Forgot your password?
New user?
New Window Opens on the Secret Life of Microbes: Scientists Develop First Microbial Profiles of Ecosystems
CAS announces the release of SciFinder Scholar for Mac OS X
Press Releases
Press Releases
| Check out our New Publishers' Select for Free Articles |
Atmospheric Environment (v.39, #8)
A hierarchical Bayesian model to estimate and forecast ozone through space and time by Nancy McMillan; Steven M. Bortnick; Mark E. Irwin; L. Mark Berliner (pp. 1373-1382).
A Bayesian hierarchical regime switching model describing the spatial–temporal behavior of ozone (O3) within a domain covering Lake Michigan during spring–summer 1999 is developed. The model incorporates linkages between ozone and meteorology. It is specifically formulated to identify meteorological regimes conducive of high ozone levels and allow ozone behavior during these periods to be different from typical ozone behavior. The model is used to estimate or forecast spatial fields of O3 conditional on observed (or forecasted) meteorology including temperature, humidity, pressure, and wind speed and direction. The model is successful at forecasting the onset of periods of high ozone levels, but more work is needed to also accurately identify departures from these periods.
Keywords: Statistical model; Space–time models; Air pollution; Ozone; Meteorology
Stable estimate of primary OC/EC ratios in the EC tracer method by S.-H. Shao-Hang Chu (pp. 1383-1392).
In fine particulate matter studies, the primary OC/EC ratio plays an important role in estimating the secondary organic aerosol contribution to PM2.5 concentrations using the EC tracer method. In this study, numerical experiments are carried out to test and compare various statistical techniques in the estimation of primary OC/EC ratios. The influence of random measurement errors in both primary OC and EC measurements on the estimation of the expected primary OC/EC ratios is examined. It is found that random measurement errors in EC generally create an underestimation of the slope and an overestimation of the intercept of the ordinary least-squares regression line. The Deming regression analysis performs much better than the ordinary regression, but it tends to overcorrect the problem by slightly overestimating the slope and underestimating the intercept. Averaging the ratios directly is usually undesirable because the average is strongly influenced by unrealistically high values of OC/EC ratios resulting from random measurement errors at low EC concentrations. The errors generally result in a skewed distribution of the OC/EC ratios even if the parent distributions of OC and EC are close to normal.When measured OC contains a significant amount of non-combustion OC Deming regression is a much better tool and should be used to estimate both the primary OC/EC ratio and the non-combustion OC. However, if the non-combustion OC is negligibly small the best and most robust estimator of the OC/EC ratio turns out to be the simple ratio of the OC and EC averages. It not only reduces random errors by averaging individual variables separately but also acts as a weighted average of ratios to minimize the influence of unrealistically high OC/EC ratios created by measurement errors at low EC concentrations. The median of OC/EC ratios ranks a close second, and the geometric mean of ratios ranks third. This is because their estimations are insensitive to questionable extreme values. A real world example is given using the ambient data collected from an Atlanta STN site during the winter of 2001–2002.
Keywords: EC tracer method; Primary OC/EC ratio estimation; Ordinary regression; Deming regression; Ratio of averages; average of ratios
Hygroscopic growth behavior of a carbon-dominated aerosol in Yosemite National Park by Christian M. Carrico; Sonia M. Kreidenweis; William C. Malm; Derek E. Day; Taehyoung Lee; Jacqueline Carrillo; Gavin R. McMeeking; Jeffrey L. Collett Jr. (pp. 1393-1404).
The influence of particulate organic material (POM) and the contribution of biomass smoke on air quality and visibility remain a paramount issue in addressing regional haze concerns in US national parks. Measurements during the Yosemite Aerosol Characterization Study (July–September 2002) indicated an aerosol dominated by POM (∼70% of identified species) and strongly influenced by biomass smoke. Here we report aerosol size hygroscopic growth measurements for dry (RH<5%) aerosol diameters of 100 and 200nm as measured with a controlled relative humidity tandem differential mobility analyzer. Hygroscopic growth was found to be negligible for relative humidity (RH) <∼40% within the sensitivity of the method. For RH>40%, particle size typically increased smoothly with RH, and overall hygroscopic growth at high RH was low to moderate in comparison to the range of values reported in the literature. For RH>80%, both monomodal and bimodal growth profiles were observed during the study, with 200nm particles more often splitting into bimodal profiles (68% of cases), indicating some degree of external mixing. Trimodal growth profiles were observed on two occasions during periods of changing meteorology and aerosol composition. For bimodal profiles for 200nm dry particles, particle diameter growth factors at RH=80% (D(RH=80%)/Do where Do is measured at RH<5%) were 1.11±0.04 and 1.29±0.08 for the more and less hygroscopic modes, respectively. Ensemble D/Do was calculated using a cubic-weighted sum of D/Do of individual modes. For 200nm particles, average ensembleD(RH=80%)/Do was 1.15±0.05 andD(RH=90%)/Do was 1.31±0.06, and were slightly large for 100nm particles. These growth factors are dramatically lower than those for typical ambient aerosol ionic components such as sulfate, nitrate and sodium salts and sulfuric acid. An inverse relationship between the POM fraction of PM2.5 and hygroscopicity was particularly evident for 200nm particles withD(RH=80%)/Do approaching ∼1.1 as the POM/ionic mass ratio exceeded 10. Linear correlations with several measurements of POM and select biomass smoke markers were −0.74< R<−0.59 for 200nm and −0.43< R<−0.25 for 100nm particles. The limited hygroscopic growth of the carbon dominated, biomass smoke influenced aerosol at Yosemite has important implications to assessing the role of POM in visibility degradation in national parks.
Keywords: PM; 2.5; Regional haze; Organic carbon; Biomass smoke; Tandem differential mobility analyzer; Secondary organic aerosol
Spatial variations of surface snow chemistry during two austral summers in western Dronning Maud Land, Antarctica by Karkas Eija Kärkäs; Teinilä Kimmo Teinilä; Aki Virkkula; Minna Aurela (pp. 1405-1416).
The glaciochemical properties of snow were measured in western Dronning Maud Land during the austral summers of 1999/2000 and 2000/2001. The surface snow samples were taken along a 350-km transect from the seaward edge of the ice shelf to the Antarctic plateau and analysed for the presence of the major inorganic components SO42−, NO3−, Cl−, NH4+, Na+, K+, Ca2+, Mg2+ and MSA. The sea-salt components and MSA decreased exponentially with increasing distance from the coast by about 64%/100km. Non-sea-salt SO42− decreased also exponentially with the distance from the ice edge by 48%/100km. No substantial trend was observed for NO3−. The pH values calculated from the ionic balance were similar to those measured in the field, although the calculated pH values were an average of 0.6 lower. The fractions of the dry deposition of the ionic concentrations were estimated and the results reveal that they are low near the Finnish research station Aboa, usually less than 10% of the total deposition.
Keywords: Antarctica; Snow; Ionic concentrations; pH; Contribution of dry deposition
Total and water-soluble trace metal content of urban background PM10, PM2.5 and black smoke in Edinburgh, UK by Mathew R. Heal; Leon R. Hibbs; Raymond M. Agius; Iain J. Beverland (pp. 1417-1430).
Toxicological studies have implicated trace metals in airborne particles as possible contributors to respiratory and/or cardiovascular inflammation. As part of an epidemiological study, co-located 24h samples of PM10, PM2.5 and black smoke (BS) were collected for 1 year at an urban background site in Edinburgh, and each sample sequentially extracted with ultra-pure water, then concentrated HNO3/HCl, and analysed for Ti, V, Cr, Mn, Fe, Ni, Cu, Zn, As, Cd and Pb. This yields a comprehensive data set for UK urban airborne trace metal. The median ( n>349) daily water-soluble metal concentration in PM2.5 ranged from 0.05ngm−3 for Ti to 5.1ngm−3 for Pb; and in PM10 from 0.18ngm−3 for Ti to 11.7ngm−3 for Fe. Median daily total (i.e. water+acid-extractable) metal concentration in PM2.5 ranged from 0.3ngm−3 for As to 27.6ngm−3 for Fe; and in PM10 from 0.37ngm−3 for As to 183ngm−3 for Fe. The PM2.5:PM10 ratio varied considerably with metal, from <17%, on average, for Ti and Fe, to >70% for V, As, Cd and Pb. The 11 trace metals constituted proportionally more of the PM10–2.5 fraction than of the PM2.5 fraction (0.9%). The proportion of water-soluble metal in each size-fraction varied considerably, from <10% water-soluble Fe and Ti in PM10−2.5, to >50% water-soluble V, Zn, As and Cd in PM2.5. Although Fe generally dominated the trace metal, water-soluble metal also contained significant Zn, Pb and Cu, and for all size and solubility fractions >90% of trace metal was comprised of Fe, Zn, Pb and Cu. Statistical analyses suggested three main sources: traffic; static combustion; and crustal. The association of metals with traffic (Cu, Fe, Mn, Pb, Zn) was consistent with traffic-induced non-exhaust “resuspension� rather than direct exhaust emission. Meteorology contributed to the wide variation in daily trace metal concentration. The proportion of trace metal in particles varied significantly with the air mass source and was highest on days for trajectories traversing over land. For Mn, Fe, Cu, Zn, As and Pb there was greater correlation of metal concentration with BS mass than with either PM10 or PM2.5 mass, suggesting that BS reflectance monitoring could be a cost-effective surrogate measure of particle metal concentration in urban background air.
Keywords: Airborne particles; Urban aerosol; ICP-MS; Air mass; Back-trajectory; Toxicology
Rate constants for atmospheric trace organics scavenging SO4− in the Fe-catalysed autoxidation of S(IV) by J. Ziajka; W. Pasiuk-Bronikowska (pp. 1431-1438).
Numerous organic compounds are released into the atmosphere in globally large quantities. Chemical transformation of these compounds under the influence of other atmospheric components has not yet been fully recognised. One of the uncertainties is the importance of the mechanism and kinetics of atmospheric reactions involving organics for the quantitative description of such phenomena as the formation of aerosols and the acidification. Rate constants for scavenging SO4− radical anions by benzene, phenol,α-pinene, myrtenal and cis-pinonic acid in aqueous phase were determined by the method described previously (Atmos. Environ. 37 (2003) 3913) yielding 2.4×109, 8.8×109, 1.2×109, 9.6×108 and 1.6×107M−1s−1, respectively. These anions were produced by autoxidation of S(IV) catalysed by Fe ions at pH 3.0 and ionic strength 0.005M. In the case of myrtenal, the formation of an aldehyde–hydrogen sulphite adduct, non-reactive with respect to oxidizing radicals, was taken into account.
Keywords: Inhibition; Phenol; Sulphoxy radicals; Sulphur dioxide; Terpenoids
Computational estimation of logarithm of n-octanol/air partition coefficient and subcooled vapor pressures of 75 chloronaphthalene congeners by Tomasz Puzyn; Jerzy Falandysz (pp. 1439-1446).
Computationally predicted data on logarithm of n-octanol/air partition coefficient (logKOA) and subcooled liquid vapor pressure (logPL) for all 75 congeners of chloronaphthalene are presented. The values oflogKOA for CNs ranged between 5.76 and 11.52 for 1-chloronaphthalene and octachloronaphthalene, respectively, while their estimated values oflogPL varied between 0.60 and −4.32. For the particular chloronaphthalene congeners within each of chloronaphthalene homologue group thelogKOA andlogPL values ranged, respectively, from 5.76 to 5.84 and from 0.53 to 0.60 for monochloro-; from 6.73 to 7.02 and from −0.71 to −0.45 for dichloro-; from 7.35 to 7.47 and from −1.06 to −0.96 for trichloro-; from 7.95 to 8.27 and from −1.75 to −1.47 for tetrachloro-; from 8.88 to 9.04 and from −2.39 to −2.26 for pentachloro-; from 9.73 to 9.89 and from −3.10 to −2.96 for hexachloro-; from 10.61 to 10.62 and from −3.54 to −3.53 for heptachloro-, and 11.52 and −4.32 for octachloronaphthalene. The data obtained were further compared with the results derived from the similar studies by means of other commonly used regression methods such as multivariate regression method and partial least squares.
Keywords: Polychlorinated naphthalenes; log; K; OA; log; P; L; QSAR/QSPR; PCR
An inverse approach for estimating the initial distribution of volatile organic compounds in dry building material by Feng Li; Jianlei Niu (pp. 1447-1455).
A model for the prediction of emission of volatile organic compounds (VOCs) from dry building material was developed based on mass transfer theory. The model considers both diffusion and convective mass transfer. In addition, it does not neglect the fact that, in most cases, the initial distribution of VOCs within the material is non-uniform. Under the condition that the initial amount of VOCs contained in the building material is the same, six different types of initial VOC distributions were studied in order to show their effects on the characteristics of emission. The results show that, for short-term predictions, the effects are significant and thus cannot be neglected. Based on the fact that the initial distribution of VOCs is very difficult to directly determine, a conjugate gradient method with an adjoint problem for estimating functions was developed, which can be used to inversely estimate the initial distribution of VOCs within the material without a priori information on the functional form of the unknown function. Simulated measurements with and without measurement errors were used to validate the algorithm. This powerful method successfully recovered all of the aforementioned six different types of initial VOC distributions. A deviation between the exact and predicted initial condition near the bottom of the material was noticed, and a twin chamber method is proposed to obtain more accurate results. With accurate knowledge of the initial distribution of VOCs, source models will be able to yield more accurate predictions.
Keywords: Mass transfer; Conjugate gradient method; Adjoint problem; Indoor air quality; Initial condition; Diffusion
A PDF micromixing model of dispersion for atmospheric flow. Part I: development of the model, application to homogeneous turbulence and to neutral boundary layer by M. Cassiani; P. Franzese; U. Giostra (pp. 1457-1469).
A Lagrangian stochastic (LS) probability density function (PDF) model has been developed to study statistics and PDF of concentration generated by continuous releases of passive substances from point and line sources in atmospheric flow. The model simulates the combined effect of turbulent mixing (macromixing) and molecular diffusivity (micromixing) on dispersion of tracers. Turbulent dispersion is modelled using an LS model; molecular diffusivity is simulated by an interaction by exchange with the conditional mean (IECM) model. A dynamical computational grid, which expands with time around the plume, has been developed to optimise computational time and memory requirements. The model has been tested with the results of a two-particle LS model in homogeneous turbulence and with wind tunnel observations in a neutral boundary layer. The proposed model can account for chemical reactions in a direct way with no closure assumptions.
Keywords: Concentration fluctuations; Micromixing modelling; Monte Carlo simulation; Turbulent dispersion; Chemical reactions
A PDF micromixing model of dispersion for atmospheric flow. Part II: application to convective boundary layer by M. Cassiani; P. Franzese; U. Giostra (pp. 1471-1479).
The Lagrangian stochastic probability density function (PDF) model developed by Cassiani et al. [Atmos. Environ. (2005) Part 1] is extended to the atmospheric convective boundary layer. The model is applied to simulate concentration statistics and PDF generated by passive releases from point and line sources in the convective boundary layer. A dynamical time-expandable grid is implemented, which optimises the computational resources required for dispersion simulations in atmospheric flow. A parameterised formulation for the micromixing time scale in convective conditions is derived. Model concentration statistics including mean field, fluctuations and concentration PDF are tested with four water tank experiments.
Keywords: Concentration fluctuations; Micromixing modelling; Monte Carlo simulation; Turbulent dispersion; Chemical reactions
Inversion of extinction data for irregularly shaped particles by M. Kocifaj; H. Horvath (pp. 1481-1495).
The retrieval of extinction data in the highly urbanized region of Bratislava city is analyzed for both, strictly non-spherical particles and volume equivalent particles. Two most typical situations are discussed in more details: the summer time dusty atmosphere consisting of strongly scattering particles with refractive index about 1.59, and winter time atmosphere with usual occurrence of ice-phase aerosols. The attention is paid to more frequent continental polar air mass, with 54% occurrence probability in the Bratislava region. The root-mean-square error of measured optical data varies from 4% to 8%. When processing extinction data it is shown that there are no significant differences between modal radii computed under assumption of randomly oriented irregularly shaped particles and for Mie equivalents. However, the differences can occur in case on non-randomly oriented particles. The modal radius of equally oriented particles in the ice cloud can be about two times larger than a modal radius of the system volume of volume identical spheres. Particle irregularity and the roughness of its surface eliminate the interference structure and ripple typical for monodisperse scattering patterns. As a consequence the subsidiary mode within size distribution function disappears. Such a mode usually occurs with Mie particles. Assuming spherical particles for the computation yields a narrower size distribution compared to assuming irregularly shaped particles.
Keywords: Aerosol size distribution; Non-spherical particles; Solar radiation; Inverse problems
A simulation study of dispersion of air borne radionuclides from a nuclear power plant under a hypothetical accidental scenario at a tropical coastal site by C.V. Srinivas; R. Venkatesan (pp. 1497-1511).
Meteorological condition in coastal regions is diurnally variable and spatially heterogeneous due to complex topography, land–sea interface, etc. A wide range of dispersion conditions is possible on a given day in the coastal regions. In case of inadvertent accidental situations, though unlikely, it would be necessary to examine the potentially severe case among different dynamically occurring local atmospheric conditions for dispersion and its range of impact around a nuclear power plant for safety analysis. In this context, dispersion of air borne radioactive effluents during a hypothetical accidental scenario from a proposed prototype fast breeder reactor (PFBR) at an Indian coastal site, Kalpakkam, is simulated using a 3-D meso-scale atmospheric model MM5 and a random walk particle dispersion model FLEXPART. A simulation carried out for a typical summer day predicted the development of land–sea breeze circulation and thermal internal boundary layer (TIBL) formation, which have been confirmed by meteorological observations. Analysis of dose distribution shows that the maximum dose for releases from a 100m stack occurs at two places within 4km distance during sea breeze/TIBL fumigation hours. Maximum dose also occurred during nighttime stable conditions. Results indicate that, on the day of present study, the highest concentrations occurred during periods of TIBL fumigation rather than during stable atmospheric conditions. Further, the area of impact (plume width at the surface) spreads up to a down wind distance of 4km during fumigation condition. Simulation over a range of 25km has shown turning of plume at the incidence of sea breeze circulation and two different dispersion patterns across the sea breeze front. These results are significant in comparison to the expected pattern shown by Gaussian plume model used for routine analysis.
Keywords: Radionuclide dispersion; Meso-scale model; FLEXPART; Sea breeze; TIBL fumigation
Relationships between aerosol optical depth and surface-layer extinction in the central part of the Upper Silesia industrial region over the period of 1983–1994 by Apoloniusz Sztyler (pp. 1513-1523).
The subject of analysis is the aerosol optical depth (AOD) in the visible part of the solar spectrum(τvis) (wavelengthλ=0.295–0.695μm). Calculation ofτvis-values was based on pyrheliometric observation data performed from 1983 to 1994 at IEIA's station (Katowice-Załęże).This paper examines the dependence ofτvis and its relation to surface-layer extinction coefficientσ (Koschmider quotient of 3.912 and visibilityV) on meteorological and anthropological factors during the observation period. Knowledge of the relationship between columnar aerosol turbidity (which can be expressed byτvis) andσ allows a more precise and accurate estimation of the aerosol effect on the radiative composition of climate. This work offers some contributions to solving this problem. Therefore, the relations have been described in the form of mathematical/statistical models of AOD (based on “momentary� and mean seasonal values), in which meteorological (and astronomical) parameters as well as the amount of industry dust emission were used as independent variables to improve exactness and credibility in the models.
Keywords: Visibility; Columnar aerosol turbidity; Aerosol optical depth models
A quantitative method for clustering size distributions of elements by Ann M. Dillner; James J. Schauer; William F. Christensen; Glen R. Cass (pp. 1525-1537).
A quantitative method was developed to group similarly shaped size distributions of particle-phase elements in order to ascertain sources of the elements. This method was developed and applied using data from two sites in Houston, TX; one site surrounded by refineries, chemical plants and vehicular and commercial shipping traffic, and the other site, 25 miles inland surrounded by residences, light industrial facilities and vehicular traffic. Twenty-four hour size-segregated (0.056< Dp (particle diameter)<1.8μm) particulate matter samples were collected during five days in August 2000. ICP-MS was used to quantify 32 elements with concentrations as low as a few picograms per cubic meter. Concentrations of particulate matter mass, sulfate and organic carbon at the two sites were often not significantly different from each other and had smooth unimodal size distributions indicating the regional nature of these species. Element concentrations varied widely across events and sites and often showed sharp peaks at particle diameters between 0.1 and 0.3μm and in the ultrafine mode ( Dp<0.1μm), which suggested that the sources of these elements were local, high-temperature processes. Elements were quantitatively grouped together in each event using Ward's Method to cluster normalized size distributions of all elements. Cluster analysis provided groups of elements with similar size distributions that were attributed to sources such as automobile catalysts, fluid catalytic cracking unit catalysts, fuel oil burning, a coal-fired power plant, and high-temperature metal working. The clustered elements were generally attributed to different sources at the two sites during each sampling day indicating the diversity of local sources that impact heavy metals concentrations in the region.
Keywords: ICP-MS; Cluster analysis; Metals; Particulate matter; Source attribution
Aerosol modelling and validation during ESCOMPTE 2001 by F. Cousin; C. Liousse; H. Cachier; B. Bessagnet; B. Guillaume; R. Rosset (pp. 1539-1550).
The ESCOMPTE 2001 programme (Atmospheric Research. 69(3-4) (2004) 241) has resulted in an exhaustive set of dynamical, radiative, gas and aerosol observations (surface and aircraft measurements). A previous paper (Atmospheric Research. (2004) in press) has dealt with dynamics and gas-phase chemistry. The present paper is an extension to aerosol formation, transport and evolution. To account for important loadings of primary and secondary aerosols and their transformation processes in the ESCOMPTE domain, the ORISAM aerosol module (Atmospheric Environment. 35 (2001) 4751) was implemented on-line in the air-quality Meso-NH-C model. Additional developments have been introduced in ORganic and Inorganic Spectral Aerosol Module (ORISAM) to improve the comparison between simulations and experimental surface and aircraft field data. This paper discusses this comparison for a simulation performed during one selected day, 24 June 2001, during the Intensive Observation Period IOP2b. Our work relies on BC and OCp emission inventories specifically developed for ESCOMPTE. This study confirms the need for a fine resolution aerosol inventory with spectral chemical speciation. BC levels are satisfactorily reproduced, thus validating our emission inventory and its processing through Meso-NH-C. However, comparisons for reactive species generally denote an underestimation of concentrations. Organic aerosol levels are rather well simulated though with a trend to underestimation in the afternoon. Inorganic aerosol species are underestimated for several reasons, some of them have been identified. For sulphates, primary emissions were introduced. Improvement was obtained too for modelled nitrate and ammonium levels after introducing heterogeneous chemistry. However, no modelling of terrigeneous particles is probably a major cause for nitrates and ammonium underestimations. Particle numbers and size distributions are well reproduced, but only in the submicrometer range. Our work points out to the need of introducing coarse dust particles to further improve the simulation of PM-10 concentrations and more accurate modelling of gas–particle interactions.
Keywords: Aerosol model; Aerosol inventory; Inorganic and organics formation; Size distribution
Total pollen counts do not influence active surface measurements by Hanns Moshammer; Herwig Schinko; Manfred Neuberger (pp. 1551-1555).
We investigated the temporal association of various aerosol parameters with pollen counts in the pollen season (April 2001) in Linz, Austria. We were especially interested in the relationship between active surface (or Fuchs’ surface) because we had shown previously (Atmos. Environ. 37 (2003) 1737–1744) that this parameter during the same observation period was a better predictor for acute respiratory symptoms in school children (like wheezing, shortness of breath, and cough) and reduced lung function on the same day than particle mass (PM10). While active surface is most sensitive for fine particles with a diameter of less than 100nm it has no strict upper cut-off regarding particle size and so could eventually be influenced also by larger particles if their numbers were high.All particle mass parameters tested (TSP, PM10, PM1) were weakly ( r approximately 0.2) though significantly correlated with pollen counts but neither was active surface nor total particle counts (CPC).The weak association of particle mass and pollen counts was due mainly to similar diurnal variations and a linear trend over time. Only the mass of the coarse fraction (TSP minus PM10) remained associated with pollen counts significantly after controlling for these general temporal patterns.
Keywords: Corona discharge monitor; Active surface; Total suspended particles; Time patterns; PM; 1; PM; 10Abbreviations; PM; 1; PM; 10; Mass concentrations of particles with diameters of less than 1; respectively; 10; μm; FEV; 1; Forced expiratory volume in 1; s; CPC; Condensation particle counts (number concentration); TSP; Total suspended particles (mass concentration)
Predicting particle number concentrations near a highway based on vertical concentration profile by Yifang Zhu; William C. Hinds (pp. 1557-1566).
This paper presents vertical profiles of particle number concentration and size distribution near the Interstate 405 Freeway. Based on vertical concentration data, averaged emission factor for total particle number concentration was determined to be 8.3e14particle−1vehicle−1mile−1. A simple analytical solution was obtained by solving the atmospheric diffusion equation and used to predict horizontal profiles of total particle number concentrations. Good agreement was achieved when model results were compared with previously published experimental data (J.Air Waste Manage. Assoc. 52 (2002b) 174–185). Model inputs are traffic, wind, and locations. With the particle emission factors, traffic compositions, and meteorological data, particle number concentrations near freeways can be quantitatively determined using the simple equation achieved in this study.
Keywords: Ultrafine particles; Atmospheric dispersion; Emission factor; Vertical concentration profile