Atmospheric Environment (v.39, #9)

This study employs a large-eddy simulation (LES) technique to predict the air exchange rate (ACH) and pollutant exchange rate (PCH) for street canyons of aspect ratio ( h / b , building-height-to-street-width) 0.5, 1.0, and 2.0 at a Reynolds number ( Re ) of 12,000 and a Schmidt number ( Sc ) of 0.72. The air ventilation and pollutant dilution of the street canyons are investigated by means of the distribution, average concentration, and retention time of pollutant in the street canyons. The flow in the street canyons at this Re and Sc lies in the wake-interference flow regime, in which the recirculations are isolated from the free-stream flow and the pollutant removal is governed by turbulent transport. Based on the LES database, the ACH and PCH of the street canyons are determined by the roof-level transient turbulence properties. The ACH increases with decreasing h / b , which is in line with the common observation of the better air ventilation in street canyons of smaller h / b . The PCH is found to be equal to the amount of pollutant emitted. Positive PCH shows that the pollutant removal rate is greater than the ground-level pollutant emission rate. As a consequence, some of the pollutant in the free surface layer, which originated from the street canyon, re-enters the canyon, lengthening the pollutant retention time in the street canyon. If pollutants from upstream sources reside in the free stream flow, the pollutant entrainment would contribute to degrading the air quality in the street canyons.
Keywords: Street canyon; Vehicular pollution; Large-eddy simulation (LES); Air exchange rate (ACH); Pollutant exchange rate (PCH);

Below-cloud ammonia scavenging in convective thunderstorms at a coastal research site in Tampa, FL, USA by Constance A. Mizak; Scott W. Campbell; Mark E. Luther; Robert P. Carnahan; Robert J. Murphy; Noreen D. Poor (1575-1584).
Sequential sampling of ammonium in wet deposition was conducted at a coastal research site in Tampa, Florida to determine and model the intra-storm variability of ammonium deposition to the Tampa Bay Estuary. Ammonium concentrations decreased with increasing precipitation depth according to an inverse power-law relationship of the form, C NH 4 = a D - b , where D is the cumulative rainfall and a and b are fitted parameters. These results show that during a precipitation event, the majority of ammonium is delivered to Tampa Bay during the initial stages of the storm. An aqueous-phase accumulation model was used to predict sequential ammonium concentrations at the Gandy Bridge site with reasonable agreement. A relationship between precipitation intensity and scavenging coefficient will facilitate the calculation of dry and wet deposition of ammonia to the Tampa Bay Estuary. The scavenging coefficients calculated with the Tampa Bay model compared well with those predicted with the Asman model.
Keywords: Ammonium; Fertilizer; Estuary; Scavenging coefficient; Sequential sampling;

A decade of weekly measurements of precipitation chemistry obtained in the US National Atmospheric Deposition Program (NADP) are combined with air concentration measurements from the Atmospheric Integrated Research Monitoring Network (AIRMoN) to estimate scavenging ratios (of concentrations in rain to like concentrations in surface air) for the ions that dominate precipitation chemistry—sulfate, nitrate, and ammonium. The intention is to provide a climatology of these ratios, for use in evaluating models that are now simulating precipitation chemistry as well as air quality. The scavenging ratios that are computed are found to be distributed approximately log-normally, with a standard deviation corresponding to about a factor of two. That is, if scavenging ratios are used to predict precipitation chemistry and wet deposition, then geometric means should be considered and the error for any given weekly prediction should be expected to be described by a standard deviation of a factor of two. (It follows that for daily predictions, the factor increases to about five.) Scavenging ratios appear to be similar for nitrate and ammonium, with those for sulfate being somewhat larger and displaying a behavior indicative of the contribution of in-cloud conversion from gaseous sulfur dioxide. Seasonal cycles are similar for most locations. The limited data available for western US stations indicate, however, that these locations yield scavenging ratios that can be significantly different from those characterizing the east. The data are in accord with the conventional understanding regarding the cloud scavenging of sulfur dioxide. For summer, each sulfur dioxide molecule appears to pass through an average of 1.3 clouds before being scavenged as sulfate. For winter, the average is 4.8 clouds.
Keywords: Deposition; Scavenging; Precipitation; Aerosols; Rainout;

Despite the recent rapid development of real-time measurements of chemical compositions in atmospheric aerosols, filter-based aerosol sampling for an extended period of time with post-sampling chemical analysis is still very common. The collected particles interact with both the in-coming gas and each other. These interactions lead to sampling artifacts. This paper investigates the role of aerosol ionic composition, inter-particle and gas-particle interactions in the sampling artifacts of PM2.5 at high relative humidity. This is achieved by comparing PM2.5 samples collected with and without a denuder/filter pack system at urban and suburban sites in Hong Kong. The samples collected with a denuder/filter pack were used to estimate the artifact-free ambient gas and particulate concentrations. Because of the prevailing high relative humidity (>70%RH) in HK, the collected particles were expected to have deliquesced and form an internal mixture, according to thermodynamic predictions. It was found that an ambient molar particulate [NH4 +]/[SO4 2−] ratio of 1.5 is a critical condition to the sampling artifact characteristics of PM2.5. The urban samples were all ammonium rich (AR; [NH4 +]/[SO4 2−] greater than 1.5) and were characterized by a high nitrate concentration and low strong-acidity. They had a significant strong acidity loss of 73%, roughly half of which was due to the neutralization of acidity by ammonia. The suburban samples were all ammonium poor (AP; [NH4 +]/[SO4 2−] less than or equal to 1.5) and were characterized by a low nitrate concentration and high strong-acidity. Sampling artifacts of the acidity, ammonium and sulfate were not important in the AP samples. In both the AR and AP samples, the individual evaporation of HNO3 and HCl were the primary reactions for the artifacts of nitrate and chloride while the concomitant evaporation of HNO3 and HCl with NH3 was insignificant. In general, the extents of the sampling artifacts of nitrate and chloride increased with increasing gas-to-particulate concentration ratio of these two species. A methodology to estimate the artifact-free ambient concentrations of nitrate and chloride from measurements of filter pack samplers (without any denuders) for deliquesced particles is proposed.
Keywords: Acidity; Ammonium; Sulfate; Nitrate; Chloride; Evaporative loss; Denuder;

Polycyclic aromatic hydrocarbons in atmospheric particles in the metropolitan area of Porto Alegre, Brazil by Juliana Braga Dallarosa; Josiane Garcia Mônego; Elba Calesso Teixeira; Josemar Luis Stefens; Flávio Wiegand (1609-1625).
The aim of the present study is to identify and quantify the main sources of polycyclic aromatic hydrocarbons (PAHs) associated with aerosols (PM10) collected at three different sampling stations: 8° Distrito, CEASA and Charqueadas. The samples were collected between November 2001 and November 2002, and the concentrations of 16 major PAHs were determined according to EPA. The filters containing particulate matter were extracted with dichloromethane in Soxhlet and the extracts were later analysed in a gaseous chromatograph coupled to a mass spectrometer (GS/MS). The average concentrations of PAHs ranged between 0.04 and 2.30 ng m−3. The analysis of principal components was applied to the chemical and meteorological variables in order to facilitate the identification of sources of PAHs emission into the atmospheric particulate. The study identified the following sources of PAHs: vehicular emissions, such as diesel oil, petrol, alcohol, and kerosene; industrial emissions, like lubricating oils; emissions from hospital waste burning, and coal burning at power plants.
Keywords: Polycyclic aromatic hydrocarbons; Principal components analysis; Vehicular emissions; Coal combustion;

PAHs and nitrated PAHs in air of five European countries determined using SPMDs as passive samplers by Hanna Söderström; Jana Hajšlová; Vladimir Kocourek; Barbara Siegmund; Anton Kocan; Mieczysław W. Obiedzinski; Mats Tysklind; Per-Anders Bergqvist (1627-1640).
The gas phase polycyclic aromatic hydrocarbons (PAHs) and nitro-PAHs in the atmosphere of five European countries (Austria, the Czech Republic, Poland, Slovakia and Sweden) were measured simultaneously during two 21-day passive sampling campaigns using semipermeable membrane devices (SPMDs). SPMD samplers, consisting of a pair of SPMDs covered by a metal umbrella, were deployed at 40 locations ranging from remote and rural to urban and industrial, at a similar time during each of the two sampling campaigns (autumn 1999, except in Poland, winter 1999, and summer 2000). The total amounts of PAHs and nitro-PAHs found in the SPMDs ranged between 5.0–1.2×103 and 1.1×10−3–4.0 ng SPMD−1  day−1, respectively. The measured environmental sampling conditions were similar between sites and, thus, the variations in the SPMD data reflected the spatial differences in gas phase concentrations of nitro-PAHs and PAHs within and between countries. The gas phase concentrations of nitro-PAHs and PAHs found in East Europe (Slovakia, the Czech Republic and Poland 1999) were 10 times higher than those measured in Sweden, Austria and Poland in 2000. In each country, the levels of PAHs and nitro-PAHs differed by one–three orders of magnitudes amongst sampling sites. The highest within-country spatial differences were found in Poland where levels of PAHs and nitro-PAHs were about one and two orders of magnitudes, respectively, higher in winter 1999 than in summer 2000, probably due to increasing emissions of coal combustion for residential heating. Differences in PAH-patterns between sites were visualized by the multivariate projection method, principal component analysis (PCA). However, no specific source patterns were found, probably since imissions rather than emissions were measured, so the PAHs detected at many sites originated from multiple sources.
Keywords: Air monitoring; Diffusive sampling; PCA; Spatial variation; Analysis;

One potential source of the potent greenhouse gas SF5CF3: the reaction of SF6 with fluorocarbon under discharge by Li Huang; Lili Zhu; Xunxi Pan; Jianliang Zhang; Bin Ouyang; Huiqi Hou (1641-1653).
In this paper, we verified Sturge's speculation (Science, 2000) that SF5CF3 might originate as a breakdown product of SF6 in high-voltage equipment via the spark discharge reactions of SF6 with a few fluorocarbons, such as CHF3 and CH2F2. However, SF5CF3 was unlikely to be formed from CF3 group attacking SF5 radical based on the results that no SF5CF3 was detected in the reactions of SF6 with CF4 or CF3Br. The reactions of SF6 with CHF3 and CH2F2 under spark discharge produced SF5CF3 and the yields of SF5CF3 were determined as a function of the initial fluorocarbon to SF6 ratio, total gas pressure and discharge time. The reaction processes of SF6 with CHF3 and CH2F2 were discussed briefly.
Keywords: Spark discharge; Source; Hydrofluorocarbons; Sulfur hexafluoride; Trifluoromethyl sulfur pentafluoride;

Modal structure and spatial–temporal variations of urban and suburban aerosols in Helsinki—Finland by Tareq Hussein; Kaarle Hämeri; Pasi P. Aalto; Pentti Paatero; Markku Kulmala (1655-1668).
Particle number size distributions were measured in the urban and suburban atmosphere of Helsinki. In the absence of a direct traffic emission influence, the temporal variation of the aerosol particle number concentrations and their particle size distribution characteristics can be generalized within the Helsinki metropolitan region. In general, the particle number size distributions are characterized by three modes: nucleation (<25 nm), Aitken (25–90 nm), and accumulation (>90 nm). Under certain conditions such as overlapping between the nucleation and Aitken modes, it is possible to characterize the particle number size distributions by two modes: ultrafine mode (<100 nm) and accumulation mode (>100 nm). Traffic combustions are considered as one of the major sources of ultrafine particles (UFP, D p < 100 nm ) in the urban atmosphere. The total particle number concentration is highest in the urban centers (as high as 140 000 cm−3, more than 90% is UFP), In general, 70–80% of the submicrometer particle number concentration is UFP in the suburban atmosphere. Close to major highways, the geometric mean diameters of the submicrometer modes are smaller and the total number concentrations can exceed 60 000 cm−3 (more than 60% is nucleation mode particles). According to the wind speed and temperature analysis, the submicrometer aerosol particles in the urban and suburban atmosphere consist of two components: UFP that is diluted with wind speeds and inversely proportional to the ambient temperature, and particles larger than 100 nm in diameter that is re-suspended with wind and proportional to the ambient temperature. The correlation analysis showed that UFP number concentrations are best correlated within the urban areas. Particles larger than 100 nm showed good correlation factors (about 0.80) within the Helsinki metropolitan area, which is an indication of similar kinds of aerosols such as regional transported particles.
Keywords: Particle number size distribution; Air quality; Ultrafine particles; Traffic; Weather;

The Ozone Weekend Effect (OWE) has become increasingly more frequent and widespread in southern California since the mid-1970s. Although a number of hypotheses have been suggested to explain the effect, there remains uncertainty associated with the root factors contributing to elevated weekend ozone concentrations. Targeting the time window of the 1997 Southern California Ozone Study (SCOS97), this paper examines traffic activity data for 14 vehicle classes at 27 weigh-in-motion (WIM) stations in southern California. Nonparametric factorial analyses of light-duty vehicle (LDV) and heavy-duty truck (HDT) traffic volumes indicate significant differences in daily volumes by day of week and between the weekly patterns of daily LDV and HDT volumes. Across WIM stations, the daily LDV volume was highest on Friday and decreased by 10% on weekends compared to that on midweek days. In contrast, daily HDT volumes showed dramatic weekend drops of 53% on Saturday and 64% on Sunday. As a result, LDV to HDT ratios increased by 145% on weekends. Nonparametric tests also suggest that weekly traffic patterns varied significantly between WIM stations located close to (central) and far from (peripheral) the Los Angeles Metro area. Weekend increases in LDV/HDT ratios were more pronounced at central WIM sites due to greater weekend declines of HDT relative to LDV traffic. The implications of these weekly traffic patterns for the OWE in southern California were investigated by estimating daily WIM traffic on-road running exhaust emissions of total organic gas (TOG) and oxides of nitrogen (NO x ) using EMFAC2002 emission factors. The results support the California Air Resource Board's (CARB's) NO x reduction hypothesis that greater weekend NO x reductions relative to volatile organic compound (VOC) emissions, in combinations with the VOC-limited ozone system, contribute to the OWE observed in the region. The results from this study can be used to develop weekend on-road mobile emission inventories for the purpose of air quality modeling.
Keywords: Weekend/weekday; Mobile source emissions; Vehicle composition; Longitudinal data;

Geostatistical investigation of ETEX-1: Structural analysis by Grégoire Dubois; Stefano Galmarini; Michaela Saisana (1683-1693).
The surface concentration data collected during the European Tracer Experiment (ETEX) are analysed for the first time by means of geostatistical techniques. The aim of the analysis is to determine the self-consistency of the data, to identify anomalous behaviours of stations with respect to the spatial and time structure of the tracer cloud measurements and to characterise the correlation structure of the measured cloud. The so-called structural analysis is usually regarded as the investigation step to be conducted prior to any interpolation procedure and mapping. Given the relevance of the ETEX dataset, one of the few existing sets of information related to controlled long-range dispersion of tracers, the analysis presented here should be regarded as a way to acquire further insight into the observed data characteristics.
Keywords: ETEX; Atmospheric dispersion; Geostatistics; Data validation; Fractals;

Seasonal characteristics and regional transport of PM2.5 in Hong Kong by Peter K.K. Louie; John G. Watson; Judith C. Chow; Antony Chen; Della W.M. Sin; Alexis K.H. Lau (1695-1710).
The Asiatic monsoon is the dominant meteorological feature of the Pearl River Delta (PRD) region in China, and influences the accumulation of locally emitted pollutants, as well as regional transport. To investigate the effect of meteorological characteristics on PM2.5 mass and chemical composition in Hong Kong, a major city of the PRD, 24-h (0000–2400 LST) PM2.5 samples were collected every sixth day from 6 November 2000 to 26 October 2001. The sampling network included a roadside site at Mong Kok (MK), an urban site at Tsuen Wan (TW), and a rural site at Hok Tsui (HT). Air parcel back trajectory and residence time analyses indicate a predominantly northeasterly transport during winter and fall in contrast to a southwesterly-to-southeasterly transport in summer. The highest seasonal PM2.5 concentrations were found in winter, followed by fall, at each of the sampling sites due to elevated organic material (OM) and ammonium nitrate (NH4NO3), consistent with the increased air parcel residence time over potential source regions. Though local mobile sources appeared to dominate the emission of carbonaceous material, a higher organic carbon/elemental carbon (OC/EC) ratio and water-soluble potassium (K+) concentration in winter imply an additional contribution from vegetative burning. Episodic pollution events were identified and investigated with back trajectories and species enrichment factors. Local and regional sources both contributed to the formation of a pollution episode on 28 February 2001. These findings reflect the degree of PM2.5 pollution in Hong Kong and provide valuable insights for planning future monitoring and modeling studies in the PRD region.
Keywords: Carbon, elemental; Carbon, organic; PM2.5; Pearl River Delta region;

Urban scale modeling of particle number concentration in Stockholm by L. Gidhagen; C. Johansson; J. Langner; V.L. Foltescu (1711-1725).
A three-dimensional dispersion model has been implemented over the urban area of Stockholm (35×35 km) to assess the spatial distribution of number concentrations of particles in the diameter range 3–400 nm. Typical number concentrations in the urban background of Stockholm is 10 000 cm−3, while they are three times higher close to a major highway outside the city and seven times higher within a densely trafficked street canyon site in the city center. The model, which includes an aerosol module for calculating the particle number losses due to coagulation and dry deposition, has been run for a 10-day period. Model results compare well with measured data, both in levels and in temporal variability. Coagulation was found to be of little importance in terms of time averaged concentrations, contributing to losses of only a few percent as compared to inert particles, while dry deposition yield particle number losses of up to 25% in certain locations. Episodic losses of up to 10% due to coagulation and 50% due to deposition, are found some kilometers downwind of major roads, rising in connection with low wind speed and suppressed turbulent mixing. Removal due to coagulation and deposition will thus be more significant for the simulation of extreme particle number concentrations during peak episodes.The study shows that dispersion models with proper aerosol dynamics included may be used to assess particle number concentrations in Stockholm, where ultrafine particles principally originate from traffic emissions. Emission factors may be determined from roadside measurements, but ambient temperature must be considered, as it has a strong influence on particle number emissions from vehicles.
Keywords: Coagulation; Dry deposition; Dispersion model; Aerosol model; Traffic; Emission factor; Particle number concentrations; Particle size distribution;

Experimental and numerical analysis of pollutant dispersion from a chimney by Nejla Mahjoub Saïd; Hatem Mhiri; Georges Le Palec; Philippe Bournot (1727-1738).
Particle image velocimetry (PIV) is used to extract and characterize the underlying organized motions, i.e. coherent structures, within the near-wake region of a turbulent round jet discharged perpendicularly from a chimney into a crossflow. This flow has been found to be quite complex owing to its three-dimensional nature and the interactions between several flow regions. Analyses of the underlying coherent structures, which play an important role in the physics of the flow, are still rare and mostly based on flow-visualization techniques. Using a PIV technique, we examined the wake regions of the chimney and plume at levels near the top of the chimney. The complex geometry of these structures in the wake of the plume as well as their interaction with the plume as it bends over after emission is discussed. In this paper we describe the Kelvin–Helmholtz vortex structures, the downwash phenomena and the effect of the height of the chimney. Extensive wind tunnel experimental results are presented and compared with numerical simulation. A good level of agreement was found between the results of flow visualization and numerical simulation.
Keywords: Wind tunnel experiments; Particle image velocimetry; Pollutant dispersion; Numerical study;

A new hypothesis on meandering atmospheric flows in low wind speed conditions by D. Oettl; A. Goulart; G. Degrazia; D. Anfossi (1739-1748).
Low wind speeds are often associated with high pollutant concentrations in the atmosphere. Dispersion modelling in such conditions is still an important challenge for scientists due to phenomena associated with low wind speeds, which are not well understood. One such phenomenon is the large horizontal oscillation of the atmosphere, which is called meandering. This study aims at providing a new hypothesis for the cause of meandering. Meandering is explained as an inherent property of atmospheric flows in low wind speed conditions, and generally no particular trigger mechanism is necessary to initiate meandering as discussed previously by several scientists (e.g. gravity waves). The hypothesis is verified by numerically and analytically solving the two-dimensional Reynolds averaged Navier–Stokes equations under the assumption of negligible Reynolds stress terms, Coriolis forces, and pressure gradients in low wind speed conditions. Meandering is shown to arise when the 2-D flow studied here is approaching or near approximate geostrophic balance, and is damped out and vanishes when the Reynolds stresses are larger. Further, the analytical solution provides an autocorrelation function for the horizontal velocity components, which was recently proposed by Anfossi et al. (Boundary Layer Meteorol. (2005), 114, 179–203) for use in low wind speed conditions. In addition, a new set of Langevin equations is proposed for simulating dispersion in low wind speed conditions.
Keywords: Low wind speeds; Meandering; Navier–Stokes equations; Dispersion; Langevin equations; Coriolis; Geostrophic balance; Reynolds stress;

Deposition of terrestrial aerosols is one of the major means of fertilizing the oligotrophic ocean waters. In order to understand the biogeochemical significance of aerosols transported from the Asian continent, and deposited over the East China Sea, we studied their chemical properties by making continuous measurements on board R/V Hakuho Maru in autumn, a season of lower dust events.Outflows of anthropogenic substances were observed over the East China Sea. The average concentration of non-sea-salt (nss)-SO4 2− in aerosols was 8.7±6.8 μg m−3, a value typical for polluted atmosphere of a highly industrialized urban area in South China. Fine mode NH4 + mainly occurred as (NH4)2SO4 and/or NH4HSO4 while in coarse mode it was formed from intermodal coagulation of ammonium particles in fine and coarse aeorosols. On the other hand, the mean concentration of NO3 (1.8±1.2 μg m−3) was lower than that in the urban atmosphere. Non-sea-salt SO4 2− were derived from combustion sources and marine biological processes, the latter accounting to 0–38%. Nitrate in coarse mode was associated with Ca2+ and Na+ perhaps due to replacement reactions.Our estimates of the annual atmospheric aerosol deposition fluxes of NH4 + and NO3 to the East China Sea were 270 Gg N yr−1 and 160 Gg N yr−1, respectively. These atmospheric inputs are comparable to the riverine inputs of the Changjiang River. The total combined nitrogen deposition could account for a biological fixation of about 2.5 Tg C (approximately 0.1–9% of the new production) in the East China Sea annually, which could actually be larger since the autumn season experience minimal dust storms.
Keywords: Marine atmosphere; Anthropogenic aerosols; Biogenic sulfur; Dry deposition fluxes; Nitrogen compounds;

Perfluorodecalin: global warming potential and first detection in the atmosphere by Keith P. Shine; Laila K. Gohar; Michael D. Hurley; George Marston; Damian Martin; Peter G. Simmonds; Timothy J. Wallington; Matt Watkins (1759-1763).
Perfluorodecalin (C10F18) has a range of medical uses that have led to small releases. Recently, it has been proposed as a carrier of vaccines, which could lead to significantly larger emissions. Since its emissions are controlled under the Kyoto Protocol, it is important that values for the global warming potential (GWP) are available. For a 50:50 mixture of the two isomers of perfluorodecalin, laboratory measurements, supplemented by theoretical calculations, give an integrated absorption cross-section of 3.91×10−16  cm2  molecule−1  cm−1 over the spectral region 0–1500 cm−1; calculations yield a radiative efficiency of 0.56 W m−2  ppbv−1 and a 100-year GWP, relative to carbon dioxide, of 7200 assuming a lifetime of 1000 years. We report the first atmospheric measurements of perfluorodecalin, at Bristol, UK and Mace Head, Ireland, where volume mixing ratios are about 1.5×10−15. At these concentrations, it makes a trivial contribution to climate change, but on a per molecule basis it is a potent greenhouse gas, indicating the need for careful assessment of its possible future usage.
Keywords: Perfluorocarbon; Radiative efficiency; Radiative forcing; GWP;