phase group (log [p L o]⩽−3.22), consistent with increased partitioning from the vapor phase with decreasing temperature.
Keywords: Polycyclic aromatic hydrocarbons; Elemental carbon; Organic carbon; Sulfate; Nitrate; Size distribution; Claremont; California;

Fine particle samples were collected at three sites in the Houston area during TexAQS 2000. Polar organic species including levoglucosan, n-alkanoic acids, n-alkenoic acids and dicarboxylic acids were quantified to characterize the composition of local ambient organic aerosols. Levoglucosan, a tracer for biomass burning, was measured as a dominant compound, followed by n-alkanoic acids. The compound distribution of higher molecular weight n-alkanoic acids (⩾C22) was consistent with biogenic emission sources. n-Alkenoic acids exhibit variable concentrations during sampling period with high ratios of octadecanoic to octadecenoic acid. The concentrations of dicarboxylic acids are compared with unapportioned organic carbon calculated by the chemical mass balancing technique. This correlation supports the conclusion that diacids are generated by photochemical reactions and contribute to the unapportioned organic mass. Data collected during a regional wood smoke episode are used to examine the influence of wood combustion to ambient particulate matter concentrations.
Keywords: PM2.5; EPA fine particulate matter supersites; Levoglucosan; n-Alkanoic acids; n-Alkenoic acids; Dicarboxylic acids;

Number concentrations of fine and ultrafine particles containing metals by Michael P. Tolocka; Derek A. Lake; Murray V. Johnston; Anthony S. Wexler (3263-3273).
Typical classification schemes for large data sets of single-particle mass spectra involve statistical or neural network analysis. In this work, a new approach is evaluated in which particle spectra are pre-selected on the basis of an above threshold signal intensity at a specified m/z (mass to charge ratio). This provides a simple way to identify candidate particles that may contain the specific chemical component associated with that m/z. Once selected, the candidate particle spectra are then classified by the fast adaptive resonance algorithm, ART 2-a, to confirm the presence of the targeted component in the particle and to study the intra-particle associations with other chemical components. This approach is used to characterize metals in a 75,000 particle data set obtained in Baltimore, Maryland. Particles containing a specific metal are identified and then used to determine the size distribution, number concentration, time/wind dependencies and intra-particle correlations with other metals. Four representative elements are considered in this study: vanadium, iron, arsenic and lead. Number concentrations of ambient particles containing these elements can exceed 10,000 particles cm−3 at the measurement site. Vanadium, a primary marker for fuel oil combustion, is observed from all wind directions during this time period. Iron and lead are observed from the east–northeast. Most particles from this direction that contain iron also contain lead and most particles that contain lead also contain iron, suggesting a common emission source for the two. Arsenic and lead are observed from the south–southeast. Particles from this direction contain either arsenic or lead but rarely both, suggesting different sources for each element.
Keywords: Real-time single-particle mass spectrometry; Ambient aerosol; Metals; Particle number concentration; Chemical composition;

Twelve months of aerosol size distributions from 3 to 560 nm, measured using scanning mobility particle sizers are presented with an emphasis on average number, surface, and volume distributions, and seasonal and diurnal variation. The measurements were made at the main sampling site of the Pittsburgh Air Quality Study from July 2001 to June 2002. These are supplemented with 5 months of size distribution data from 0.5 to 2.5 μm measured with a TSI aerosol particle sizer and 2 months of size distributions measured at an upwind rural sampling site. Measurements at the main site were made continuously under both low and ambient relative humidity. The average Pittsburgh number concentration (3–500 nm) is 22,000 cm−3 with an average mode size of 40 nm. Strong diurnal patterns in number concentrations are evident as a direct effect of the sources of particles (atmospheric nucleation, traffic, and other combustion sources). New particle formation from homogeneous nucleation is significant on 30–50% of study days and over a wide area (at least a hundred kilometers). Rural number concentrations are a factor of 2–3 lower (on average) than the urban values. Average measured distributions are different from model literature urban and rural size distributions.
Keywords: Number concentration; Aerosol size distribution; Urban air quality; Pittsburgh Air Quality Study; PAQS;

Measurements of aerosol size distributions and hygroscopicity were combined to infer the size-resolved composition of the ambient aerosol during the Houston Supersite campaign from June through October, 2001. The 3000+ distributions used in this analysis were measured at the Aldine site north and typically downwind of the greater Houston metropolitan area. Size distributions spanning the diameter range from 0.025 to 0.700 μm, and hygroscopic behavior at eight logarithmically spaced dry diameters from 0.025 to 0.344 μm were analyzed. At smaller dry sizes, the aerosol hygroscopic growth factor distributions were typically monomodal and peaked at or near a growth factor of 1.0, indicating predominantly non-hygroscopic aerosols. Particles larger than ∼0.100 μm exhibited bimodal growth patterns, with increasing importance of the hygroscopic mode with increasing dry particle size.Hygroscopic growth distributions were used to partition the aerosol into pure insoluble, mixed insoluble, mixed soluble, and pure soluble categories. This categorization scheme was used to analyze in detail four multi-day episodes. During two of these episodes new particle formation events were observed daily. The recently formed particles were only sparingly hygroscopic, suggesting they were composed primarily of insoluble material. To contrast these periods during which pronounced diurnal variability was observed, a third period was chosen because the aerosol concentration and composition varied little. The fourth period analyzed spanned the passage of a cold front, which had a pronounced influence on the aerosol distributions.
Keywords: Aerosol; DMA; TDMA; Hygroscopic; Hygroscopicity; Organic; Houston; Nucleation;

Mass balance closure and the Federal Reference Method for PM2.5 in Pittsburgh, Pennsylvania by Sarah L Rees; Allen L Robinson; Andrey Khlystov; Charles O Stanier; Spyros N Pandis (3305-3318).
Daily ambient aerosol samples were taken in Pittsburgh, Pennsylvania from the summer 2001 to the winter 2002 as part of the Pittsburgh Air Quality Study (PAQS). The study measured PM2.5 mass by the Federal Reference Method (FRM) and the PM2.5 chemical composition by a variety of filter-based and continuous instruments. This paper examines the mass balance between the FRM-measured mass and the sum of the aerosol chemical components. For the 7-month study period, the average FRM-measured mass is 11% greater than the sum of the mass of the aerosol chemical components. This mass balance discrepancy varies seasonally, with the average FRM-measured mass 17% greater than the sum of the chemical components for the summer months, with discrepancies as large as 30% during certain periods. Meanwhile, the FRM-measured mass was at or slightly below the sum of the chemical components for the winter months.The mass balance discrepancy and its seasonal shift cannot be explained by measurement uncertainty; instead the discrepancy is due to combination of retained aerosol water on the conditioned FRM filters and volatilization losses. The relative importance of these different effects varies with aerosol composition and causes the observed seasonal variation in the mass balance. The contribution of the aerosol water to the FRM-measured mass is estimated using continuous measurements of aerosol water at the site; volatilization losses are estimated from other filter-based instruments. Water contributes 16% of the FRM mass in the summer, and 8% of the FRM mass in the winter; it also appears responsible for episodes where the FRM-measured mass is significantly greater than the sum of components. Retention of water is greatest during acidic conditions, which commonly occur during the summer months. Volatilization losses are estimated at 5% of the FRM mass during the summer, and 9% for the winter. Volatilization losses appear to be most significant on days dominated by organic aerosol, or winter days with relatively high nitrate concentration. Accounting for the effects of water and volatilization losses closes the mass balance between the FRM and the sum of the chemical components, providing insight into the FRM measurements.
Keywords: Federal Reference Method; Mass balance; Aerosol water; Aerosol sampling; Aerosol composition; Aerosol volatilization;

Laser-induced breakdown spectroscopy (LIBS) was used to measure the distribution of seven species in individual ambient aerosol particles during an 8-day period from 26 August to 2 September 2002 at the Pittsburgh Aerosol Supersite. Particle hit rates were on the order of 10−4–10−5 for Al, Ca, Cr, Cu, Mg, Mn, and Na. Weekly average concentrations between 29 and 720 ng m−3 are reported along with conservative threshold detection limits for individual particles between 15 and 184 fg, depending on the element. Hourly concentrations are reported for Ca, Mg, and Na; Mg concentrations are found to be somewhat correlated with both Ca and Na, while Ca and Na appear uncorrelated. A representative example of measured Mg particle masses illustrates that the detection threshold poses a limitation in this data set, which could be rectified in future implementations. Finally, the presence of multi-element particles in the data set suggest the use of high-sensitivity, wide-range echelle spectrometers for particle source apportionment and determination of associations between elements.
Keywords: Particulate matter; Aerosol composition; Laser-induced breakdown spectroscopy; LIBS; PM2.5;

The need for continuous personal monitoring for exposure to particulate matter has been demonstrated by recent health studies showing effects of PM exposure on time scales of less than a few hours. Filter-based methods cannot measure this short-term variation of PM levels, which can be quite significant considering human activity patterns. The goal of this study was to evaluate the active-flow personal DataRAM for PM2.5 (MIE pDR-1200; Thermo Electron Corp., Franklin, MA) designed as a wearable monitor to continuously measure particle exposure. The instrument precision was found to be good (2.1%) and significantly higher than the passive pDR configuration tested previously. A comparison to other proven continuous monitors resulted in good agreement at low relative humidities. Results at higher humidity followed predictable trends and provided a correction scheme that improved the accuracy of pDR readings. The pDR response to particle size also corresponded to previously observed and theoretical errors. The active flow feature of the pDR allows collection of the sampled particles on a back-up filter. The 24-h mass measured on this filter was found to compare very well with a Federal Reference Method for PM2.5 mass.
Keywords: Personal exposure; Personal monitor; PM2.5; DataRAM; pDR; PM; Particulate matter; Nephelometer;

Quality control of semi-continuous mobility size-fractionated particle number concentration data by Rong Chun Yu; Hee Wen Teh; Peter A. Jaques; Constantinos Sioutas; John R. Froines (3341-3348).
Fine and ultrafine particles have been postulated to play an important role in the association between ambient particulate matters and adverse health effects. As part of the EPA Supersite Program, the Southern California Particle Center & Supersite has conducted a series of monitoring campaigns that contribute to a better understanding of the sources, chemical composition and physical state of ambient aerosols. The Scanning Mobility Particle Sizer (SMPS) was deployed to semi-continuously measure mobility size-fractioned particle number concentrations. As part of the quality control efforts, we developed a two-stage graphic and statistical procedure to label and identify potentially discordant observations. The first stage considered the entire size-fractionated data by date-time as a whole to plot total concentration (TC) vs. coefficient of variation (CV), both in log scale. TC represents the magnitude of overall concentration for a size distribution; while CV represents the relative variability. This plot was used to partition all size distributions into four to five distinct regions. In each region, a generalized extreme studentized deviate (ESD) and a modified Z-score procedure were applied to identify potential discordant outliers. We have found that the majority of particle size distributions are concentrated within a ‘normal’ region, with TC ranging from 102 to 105  cm−3 and CV varying between 20% and 200%. Size distributions that are contaminated with discordant outliers are displayed distinctly from the ‘normal’ region and form four to five clusters in the Log TC–Log CV plot. The pattern of clusters in the plot is consistent among the four sampling sites in this study, suggesting the robustness of this technique. The generalized ESD and modified Z-score effectively identify discordant outliers and reveal that the pattern of clustering outliers are consistent within each distinct region. It has, thus, been concluded that the new approach is a useful quality control tool to identify potential discordant outliers in SMPS data.
Keywords: Scanning mobility particle sizer (SMPS); Quality control; Ultrafine particle; Outlier; Generalized extreme studentized deviate method; Modified Z-score method;

Daily integrated PM2.5 (particulate matter ⩽2.5 μm in aerodynamic diameter) composition data including eight individual carbon fractions collected at the Jefferson Street monitoring site in Atlanta were analyzed with positive matrix factorization (PMF). Particulate carbon was analyzed using the thermal optical reflectance method that divides carbon into four organic carbon (OC), pyrolized organic carbon (OP), and three elemental carbon (EC) fractions. A total of 529 samples and 28 variables were measured between August 1998 and August 2000. PMF identified 11 sources in this study: sulfate-rich secondary aerosol I (50%), on-road diesel emissions (11%), nitrate-rich secondary aerosol (9%), wood smoke (7%), gasoline vehicle (6%), sulfate-rich secondary aerosol II (6%), metal processing (3%), airborne soil (3%), railroad traffic (3%), cement kiln/carbon-rich (2%), and bus maintenance facility/highway traffic (2%). Differences from previous studies using only the traditional OC and EC data (J. Air Waste Manag. Assoc. 53(2003a)731; Atmos Environ. (2003b)) include four traffic-related combustion sources (gasoline vehicle, on-road diesel, railroad, and bus maintenance facility) containing carbon fractions whose abundances were different between the various sources. This study indicates that the temperature resolved fractional carbon data can be utilized to enhance source apportionment study, especially with respect to the separation of diesel emissions from gasoline vehicle sources. Conditional probability functions using surface wind data and identified source contributions aid the identifications of local point sources.
Keywords: Thermal optical method; Carbon fraction; Positive matrix factorization; Source apportionment; Conditional probability function;

Analysis of motor vehicle emissions in a Houston tunnel during the Texas Air Quality Study 2000 by Gary R. McGaughey; Nimish R. Desai; David T. Allen; Robert L. Seila; William A. Lonneman; Matthew P. Fraser; Robert A. Harley; Alison K. Pollack; Jason M. Ivy; James H. Price (3363-3372).
Measurements from a Houston tunnel were used to develop fuel consumption-based emission factors for CO, NO x , and non-methane organic compound (NMOC) for on-road gasoline vehicles. The Houston NO x emission factor was at the low range of emission factors reported in previous (primarily pre-1996) tunnel studies while the NMOC emission factor was slightly higher than that reported in the previous tunnel studies.
Keywords: Tunnel Study; Vehicle emission factors; On-road gasoline vehicles; Hydrocarbon speciation;