Atmospheric Environment (v.41, #17)

Aqueous-phase nitration of phenol by N2O5 and ClNO2 by Mathew R. Heal; Mark A.J. Harrison; J. Neil Cape (3515-3520).
Nitrophenols are present in the atmospheric gas phase and in cloud and rainwater. Their formation via aqueous-phase reactions of phenol with the nitronium ion, NO2 +, arising from N2O5 and ClNO2 partitioning into the aqueous phase, has been proposed but not verified experimentally. Here, we demonstrate for the first time that gaseous N2O5 and ClNO2 partitioning into dilute aqueous solutions of phenol yields 2- and 4-nitrophenol (and 4-nitrosophenol), but no dinitrophenol isomers. The rate of nitration does not vary significantly between 5 and 20 °C, presumably because of opposing temperature dependences in Henry's law partitioning and reaction rate coefficients. The rate coefficient for reaction of NO2 + with phenol could not be directly quantified but is evidently large enough for this reaction to compete effectively with the reaction between NO2 + and water and to provide a feasible route to nitrophenol production in the atmosphere.
Keywords: Tropospheric chemistry; Heterogeneous chemistry; Nitrosation; Nitronium ion;

A major problem with the analysis and investigation of combustion aerosols in the real-world environment is related to strong stochastic variations of the external and environmental parameters and factors (e.g., atmospheric turbulence, traffic fluctuations, etc.). Therefore, this paper develops new powerful statistical methods based on the canonical correlation analysis and the moving average technique, applied to combustion aerosols near a busy road. As a result, a new physical insight into the evolution of combustion aerosols and possible sources of nano-particle modes is presented and discussed. Several new particle modes are identified, analysed and associated either with trucks or cars on the road. In particular, liquid and solid particle modes are identified, and the mechanism of thermal fragmentation of solid nano-particles is used for the interpretation of the obtained results.
Keywords: Combustion aerosols; Statistical analysis; Canonical correlations; Sources of particles;

Multi-channel statistical analysis of combustion aerosols. Part II by D.K. Gramotnev; G. Gramotnev (3535-3545).
This paper develops a new method of statistical analysis of interaction and transformation between different modes in the particle size distribution in atmospheric aerosols in the presence of strong stochastic fluctuations of the environmental and meteorological parameters. Fast processes of mode transformation are investigated in combustion aerosols near a busy road on the basis of the fragmentation mechanism of particle evolution. A unique anti-symmetric correlation pattern between different modes is described and explained by means of the formulated fragmentation theorem. This provides yet another confirmation of the fragmentation mechanisms of aerosol evolution.
Keywords: Combustion aerosols; Statistical analysis; Aerosol evolution; Particle modes; Particle formation;

At a suburban location in southern Korea, the optical properties of the atmosphere were characterized over a horizontal light path of 1.5 km (in two ways) from 22 October to 19 November 2002. This was achieved by measuring light extinction with a long path differential optical absorption spectrometer system in the ultraviolet and visible wavelength region. The extinction coefficients were obtained relatively as a ratio of a target air spectrum to a defined reference spectrum measured over the same light path (290–760 nm). To assess the measured extinction coefficients, the extinction coefficients at 550 nm were compared to those measured with a commercial long-path transmissometer.To avoid the absorption of known gases (H2O, NO2, and O3), extinction coefficients at the spectral bands of 325, 394, 472, 550, 580, 680, and 753 nm, with 3 nm window widths, were selected and analyzed for the purpose of the study of the aerosol properties. Importantly, the atmospheric physical properties during the episodes (such as a biomass burning or a dust storm) were investigated by means of the Angstrom parameters and the mass ratio of fine-to-coarse particles. In addition, it was found that the Angstrom exponents decrease monotonously for relative humidity above 50%.
Keywords: Angstrom exponent; Turbidity; Mass ratio of fine-to-coarse particles; Biomass burning; Dust storm;

An analysis on abnormally low ozone in the upper troposphere over subtropical East Asia in spring 2004 by C.Y. Chan; Y.S. Li; J.H. Tang; Y.K. Leung; M.C. Wu; L.Y. Chan; C.C. Chang; S.C. Liu (3556-3564).
Abnormally low ozone (O3) mixing ratios were observed by electrochemical concentration cell (ECC) ozonesondes in the upper troposphere over subtropical East Asia in spring 2004, a season when high tropospheric O3 is usually observed in the region. Low O3 with a lowest mixing ratio of 13 ppbv, less than a fourth of the respective seasonal average of 60–100 ppbv, was observed at 11–18 km above ground over Hong Kong (22.31°N, 114.17°E), Sanya (18.23°N, 109.52°E) and Taipei (24.98°N, 121.43°E). The origin of the low O3 was investigated using meteorological evidence, satellite imagery and three-dimensional backward air trajectory. We found for the first time that the low O3 resulted from deep convective pumping of low O3 maritime air masses near the center of typhoon Sudal from the boundary layer of the tropical region to the east of the Philippines to the upper troposphere. The low O3 air masses were then transported to the higher latitudes far ahead of the typhoon following the long-range transport driven by the circulations associated with the typhoon and the northern Hadley cell. The findings of this study highlight that more research efforts are needed to understand the effect of the circulation associated with tropical cyclones on the distribution and budget of O3 and other trace gases in the troposphere.
Keywords: Ozone; China; Typhoon; Deep convection; Hadley circulation;

The number of airborne pollen produced from plants is visible as a haze mixed with urban air pollution in an urban area when atmospheric conditions are proper for pollination of pollen from April to May in Korea. The big loading of airborne pollen can cause further visibility degradation in an urban area. In order to investigate physico-chemical characteristics of visibility impairment by airborne pollen, chemical aerosol measurements, optical aerosol monitoring, and scenic monitoring were performed during the intensive monitoring period from April 19 to May 2, 2005 in the urban area of Seoul, Korea. The particles collected on filters were examined with a scanning electron microscope (SEM) interfaced with an energy dispersive X-ray analysis to characterize size, elemental composition, and count of airborne pollen. During the airborne pollen period, the daily averaged number concentrations of airborne pollen; P PM 2.5 and P PM 10 were calculated to be 8.4±6.9 and 113.7±91.1 m−3, respectively. The daily averaged light extinction coefficient (b ext), light scattering coefficient for open path (b scat), light scattering coefficient for dry particle in the fine regime (b scat,fine), and light absorption coefficient in the fine regime (b abs,fine) were measured to be 459±267, 357±214, 263±165, and 44±30 Mm−1, respectively. And mass concentration of PM2.5 and PM10 were measured to be 46.5±29.1 and 97.0±41.7 μg m−3. The average light absorption coefficient by airborne pollen was estimated to be about 30 M m−1 and the average light scattering coefficient by airborne pollen was estimated to be 67±57 Mm−1. During the airborne pollen period the average contribution of airborne pollen to visibility impairment was roughly estimated to be 19–25%.
Keywords: Visibility; Airborne pollen; Optical property; Light extinction;

Chemical composition of aerosols during a major biomass burning episode over northern Europe in spring 2006: Experimental and modelling assessments by Sanna Saarikoski; Markus Sillanpää; Mikhail Sofiev; Hilkka Timonen; Karri Saarnio; Kimmo Teinilä; Ari Karppinen; Jaakko Kukkonen; Risto Hillamo (3577-3589).
The long-range transported smokes emitted by biomass burning had a strong impact on the PM2.5 mass concentrations in Helsinki over the 12 days period in April and May 2006. To characterize aerosols during this period, the real-time measurements were done for PM2.5, PM2.5–10, common ions and black carbon. Moreover, the 24-h PM1 filter samples were analysed for organic and elemental carbon (OC and EC), water-soluble organic carbon (WSOC), ions and levoglucosan. The Finnish emergency and air quality modelling system SILAM was used for the forecast of the PM2.5 concentration generated by biomass burning. According to the real-time PM2.5 data, the investigated period was divided into four types of PM situations: episode 1 (EPI-1; 25–29 April), episode 2 (EPI-2; 1–5 May), episode 3 (EPI-3; 5–6 May) and a reference period (REF; 24 March–24 April). EPI-3 included a local warehouse fire and therefore it is discussed separately. The PM1 mass concentrations of biomass burning tracers—levoglucosan, potassium and oxalate—increased during the two long-range transport episodes (EPI-1 and EPI-2). The most substantial difference between the episodes was exhibited by the sulphate concentration, which was 4.9 (±1.4) μg m−3 in EPI-2 but only 2.4 (±0.31) μg m−3 in EPI-1 being close to that of REF (1.8±0.54 μg m−3). The concentration of particulate organic matter in PM1 was clearly higher during EPI-1 (11±3.3 μg m−3) and EPI-2 (9.7±4.0 μg m−3) than REF (1.3±0.45 μg m−3). The long-range transported smoke had only a minor impact on the WSOC-to-OC ratio. According to the model simulations, MODIS detected the fires that caused the first set of concentration peaks (EPI-1) and the local warehouse fire (EPI-3), but missed the second one (EPI-2) probably due to dense frontal clouds.
Keywords: Aerosols; Biomass burning; Chemical composition; Long-range transport;

Optimization of the molecular geometry of 209 polychlorinated diphenyl ethers (PCDEs) and diphenyl ether (DE) were carried out at the B3LYP/6-31G* level with Gaussian 98 program. Based on the theoretical linear solvation energy relationship (TLSER) model, the obtained structural parameters were taken as theoretical descriptors to establish the first novel QSPR model (I) for predicting subcooled liquid vapor pressure (P L) of PCDEs. The model (I) achieved in this work contains three variables: most negative atomic partial charge in molecule (q ), dipole moment of the molecules (μ) and mean molecular polarizability (α), of which r 2=0.988, SD=0.134. Meanwhile, it was found that there was significant correlation between the Cl substitution positions and some structural parameters. Consequently, the numbers of Cl substitution positions were also taken as theoretical descriptors to establish the second novel QSPR model (II) for predicting P L of all PCDE congeners, of which r 2=0.991, SD=0.112. In addition, the t-values of two models are both large and the variation inflation factors (VIF) of variables herein are all less than 5.0, suggesting significant statistics of the P L predicting models. And the results of cross-validation test also showed that the second model exhibited optimum stability and better predictive power. Moreover, results of prediction were further compared with data from similar studies by semi-empirical PM3 method, molecular connectivity indices (MCIs) method and molecular descriptors method.
Keywords: Polychlorinated diphenyl ethers (PCDEs); Subcooled liquid vapor pressure (P L); Quantitative structure–property relationship (QSPR); Density functional theory (DFT); Position of Cl substitution (PCS) method;

Variability of atmospheric pesticide concentrations between urban and rural areas during intensive pesticide application by Anne Scheyer; Stéphane Morville; Philippe Mirabel; Maurice Millet (3604-3618).
Intensive pesticide use leads to the contamination of water, soil and atmosphere. Atmospheric transport is responsible for pesticide dispersal over long distances. In this study, we evaluate the local dispersal of pesticides from agricultural to urban areas. For this purpose, three high-volume samplers, each equipped with a glass fiber filter and XAD-2 resin for the sampling of particulate and gas phase have been placed in a south-west transect (predominant wind direction) characteristic of rural and urban areas. The urban site (Strasbourg centre) is situated in the middle of two rural sites. Samples were taken simultaneously at three sites during pesticide treatments in autumn and spring 2002–2003. Sampling took place for 24 h at a flow rate of 10–15 m3  h−1. The pesticides studied were those commonly used in the Alsace region for all crops (maize, cereal, vines …). Many of the pesticides analysed in atmospheric samples were not detected or observed very episodically at very low concentrations. For metolachlor, alachlor, trifluralin, atrazine and diflufenican, higher concentrations were observed, essentially during the application of these compounds. Moreover, some “spraying peaks” were observed for alachlor in the south rural site (near crops) at a level of 31 ng m−3 on 16–17 May 2003. These results show site and time dependence of atmospheric contamination by pesticides. A limited dispersal was also observed especially in the urban area during the application periods of pesticides.
Keywords: Pesticides; Atmosphere; Spatial and temporal variations of concentrations; Pesticide applications periods;

Copper complexation in coastal rainwater, southeastern USA by Melanie Witt; Stephen Skrabal; Robert Kieber; Joan Willey (3619-3630).
Complexation of dissolved copper (Cu) was studied in Atlantic coastal rainwater using adsorptive cathodic stripping voltammetry (ASCV) with salicylaldoxime as a competing ligand at pH 7.8. Detectable concentrations of strong Cu-complexing ligands possessing conditional stability constants of 1013–1016 were observed in over 80% of the rain events occurring over the course of 2 year. In 11 of the 23 samples analysed, total dissolved Cu concentrations were higher than those of dissolved ligands, indicating that a significant fraction of the Cu occurred as the free ion and as weaker complexes. In the remaining samples, ligand concentrations were equal to or greater than Cu concentrations, indicating virtually complete (>99%) complexation of the ambient Cu. By varying the analytical detection window, two classes of ligands with differing conditional stability constants were detected in selected rain samples suggesting that the Cu ligands most likely represent a spectrum of organic compounds. Back trajectory analysis indicated that continentally dominated rain samples contained higher concentrations of Cu and organic ligands relative to storms of marine origin, suggesting a strong terrestrial and/or anthropogenic source of both Cu and ligands in rain at this location. Variability in Cu speciation may impact a variety of atmospheric redox reactions because free and complexed forms of the metal have very different reactivities.
Keywords: Copper; Trace metals; Atmosphere; Speciation; Precipitation; Trajectories;

Air quality and organic compounds in aerosols from a coastal rural area in the Western Iberian Peninsula over a year long period: Characterisation, loads and seasonal trends by Tiago Oliveira; Casimiro Pio; Célia Alves; Armando Silvestre; Margarita Evtyugina; Joana Afonso; Alexandre Caseiro; Michel Legrand (3631-3643).
Ambient samples of fine organic aerosol collected from a rural area (Moitinhos) in the vicinity of the small coastal Portuguese city of Aveiro over a period of more than one year have been solvent-extracted and quantitatively characterised by gas chromatography–mass spectrometry. Particles were also analysed with a thermal-optical technique in order to determine their elemental and organic carbon content. In addition, meteorological sensors and real-time black carbon, ozone and carbon monoxide monitors were used. Particulate matter values were higher than background levels in continental Europe. A patent seasonal variation for organic and elemental carbon concentrations was observed, presumably related to stronger local primary emissions and to limited vertical dispersion. The higher levels were most likely a result of residential wood burning, since black carbon and carbon monoxide maximised during late evening hours in wintertime. Of the bulk of elutable organics, more than a half, on average, was present as acidic fraction. Alcohols, aliphatic and polyaromatic hydrocarbons represented together, more than 30% of the elutable mass, also showing a marked seasonal pattern with a minimum in summer and a maximum in winter. The winter increase was more evident for resinic acids, phytosterols, n-alkanoic acids and polycyclic aromatic hydrocarbons.
Keywords: Ozone; Carbon monoxide; Black carbon; Organic carbon; Aerosols; Organic compounds;

The effect of different ventilation modes on in-vehicle carbon monoxide exposure by L. Abi Esber; M. El-Fadel; I. Nuwayhid; N. Saliba (3644-3657).
In-vehicle carbon monoxide (CO) concentration profiles were monitored in a passenger vehicle driven along a heavily traveled route of a commercial/residential area of Beirut, Lebanon, under several ventilation modes. Trips were conducted during morning rush hours in spring and summer time. Concomitant monitoring of car-exterior CO level, ambient CO level and wind speed was also undertaken. The highest mean CO exposure was experienced for the “windows closed, vents closed” and “windows closed, AC on recirculation” ventilation settings, with mean CO levels of 37.4 and 30.8 ppm, respectively, exceeding the 1-h air quality guidelines. The exposure was less significant for other ventilation modes with respective mean values of 10.8 – 19 ppm . Mean car-exterior CO levels were lower than the 1-h air quality guidelines, but exceeded the 8-h CO exposure guidelines. Ambient CO levels were low and non-representative of the personal exposure of individuals neither inside nor in the vicinity of road vehicles. In-vehicle CO levels revealed moderate to good correlations to out-vehicle CO levels for ventilation modes allowing for outdoor air intake, and no correlation to ambient CO levels and wind speed. Infiltration as a result of indoor–outdoor air exchange and intrusion from engine combustion/exhaust infiltration constituted the main sources of observed in-vehicle CO levels.
Keywords: In-vehicle exposure; Carbon monoxide; Ventilation mode; Beirut; Lebanon;

Evaluation of a simplified top-down model for the spatial assessment of hot traffic emissions in mid-sized cities by Devis Tuia; Margarita Ossés de Eicker; Rainer Zah; Mauricio Osses; Erika Zarate; Alain Clappier (3658-3671).
Traffic emission estimation in developing countries is a key-issue for air pollution management. In most cases, comprehensive bottom-up methodologies cannot be applied in mid-sized cities because of the resource cost related to their application. In this paper, a simplified emission estimation model (SEEM) is evaluated. The model is based on a top-down approach and gives annual global hot emission. Particular attention is paid to the quality of the input traffic data. The quality of results is assessed by application of the SEEM model in the Chilean Gran Concepción urban area and by comparison with a bottom-up approach that has been led for the year 2000. The SEEM model estimates emissions with an accuracy of about 20 % and is related to important resource savings. The results of the SEEM model are then distributed in space with a disaggregation approach and using GIS techniques. The relevancy of the disaggregation approach is evaluated among several possibilities through statistical methods. A spatial disaggregation using principal roads density gives the best results in terms of emissions repartition and gives a globally accurate image of the distribution of hot emissions in a mid-sized city.
Keywords: Traffic; Hot emissions; Top-down models; Data quality; Estimation accuracy; Spatial disaggregation; Chile; GIS;

Developing intake fraction estimates with limited data: Comparison of methods in Mexico City by Gretchen Stevens; Benjamin de Foy; J. Jason West; Jonathan I. Levy (3672-3683).
In order to estimate the health benefits of reducing mobile source emissions, analysts typically use detailed atmospheric models to estimate the change in population exposure that results from a given change in emissions. However, this may not be feasible in settings where data are limited or policy decisions are needed in the short term. Intake fraction (iF), defined as the fraction of emissions of a pollutant or its precursor that is inhaled by the population, is a metric that can be used to compare exposure assessment methods in a health benefits analysis context. To clarify the utility of rapid-assessment methods, we calculate particulate matter iFs for the Mexico City Metropolitan Area using five methods, some more resource intensive than others. First, we create two simple box models to describe dispersion of primary fine particulate matter (PM2.5) in the Mexico City basin. Second, we extrapolate iFs for primary PM2.5, ammonium sulfate, and ammonium nitrate from US values using a regression model. Third, we calculate iFs by assuming a linear relationship between emissions and population-weighted concentrations of primary PM2.5, ammonium nitrate, and ammonium sulfate (a particle composition method). Finally, we estimate PM iFs from detailed atmospheric dispersion and chemistry models run for only a short period of time. Intake fractions vary by up to a factor of five, from 23 to 120 per million for primary PM2.5. Estimates of 60, 7, and 0.7 per million for primary PM, secondary ammonium sulfate, and secondary ammonium nitrate, respectively, represent credible central estimates, with an approximate factor of two uncertainty surrounding each estimate. Our results emphasize that multiple rapid-assessment methods can provide meaningful estimates of iFs in resource-limited environments, and that formal uncertainty analysis, with special attention to model biases and uncertainty, would be important for health benefits analyses.
Keywords: Intake fraction; Mexico City; Exposure assessment; Risk assessment; Particulate matter; Mobile source;

A numerical study of an autumn high ozone episode over southwestern Taiwan by Chuan-Yao Lin; Zifa Wang; Charles C.-K. Chou; Chih-Chung Chang; Shaw C. Liu (3684-3701).
Elevated ozone concentration is one of the current major environmental concerns in Taiwan. The spatial distribution and seasonal variations of ground level ozone over Taiwan are investigated by using air quality network stations of Taiwan Environmental Protection Administration (TEPA). Data shows that high ozone episodes frequently occur over southwest Taiwan during autumn. In this season, shallow northeasterly winds prevail after frontal passage and are diverted by the Central Mountain Range (CMR) because of its mean altitude of about 2.5 km. The windward side in northern Taiwan is usually associated with cloudy days, whereas sunny days with weak wind speeds usually occur on the lee side of the CMR over southwest Taiwan due to topographical blocking. Numerical results indicate that anthropogenic emissions from the north of Kaohsiung could contribute as much as 41% of ozone for the Kaohsiung metropolitan area and 24% for the inland rural Pingtung area during the northerly flow. It is concluded that the contribution of the emissions from the north of Kaohsiung is significant and cannot be ignored. The northerly air masses, which flows over the western plain during daytime, picks up ozone and its precursors which are transported to southwestern Taiwan. After a sea breeze develops, strong onshore flow transports significant amounts of ozone and precursors to the inland rural areas resulting in the high ozone episodes that frequently occur over southwestern Taiwan during the autumn season.
Keywords: Ozone episode; Numerical study; Taiwan;

Total gaseous mercury (Hg) fluxes from large (7.3×5.5×4.5 m, L×W×D) climate-controlled gas exchange mesocosms (Ecologically Controlled Enclosed Lysimeter Laboratories or EcoCELLs) containing tallgrass prairie soil–plant monoliths were measured from 2002 to 2005. EcoCELL Hg fluxes (calculated based on the difference in air Hg concentrations inside mesocosms and in incoming air, soil area of the monoliths, and airflow through the system) indicated a net annual emission of 102 μg m−2, while soil Hg fluxes measured simultaneously using a dynamic flux chamber were an order of magnitude lower. Since Hg fluxes measured from empty EcoCELLs in winter and when housing the soil–plant monoliths at the same time of year were similar, we hypothesized that the Hg signal generated by the tallgrass prairie soil–plant monoliths was too low to be detected using the EcoCELL technology. Because mesocosm Hg exchange was correlated with solar radiation and temperature, with the largest emissions occurring at midday and in the summer, we also hypothesized that the flux from mesocosm infrastructure would change over time. Limited by the ongoing experiment, the EcoCELLs were manipulated to test the above hypotheses. When monoliths were completely covered and excluded from the exchange with the surrounding air, mesocosm Hg exchange was unaffected. Furthermore, removal of vegetation at the end of each growing season did not affect mesocosm Hg fluxes. Tests with changing mesocosm airflow also indicated that the signal from the tallgrass prairie monoliths was not being measured. These results suggest that, although EcoCELLs performed well in a study using Hg contaminated soils and have been successfully applied to understand processes controlling Hg fluxes, there are limitations of this technology for quantifying Hg exchange from background substrates. Prior to the use of similar systems the detection limit and Hg exchange from an empty system need to be carefully quantified.
Keywords: Mercury flux; Mesocosms; Chamber blank; Background system;

Characterization of aging wood chip combustion aerosol in an environmental chamber by Ari P. Leskinen; Jorma K. Jokiniemi; Kari E.J. Lehtinen (3713-3721).
Aging of aerosol from wood chip combustion in a stoker burner was monitored in an outdoor environmental chamber for 19–27 h in order to study the size, volatility and organic carbon (OC) content of the combustion aerosol particles during aging. A scanning mobility particle sizer, a volatility tandem differential mobility analyzer (VTDMA), and a thermal–optical carbon analyzer were utilized. The VTDMA and carbon analyses were performed at the beginning, after 17–24 h of aging and at one intermediate point. The size decrease of freshly emitted particles was 6–10% when heated to 360 ∘ C , and was found to depend on the experiment start time. For particles aged for 24 h, a 74–86% decrease in particle size at 360 ∘ C was observed. The more volatile OC fraction and the total OC fraction in the particles increased and the less volatile OC fraction decreased with aging. This suggests that during aging more volatile compounds condense on or heavier compounds photodegrade into lighter ones in the particles. Occasionally, new particle formation and growth were observed in the following day. The new particles were found to be composed mainly of volatile material.
Keywords: Aerosol; Wood combustion; Aging; Environmental chamber; Tandem differential mobility analyzer;

Multi-zonal air flow rates in residences in Boston, Massachusetts by Robin E. Dodson; Jonathan I. Levy; James P. Shine; John D. Spengler; Deborah H. Bennett (3722-3727).
In spite of the importance of interzonal air flow for indoor air quality assessment, few studies have characterized these flows. As part of the Boston Exposure Assessment in Microenvironments (BEAM) Study, air flow rates were estimated within 45 residences in the Boston area, most over two seasons. Thirty-five residences had basements, 11 of which also had attached garages, and 10 other residences had common apartment hallways. Air flow rates between zones were calculated using tracer gases (PFTs and SF6) and mass-balance models. Mean air flow rates from the basement to the occupied zone were significantly higher in the winter (174 m3  h−1) than in the summer (67 m3  h−1). The mean percent of the total air flow within the occupied zone of the residence from the basement was 26% (SD=34%) in the summer and 47% (SD=26%) in the winter while the mean percent from apartment hallways was 22% (SD=33%). Residences with garages attached to the basement had higher air flow rates to the adjacent zone (means from 50 to 887 m3  h−1) than those with garages attached directly to the occupied zone (means from 1 to 65 m3  h−1). These data provide a basis for modeling the contribution of indoor sources to concentrations in occupied zones.
Keywords: Attached garages; Basement; Apartment; Mass-balance model; Air flow rates;