Comparison of Simulations with Field Data - aaqr

Aerosol and Air Quality Research, 13: 3–12, 2013 Copyright © Taiwan Association for Aerosol Research ISSN: 1680-8584 print / 2071-1409 online doi: 10.4209/aaqr.2011.01.0004

Distribution of Nanoparticles near a Major U.S. and Canada Trade Bridge: Comparison of Simulations with Field Data Kambiz Nazridoust1, Goodarz Ahmadi1, Chaosheng Liu1, Andrea R. Ferro2, Timothy R. McAuley3, Peter A. Jaques4, Philip K. Hopke4* 1

Department of Mechanical & Aeronautical Engineering, Clarkson University, Potsdam, N.Y. 13699, USA Department of Civil & Environmental Engineering, Clarkson University, Potsdam, N.Y. 13699, USA 3 Consulting for Health, Air, Nature, & a Greener Environment (CHANGE), Queensbury, N.Y. 12804, USA 4 Center for Air Resources Engineering and Science, Clarkson University, Potsdam, N.Y. 13699, USA 2

ABSTRACT Dispersion of ultrafine particles arising from traffic emissions on a major international bridge (the Peace Bridge) between U.S. and Canada was studied during the summer of 2004. A computational model for evaluating the transport and dispersion of vehicular emissions from the Peace Bridge Complex (PBC) into the downwind neighborhood was developed to improve the estimation of ultrafine particle number concentrations in this area of Buffalo, New York. An unstructured computational grid of the Peace Bridge and its vicinity was generated, and the mean airflow was simulated using the standard k-ε turbulence model in the FLUENTTM code (ANSYS, Inc, Canonsburg, PA). A Discrete Random Walk (DRW) model was used to simulate the instantaneous turbulence fluctuating velocity. A Lagrangian particle-tracking model was used to simulate the transport and dispersion of particles from the motor vehicles on the bridge and in the Peace Bridge Plaza area. The particle transport model accounts for the drag and Brownian forces acting on the particle, as well as the gravitational sedimentation effects. These results were compared with a series of particle size distribution measurements made over the region of interest. For particulate emissions measured in the size range of 16 to 166 nm, the simulated sizefractionated particle concentrations show agreement with the field measurements with estimated errors of approximately 15%. These results suggest that CFD modeling could provide the basis for reasonable estimates of the exposure from specific major roads in the downwind area. Keywords: Particle number concentrations; Size-distribution; Peace bridge; Computer model; Lagrangian.

INTRODUCTION Emissions from motor vehicles are known to be a major source of particulate matter in urban areas (Fenger, 1999; Sharma and Khare, 2001; Perrino et al., 2002). Spatial and temporally-resolved concentrations of these particles are dependent on the transport of the emissions from the vehicles on major roadways into the downwind neighborhood and the loss of particles through deposition during transport. Exposure to the particles would then depend on the locations and activity patterns of the human receptors. Given the potential for variability in the exposure from a major roadway compared to the local sources, it is difficult to directly relate local concentrations to the major roadways such that the health effects of those emissions can be assessed based on

*

Corresponding author. Tel.: +1315-268 3861; Fax: +1-315-268-4410 E-mail address: [email protected]

local area rates of illness such as childhood asthma (Lwebuga-Mukasa and Pszonak, 2001). Prior efforts have been made to use computational fluid dynamics (CFD) modeling of pollutant transport in urban areas, generally focusing on urban street canyons. Yamartino and Wiegand (1986) developed and evaluated simple models for the flow, turbulence, and pollutant concentration fields within an urban street canyon. Pollutant transport and dispersion in street canyons were analyzed by Baik and Kim (1999, 2002). Leuzzi and Monti (1998) performed particle trajectory simulations of dispersion around a building using a Lagrangian stochastic (LS) model. Computer simulations of particle transport and deposition near small buildings were reported by Ahmadi and Li (2000) and Liu and Ahmadi (2006). Chang and Meroni (2001) performed numerical and physical modeling of the bluff body flow and dispersion in urban street canyons. Baker (2001) considered the flow velocities and the dispersion of pollutants in the wake of a number of different types of ground vehicles. A series of simulations for different street canyons were reported by Anquetin et al. (1999), Walton et al. (2002), and Walton and

4

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

Cheng (2002). Nazridoust and Ahmadi (2006) compared the predictions of different computational models with the experimental data for street canyons. Computer modeling of airflow and particle transport near Peace Bridge was first reported by Liu and Ahmadi (2005). In that work, different turbulence models were tested, but they used hypothetical meteorological and emissions estimates as input to the modeling. Subsequently, the data have become available (McAuley et al., 2010a) and now an effort has been made to verify the CFD model for a typical residential area downwind of a major traffic source against field measurements. The Peace Bridge connecting Fort Erie, Ontario to Buffalo, New York was the second busiest U.S.-Canada bordercrossing between 1997 and 2006 with about 1.2 million average annual commercial vehicles (NAFTA, 2007). Emissions from the typically continuous traffic, including heavy diesel trucks, are transported by prevailing winds into an adjacent residential community (Ferro and Jaques, 2007; Ogueli et al., 2007; McAuley et al., 2010a, b) that has been reported to have an unusually high level of asthma (Lwebuga-Mukasa and Pszonak, 2001; Oyane et al., 2004). Studies related to experimental and theoretical analyses of particle transport and deposition in turbulent flows have been presented by Hinze (1975), Hinds (1982), Wood (1981), and Papavergos and Hedley (1984). Papavergos and Hedley evaluated the limitations and assumptions involved in the application of past theories in aerosol modeling to engineering design, and discussed their accuracy with respect to experimental data. More recently, the use of computational models for simulating aerosol transport in airflow has attracted considerable attention. Computer models for particle trajectory analysis and the corresponding particle deposition rate was described by Li and Ahmadi (1992, 1993), Li et al. (1994), and He and Ahmadi (1999), among others. A number of authors have considered experimental studies of pollutant transport in urban areas. Namdeo et al. (1999) measured levels of airborne fine and coarse particles in an urban street canyon in Nottingham, UK in order to investigate dispersion of traffic-related particulate pollution. Particle number size distributions from 3 to 800 nm in an urban area in Germany were observed and compared with simple numerical simulations (Wehner et al., 2002). In addition to site measurements of pollutants, extensive series of wind tunnel experiments were conducted by various authors. Mirzai et al. (1994) and Higson et al. (1994) performed wind tunnel experiments concerning dispersion of gaseous pollutants around an isolated building. Gerdes and Olivari (1999) investigated pollutant dispersion in an urban street canyon in a wind tunnel. A series of wind tunnel tests investigating dispersion of pollutants in a model of an urban area were conducted by Pearce and Baker (1999). Baker and Hargreaves (2001) performed a wind tunnel study of pollution dispersion in the wake of a moving vehicle in a cross-wind and used the results to assess the validity of a numerical model. Results from wind tunnel simulations of gaseous tracer dispersion in the convective boundary layer capped by a temperature inversion were presented by Fedorovich and Thäter (2002). In the present work, transport and dispersion of nano-

particulate pollutants that are emitted from motor vehicle exhaust in the Peace Bridge area are studied using a computational simulation model. To assess potential exposures to residents in the surrounding community of the Peace Bridge Plaza Complex (PBC), particular attention was given to pollutant deposition on the buildings and the ground level in the residential areas that are located downstream of the Peace Bridge Plaza. The mean turbulent airflow conditions were evaluated using the standard k-ε model of the FLUENT code (ANSYS, Inc, Canonsburg, PA). In contrast to the earlier work of Liu and Ahmadi (2005), particular attention was given to the evaluation of particle concentration at different sites. The instantaneous fluctuating turbulent velocity field was generated by a Discrete Random Walk model. The particle equation of motion included drag, lift and Brownian forces in addition to the gravitational sedimentation effects. The Brownian force is simulated as a white noise process. The size range of the simulation was truncated at 166 nm above which the particles are primarily the secondary accumulation mode particles transported into the area. Several computer simulations of the transport and deposition of particles near the Peace Bridge were performed. For particulate emissions in the size range of 16 to 166 nm, the corresponding simulated particle concentrations at specific locations are compared with field measurement data from these sites. Although the particle size distribution measurements extended to 6 nm, only the data for sizes > 16 nm were used so that the aperiodic nucleation events were excluded. FLOW AND PARTICLE TRANSPORT SIMULATIONS For steady state and incompressible airflow in this model, the governing equations are the continuity and the averaged Navier-Stokes equations (Hinze, 1975). Turbulent flow was modeled using the standard k-ε turbulence model as well as the Reynolds stress transport (RST) model (Liu and Ahmadi, 2005). While for near-wall flow conditions and particle deposition analysis, the RST model is more suitable, for particle dispersion and concentration analysis in the bulk flow, the k-ε turbulence model which converges faster is preferable. For the ultrafine particles, a Lagrangian trajectory analysis was performed. The discrete phase model of FLUENT code (ANSYS, Inc, Canonsburg, PA) was used and the dispersion of particles in the domain is studied. Here the stochastic tracking (random walk) model, which included the effect of instantaneous turbulent velocity fluctuations on particle trajectories, was used in the analysis. Detail procedures for particle transport and dispersion applied in this study were described by Liu and Ahmadi (2005, 2006). Additional details of the procedure for simulating the instantaneous velocity are provided in FLUENT User’s manual (ANSYS, Inc, Canonsburg, PA) users guide (1998). FIELD STUDY A field study was performed in the Peace Bridge Plaza

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

Complex area. The results were presented by Ferro and Jaques (2007) and McAuley et al. (2010a, b). A brief description will be provided here. An Engine Exhaust Particle Sizer™ (EEPS™; TSI Inc.) housed in a mobile lab, was used to directly measure the particle number concentrations over the size range of 5.6 to 560 nm at a number of sites near the Peace Bridge Complex (PBC) during June 2004 (Fig. 1). The EEPS™ measures particle size distributions continuously at 0.1 second time resolution, and has been characterized and reviewed in more detail (Johnson et al., 2004, Liu et al., 2005). The particle concentrations were time averaged, compiled, and categorized into particle size bins of 16 to 166 nm for comparisons with the computational model results. In the field study of the emissions, wind data from the meteorological tower at the Great Lakes Center (GLC) of Buffalo State University were acquired at a location about 650 meters upwind of the PBC and adjacent to the shoreline of the Niagara River. Additionally, wind conditions were measured directly in the field with a portable system housed on the mobile lab. Particle number concentration measurements were made within a one kilometer area between the north and east of the Peace Bridge Complex (PBC) that was predominantly downwind, and emissions transport between the PBC, and sites R1, R2 and M2 are evaluated in this paper. In the field study, the investigators also presented an average 3-year footprint of wind direction and speed in the PBC area for June, showing equal influences of wind from both the west-southwest, and southwest (Ferro and Jaques, 2007). A relatively stable measured west-southwest wind at about 3 m/s, was used in this study as the boundary conditions for simulating the mean airflow velocity. Three-dimensional contour plots of particle size

5

distributions for the GLC and R1 site on June 24th are presented for comparisons with the model (Figs. 2(a) and 2(b)). For the simulations, the measured pollutant concentrations at site R1, which is closest to the Peace Bridge complex, are assumed to represent the emissions from the traffic on the Peace Bridge and from the PBC. The concentrations at R2 and M2 sites were computed and the estimates were compared with the measured data. SIMULATION RESULTS AND COMPARISON WITH FIELD DATA The entire Peace Bridge and part of the residential areas in South-West Buffalo were modeled. The size of the computational domain was 1800 m × 1500 m × 120 m. The GAMBIT code (FLUENT User Guide, 1998) was used to generate an unstructured grid for the Peace Bridge and its vicinity for the computer simulation. A total of 606,592 tetrahedral cells were used in these simulations. Fig. 3(a) shows the geometric features of the computational domain. Here, the width of the Niagara River is 815 m and an area of 1500 m × 775 m of the City of Buffalo (US) near the bridge was modeled. The surface of Niagara River is assumed to be the reference zero elevation level. The elevation of the bridge deck from the surface of the river is 32 m, and the height of the ground levels on the Canadian and US side are 19 m and 26 m, respectively. While some of the buildings near the Peace Bridge Plaza are to the scale, to make the computational simulation manageable, the residential buildings in Buffalo’s West Side were grouped into 113 blocks with typical sizes of 115 m × 50 m × 10 m and 115 m × 50 m × 8 m. Fig. 3(b) shows the computational grid of the Peace Bridge and the Customs Plaza on the US side. The details of the grid for a vertical section near the

Fig. 1. GPS map of measurement sites tracking PM in Buffalo West-Side near the Peace Bridge Complex (PBC). The arrows indicate that those streets are one way streets moving in the direction of the arrow.

6

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

Fig. 2. EEPS™ measurements on June 24 for: a) Background site; and b) R1, upwind site.

Fig. 3. a) Geometric features of the computational domain. b) Computational surface grid for Peace Bridge and its vicinity; and c) a cross section of the computational grid.

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

bridge are shown in Fig. 3(c). It is seen that the grid is comparatively denser near the road and building surfaces. As noted before, the mean wind velocity of 3 m/s from west-southwest for the duration of field study obtained from the meteorological tower at the Great Lakes Center (GLC) and also measured at the mobile lab was used in the simulations. During this period the wind speed and direction were roughly constant. A temperature of 283 K, μ = 1.84 × 10–5 N·s/m2, and ρ = 1.125 kg/m3 for the air were also assumed. The Reynolds number based on the heights of the buildings was about 1.4 × 106, and the air was in a state of turbulent motion. In the earlier work of Liu and Ahmadi (2005) the Reynolds Stress Transport (RST) model was used to account for anisotropy of turbulence. In the present study which is focused on dispersion and concentration of particulate emission the standard k-ε turbulent model of the FLUENT code (ANSYS, Inc, Canonsburg, PA) was used for evaluating the local airflow field in the region. In the analysis, a Lagrangian particle tracking was used and a particle-to-fluid density ratio of S = 2000 was assumed. Furthermore, it was assumed that when a particle touches a solid surface, it sticks without rebound. Fig. 4(a) displays the mean airflow velocity contours near the Peace Bridge and surrounding area. Low velocity regions are observed in the wake of the bridge for a cross wind and between the buildings on the US side. The corresponding mean wind velocity vector fields near the Peace Bridge and Peace Bridge Plaza are shown in Fig. 4(b). The velocity field near the buildings are quite complex with re-circulation regions forming around the buildings. Additional airflow simulation results for different wind speeds can be found in Liu and Ahmadi (2005). To estimate transport and dispersion of particulate pollutant emissions from the traffic in the Peace Bridge area, a series of Lagrangian particle trajectory analyses were performed for the different size particles. For different diameters, ensembles of 8,520, 4,560 and 18,560 particles were assumed to be emitted, respectively, from the level of the road on Peace Bridge, Peace Bridge Plaza. To account for the far field background, particles are injected from south and west boundaries of the computational domain with a uniform distribution. The numbers of deposited particles that pass through different surfaces near different sites of interest were evaluated. Continuous particle size data collected by the EEPS™ at the GLC (Fig. 2(a)) showed that background particles were consistently in the range of about 40 to 90 nm. In contrast, measurements immediately downwind of the PBC at R1 (Fig. 2(b)) show high variability with time, likely the result of variations in the traffic densities at various distances from the plaza (i.e., between about 45 to 150 m, generally depending on whether trucks were heading to or coming from Canada). For example, frequent occurrences of nuclei particles (5 to 15 nm) measured upwind at R1 were likely from idling and slow-moving trucks heading to Canada (45 to 60 meters), particles in the 15 to 30 nm mode are likely from trucks somewhat further away (e.g., 60–75 meters), while particles in a larger mode (40–90 nm) appear to

7

directly evolve from the 15–30 nm mode. This latter mode was likely to be emissions from trucks more distant within the PBC area source. It is also observed that as with the GLC measurements, this larger mode continuously appears, thus demonstrating the contribution of both background and PBC sources to R1, R2, and M2. The corresponding particle concentrations at R1, the GLC, and the empirical and predicted values at R2 and M2 are shown in Fig. 5. It can be seen that R1 has the highest particle concentrations for both 16 and 46 nm particles (Fig. 5(a)). The concentration at R1 decreases as particle size increases. In general, the background concentration was much lower than that at R1 with the peak background number concentration occurring at 76 nm (Fig. 5(a)). The Lagrangian particle tracking method was used to estimate particle number concentrations at sites R2 and M2 using the field measurement data at R1 as the source concentrations and the GLC as the far field background site. Because measurements directly adjacent to the source were not possible, R1, being closest, was treated as the source-site for the model. The far field wind entering the southern and western inlets of the computational domain were assumed to transport the background concentration that was measured on the windward side of the PBC. The percentages of particles of different sizes that cross planes at the R2 and M2 sites perpendicular to the mean wind direction were then evaluated. Based on the particle concentrations of different sizes at R1 and the fraction that are transported, the corresponding concentrations at sites R2 and M2 were evaluated. Figs. 5(b) and 5(c), respectively, compare the simulation results with the measured concentrations at R2 and M2. In general, the simulation results are within 1% to 17% of the experimental values. The errors in the experimental values are of the order of 10% at higher particle number concentrations and increasing as the particle number decreases. The simulation results for 16 nm particles substantially underestimated the concentration measured at site R2 while overestimating the concentration of 46 nm particles. These results may reflect the inability of the model to fully account for the evolution of particle sizes for diesel exhaust emissions (e.g., Zhu et al., 2002a, b; Zhang et al., 2004). It should also be emphasized that, while all of the field measurements were for a single day, the field measurements were conducted at different times of day, and, thus, the averaged results are subject to temporal variations of traffic and wind speed. Therefore, part of the deviations between the simulations and the data could be attributed to the minor changes in emissions from the traffic over time since this date was chosen because there was very little variability in the wind speed and direction. Figs. 6(a) and 6(b) show the distributions of estimated deposited pollutant particles of various sizes on different surfaces, respectively, for emissions from the Peace Bridge traffic and from the far field background. The capture efficiency shown in this figure was defined as the percentage of the emission that is capture of the respected area. Here, “US Bank” refers to the river bank on the US side, and “US Ground” denotes the ground on the US side which is

8

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

Fig. 4. Mean airflow velocity: a) contours near the PB; and b) vector field near the PB. not covered by the buildings. It is observed that the estimated deposition mainly occurred on the residential buildings in the Buffalo-West Side. These figures also show the deposition patterns are similar for particles in the size ranges studied. Although particles with diameters between 16 and 160 nm have a wide range of diffusivities, the deposition is essentially uniform. Thus, it appears that the turbulent diffusion transport is the primary driver of deposition rather than the individual particle properties.

CONCLUSIONS Computer simulations for analyzing the transport, dispersion, and deposition of PM emissions from PBC were performed to produce a series of Lagrangian simulations for particles of different sizes in the range of 16 to 166 nm. The simulation results were compared with field measurement data. The model includes all the relevant forces acting on the particle including the Brownian motion as well as the

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

9

Fig. 5. a) Compiled particle data for R1 and the GLC near PBC on June 24. Comparison of: b) measured data and simulated PM concentrations for site R2 on June 24, 2004; and c) measured data and simulated PM concentrations for site M2 on June 24, 2004. effect of instantaneous turbulence fluctuations. Based on the observed results, the following conclusions can be drawn: a) Computer simulations provided estimates within 15% accuracy for particle transport and deposition of traffic related emissions. The simulation results for particulate concentrations at distances downwind from the PBC are in agreement with the field measurements with differences of the order of 15%. b) A fraction of PM emissions from the Peace Bridge Complex (PBC) were deposited on the buildings and streets in the residential areas of Buffalo’s

West Side. Thus, populations living in this area could be exposed to high fractions of particles associated with traffic from a major nearby highway. c) The deposition patterns for various particle sizes smaller than 200 nm are similar. Thus, the results of this type of simulation could be used to extrapolate fixed near-road monitor measurements into site specific concentrations that could be combined with timeactivity data to produce exposure estimates for use in health effects models that can relate illness to specific major highway traffic.

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

10

Fig. 6. Particle deposition patterns of particles originating from the: a) PBC; and b) far field. ACKNOWLEDGEMENTS

REFERENCES

This work was supported by the U.S. Department of Energy (NETL), and the NYSTAR through the Center for Advanced Material Processing (CAMP) of Clarkson University. Additional Financial support for this research was by the Health Effects Institute (HEI), Agreement 04-4, and supplemental support for the MAPL and supplies by Clarkson University’s Biology Department. The views expressed in this paper are those of the authors and do not necessarily reflect the views of HEI or its sponsors. The Center for Air Resources Engineering and Science (CARES) of Clarkson University provided several field instruments.

Ahmadi, G. and Li, A. (2000). Computer Simulation of Particle Transport and Deposition near a Small Isolated Building. J. Wind Eng. Ind. Aerodyn. 84: 23–46. Anquetin, S., Guilbaud, C. and Chollet, J.P. (1999). Thermal Valley Inversion Impact on the Dispersion of a Passive Pollutant in a Complex Mountainous Area. Atmos. Environ. 33: 3953–3959. Baik, J.J. and Kim, J.J. (1999) A Numerical Study of Flow and Pollutant Dispersion Characteristics in Urban Street Canyons. J. Appl. Meteorol. 38: 1576–1589. Baik, J.J. and Kim, J.J. (2002). On the Escape of Pollutant from Urban Street Canyons. Atmos. Environ. 36: 527–

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

536. Baker, C.J. (2001). Flow and Dispersion in Ground Vehicle Wakes. J. Fluids Struct. 15: 1031–1060. Baker, C.J. and Hargreaves, D.M. (2001). Wind Tunnel Evaluation of a Vehicle Pollution Dispersion Model. J. Wind Eng. Ind. Aerodyn. 89: 187–200. Chang, C.H. and Meroney, R.H. (2001). Numerical and Physical Modeling of Bluff Body Flow and Dispersion in Urban Street Canyons. J. Wind Eng. Ind. Aerodyn. 89: 1325–1334 Fedorovich, E. and Thäter, J. (2002). A Wind Tunnel Study of Gaseous Tracer Dispersion in the Convective Boundary Layer Capped by a Temperature Inversion. Atmos. Environ. 36: 2245–2255. Fenger, J. (1999). Urban Air Quality. Atmos. Environ. 33: 4877–4900. Ferro, A.R. and Jaques, P.A. (2007). Characterization of Primary and Secondary Particles and Associated Personal Exposures near a Major International Trade Bridge between the U.S. and Canada, Health Effects Institute, a Pilot Study Report. FLUENT User Guide (1998). FLUENT Inc., Lebanon, NH. Gerdes, F. and Olivari, D. (1999). Analysis of Pollutant Dispersion in an Urban Street Canyon. J. Wind Eng. Ind. Aerodyn. 82: 105–124. He, C. and Ahmadi, G. (1999). Particle Deposition in a nearly Developed Turbulent Duct Flow with Electrophoresis. J. Aerosol Sci. 30: 739–758. Higson, H.L., Griffiths, R.F., Jones, C.D. and Hall, D.J. (1994). Concentration Measurements around an Isolated Building: A Comparison between Wind Tunnel and Field Data. Atmos. Environ. 28: 1827–1836. Hinds, W.C. (1999). Aerosol Technology, Properties, Behavior, and Measurement of Airborne Particles, 2nd Edition, John Wiley, New York. Hinze, J.O. (1975). Turbulence, McGraw-Hill, New York. Leuzzi, G. and Monti, P. (1998). Particle Trajectory Simulation of Dispersion around a Building. Atmos. Environ. 32: 203–214. Li, A. and Ahmadi, G. (1992). Dispersion and Deposition of Spherical Particles from Point Sources in a Turbulent Channel Flow. Aerosol Sci. Technol. 16: 209–226. Li, A., Ahmadi, G., Bayer, R.G. and Gaynes, M.A. (1994). Aerosol Particle Dispersion and Deposition in an Obstructed Turbulent Duct Flow. J. Aerosol Sci. 25: 91– 112. Liu, C. and Ahmadi, G. (2005). Computer Simulation of Pollutant Transport and Deposition near Peace Bridge. Part. Sci. Technol. 23: 109–127. Liu, C. and Ahmadi, G. (2006). Transport and Deposition of Particle near a Building Model. Build. Environ. 41: 828–836. Lwebuga-Mukasa, J.S. and Pszonak, R. (2001). Patterns of Inpatient and Outpatient Care for Asthma in Erie and Niagara Counties, Western New York State. J. Asthma 38: 155–160. McAuley, T.R., Ferro, A.F., Hopke, P.K. and Jaques, P.A. (2010a). Spatial Measurements of Nanoparticles Using an Engine Exhaust Particle SizerTM within a Local

11

Community Downwind of a Major International Trade Bridge in Buffalo, N.Y. Aerosol Sci. Technol. 44: 1096– 1104. McAuley, T.R., Fisher, R., Zhou, X., Jaques, P.A. and Ferro, A.R., (2010b). Relationships of Outdoor and Indoor Ultrafine Particles at Residences Downwind of a Major International Border Crossing in Buffalo, NY. Indoor Air 20: 298–308. Mirzai, M.H., Harvey, J.K. and Jones, C.D. (1994). Wind Tunnel Investigation of Dispersion of Pollutants due to Wind Flow around a Small Building. Atmos. Environ. 28: 1819–1826. NAFTA Safety Stats (2007). Number of Incoming Border Crossings by State Port of Entry Incoming Trucks, U.S.Canada Border). http://ai.fmcsa.dot.gov/international/bo rder.asp. Namdeo, A.K., Colls, J.J. and Baker, C.J. (1999). Dispersion and Re-suspension of Fine and Coarse Particulates in an Urban Street Canyon. Sci. Total Environ. 235: 3–13. Nazridoust, K. and Ahmadi, G. (2006). Airflow and Pollutant Transport in Street Canyons. J. Wind Eng. Ind. Aerodyn. 94: 491–522. Ogulei, D., Hopke, P.K., Ferro, A.R. and Jaques, P.A, (2007). Factor Analysis of Submicron Particle Size Distributions near a Major United States-Canada Trade Bridge. J. Air Waste Manage. Assoc. 57: 190–203. Oyane, T.J., Rogerson, P. and Lwebuga-Mukasa, J.S. (2004). Geographic Clustering of Adult Asthma Hospitalization and Residential Exposure to Pollution at the United State-Canada Boarder Crossing. Am. J. Publ. Health 94: 1250–1257. Papavergos, P.G. and Hedley, A.B. (1984). Particle Deposition Behavior from Turbulent Flows. Chem. Eng. Res. Des. 62: 275–295 Pearce, W. and Baker, C.J. (1999). Wind Tunnel Tests on the Dispersion of Vehicular Pollutants in an Urban Area. J. Wind Eng. Ind. Aerodyn. 80: 327–349. Perrino, C., Catrambone, M., Di Menno Di Bucchianico, A. and Allegrini, I. (2002) Gaseous Ammonia in the Urban Area of Rome, Italy and Its Relationship with Traffic Emissions. Atmos. Environ. 36: 5385–5394. Perry, N.W. (2000) I-190 & Peace Bridge, Buffalo Overview. webpage http://www.empirestateroads.com/week/week15. html Sharma, P. and Khare, M. (2001). Modelling of Vehicular Exhausts – A Review. Transp. Res. D Trans. Environ. 6: 179–198. Walton, A., Cheng, A.Y.S. and Yeung, W.C. (2002a). Largeeddy Simulation of Pollution Dispersion in an Urban Street Canyon––Part I: Comparison with Field Data. Atmos. Environ. 36: 3601–3613. Walton, A. and Cheng, A.Y.S. (2002b). Large-eddy Simulation of Pollution Dispersion in an Urban Street Canyon – Part II: Idealised Canyon Simulation. Atmos. Environ. 36: 3615–3627. Wehner, B., Birmili, W., Gnauk, T. and Wiedensohler, A. (2002). Particle Number Size Distributions in a Street Canyon and Their Transformation into the Urban-air Background: Measurements and a Simple Model Study.

12

Nazridoust et al., Aerosol and Air Quality Research, 13: 3–12, 2013

Atmos. Environ. 36: 2215–2223. Wood, N.B. (1981). A Simple Method for Calculation of Turbulent Deposition to Smooth and Rough Surfaces. J. Aerosol Sci. 12: 275–290. Yamartino, R.J. and Wiegand, G. (1986). Development and Evaluation of Simple Models for the Flow, Turbulence and Pollutant Concentration Fields within an Urban Street Canyon. Atmos. Environ. 20: 2137–2156. Zhang, K.M., Wexler, A.S., Zhu, Y.F., Hinds, W.C. and Sioutas, C. (2004). Evolution of Particle Number Distribution near Roadways, Part II: The ‘Road-toAmbient’ Process. Atmos. Environ. 38: 6655–6665.

Zhu, Y., Hinds, W. C., Kim, S., Shen, S. and Sioutas, C. (2002a). Study of Ultrafine Particles near a Major Highway with Heavy-duty Diesel Traffic. Atmos. Environ. 36: 4323–4335. Zhu, Y.F., Hinds, W.C., Kim, S. and Sioutas, C. (2002b). Concentration and Size Distribution of Ultrafine Particles near a Major Highway. J. Air Waste Manage. Assoc. 52: 1032–1042. Received for review, January 27, 2011 Accepted, October 19, 2012