Atmospheric Environment (v.39, #11)

An ozone (O3) episode that occurred in Hong Kong during 18–22 August 1999 was investigated. This episode lasted nearly 2 days with Typhoon Sam located at 500 km away from Hong Kong in the southeast. The maximum ground-level concentration of O3 reached 334 μg m−3, which violated the Hong Kong Air Quality Objective for O3 (240 μg m−3, 1 h). The visibility was reduced to 3.2 km and UV radiation was reduced by 50%. Investigations of local meteorological factors revealed that cloudless sky with high temperature and weak wind was responsible for the formation of O3, which was trapped by sea-breeze circulations developed under stable atmospheric conditions. Backward trajectories indicated that transboundary transport induced by synoptic northwest wind due to the cyclone system played a critical role in this episode. It carried aged air mass of O3 as well as its precursors from areas with heavy emissions. The PATH (Pollutants in the Atmosphere and their Transport over Hong Kong) model system was used to simulate the flow field and O3 concentration. The computed results were basically consistent with field observations. Comparison with measurements showed that the model generally performed well in simulating flow structure over coastal complex terrain. The computed O3 peaks at most stations were underestimated by 10%, which might be due to the uncertainty in NO x and VOC emission inventory in the study region. The H2O2/HNO3 ratio of 0.8–1.2 was identified to be the transition value for separating NO x - and VOC-sensitive regimes of O3 chemistry. A process analysis further revealed the importance of horizontal transport in all physical and chemical evolution phases. Sensitivity study suggested that regional transport contributed to 60–90% of O3 level at urban and rural areas of Hong Kong during the study period.
Keywords: Ozone; Air quality modeling; Photochemical smog; Transboundary transport;

PM2.5 exposure assessment of the population in Lower Manhattan area of New York City after the World Trade Center disaster by S.P. Ng; C. Dimitroulopoulou; A. Grossinho; L.C. Chen; M. Kendall (1979-1992).
On 11 September 2001, the explosion and the collapse of the World Trade Center (WTC) Twin Towers in New York City (NYC), USA, generated a massive release of dust and inhalable toxic substances to the atmosphere as a result of the pulverization of various building materials, furniture, and computers. Many concerns were raised as Particulate Matter (PM) levels in Lower Manhattan might not meet the current National Ambient Air Quality Standards (NAAQS) (65 μg m−3). The current study aims to provide a first estimate of the scale of population exposures during this episode. Data collected from existing monitoring stations in September showed the occurrence of a series of high peaks of PM2.5 registered in the Lower Manhattan area after the 11 September. An interpolation technique was used within a Geographical Information Systems (GIS) environment to estimate outdoor PM2.5 concentrations over NYC. Monthly average of 24 h outdoor PM2.5 concentration of Lower Manhattan was 20.2 μg m−3 and did not exceed the NAAQS value. PM2.5 concentrations in indoor micro-environments were simulated by a deterministic micro-environmental model (INTAIR) and linear regression equations. Three typical population groups were identified for the NYC area—home-makers, office/shop-workers, and students/children—and their 12 h nighttime and daytime exposures were estimated from 14 September until the end of September, either as mean exposure (daytime and nighttime) or as exposure weighted by residential population distribution (nighttime only). Average nighttime and daytime exposures of the Lower Manhattan population were calculated to be 37.3 and 23.6 μg m−3, respectively (daily average: 30.45 μg m−3), in which the various group movements and activities, smoking habits of individuals, and special population movements due to access restrictions and risk avoidance behaviors were considered. Within the study period, assuming the real nighttime population distribution followed the residential population pattern, approximate one quarter of the population was exposed to 20–30 μg m−3 PM2.5; one half of the Lower Manhattan population was exposed to 10–20 and 30–60 μg m−3; around one fifth of the population was exposed to over 60 μg m−3 during nighttime. The results indicated that although the outdoor PM2.5 concentration was lower than the NAAQS value, personal exposure levels, which were generally higher than the outdoor PM2.5 concentration, might still be a reason of concern.
Keywords: Exposure assessment; World Trade Center (WTC); Geographic Information Systems (GIS); Particulate matter (PM); Micro-environment;

Typical day patterns of CO measured at four different stations in the Naples urban area were studied in order to evaluate the average residence time of passive vehicular pollutants on urban roads. The mass balance equation was written under the assumption that each road would act as a completely mixed volume. Average hourly concentration data of CO measured between 0:00 and 5:00 a.m. on working days (Monday–Friday) from 1/1/2001 to 31/12/2002 were regressed with the model equation. Average residence time is in the range 2.5–4.4 h.
Keywords: Air quality; Urban area; Residence time; Street canyon; CO;

This paper assesses the problem of the calculation of an averaging uncertainty for measurements in the framework of the European Air Quality Directives. Current environmental legislation establishes maximum uncertainties associated with a defined period of measurements. The difficulties of an ‘a priori’ determination of uncertainty contributions associated with both the averaging of measurements and an incomplete time series are discussed. The definition of an overall uncertainty, which includes budget contributions from the afore-mentioned factors, is not helpful as a regulatory parameter for quality. Alternatively, it should be used as additional information to associate with the average measurement value. This should be taken into consideration in future revisions of the European ambient air quality legislation.
Keywords: Air quality directives; Time series; Uncertainty; Limit values; Data quality objectives;

Seasonal and shorter-term variations in particulate atmospheric nitrate in Baltimore by Seung S. Park; John M. Ondov; David Harrison; Narayanan P. Nair (2011-2020).
Fine particle nitrate concentrations were measured at 10-min intervals for approximately 9.5 months beginning on 14 February 2002, at the Baltimore Supersite Ponca St. location using an R&P 8400N semi-continuous monitor. The measurement results were used to characterize seasonal and shorter-term excursions in nitrate concentrations and determine their influence on PM2.5 concentrations. Over the 9.5-month study period, nitrate levels of 1.7±1.6 μg m−3 accounted for 11.4% of the PM2.5 mass. Monthly averages ranged from 0.8 μg m−3 in August to 2.9 μg m−3 in November, and accounted for 4.7–17.3% of monthly PM2.5 mass. Hourly averages, however, were often larger, especially in the colder months, owing to numerous relatively short-term transients, where hourly nitrate concentrations exceeded 5.0 μg m−3. These often occurred along with NO x and ultrafine particle transients during the morning commute hours.A total of 275 short-term transients (31.7% of the total) exceeding 1.0 μg m−3 were identified. These were associated with one of three sets of conditions. The first, most typical (177, i.e. 64.4% of the 275 incidences), is characterized by high NO x typically between 0500 and 0800EST and is attributed to early morning commute traffic activity. The second type occurred during the afternoon due to photochemical activity. The excursions in the afternoon occurred infrequently (only 9.5% of all the observed transients) during the study period and were characterized by less elevated nitrate concentrations than morning and nighttime transients. The third (72, i.e. 26.2% of the 275 transients) occurred at night, typically between 2000EST and 0200EST.Multiple linear regression analysis between nitrate excursions and volume size distributions indicates that particulate nitrate observed is closely related to the near accumulation (0.1–0.2 μm) and droplet modes (0.5–1.0 μm) in the morning hours, and associated with the droplet (0.5–1.0 μm) and coarse modes (1.0–2.5 μm) for nighttime transients, suggesting that processes governing particulate nitrate formation depend on time-of-day.
Keywords: Baltimore PM Supersite; PM2.5 semi-continuous nitrate; Frequency distribution; Short-term transients;

The Multiple-scale Atmospheric Transport and CHemical modelling system (MATCH)—driven by meteorological data from the ECMWF—has been applied to a model domain covering Southeast Asia to complete a simulation extending over the full year of 2000. The current paper presents an evaluation of the model performance using archived chemical and meteorological data collected in the region during the year 2000. The calculated sulphate concentrations (on atmospheric aerosols and in precipitation) compare reasonably with observations while the atmospheric SO2 mixing ratios show worse correspondence. This latter mismatch is attributed to local variations in the measured SO2 concentrations that are not resolved in the regional model and possible miss-location of the emissions in our model. It can also be pointed out that different laboratories measuring SO2 at the same site occasionally report SO2 concentrations that differs by an order of magnitude or more.The seasonal variations of the modelled species are less than initially expected but generally in accordance with the measurements available. Most of the Malaysian cities have comparatively low concentrations of sulphate in precipitation. This is supported both by the model results and by independent measurements. From the model simulations and the measurements, it is concluded that the sulphur deposition is still relatively low (i.e. <0.5 g sulphur m−2  year−1) in most of rural Malaysia. This is also the case in Myanmar, Laos, central Vietnam, Kampuchea and southern Thailand. The situation in the vicinity of the large cities in the region is, however, much worse and the deposition is similar, or larger, than estimated critical loads.
Keywords: Acid deposition; Air pollution EANET; Malaysia; MATCH; Thailand;

In Santiago de Chile air quality standards are frequently exceeded during both summer and winter time. The evaluation of the temporal and spatial evolution of the chemical and physical characteristics of these air pollution problems requires the application of three-dimensional models. In this work, the Chilean Air Pollution Dispersion Model (CADM), a three-dimensional Eulerian chemical transport model is presented. The model is applied for the simulation of air pollution dispersion during summer. Carbon monoxide (CO) dispersion is studied in order to evaluate advection and diffusion as simulated by CADM and to characterise dispersion patterns in the Santiago basin. Horizontal and vertical dispersion patterns are presented, explaining how pollutants are ventilated out of the Santiago basin. Also, the role of turbulence on the pollutant distribution is discussed. Furthermore, an attempt is made to explain how pollutants are accumulated in the basin during night in the nocturnal stable planetary boundary layer and what mechanisms affect the transport of pollutants under these conditions. Comparing results from model simulations with observations, it can be shown that the model is able to represent the diurnal cycles of CO reasonably well. It can be concluded that CADM is suited for the application of air pollution dispersion in the Santiago basin in particular and complex terrain in general.
Keywords: Santiago de Chile; Complex terrain; Carbon monoxide; Nocturnal boundary layer; Divergence; Dispersion;

A primary nitrogen dioxide (NO2) emissions inventory has been developed for road transport sources in London based on recent exhaust emissions measurements of nitrogen oxides (NO X =NO+NO2) and NO2 from a range of vehicle types and technologies. The emissions inventory estimate of the mean primary volumetric NO2:NO X ratio (NO2 fraction) on road links in London was 10.2%. The emissions inventory highlights important spatial gradients in primary NO2 emissions across London. On major roads such as motorways, where the vehicle speed is high (>80 km h−1), the primary NO2 fraction is estimated to be less than 5.0% by volume. By contrast, on the congested roads of Central London, where emissions from diesel vehicles such as taxis and buses are high because of their high flows in this area of London, the primary NO2 fraction is typically 12.0% by volume. These differences have important implications for the estimation of ambient NO2 concentrations close to roads in London, and highlight why it is likely to be more difficult to meet international limits for ambient NO2 concentrations in Central London compared with other areas of the capital. The emissions inventory estimates of primary NO2 fraction have been compared with estimates derived from 41 roadside measurement sites in London, the latter being based upon an analysis of hourly mean NO X and NO2 concentrations. The results show that the mean primary NO2 fraction of 11.3% derived from ambient measurements agrees well with the mean estimate of 10.2% derived from the emissions inventory. The range of primary NO2 emission fractions derived from the ambient measurements is, however, much wider than that derived from the emissions estimates. Some of the possible reasons for the disagreement are discussed. It is concluded that a larger database of emissions measurements from a range of vehicle types and pollution control technologies is required to develop a more robust emissions inventory for primary NO2 in urban areas.
Keywords: Nitrogen dioxide; Emission factor; Air quality assessment; London;

The flux of isoprene from a willow coppice plantation and the effect on local air quality by M. Olofsson; B. Ek-Olausson; N.O. Jensen; S. Langer; E. Ljungström (2061-2070).
Isoprene fluxes from a Salix viminalis (willow) plantation in western Sweden were measured using the relaxed eddy accumulation (REA) technique. Fluxes of up to 0.23 μg m−2  s−1 could be observed. A standard emission factor at 303 K and a PAR flux of 1000 μ mol m−2 s−1 was estimated to 0.98 μg m−2  s−1 by using the G93 algorithm. The chemistry of an air parcel passing over a willow coppice plantation was investigated utilising a Lagrangian box model in which the measured isoprene fluxes were used as input data. Dispersion after the field was accounted for by a procedure based on the Gaussian plume model. The calculations indicate that, in most cases, the isoprene emissions have a small effect on the local air quality.
Keywords: Salix viminalis; REA; Micrometeorology; Modelling; Isoprene emission;

Under the French national research program PRIMEQUAL, measurements of outdoor and indoor pollution have been performed in eight school buildings in La Rochelle (France) and its suburbs. The school buildings were either naturally ventilated by opening the windows or mechanically ventilated with minimum fresh air, and demonstrated various permeabilities. Ozone, nitrogen oxides (NO and NO2), and particulate matter (PM) (15 size intervals ranging from 0.3 to 20 μm) concentrations were monitored continuously indoors and outdoors for two 2-week periods. The indoor relative humidity, temperature, CO2 concentration (room occupancy), window openings and permeability of the building were also measured. Principal component analysis (PCA), a multivariate observation-based statistical method, was used to determine the parameters influencing the relationship between the outdoor and indoor concentration levels. After a brief description of the experimental data and methodology, the paper presents a detailed analysis of the PCA diagrams. This analysis leads to distinguish between positively correlated, negatively correlated and non-correlated variables. The main conclusions arising from the study are: (1) the influence of the room occupancy on the particle concentrations indoors changes with different particle sizes, (2) the building air-tightness and the outdoor concentration level greatly influence the indoor/outdoor (I/O) concentration ratios of ozone, and (3) indoor ozone and particles concentrations are negatively correlated, which may be the result of complex homogeneous and/or heterogeneous processes.
Keywords: Schools; Principal component analysis; I/O ratios; Particulate matter; Nitrogen dioxide; Nitric oxide; Ozone; Building permeability;

Measurements of dry deposition flux to surrogate surfaces were made in Qingdao in July 2001 and March 2002, and airborne concentration measurements of Fe, Al, Cu, Pb, Zn and Cd were made from April 2001 to May 2002 to determine atmospheric inputs of pollutants to the coastal waters. Size-dependent particle dry deposition velocities were obtained using Williams’ model with meteorological inputs from past observations, taking into account the particle growth in the humid region near the air/sea interface. Sensitivity tests show that the model provides deposition velocities comparable with recent reference values. A comparison of the modeled dry deposition fluxes with measurements and GESAMP (Group of Experts on Scientific Aspects of Marine Pollution) suggestions demonstrated that the current Williams’ model produced reasonable results. Using the averages of measured concentrations of six metal elements, the dry deposition fluxes for four seasons in the Qingdao area were calculated. The results showed that the deposition fluxes of crustal elements Al, Fe and Mn in spring and winter account for more than 70% of the whole year's deposition, while fluxes of Cu, Pb and Zn in autumn and winter contribute more than 70% of the whole year's deposition.
Keywords: Dry deposition fluxes; Water surface; Surrogate surface; Williams’ model; Metal elements;

Radiative effects of natural aerosols: A review by S.K. Satheesh; K. Krishna Moorthy (2089-2110).
In recent years, there has been a substantial increase in interest in the influence of anthropogenic aerosols on climate through both direct and indirect effects. Several extensive investigations and coordinated field campaigns have been carried out to assess the impact of anthropogenic aerosols on climate. However, there are far fewer studies on natural aerosols than on anthropogenic aerosols, despite their importance. Natural aerosols are particularly important because they provide a kind of base level to aerosol impact, and there is no effective control on them, unlike their anthropogenic counterparts. Besides, on a global scale the abundance of natural aerosols is several times greater than that of the major anthropogenic aerosols (sulphate, soot and organics). The major natural aerosol components are sea salt, soil dust, natural sulphates, volcanic aerosols, and those generated by natural forest fires. As with anthropogenic aerosols, the abundance of natural aerosols such as soil dust is also increasing, due to processes such as deforestation, which exposes more land areas which may then interact directly with the atmosphere, and due to other human activities. Since a major fraction of the natural aerosol (sea salt and natural sulphate) is of the non-absorbing type (and hygroscopic), it partly offsets the warming due to greenhouse gases as well as that due to absorbing aerosols (e.g., soot). The mineral dust transported over land and ocean causes surface cooling (due to scattering and absorption) simultaneously with lower atmospheric heating (due to absorption); this could in turn intensify a low-level inversion and increase atmospheric stability and reduce convection. To accurately predict the impact of dust aerosols on climate, the spatial and temporal distribution of dust is essential. The regional characteristics of dust source function are poorly understood due to the lack of an adequate database. The reduction of solar radiation at the surface would lead to a reduction in the sensible heat flux and all these will lead to perturbations in the regional and global climate. Enhanced concentration of sea salt aerosols at high wind speed would lead to more condensation nuclei, increase in the cloud droplet concentration and hence cloud albedo. Even though direct radiative impacts due to sea salt and natural sulphate are small compared to those due to anthropogenic counterparts, their indirect effects (and the uncertainties) are much larger. There is a considerable uncertainty in sea salt aerosol radiative forcing due to an inadequate database over oceans. The presence of natural aerosols may influence the radiative impact of anthropogenic aerosols, and it is difficult to separate the natural and anthropogenic aerosol contributions to radiative forcing when they are in a mixed state. Hence it is necessary to document the radiative effects of natural aerosols, especially in the tropics where the natural sources are strong. This is the subject matter of this review.
Keywords: Aerosols; Climate change; Radiative forcing; Radiation budget;

Air quality measurements were made on interstate highways in the Minneapolis metropolitan area. Gas and aerosol concentrations were measured on weekdays and weekends. By exploiting the difference in the relative volumes of heavy duty (HD) diesel and light duty (LD) spark ignition (SI) vehicles on weekdays and weekends, we were able to estimate apportioned fuel specific emissions.The on-road, apportioned, fuel specific particle number emissions factors, estimated from condensation particle counter (CPC) measurements were 1.34±0.2×1016 particles kg−1 for diesels and 7.1±1.6×1015 particles kg−1 for spark ignition vehicles. Estimates from the scanning mobility particle sizer (SMPS) measurements were 2.1±0.3×1015 particles kg−1 for diesels and 3.9±0.6×1014 particles kg−1 for SI vehicles. The difference between CPC and SMPS measurements is mainly due to different lower size detection limits of the instruments, ∼3 and ∼10 nm, respectively.On a weekly weighted basis and on weekdays, the majority of particle number was attributed to HD diesel traffic. Weekend production of particles can be primarily attributed to light duty SI automobiles. On a per vehicle basis, HD vehicles produced substantially greater numbers of particles. On a fuel specific basis, HD vehicles produce slightly higher concentrations of particles than light duty vehicles. The relative contribution of LD vehicles to particle number emissions increased as particle size decreased.The HD apportioned size distributions were similar to size distributions measured during other on-road and laboratory studies. The LD apportioned size distribution was bounded by laboratory and on-road size distributions. Our work is representative of summer, highway cruise conditions. It is likely that under cold start and high load operating conditions LD emissions will increase relative to HD emissions.
Keywords: Nanoparticles; On-road measurements; Vehicle emission factors; Particle size distribution; Particulate matter;

Sea salt generation, dispersion and removal on the regional scale by V.L. Foltescu; S.C. Pryor; C. Bennet (2123-2133).
The MATCH (Multi-scale Atmospheric Transport and Chemistry) modelling system has been used as the modelling platform for regional-scale sea salt simulations. A new model version for sea salt aerosols has been constructed and assessed. This study presents the model concept and shows that the model is capable of simulating sea salt on a regional scale. The model agrees well with observations, particularly in terms of the sea salt mass concentrations at certain locations in Northern Europe. The observed wet fluxes are generally under-predicted due in part to systematic under-prediction of precipitation by the meteorological model used to drive MATCH.
Keywords: Sea salt; Model; Emission; Dry deposition; Wet deposition;

Concentration fluctuations and relative dispersion PDF by E. Ferrero; L. Mortarini (2135-2143).
In this paper, we consider the problem of the relative dispersion of particle pairs released in a homogeneous isotropic stationary turbulent field. A one-dimensional two-particle Lagrangian stochastic model is considered. Two Langevin equations for the particles separation ( Δ ) and barycentre ( Z ) are presented and the results of the model simulations are discussed. The small-scale turbulence structure is analysed by reproducing the Δ and Z mean square trends. These are compared with the theoretical predictions and with a new formula to verify the Richardson's t 3 -law and the existence of an intermediate subrange, respectively, whose extension depends on the initial separation. Concerning the separation probability density function (PDF), two different forms are found for small and long times, respectively, according to the classical turbulence theory and the results of previous Lagrangian stochastic models. The mean concentrations and concentration fluctuations predicted by the model are compared with a new formula based on the Richardson separation PDF and with the formula based on the Gaussian PDF.
Keywords: Concentration fluctuations; Lagrangian stochastic model; Relative dispersion;

We present a general model where the logarithm of hourly concentration of an air pollutant is modelled as a sum of non-linear functions of traffic volume and several meteorological variables. The model can be estimated within the framework of generalised additive models.Although the model is non-linear, it is simple and easy to interpret. It quantifies how meteorological conditions and traffic volume influence the level of air pollution. A measure of relative importance of each predictor variable is presented.Separate models are estimated for the concentration of PM 10 , PM 2.5 , the difference PM 10 – PM 2.5 , NO 2 and NO x at four different locations in Oslo, based on hourly data in the period 2001–2003. We obtain a reasonably good fit, in particular for the largest particles, PM 10 and PM 10 – PM 2.5 , and for NO x . The most important predictor variables are related to traffic volume and wind. Further, relative humidity has a clear effect on the PM variables, but not on the NO variables. Other predictor variables, such as temperature, precipitation and snow cover on the ground are of some importance for one or more of the pollutants, but their effects are less pronounced.
Keywords: Air quality modelling; Urban air quality; Particulate matter; Nitrogen oxides; Forward validation;

Retention efficiencies of nitric acid, hydrochloric acid and ammonia were measured for different filters, with particular emphasis on cellulose (CE) and cellulose acetate-nitrate (CA) materials. Gases were produced either by nebulising aqueous solutions or by a novel technique based on the desorption from ammonium salts deposited on quartz fibre (QF) filters. Efficiencies for pure acidic gases and ammonia on CE and CA ranged from very low (⩽3.6%) to low (∼10% for HNO3 on CE). In contrast, if acidic gases and ammonia were supplied in equimolar concentrations, they were retained (almost) completely on CE, with high efficiency on CA (60–80% for NH3+HNO3; 20–45% for NH3+HCl), also with high efficiency on glass fibre filter, but with very low efficiency on QF and Teflon (Tf) filters (<1%). For CA, retention efficiencies were found to increase with increasing relative humidity and to decrease with decreasing mean pressure at which the filters were exposed to the gases. Once retained on CA filters, the retained gases may be lost again during subsequent exposure to clean air.
Keywords: Aerosol filters; Sampling artefacts; Gas retention; Nitric acid; Hydrochloric acid; Ammonia;

Free amino acids in marine aerosols over the western North Pacific Ocean by Kiyoshi Matsumoto; Mitsuo Uematsu (2163-2170).
Atmospheric input of fixed nitrogen species to the ocean has attracted considerable attention from the viewpoint of the oceanic biogeochemical cycle of nitrogen, although few measurements of organic nitrogen compounds in atmospheric aerosols have been extensively conducted over remote ocean areas. In this study, we report the geographical distribution of dissolved free amino acids (DFAA) in the water-soluble fraction of two size-segregated marine aerosols over the western North Pacific. The concentrations of DFAA showed higher values over the region north of 30°N, whereas they clearly decreased south of 30°N. Approximately 59–96% of DFAA was found in fine-mode particles. Long-range transport from continental sources could largely contribute to DFAA in marine aerosols over the remote North Pacific.
Keywords: Free amino acids; Organic nitrogen; Marine aerosol; Remote ocean; Western Pacific;