Atmospheric Environment (v.45, #32)

Attributing health effects to apportioned components and sources of particulate matter: An evaluation of collective results by Lindsay Wichers Stanek; Jason D. Sacks; Steven J. Dutton; Jean-Jacques B. Dubois (5655-5663).
It has been hypothesized that the composition of particulate matter (PM) may be a better predictor of health effects than PM mass alone. The regional differences in PM composition and the heterogeneity in PM risk estimates in large multi-city epidemiologic studies are consistent with this hypothesis. Since 2005, efforts have been made to relate apportioned components and sources of PM with human health outcomes in epidemiology, controlled human exposure and toxicology studies. We reviewed published studies that: 1) focused on short-term exposure to PM; 2) included at least five components of PM; 3) grouped them into factors or sources; and 4) used quantitative methods to examine the relationship between the factors or sources and health effects. We then examined whether specific groups of PM components or sources of PM are consistently linked to specific health effects across studies. Collectively, these studies suggest that cardiovascular effects may be associated with PM2.5 from crustal or combustion sources, including traffic, but at this time, no consistent relationships have emerged. Fewer studies evaluated respiratory health effects, and the evidence for associations was limited. Apportionment methods have linked a variety of health effects to multiple groups of PM components and sources of PM, but the collective evidence has not yet isolated factors or sources that would be closely and unequivocally related to specific health outcomes.► PM chemical composition may better predict health effects than PM mass or size. ► We reviewed 29 studies that attempted this by using apportionment methods. ► Were specific sources or factors consistently linked to specific health effects? ► Apportionment methods have identified links, but we found little if any consistency.
Keywords: Particulate matter; Health effect; Source apportionment; Particle pollution; Epidemiology; Toxicology;

Mercury and its related compounds are widely recognized as global pollutants. The accurate atmospheric modeling of its transport and fate has been the subject of much research throughout the last decade. Atmospheric gas, aqueous and heterogeneous chemistry are expected to occur for Hg-containing species and accurate implementation of their chemical parameters is essential for realistic modeling of mercury cycling. Although significant progress has been made, the current state of knowledge of mercury chemistry exhibits numerous uncertainties. The objective of this two-part review is to explore the sources of uncertainty from the viewpoint of mercury chemistry. In this first part, we assess the discrepancy that exists in the currently available mercury kinetic parameters for the gas and aqueous phases. Theoretical and experimental approaches of rate constant determination exhibit various levels of limitation and accuracy. We present an overview of the available techniques and the assumptions and shortcomings associated with these methods in order to assist the atmospheric modellers. We review specific mercury oxidation and reduction reactions that have been investigated and are commonly implemented in mercury models with respect to the uncertainties associated with them. We reveal that for most of these mercury reactions our current state of knowledge reflects a lack of proper understanding of their mechanisms. Atmospheric heterogeneity is a topic of great importance and we elaborate upon it in part II of this review.► We assess discrepancies in the existing mercury gas and aqueous phase rate constants. ► Assumptions in experimental and theoretical methods of rate determination are discussed. ► Uncertainties for specific mercury redox reactions are evaluated. ► Our current state of knowledge lacks proper understanding of most mercury reaction mechanisms. ► The available rate constants are not realistic due to atmospheric heterogeneity.
Keywords: Mercury cycling; Uncertainties; Experimental approach Theoretical calculations; Rate constants; Oxidation; Reduction; Heterogeneous chemistry;

Parametric study on particle size and SOF effects on EGR cooler fouling by Kwang Seok Hong; Kyo Seung Lee; Soonho Song; Kwang Min Chun; Doyoung Chung; Sunki Min (5677-5683).
Recent diesel engine technologies, developed for enhanced regulation of exhaust emissions, are characterized by high exhaust gas recirculation (EGR) rates and high-pressure fuel injection. The use of high EGR rates, by which high temperatures can be avoided in a cylinder, is an effective method for reduction of nitrogen oxide (NO x ) emissions. High-pressure fuel injection leads to smaller soot particles, which decreases mass-based soot emission. These technologies, however, also have effects on particulate matter (PM) characteristics that are closely related to EGR cooler fouling. High-pressure fuel injection, which makes smaller soot particles, increases the concentration of soot particles, and a high EGR rate, which lowers the temperature in the cylinder, increases the soluble organic fraction (SOF). In this study, we evaluated the effects of these changes in PM characteristics on EGR cooler fouling. Instead of an engine-based experiment, in which a parametric study is nearly impossible, a laboratory experiment was performed to separate variables. A soot generator was used to make model exhaust gas because the variables could be controlled separately (e.g., mean particle size and concentration of soot particles) and this improved the reproducibility of the experiments. Additionally, n-dodecane, a model compound representing diesel fuel, was vaporized and injected into the exhaust gas to test the effects of SOF on cooler fouling. For particle sizes ranging from 41 to 190 nm in diameter, the deposition fraction was inversely proportional to particle size. Thus, smaller soot particles in the exhaust gas were more likely to cause formation of thermophoretic deposits on the wall of the EGR cooler. At an EGR gas temperature of 350 °C, the deposition fraction was greatest (84%) for the smallest particle size of 41 nm, whereas the deposition fraction was least (7%) for the largest particle size, 190 nm. The performance degradation of the EGR cooler showed a similar trend to the measured deposit mass. As coolant temperature decreased, the effect of n-dodecane injection on the growth of PM deposits increased, which significantly reduced the overall thermal conductivity of the EGR cooler. This result showed that ‘wet soot’ caused greater contamination of the EGR cooler.► We evaluated the effects of PM on EGR cooler fouling. ► A soot generator and n-dodecane were used where mean particle size and concentration of soot particles could be controlled. ► The deposition and the performance degradation were inversely proportional to particle size. ► The n-dodecane injection reduced the overall thermal conductivity of the EGR cooler. ► This result showed that ‘wet soot’ caused greater contamination of the EGR cooler.
Keywords: Diesel engine; EGR; EGR cooler fouling; Soot; SOF;

Human exposure to fine particulate matter (PM2.5) is associated with short and long term adverse health effects. The amount of ambient PM2.5 that infiltrates indoor locations such as residences depends on air exchange rate (ACH), penetration factor, and deposition rate. ACH varies by climate zone and thus by geographic location. Geographic variability in the ratio of exposure to ambient concentration is estimated based on comparison of three modeling domains in different climate zones: (1) New York City; (2) Harris County in Texas, and (3) a six-county domain along the I-40 corridor in North Carolina. Inter-individual variability in exposure to PM2.5 was estimated using the Stochastic Human Exposure and Dose Simulation for Particulate Matter (SHEDS-PM) model. ACH is distinguishably the most sensitive input for both ambient and non-ambient exposure to PM2.5. High ACH leads to high ambient exposure indoors but lower non-ambient exposure, and vice versa. For summer, the average ratio of exposure to ambient concentration varies by 13 percent among the selected domains, mainly because of differences in housing stock, climate zone, and seasonal ACH. High daily average exposures for some individuals are mainly caused by non-ambient exposure to smoking or cooking. The implications of these results for interpretation of epidemiological studies are discussed.► Most human exposure to ambient fine particulate matter (PM2.5) occurs indoors. ► Ambient exposure to concentration (E a/C) for PM2.5 averages 0.52–0.60. ► Average E a/C varies with geographic differences in residential air exchange rates. ► The coefficient of variation of inter-individual variability in E a/C is 0.25. ► Variations in E a/C are not accounted for in many health effects studies.
Keywords: Exposure; Particulate matter; Variability; Air exchange rate; Penetration factor; Deposition rate;

Estimation of the concentrations of primary and secondary organic carbon in ambient particulate matter: Application of the CMB-Iteration method by Guo-Liang Shi; Ying-Ze Tian; Yu-Fen Zhang; Wen-Yuan Ye; Xiang Li; Xue-Xi Tie; Yin-Chang Feng; Tan Zhu (5692-5698).
A new method (the CMB-Iteration method) was developed and applied to estimate primary organic carbon (POC) and secondary organic carbon (SOC) concentrations in ambient particulate matter. In addition to the concentrations, this model also calculates the source contributions to POC and SOC. For each source category, the estimated source contribution is continuously calculated until the iteration minimum is achieved, i.e., the ratio of the estimated source contribution at the final step iteration to previous step iteration is less than 0.01. The application of this method is used to analyze a reported database. The result of the CMB-Iteration method is consistent with in the reported result in the literature. Additionally, synthetic datasets were analyzed using the CMB-Iteration method, and acceptable results were obtained. Finally, this method is used for a dataset collected from Taiyuan, China across different seasons (winter and summer). The calculated concentrations of SOC are 14.3 and 10.7 μg/m3 in winter and summer, respectively, suggesting that there are high secondary OC particles in this highly polluted city (Taiyuan).► A new receptor model (CMB-Iteration method) was developed. ► CMB-Iteration method can estimate the POC and SOC concentrations. ► Some datasets were studied and acceptable results were obtained. ► SOC concentration in winter was higher than that in summer in Taiyuan.
Keywords: CMB-Iteration method; Secondary organic carbon; Source apportionment; Receptor model;

Wet depositional fluxes of 210Pb- and 7Be-bearing aerosols at two different altitude cities of North Pakistan by N. Ali; E.U. Khan; P. Akhter; M.A. Rana; M.U. Rajput; N.U. Khattak; F. Malik; S. Hussain (5699-5709).
The natural radionuclides 210Pb and 7Be have been widely used as tracers of aerosols for the study of pollution and atmospheric transport processes. The depositional fluxes of atmospheric 210Pb and 7Be-bearing aerosols in precipitation samples have been continuously measured from October 2006 to April 2010 at a high altitude city of Murree (33° 85′ N, 73° 41′ E, Alt.2300 m a.s.l) and low altitude city of Islamabad (33° 38′ N, 73° 09′ E, Alt. 534 m a.s.l), Pakistan, using a high resolution spectrometer.The specific activity values of 210Pb and 7Be measured in rain samples collected monthly from open lawn of Murree ranged from 2–87 mBq l−1 and 80–1680 mBq l−1 with their mean values of 30 ± 11 mBq l−1 and 573 ± 201 mBq l−1, respectively. Similarly the mean specific concentration values of 210Pb and 7Be in open lawn rain samples of Islamabad are 123 ± 12 mBq l−1 and 442 ± 321 mBq l−1 respectively. The specific concentrations of 210Pb and 7Be in throughfall samples collected at Murree site only are in the range from 5–34 mBq l−1 and 110–1177 mBq l−1 with the mean values of 17 ± 6 mBq l−1 and 442 ± 321 mBq l−1 respectively. The annual deposition fluxes of these radionuclides at Murree varied from 1 to 129 Bq m−2  y−1 and 17 to 2202 Bq m−2  y−1 in the open lawn rain samples and from 1 to 35 Bq m−2  y−1 and 9 to 975 Bq m−2  y−1 in throughfall samples respectively. Similarly the annual mean flux values of both radionuclides in open lawn rain samples of Islamabad are 1137 Bq m−2  y−1 and 3801 Bq m−2  y−1 respectively. Observed seasonal variations of deposition of 210Pb and 7Be are explained in terms of different environmental features. The role of plantation in the interception of radionuclides in throughfall samples is analyzed and discussed. The effect of the altitude on the deposition processes at the investigated site is studied using the comparison of present results with the relevant published results and analysis.► 210Pb and 7Be fluxes in open rain were higher than that of throughfall samples. ► Seasonal dependence of radionuclides has been observed. ► 210Pb and 7Be can not be used as independent atmospheric tracers. ► 210Pb fluxes at Islamabad are higher than at Murree.
Keywords: Environment; Aerosols; 210Pb; 7Be; Scavenging; Depositional flux; Gamma-spectrometry;

Chemical composition of urban airborne particulate matter in Ulaanbaatar by Masataka Nishikawa; Ichiro Matsui; Dashdondog Batdorj; Dulam Jugder; Ikuko Mori; Atsushi Shimizu; Nobuo Sugimoto; Katsuyuki Takahashi (5710-5715).
Atmospheric pollution caused by airborne particulate matter in the winter season in Ulaanbaatar, Mongolia is a very serious problem. However, there is a complete lack of scientific observation data to define the situation prior to any remediation. PM10 and PM2.5 average monthly values obtained by continuous monitoring showed the concentrations of particles of both size categories exceeded 100 μg m−3 during November to February (winter). PM10 particles were sampled with filters in January (i.e. during the heating period) and June (i.e.non-heating period) of 2008 in central Ulaanbaatar. To determine the composition of urban airborne particulate matter we analyzed a range of ionic components, multiple elements including heavy metals, and organic and inorganic carbon (soot). We also measured the stable carbon isotope ratio of the soot. Total carbon (sum of organic carbon and inorganic carbon) accounted for 47% of the mass of the PM10 during the heating period and 33% during the non-heating period, and was the largest component of urban airborne particulate matter in Ulaanbaatar. Stable isotope ratios (δ13C) of soot generated during the heating period (−23.4 ± 0.2‰) approximated the ratios for coal used in Ulaanbaatar (−21.3 to −24.4‰), while the ratios during the non-heating period (−27.1 ± 0.4‰) were clearly different from the coal values. In the heating period, a very high correlation was observed between soot and organic carbon, SO4 2–, NO3 , F, Zn, As, and Pb, and we concluded that they were derived from coal combustion along with soot. In addition, the concentrations and their ratios relative to each other of Al, Fe, Ca, K, Na, Mg, and Mn hardly differed between the heating period and the non-heating period, and it was concluded that they were derived from soil dust.Display Omitted► PM10 and PM2.5 monthly concentrations in Ulaanbaatar exceed 0.1 mg m−3 in the winter. ► Total carbon (organic (OC) + inorganic (EC)) was largest component of PM. ► A very high correlation between EC, OC and others in PM was observed. ► The stable carbon isotope ratios of EC approximated the ratios for Mongolian coal.
Keywords: Ulaanbaatar; Urban airborne particulate matter; Chemical composition; Stable carbon isotope ratio; Coal; Soil dust;

Episodes of extremely high concentrations of tropospheric ozone in the urban environment in Bor - Serbia by Milica Arsić; Djordje Nikolić; Predrag Djordjević; Ivan Mihajlović; Živan Živković (5716-5724).
This paper presents results of measuring episodes of extremely high ozone concentrations in urban area of the town of Bor in Serbia. The measured ozone concentration of 3000 μgm−3 in two episodes, which lasted 3–5 days in November 2010, has not been recorded in the literature in this field. Sensitivity of generated ozone to the NOx was determined, since the concentrations of NOx in the same period were also high. In the morning when working hours start, episodes of increased NOx concentrations appear after several hours, leading to episodes of increased ozone concentrations reaching up to twenty times more than the threshold value, defined as harmful to vegetation and human health.► In this study we investigated very high episodes of the ground-level ozone. ► We examined impact of the primary pollutants on the ozone formation. ► Influence of the metorological paramethers on the ground-level ozone was analyzed. ► Measured mean hourly values of ozone varied in a large interval (7.38–3213 μgm−3). ► Strong NOx sensitivity of the generated ozone, above 120 μgm−3, has been proposed.
Keywords: Episodes of ozone concentrations; Tropospheric ozone; NOx; Urban environment; Urban air quality; Photochemical smog;

Traffic aerosol emission velocity derived from direct flux measurements over urban Stockholm, Sweden by M. Vogt; E.D. Nilsson; L. Ahlm; E.M. Mårtensson; H. Struthers; C. Johansson (5725-5731).
Size-resolved aerosol vertical number fluxes were measured using the eddy covariance method, 105 meters above the ground over the city of Stockholm, Sweden, between 1st April 2008 and 15th April 2009. The size range of the measurements cover particles from 0.25 to 2.5 μm diameter (D p). Emission velocities (v e) were calculated for the same size range and were found to be well correlated with friction velocity ( u ∗ ) and CO2 fluxes ( F CO 2 ). These variables were used to parameterize the emission velocity as v e = ( 129.14 ( 0.01 D p 2 + 0.06 D p + 0.008 ) + 1.22 ) u ∗ F CO 2 where v e and u ∗ are given in [m s−1], D p in [μm], and F CO 2 in [mmol m−2s−1].The parameterization reproduces the average diurnal cycle from the observations well for particles sizes up to 0.6 μm D p. For larger particles the parameterization tends to over predict the emission velocity. These larger particles are not believed to be produced by combustion and therefore not well represented by F CO 2 , which represents the traffic source through its fossil fuel consumption and the related CO2 emissions.► Size-resolved vertical aerosol number fluxes were measured. ► A parameterization for emission velocity was obtained. ► Particle diameter, friction velocity and CO2 flux were used for the parameterization. ► The equation reproduces the observed diurnal cycle and individual half hours.
Keywords: Primary aerosol emissions; Carbon dioxide emissions; Traffic aerosol; Urban aerosol; Traffic activity; Emission factors; Eddy covariance; Aerosol flux;

In response to the absence of monitoring data, the Air Pollution Model (TAPM), a 3D prognostic model that predicts both meteorology and air pollution concentrations, was used to investigate the transport of ammonia across part of the Sunshine Coast region (South East Queensland, Australia), following its emission to the atmosphere from 41 poultry farms (egg and broiler) in the district, which cumulatively house approximately 9.8 million fowl. The study estimated the total amount of ammonia that is deposited directly into the Pumicestone Passage waterbody (63 km2) and onto the surrounding catchment (1184 km2), through both wet and dry deposition processes over a twelve-month period. The results indicate that wet deposition is the dominant deposition process into the waterbody (89%) and catchment area (94%). Most of the ammonia deposition is predicted to occur within a relatively short distance from the farms, due to higher concentrations at these locations. In the base case simulation, the estimated 1823 tonnes of annual ammonia emissions were predicted to result in the direct deposition of approximately 2.3 tonnes of ammonia into the Pumicestone Passage waterbody and approximately 53.5 tonnes onto the Pumicestone Passage catchment land surface, where there was potential for its subsequent run-off into the waterbody. This annual loading into the waterbody is not insignificant and is likely to contribute to the formation of algal blooms. The fate of the remaining 91% of the ammonia estimated to be emitted from the farms was not accounted for in the deposition totals. This ammonia most likely remains suspended in the atmosphere and is transported outside of the simulation study area.► Transport and fate of gaseous ammonia from poultry production was modelled. ► Ammonia deposition to a coastal waterbody (Pumicestone Passage) was estimated. ► Total annual deposition to the waterbody was 2.3 tonnes, comparable to other studies. ► Total annual deposition to the waterbody catchment area was 54 tonnes. ► Wet deposition accounted for 96% of all ammonia deposited in the study area.
Keywords: Ammonia; Poultry; Emission; Transport; Deposition; Australia;

Nitrogen dioxide (NO2) exchange between the atmosphere and five European tree species was investigated in the laboratory using a dynamic branch enclosure system (consisting of two cuvettes) and a highly specific NO2 analyzer. NO2 measurements were performed with a sensitive gas phase chemiluminescence NO detector combined with a NO2 specific (photolytic) converter, both from Eco-Physics (Switzerland). This highly specific detection system excluded bias from other nitrogen compounds. Investigations were performed at two light intensities (Photosynthetic Active Radiation, PAR, 450 and 900 μmol m−2 s−1) and NO2 concentrations between 0 and 5 ppb. Ambient parameters (air temperature and relative humidity) were held constant. The data showed dominant NO2 uptake by the respective tree species under all conditions. The results did not confirm the existence of a compensation point within a 95% confidence level, though we cannot completely exclude emission of NO2 under very low atmospheric concentrations. Induced stomatal stricture, or total closure, by changing light conditions, as well as by application of the plant hormone ABA (Abscisic Acid) caused a corresponding decrease of NO2 uptake. No loss of NO2 to plant surfaces was observed under stomatal closure and species dependent differences in uptake rates could be clearly related to stomatal behavior.► A highly specific and sensitive NO2 analyzer provided identification of NO2 exchange. ► The results did not confirm the existence of a compensation point. ► NO2 deposition velocity was strongly correlated to stomatal conductance. ► Species dependent different uptake rates could be clearly related to stomatal behavior. ► No loss of NO2 to plant surfaces was observed.
Keywords: Nitrogen dioxide; Vegetation; Gas exchange; Deposition; Stomatal uptake;

Long-term study of cloud condensation nuclei (CCN) activation of the atmospheric aerosol in Vienna by J. Burkart; G. Steiner; G. Reischl; R. Hitzenberger (5751-5759).
During a total of 11 months, cloud condensation nuclei (CCN at super-saturation S 0.5%) and condensation nuclei (CN) concentrations were measured in the urban background aerosol of Vienna, Austria. For several months, number size distributions between 13.22 nm and 929 nm were also measured with a scanning mobility particle spectrometer (SMPS). Activation ratios (i.e. CCN/CN ratios) were calculated and apparent activation diameters obtained by integrating the SMPS size distributions. Variations in all CCN parameters (concentration, activation ratio, apparent activation diameter) are quite large on timescales of days to weeks. Passages of fronts influenced CCN parameters. Concentrations decreased with the passage of a front. No significant differences were found for fronts from different sectors (for Vienna mainly north to west and south to east). CCN concentrations at 0.5% S ranged from 160 cm−3 to 3600 cm−3 with a campaign average of 820 cm−3. Activation ratios were quite low (0.02–0.47, average: 0.13) and comparable to activation ratios found in other polluted regions (e.g. ). Apparent activation diameters were found to be much larger (campaign average: 169 nm, range: (69–370) nm) than activation diameters for single-salt particles (around 50 nm depending on the salt). Contrary to CN concentrations, which are influenced by source patterns, CCN concentrations did not exhibit distinct diurnal patterns. Activation ratios showed diurnal variations counter-current to the variations of CN concentrations.► CCN concentrations (at 0.5% super-saturation) are highly variable. ► No seasonal differences were found. ► CN show typical diurnal trends, while CCN do not. ► Activation ratios were found to be low (0.02–0.47, average: 0.13) but equally highly variable.
Keywords: CCN; Urban aerosol; Activation ratio; Activation diameter;

Multi Axis Differential Optical Absorption Spectroscopy (MAX-DOAS) measurements were taken during 12 hazy days in December 2006 in Fresno, USA. We obtained vertical distributions of aerosol extinction coefficients (AECs) and aerosol optical depths (AODs) at UV wavelengths of 353 and 380 nm by applying a recently developed aerosol retrieval algorithm based on O4 slant column densities (SCDs) from MAX-DOAS measurements. A linear correlation coefficient (R) of 0.74 (0.73) was obtained between surface PM2.5 concentrations and AECs at 353 (380) nm within the 0–1 km layer under hazy conditions. AODs obtained from MAX-DOAS measurements were validated by comparison with those measured by a collocated sunphotometer. AODs derived at both wavelengths were overestimated under the hazy atmospheric conditions compared with those retrieved from sunphotometer measurements. A coefficient of determination (R 2) of 0.58 (0.79) was obtained for AODs at 353 (380) nm from MAX-DOAS and sunphotometer during the hazy period, and AODs obtained using MAX-DOAS agreed with sunphotometer data within 50% and 40% for the different wavelengths, respectively.► AODs and AEC vertical profiles were obtained using MAX-DOAS during a hazy period. ► AECs at the 0–1 km layer were compared to surface PM2.5 concentrations. ► Correlations between AECs at 380 nm were comparable to those at 353 nm. ► AODs at 353 and 380 nm were overestimated compared to sunphotometer data. ► Better correlations between AODs at 380 nm were observed than those at 353 nm.
Keywords: Aerosol remote sensing; MAX-DOAS; O4; Haze effect;

Comparative study of the atmospheric chemical composition of three South American cities by Pérola C. Vasconcellos; Davi Z. Souza; Simone G. Ávila; Maria P. Araújo; Edson Naoto; Kátia H. Nascimento; Fernando S. Cavalcante; Marina Dos Santos; Patricia Smichowski; Eduardo Behrentz (5770-5777).
PM10 samples were collected in 2008 at three sites in South America in the framework of an international project (South American Emissions Megacities, and Climate; SAEMC). The concentration of metals, metalloids, ion and organic compounds of most PM10 samples collected at three sites (Buenos Aires (BAI), Bogotá (BOG) and São Paulo (SPA)) is below the air quality standard of the respective countries. At the sites n-alkanes and carbon preference index distribution indicated the influence of petroleum residues derived from vehicular emissions. Most PAH detected are attributed to light-duty gasoline vehicles and to stationary sources. At all sites benzo[a]pyrene equivalent values mean a significant cancer risk. Sulfate, nitrate, ammonium, calcium and sodium are the most abundant water-soluble ions at the three sites. Ammonium sulfate is likely the form presented for these species formed by photochemical reactions of precursors emitted mainly by vehicles. At BAI and SPA, formate/acetate ratios indicated the contribution of photochemical reactions; on the contrary, at BOG site, acetate is predominant, indicating strong contribution of vehicular emissions. São Paulo samples showed the highest concentrations of elements among all the sites. None of the toxic or potentially toxic elements exceed the guideline values of the World Health Organization. At BAI site earth crust seems to be the major source of Fe and Mn; at SPA, anthropogenic source is responsible for Pb and Zn presences. Traffic related element is well correlated at the three sites.► Atmospheric composition of three cities in South America was investigated. ► Organic and inorganic chemical composition of the particulate matter was useful to study the anthropogenic and natural markers. ► Benzo[a]pyrene equivalent values meant a significant cancer risk. ► Formate/acetate ratios indicated the strong contribution of photochemical reactions.
Keywords: Atmospheric pollution; SAEMC project; Atmospheric particulate matter; Organic pollutants; Water-soluble ions; Metals and metalloids;

The quantitative values of the urban warming effect over city stations in the Korean peninsula were estimated by using the warming mode of Empirical Orthogonal Function (EOF) analysis of 55 years of temperature data, from 1954 to 2008. The estimated amount of urban warming was verified by applying the multiple linear regression equation with two independent variables: the rate of population growth and the total population.Through the multiple linear regression equation, we obtained a significance level of 0.05% and a coefficient of determination of 0.60. This means that it is somewhat liable to the estimated effects of urbanization, in spite of the settings of some supposition. The cities that show great warming due to urbanization are Daegu, Pohang, Seoul, and Incheon, which show values of about 1.35, 1.17, 1.16, and 1.10 °C, respectively. The areas that showed urban warming less than 0.2 °C are Chupungnyeong and Mokpo. On average, the total temperature increase over South Korea was about 1.37 °C; the amount of increase caused by the greenhouse effect is approximately 0.60 °C, and the amount caused by urban warming is approximately 0.77 °C.► The quantitative estimation of the urban warming based on the Empirical Orthogonal Function. ► Urban warming of 0.77 °C for 55 years over South Korea. ► Greenhouse warming of 0.60 °C for 55 years over South Korea. ► Urban warming of about 56% of total temperature increase for 55 years.
Keywords: Urban warming; Global warming; Empirical Orthogonal Function; Korea; Urbanization; Mean temperature;

The enhancement of PM2.5 mass and water-soluble ions of biosmoke transported from Southeast Asia over the Mountain Lulin site in Taiwan by Chung-Te Lee; Ming-Tung Chuang; Neng-Huei Lin; Jia-Lin Wang; Guey-Rong Sheu; Shuenn-Chin Chang; Sheng-Hsiang Wang; Hill Huang; Horng-Wen Chen; Yuan-Liang Liu; Guo-Hau Weng; Hsin-Yo Lai; Shao-Peng Hsu (5784-5794).
Biomass burning (BB) in Southeast Asia (Indochina and southern China) occurs frequently in March and April every year. The burning plume is ordinarily transported eastward by the prevailing westerly, further affecting downstream air quality in East Asia. In this study, atmospheric aerosols were collected at the downstream Lulin Atmospheric Background Station (LABS, 2862 m a.s.l., central Taiwan) from April 2003 to April 2009. Results show that monthly means of PM2.5 were highest during the BB period, especially in March. The PM2.5 mean for BB activity was 17.5 μg m−3, while the daily PM2.5 mean can sometimes be above 40 μg m−3. The background PM2.5 level in free troposphere of the West Pacific was at 3.7 ± 1.8 μg m−3. This mean is roughly the same regardless of the air masses moving from China, Pacific Ocean, and South China Sea toward LABS. In addition, the highest PM2.5 level occurred in 2004, making it the most active year of BB for the whole observation period. Greater amounts of nitrate and potassium ions were observed in the PM2.5 collected during the BB period compared to the non-BB (NBB) period. Linear regression analysis on PM2.5 water-soluble ions shows a moderate correlation (R 2 = 0.59) between non-sea-salt potassium and nitrate ions during the BB period. Furthermore, for all trajectory source origins, ammonium ion had the best correlation (R 2 = 0.84) with non-sea-salt sulfate when the air masses were influenced by anthropogenic sources during the NBB period. The enhancement ratios of nitrate ion during the BB period could reach 6.7 and 9.7 relative to air masses from the BB source region and from the pristine area during the NBB period, respectively. During the study period, ammonia gas was found to be insufficient to neutralize sulfuric and nitric gases. Therefore, most aerosols were more acidic than basic. Our long-term observation of atmospheric aerosols with inter-annual variability is valuable in providing data for verifying BB source inventory and model performance in East Asia.► This paper reports six years’ aerosol data collected at Mt. Lulin (2862 m a.s.l.). ► It is the only site that receives aerosol from Southeast Asia in the West Pacific. ► The PM2.5 at around 4 μg m−3 can represent global background 3 km above ground. ► The observation found the most active biomass burning was in 2004. ► Great enhancement of PM2.5 and water-soluble ions observed for the transported biosmoke.
Keywords: Biomass burning; PM2.5; Water-soluble ions; Long-range transport; High-elevation site;

Development and evaluation of an air quality modeling approach to assess near-field impacts of lead emissions from piston-engine aircraft operating on leaded aviation gasoline by Edward Carr; Mark Lee; Kristen Marin; Christopher Holder; Marion Hoyer; Meredith Pedde; Rich Cook; Jawad Touma (5795-5804).
Since aviation gasoline is now the largest remaining source of lead (Pb) emissions to the air in the United States, there is increased interest by regulatory agencies and the public in assessing the impacts on residents living in close proximity to these sources. An air quality modeling approach using U.S. Environmental Protection Agency’s (EPA) American Meteorological Society/Environmental Protection Agency Regulatory Model (AERMOD) was developed and evaluated for estimating atmospheric concentrations of Pb at and near general aviation airports where leaded aviation gasoline (avgas) is used. These detailed procedures were made to accurately characterize emissions and dispersion leading to improved model performance for a pollutant with concentrations that vary rapidly across short distances. The new aspects of this work included a comprehensive Pb emission inventory that incorporated sub-daily time-in-mode (TIM) activity data for piston-engine aircraft, aircraft-induced wake turbulence, plume rise of the aircraft exhaust, and allocation of approach and climb-out emissions to 50-m increments in altitude. To evaluate the modeling approach used here, ambient Pb concentrations were measured upwind and downwind of the Santa Monica Airport (SMO) and compared to modeled air concentrations. Modeling results paired in both time and space with monitoring data showed excellent overall agreement (absolute fractional bias of 0.29 winter, 0.07 summer). The modeling results on individual days show Pb concentration gradients above the urban background concentration of 10 ng m−3 extending downwind up to 900 m from the airport, with a crosswind extent of 400 m. Three-month average modeled concentrations above the background were found to extend to a maximum distance of approximately 450 m beyond the airport property in summer and fall. Modeling results show aircraft engine “run-up” is the most important source contribution to the maximum Pb concentration. Sensitivity analysis shows that engine run-up time, Pb concentration in avgas, and the fraction of piston-engine aircraft that are twin-engine are the most important parameters in determining near-field Pb concentrations. Year-long air quality modeling for 2008 and sensitivity analysis for the maximum 3-month average concentration period suggest the potential for 3-month average Pb concentrations that exceed the current National Ambient Air Quality Standard for Lead (0.15 μg m−3). The modeling methodology used in this analysis is generally transferable to other general aviation single runway airports in coastal environments of which there are over 1700 in the United States. This modeling approach can also be used to evaluate the air quality improvements from various emission reduction measures.► Developed characterization of lead source strength and temporal variability. ► Determined spatial extent of elevated lead concentrations in near airport environment. ► Compared near airport lead concentrations with day-specific modeling results. ► Established importance of engine run-up emissions to ambient lead concentrations.
Keywords: Lead; Piston-engine aircraft; Modeling; Monitoring; General aviation airports;

Development of a passive sampler for gaseous mercury by M.S. Gustin; S.N. Lyman; P. Kilner; E. Prestbo (5805-5812).
Here we describe work toward development of the components of a cost effective passive sampling system for gaseous Hg that could be broadly deployed by nontechnical staff. The passive sampling system included an external shield to reduce turbulence and exposure to precipitation and dust, a diffusive housing that directly protects the collection surface during deployment and handling, and a collection surface. A protocol for cleaning and deploying the sampler and an analytical method were developed. Our final design consisted of a polycarbonate external shield enclosing a custom diffusive housing made from expanded PTFE tubing. Two collection surfaces were investigated, gold sputter-coated quartz plates and silver wires. Research showed the former would require extensive quality control for use, while the latter had interferences with other atmosphere constituents. Although the gold surface exhibited the best performance over space and time, gradual passivation would limit reuse. For both surfaces lack of contamination during shipping, deployment and storage indicated that the handling protocols developed worked well with nontechnical staff. We suggest that the basis for this passive sampling system is sound, but further exploration and development of a reliable collection surface is needed.► A method of deployment and handling for a passive sampling surface was developed. ► Gold and silver collection surfaces for gaseous mercury were investigated. ► The sampling system developed was robust. ► Gold and silver surfaces may become passivated over time.
Keywords: Method development; Total gaseous mercury; Tubular passive sampler; Gold surface; Silver surface;

In order to assess the contribution of natural sources (sea salt and mineral dust) to PM10 levels in North-West Germany, a one year measurement project was conducted at two sites between April 2008 and March 2009. The sites were located in an urban and a regional background area. A Positive Matrix Factorization (PMF) based source apportionment study was carried out using chemical composition data of PM10 and PM1 filter samples from 79 selected days in a pooled dataset yielding eight source related factors. High to moderate urban versus regional correlations are obtained for factors denoted as aged marine aerosol, aged mineral dust, secondary sulfate and fossil fuel combustion. The factors identified as secondary nitrate, biomass combustion, re-suspended road dust and industry do not correlate significantly. Since the PMF factors do not represent the natural sources in the meaning of the EU air quality directive, tracer methods based on sodium, chloride and calcium are proposed to infer the PM10 concentrations of natural sea salt, mineral dust background, and the impact of long-range dust intrusions on PM10 concentrations, respectively. These tracer methods are viewed suitable for application in routine source apportionment within the air quality monitoring network.► PM10 shares by sea salt and mineral dust are quantified. ► Various mass closure approaches and PMF results are compared. ► Tracer models are more suitable than PMF to quantify natural contributions. ► Methodologies to routinely monitor the source contributions are proposed. ► For NW Germany sea salt has a minor effect on daily PM10 limit exceedances.
Keywords: Source apportionment; Natural sources; Positive matrix factorization; Aged sea salt; Mineral dust;

Long-term precipitation trend as a function of isentropic variability by Robert C. Balling; Kimberly DeBiasse; Matthew B. Pace; Randall S. Cerveny; David M. Brommer (5822-5827).
Analysis of trends in climatological isentropes provides a means of evaluating seasonal precipitation trends. Past research has indicated that isentropic patterns are linked to moisture influx and adiabatic uplift. We calculate trends across the summer 315 K isentropic surface for the conterminous United States over the last fifty years. We identify positive correlations between time and the pressure levels of the 315 K isentropic surface in the southern Plains states and the New England area. Such correlations suggest long-term rising of pressures (lowering heights) for the isentropic surface (and therefore enhanced moisture inflow) into those regions. Weaker correlations with time and the isentropic surface are found in the Southwest and in the Northwest. Using two well-accepted precipitation databases, we compare the map pattern correlations between the trends in isentropic surfaces and the trends in both precipitation daily totals and in hourly durations. A strong correlation (+0.67) exists between the trends in the isentropic surface and the daily precipitation and indicates that the precipitation increases are likely influenced by a stronger moisture influx. A weaker positive correlation (+0.44) is evident between the trends in daily precipitation intensity and the trends in the 315 K isentropic surface. However, the spatial relationship between the trends in the isentropic surface and the trends in the hourly precipitation duration is only −0.07 indicating no significant accounting of trends in duration by corresponding isentrope surface trends. Continued study of the relatively neglected analysis of climatological isentropes will add to our understanding of the physical changes in precipitation patterns across the United States.► Positive trend in isentropic pressure levels exists in South US and New England. ► Relationship exists between trends in isentropic surface and daily precipitation. ► No relationship exists between isentropic surface and hourly precipitation duration.
Keywords: Isentropic surface; Precipitation characteristics; Climate change;

Passive sampling of volatile organic compounds (VOCs) in air has received increasing attention in recent years. However, in order to use passive sampling as a reliable sampling technique a compound and sampler specific uptake rate is needed. Therefore, the scope of our study was threefold. First, uptake rates for 25 VOCs were determined under real indoor and outdoor conditions using axial-sampling tube-type samplers filled with Tenax TA, and active (pumped) sampling as a reference technique. Secondly, the mechanisms of passive sampling were investigated by comparing the experimentally determined uptake rates (0.13–0.46 ml min−1) to the ideal uptake rates, calculated based on Fick’s first law of diffusion and sampler geometry. Sampling efficiency SE, defined as the ratio between the experimental and ideal uptake rate, was introduced as a correction factor and showed that ideal uptake rates may underestimate VOC concentrations by a factor up to 4. This compound dependent SE is explained in terms of the partitioning coefficient K, i.e. the compound’s Tenax TA to air concentration equilibrium ratio. Compounds with a low K-value showed the most pronounced non-ideal sorptive behavior. Third, the experimentally determined uptake rates were used to determine VOC concentrations (between 12 and 311 μg m−3) in 6 homes for the elderly in Antwerp (Belgium). This study provides unique data for indoor air quality at care centers in Flanders.► Passive sampling on axial tube-type Tenax TA sorbent samplers is investigated. ► Uptake rates for 25 VOCs compounds are determined in real environments. ► Experimental uptake rates are a factor of 1.4–3.8 lower than the ideal ones. ► A positive correlation between sampling efficiency and log (K Tenax/air) is found. ► First data on VOC concentration levels in homes for the elderly are presented.
Keywords: Passive sampling; VOCs; Uptake rates; Indoor air; Homes for the elderly; Flanders;

An integrated experimental methodology has been applied to measure number and size distributions of particles in the 5–560 nm size range in the wake of a diesel car running at different speeds. Measurements were made at both ground-fixed (0.10 and 0.25 m above the ground level) and on-board (in 12 different sampling locations behind the moving car) measurement configurations using a fast response differential mobility spectrometer (Cambustion DMS50) with a sampling frequency up to 10 Hz. Results from both the experimental campaigns were analysed to understand the dynamics, dispersion and transport of nanoparticle emissions in the wake of a moving vehicle. Temporal changes in results were divided into three main stages (pre-evolution, evolution and post-evolution) after the release of exhaust emissions from the tailpipe. Evolution stage is of most interest where all the changes to particle number and size distribution occurred. Up to four evolution sub-stages were observed, each showing distinct evolution patterns of particle size distributions, depending on the particular experimental run. In agreement with previous studies, dilution was found to be the dominant process throughout all the evolution stages. The first evolution sub-stage was common to all the measurements, and consisted of an initial particle number concentrations and distributions change due to rapid (less than 1 s) nucleation followed by a rapid increase of accumulation mode particle number concentrations. After this first sub-stage the presence of vehicle wake with recirculating particles and the possible influence of other transformation processes lead to complex interactions. Results from the two experimental datasets clearly confirm the presence of two separate groups of particles: (i) new particles, which are freshly emitted and come directly from the tailpipe and (ii) relatively aged particles, which are entrained within the recirculation vortices of the vehicle wake and reside there for a longer time. The two groups have different characteristics and interact with each other. This interaction has often been overlooked in past studies about local scale dispersion of nanoparticle from moving vehicles.► Number concentrations of particles in the 5–560 nm size range were measured. ► Both on-board and ground-fixed sampling in the wake of a moving diesel car. ► 10 Hz sampling frequency allowed analysis of fast processes acting on nanoparticles. ► Particles entrained in the car wake differed from new freshly emitted particles. ► Complex interactions between wake and emitted plume were analysed and discussed.
Keywords: Fast response measurements; Nanoparticles dispersion; Number concentrations; Size distribution; Ultrafine particles; Vehicle wake;

Long-term trends in atmospheric reactive nitrogen across Canada: 1988–2007 by Antoni L. Zbieranowski; Julian Aherne (5853-5862).
The long-term trends in atmospheric reactive nitrogen (Nr) species at 12 Canadian Air and Precipitation Monitoring Network (CAPMoN) stations (9 with air and precipitation observations) across Canada were evaluated during the period 1988–2007. The non-parametric Mann–Kendall test was used to determine monotonic trends in the annual average concentrations of gaseous nitric acid (HNO3), particulate nitrate ( pNO 3 − ) , particulate ammonium ( pNH 4 + ) , wet ammonium ( NH 4 + ) and wet nitrate ( NO 3 − ) in response to emission reductions primarily driven by the Canada–United States Air Quality Agreement. Annual air concentrations (1988–2007) of pNH 4 + and HNO3 significantly decreased at all CAPMoN stations, while pNO 3 − concentrations increased at 6 of 9 stations. Precipitation NH 4 + had no significant or consistent trend; in contrast precipitation NO 3 − concentrations significantly decreased at 9 of the 11 stations and increased at one (non-significant). Normalized temporal sequences showed consistent patterns across Canada for several Nr species. Annual average air concentrations of pNH 4 + and HNO3 had synchronous time-series with consistently decreasing concentrations across all CAPMoN stations (1988–2007), in contrast, pNO 3 − had a complex temporal pattern, characterised by an initial period of no change (1988–1993), followed by a period of steep increase (1993–2002) and then a period of steep decrease (2002–2007). The period of steep decrease started around 2002 and was mirrored by all Nr species (except wet NH 4 + ) at all CAPMoN stations. The steep decrease was consistent with the observed decrease in NOx emissions from power plants and on-road vehicles in the United States and Canada. Legislated emission reductions have resulted in significant decreases in Nr concentrations except for NH 4 + , as such uncontrolled ammonia emissions continue to be a significant source of Nr.► Atmospheric NO 3 − , pNH 4 + and HNO3 significantly decreased across Canada (1988–2007). ► No trend in atmospheric NH 4 + concentrations across Canada (1988–2007). ► Decreases in atmospheric reactive nitrogen follow NOx and SO2 emission reductions. ► Greatest decreases in reactive nitrogen concentrations in southern Ontario. ► Reduced NOx emissions are predicted to decrease environmental impacts of N deposition.
Keywords: Air and precipitation concentrations; Atmospheric deposition; Mann–Kendall; Monotonic trend; Canadian Air and Precipitation Monitoring Network (CAPMoN);

NO and NO2 interconversion downwind of two different line sources in suburban environments by A.M. Chaney; D.J. Cryer; E.J. Nicholl; P.W. Seakins (5863-5871).
Nitrogen monoxide (NO), nitrogen dioxide (NO2) and ozone (O3) have been measured downwind (0–450 m) from two line sources in a suburban environment in a set of initial experiments designed to investigate the role of chemistry in determining the partitioning of NOx into NO and NO2 and the ability of a commercial modelling package (ADMS) to reproduce downwind speciated NOx concentrations. NO concentrations decrease more rapidly than NO2 as NO is removed both by dilution and by chemical reaction with ozone. Whilst NO is the dominant form of NOx at the roadside, NO2 concentrations decay more slowly (due to chemical production and higher background levels) and NO2 typically forms the majority of NOx by 50–100 m from the source. For a majority of modelling studies the input NO and NO2 concentrations were set to match the experimental concentrations at the initial site. The measured decays of NO and NO2 can be accurately reproduced by the ADMS dispersion model only when chemical transformation is taken into account. When NOx is estimated directly from traffic flow measurements, the modelled NO2 tends to underestimate the measured value.Display Omitted► [NO] and [NO2] monitored downwind of line source. ► Measurements show evidence of chemical transformation determining [NO2]. ► Concentration decays can be modelled successfully if fast chemistry is included. ► Results support the use of such models for legislative or other purposes.
Keywords: Line sources; NO and NO2 concentrations; Chemical transformation; Dispersion modelling;

Uptake rate behavior of tube-type passive samplers for volatile organic compounds under controlled atmospheric conditions by Christophe Walgraeve; Kristof Demeestere; Jo Dewulf; Katrijn Van Huffel; Herman Van Langenhove (5872-5879).
A systematic laboratory research was performed to investigate the diffusive uptake of four target compounds (limonene, toluene, ethyl acetate, hexane) on axial-sampling sorbent tubes filled with Tenax TA. In a controlled atmosphere (temperature: 21.0 ± 1.8 °C; relative humidity set at different values between 5 and 80%), these passive samplers were exposed for 1, 3 and 7 days to a constant target compound concentration between 8 and 85 ppbv. Simultaneous active sampling and TD-GC-MS analysis was performed to determine VOCs concentrations. A power-law relationship between the sorbed mass on the sampler and exposure dose (11–179 ppmv min) was observed for all compounds. Compounds with a lower Tenax to air partitioning coefficient, i.e. with a lower affinity for Tenax showed a larger deviation from the expected linear relationship, assuming theoretical sampling rates on an ideal sorbent. Results showed that experimentally determined uptake rates (0.90–2.12 ng (ppmv min)−1) may be up to a factor of 2 lower than the theoretical ideal sampling rates, calculated from the molecular diffusion coefficients and the geometrical parameters of the sampler. The applicability of internal standard calibration for passive sampling was evaluated and a practical methodology is proposed with ethylbenzene-d10 as an internal standard.► Four air pollutants are studied: limonene, toluene, ethyl acetate and hexane. ► The diffusive uptake process is studied quantitatively and mechanistically. ► Sampling dose and thermodynamic sorption constants affect the uptake behavior. ► Ideal uptake rates underestimate TWA concentrations for Tenax TA-filled samplers. ► A method is proposed to apply internal standard calibration for passive sampling.
Keywords: Volatile organic compounds; Passive sampling; Uptake rate; Partitioning coefficient; Internal standard calibration;

This paper explains the first emission inventory for BC and OC for Danish sources related to fuel combustion for the historical 1990–2008 period and the forecast years 2009–2030. The core basis for the BC/OC inventory is the Danish TSP emission inventories and projections. The relevant BC/OC emission factor information is based primarily on the European models GAINS and COPERT IV. Residential sources are the largest contributor to TSP, BC and OC emissions, representing in 2008 70%, 62 % and 83% of the total, followed by road transport exhaust and non exhaust, other mobile sources and other stationary sources. The total Danish emissions of TSP, BC and OC decrease by 14%, 28% and 12%, respectively, from 1990 to 2030. The emissions for Greenland and the Faroe Islands are less than 1% of the Danish total.For residential wood combustion a small number of detailed BC/OC measurements are available from the literature. These emission data are, however, disregarded in the Danish case, either due to lack of combustion unit representation, or data being related to wood type not accounted for in the Danish fuel data. To improve the inventory in this part, more emission measurements are needed especially for new stoves and boilers, which will become increasingly dominant in the future.For private stoves as such, it is desirable to develop a functional filter technology in order to bring down the particulate emissions in the future just as efficiently as for road transport and larger non road mobile machinery types.The work presented in this paper may serve as an input for policy makers dealing with the environmental impacts from combustion related BC and OC emissions. Further the calculated results can be used for air dispersion modelling purposes following a spatial distribution of the detailed emission inventory data.► Technology based BC/OC emissions are calculated for Denmark, Greenland/Faroe Islands. ► Danish TSP[BC, OC] emissions decrease by 14%[28 %, 12 %] from 1990 to 2030. ► For residential plants the TSP[BC, OC] emission shares were 70%[62 %, 83 %] in 2008. ► Emission contribution from Greenland and Faroe Islands is less than 1%. ► Stove and small diesel engine emissions may decline if filters are further developed.
Keywords: BC/OC emissions; TSP; Stationary sources; Road transport; Non road mobile machinery;

Water uptake properties of internally mixed sodium halide and succinic acid particles by Lorena Miñambres; Estíbaliz Méndez; María N. Sánchez; Fernando Castaño; Francisco J. Basterretxea (5896-5902).
Sea salt aerosols include appreciable fractions of organic material, that can affect properties such as hygroscopicity, phase transition or chemical reactivity. Although sodium chloride is the major component of marine salt, bromide and iodide ions tend to accumulate onto particle surfaces and influence their behaviour. The hygroscopic properties of internally mixed submicrometric particles composed of succinic acid (SA) and NaX (where X = F, Cl, Br or I) have been studied by infrared absorption spectroscopy in an aerosol flow cell at ambient temperature for different relative succinic acid/NaX compositions. The results show that deliquescence relative humidities of SA/NaF and SA/NaCl are equal to those of the pure sodium halides. SA/NaBr particles, on the other hand, deliquesce at lower relative humidities than pure NaBr particles, the effect being more marked as the SA/NaBr mass ratio approaches unity. The SA/NaI system behaves as a non-deliquescent system, absorbing liquid water at all relative humidities, as in pure NaI. Succinic acid phase in the particles has been spectroscopically monitored at given values of both RH and SA/NaX solute mass ratio. The different hygroscopic properties as the halogen ion is changed can be rationalized in terms of simple thermodynamic arguments and can be attributed to the relative contributions of ion–molecule interactions in the solid particles. The observed behaviour is of interest for tropospheric sea salt aerosols mixed with organic acids.► Succinic acid (SA)/sodium halide particle water uptake studied by IR spectroscopy. ► SA/NaBr particles deliquesce at lower relative humidities than pure NaBr. ► SA/NaI particles behave as non-deliquescent. ► Different behaviour attributed to ion–molecule interactions in particles.
Keywords: Sodium halides; Organic acids; Tropospheric marine aerosol; Deliquescence; Infrared absorption spectroscopy; Surface chemistry;

Inverse modeling analysis of soil dust sources over East Asia by Bonyang Ku; Rokjin J. Park (5903-5912).
Soil dust is the dominant aerosol by mass concentration in the troposphere and has considerable effects on air quality and climate. Parts of East Asia, including southern Mongolia, northern China, and the Taklamakan Desert, are important dust source regions. Accurate simulations of dust storm events are crucial for protecting human health and assessing the climatic impacts of dust events. However, even state-of-the-art aerosol models still contain large uncertainties in soil dust simulations, particularly for the dust emissions over East Asia. In this study, we attempted to reduce these uncertainties by using an inverse modeling technique to simulate dust emissions. We used the measured mass concentration of particles less than 10 μm in aerodynamic diameter (PM10) in the surface air over East Asia, in combination with an inverse model, to understand the dust sources. The global three-dimensional GEOS-Chem chemical transport model (CTM) was used as a forward model. The inverse model analysis yielded a 76% decrease in dust emissions from the southern region of the Gobi Desert, relative to the a priori result. The a posteriori dust emissions from the Taklamakan Desert and deserts in eastern and Inner Mongolia were two to three fold higher than the a priori dust emissions. The simulation results with the a posteriori dust sources showed much better agreement with these observations, indicating that the inverse modeling technique can be useful for estimation of the optimized dust emissions from individually sourced regions.► Inverse modeling study using surface observations for soil dust sources. ► The a priori dust source is 38.2 Tg over Asia in April 2001. ► The a posteriori source is 35.4 Tg, slightly lower than the a priori value. ► The spatial pattern of the a posteriori differs significantly from the a priori.
Keywords: Inverse modeling; Soil dust; Dust emission; Particulate matter; Chemical transport model;

Emission estimates of particulate matter (PM) and trace gases (SO2, NO and NO2) from biomass fuels used in rural sector of Indo-Gangetic Plain, India by T. Saud; T.K. Mandal; Ranu Gadi; D.P. Singh; S.K. Sharma; M. Saxena; A. Mukherjee (5913-5923).
In this paper, we present the experimentally determined emission factors and emission estimates of particulate matter (PM), SO2, NO and NO2 emitted from biomass fuels used as energy in rural area of Indo-Gangetic Plain (IGP), India. Biomass fuel samples were collected at district level from this region. The burning of the collected biomass fuels is performed by using the modified dilution sampler based on studies done by . In this study, the emission factor represents the total period of burning including pyrolysis, flaming and smoldering. The average emission factor of PM from dung cake, fuel-wood and crop residue over Delhi, Uttar Pradesh, Punjab, Haryana, Uttarakhand and Bihar are estimated as 16.26 ± 2.29 g kg−1, 4.34 ± 1.06 g kg−1 and 7.54 ± 4.17 g kg−1 respectively. Similarly, the average emission factor of SO2, NO and NO2 from dung cake, fuel-wood and crop residue over this region are also determined (SO2: 0.28 ± 0.09 g kg−1, 0.26 ± 0.10 g kg−1 and 0.27 ± 0.11 g kg−1, NO: 0.27 ± 0.21 g kg−1, 0.41 ± 0.25 g kg−1 and 0.54 ± 0.50 g kg−1 and NO2: 0.31 ± 0.23 g kg−1, 0.35 ± 0.28 g kg−1 and 0.54 ± 0.47 g kg−1 respectively). The emission of PM, SO2, NO and NO2 from biomass fuels used as energy in rural household over, IGP are also estimated in this paper. The result shows the regional emission inventory from Indian scenario with spatial variability.► Emission factor of particulate matter, SO2 and NOx from biomass over IGP, India. ► Refinement of budget estimation of particulate matter, SO2 and NOx over IGP, India. ► New budget of particulate matter, SO2 and NOx from biomass fuel in India.
Keywords: Biomass burning; Indo-Gangetic Plain; Particulate matter; Trace gases;

Dust storm contributions to airborne particulate matter in Reykjavík, Iceland by Throstur Thorsteinsson; Guđrún Gísladóttir; Joanna Bullard; Grant McTainsh (5924-5933).
Episodes of high levels of particulate matter (PM) in Reykjavík occur several times a year. The main sources of daily variation in PM are traffic or highly localized (e.g. construction) sources, however several episodes have been identified where these are not the cause. Examining PM10 (diameter < 10 μm) levels around the time when dust storms are seen on satellite images, and verifying that the weather conditions are favorable for the duration of the high levels of PM (>50–100 μg m−3; 30-min average), demonstrates that dust storms are the source of these increased levels of PM10. Since satellite coverage is sparse, visual confirmation of many such peaks in PM10 cannot be achieved. The level of pollution measured in Reykjavík during dust storms indicates that at least 200 kg s−1 of PM10 sized material is being eroded and transported away from sand plains ∼110 km away – this equates to an emission rate of 35 g m2 h−1. The source regions for dust storms in Iceland are the sandur areas on the southern coast of Iceland, and regions close to the glaciers. With climate warming, and fast retreating glaciers, the potential source regions in Iceland are rapidly increasing.► We show that dust storms are an important contributor to PM pollution in Reykjavik. ► With climate warming potential source regions for dust storms are increasing. ► The emission rate of PM10 from sand plains 110 km away is estimated 35 g m2 h−1.
Keywords: Particulate matter; Air pollution; Climate; Sandstorms; Iceland;

Using the Generalised Additive Model to model the particle number count of ultrafine particles by S. Clifford; S. Low Choy; T. Hussein; K. Mengersen; L. Morawska (5934-5945).
In this paper, we compare the Generalised Linear Model (GLM) and Generalised Additive Model (GAM) for modelling the particle number concentration (PNC) of outdoor, airborne ultrafine particles in Helsinki, Finland. We examine temporal trends in PNC and examine the relationship between PNC and rainfall, wind speed and direction, humidity, temperature and solar insolation. Model choice is via the Akaike Information Criterion (AIC).We have shown that the Generalised Additive Model provides a better fit than the equivalent Generalised Linear Model (ELM) when fitting models with the same covariates with equivalent degrees of freedom (AIC and BIC for the GAM are 10266.52 and 10793.04, AIC and BIC for the ELM are 10297.19 and 10885.97, both have an R 2 value of 0.836). We also present results that show that modelling both temporal trends and the effect of rainfall, wind speed and direction, humidity, temperature and solar insolation yields a better fitting model, according to the AIC, than either temporal trends or meteorological conditions by themselves.The model is applicable to any longitudinal monitoring-type measurement campaign where long time series are recorded. Use of this technique may be inappropriate for very short measurement campaigns. Attempting to fit a representative daily trend to one or two days’ measurements may lead to a high degree of uncertainty; inclusion of a yearly trend requires having at least a year’s worth of data with few gaps, particularly large gaps. In such a situation, the temporal trends may end up being penalised to zero and the model reverts to one largely influenced by meteorology.► A Generalised Additive Model (GAM) was used to model outdoor ultrafine particle number concentration (PNC) in Helsinki. ► The GAM provided a better fit to measured PNC than the linear model (LM). ► The GAM also allowed for qualitatively and quantitatively different statistical inferences to be made. ► A GAM based on both meteorological and temporal data provided a better fit to measured PNC than only one of these.
Keywords: GAMs; Generalised Additive Model; Air pollution; mgcv; Regression; R; Semi-parametric regression; Particle number concentration; Smoothing splines; Splines; Penalised splines;

This paper reports the results of a parametric CFD study on idealized city models to investigate the potential of slope flow in ventilating a city located in a mountainous region when the background synoptic wind is absent. Examples of such a city include Tokyo in Japan, Los Angeles and Phoenix in the US, and Hong Kong. Two types of buoyancy-driven flow are considered, i.e., slope flow from the mountain slope (katabatic wind at night and anabatic wind in the daytime), and wall flow due to heated/cooled urban surfaces. The combined buoyancy-driven flow system can serve the purpose of dispersing the accumulated urban air pollutants when the background wind is weak or absent. The microscopic picture of ventilation performance within the urban structures was evaluated in terms of air change rate (ACH) and age of air. The simulation results reveal that the slope flow plays an important role in ventilating the urban area, especially in calm conditions. Katabatic flow at night is conducive to mitigating the nocturnal urban heat island. In the present parametric study, the mountain slope angle and mountain height are assumed to be constant, and the changing variables are heating/cooling intensity and building height. For a typical mountain of 500 m inclined at an angle of 20° to the horizontal level, the interactive structure is very much dependent on the ratio of heating/cooling intensity as well as building height. When the building is lower than 60 m, the slope wind dominates. When the building is as high as 100 m, the contribution from the urban wall flow cannot be ignored. It is found that katabatic wind can be very beneficial to the thermal environment as well as air quality at the pedestrian level. The air change rate for the pedestrian volume can be as high as 300 ACH.► Parametric CFD is used to study the effect of slope flow on city ventilation. ► The katabatic flow is conducive to mitigate the urban heat island. ► In our simulation, when the building is lower than 60 m, the slope wind dominates. ► When the building is as high as 100 m, the urban wall flow is important. ► The pedestrian air change rate can be as high as 300 ACH.
Keywords: Slope flow; Ventilation; Street canyon; Age of air; Wall flow;