Atmospheric Environment (v.108, #C)

Multi-wavelength optical determination of black and brown carbon in atmospheric aerosols by D. Massabò; L. Caponi; V. Bernardoni; M.C. Bove; P. Brotto; G. Calzolai; F. Cassola; M. Chiari; M.E. Fedi; P. Fermo; M. Giannoni; F. Lucarelli; S. Nava; A. Piazzalunga; G. Valli; R. Vecchi; P. Prati (1-12).
In this paper, a new way to apportion the absorption coefficient (babs) of carbonaceous atmospheric aerosols starting from a multi-wavelength optical analysis is shown. This methodology can disentangle and quantify the contribution to total absorption of equivalent black carbon (EBC) emitted by wood burning (EBCWB) and fossil fuel (EBCFF) as well as brown carbon (BrC) due to incomplete combustion. The method uses the information gathered at five different wavelengths in a renewed and upgraded version of the approach usually referred to as Aethalometer model. Moreover, we present the results of an apportionment study of carbonaceous aerosol sources performed in a rural area and in a coastal city, both located in the North-West of Italy. Results obtained by the proposed approach are validated against independent measurements of levoglucosan and radiocarbon. At the rural site the EBCWB and EBCFF relative contributions are about 40% and 60% in winter and 15% and 85% in summer, respectively. At the coastal urban site, EBCWB and EBCFF are about 15% and 85% during fall. The OC contribution to the wood burning source at the rural site results approximately 50% in winter and 10% in summer and about 15% at the coastal urban site in fall. The new methodology also provides a direct measurement of the absorption Ångström exponent of BrC (αBrC) which resulted αBrC = 3.95 ± 0.20.Display Omitted
Keywords: Carbonaceous aerosol; Light absorption; Source apportionment;

Enhanced PM10 bounded PAHs from shipping emissions by S. Pongpiachan; M. Hattayanone; C. Choochuay; R. Mekmok; N. Wuttijak; A. Ketratanakul (13-19).
Earlier studies have highlighted the importance of maritime transport as a main contributor of air pollutants in port area. The authors intended to investigate the effects of shipping emissions on the enhancement of PM10 bounded polycyclic aromatic hydrocarbons (PAHs) and mutagenic substances in an industrial area of Rayong province, Thailand. Daily PM10 speciation data across two air quality observatory sites in Thailand during 2010–2013 were collected. Diagnostic binary ratios of PAH congeners, analysis of variances (ANOVA), and principal component analysis (PCA) were employed to evaluate the enhanced genotoxicity of PM10 during the docking period. Significant increase of PAHs and mutagenic index (MI) of PM10 were observed during the docking period in both sampling sites. Although stationary sources like coal combustions from power plants and vehicular exhausts from motorway can play a great role in enhancing PAH concentrations, regulating shipping emissions from diesel engine in the port area like Rayong is predominantly crucial.
Keywords: Industrial estate; Mutagenicity; Polycyclic aromatic compounds; Vehicle emissions; Thailand;

Characteristics of flow and reactive pollutant dispersion in urban street canyons by Soo-Jin Park; Jae-Jin Kim; Minjoong J. Kim; Rokjin J. Park; Hyeong-Bin Cheong (20-31).
In this study, the effects of aspect ratio defined as the ratio of building height to street width on the dispersion of reactive pollutants in street canyons were investigated using a coupled CFD-chemistry model. Flow characteristics for different aspect ratios were analyzed first. For each aspect ratio, six emission scenarios with different VOC–NOX ratios were considered. One vortex was generated when the aspect ratio was less than 1.6 (shallow street canyon). When the aspect ratio was greater than 1.6 (deep street canyon), two vortices were formed in the street canyons. Comparing to previous studies on two-dimensional street canyons, the vortex center is slanted toward the upwind building and reverse and downward flows are dominant in street canyons. Near the street bottom, there is a marked difference in flow pattern between in shallow and deep street canyons. Near the street bottom, reverse and downward flows are dominant in shallow street canyon and flow convergence exists near the center of the deep street canyons, which induces a large difference in the NOX and O3 dispersion patterns in the street canyons. NOX concentrations are high near the street bottom and decreases with height. The O3 concentrations are low at high NO concentrations near the street bottom because of NO titration. At a low VOC–NOX ratio, the NO concentrations are sufficiently high to destroy large amount of O3 by titration, resulting in an O3 concentration in the street canyon much lower than the background concentration. At high VOC–NOX ratios, a small amount of O3 is destroyed by NO titration in the lower layer of the street canyons. However, in the upper layer, O3 is formed through the photolysis of NO2 by VOC degradation reactions. As the aspect ratio increases, NOX (O3) concentrations averaged over the street canyons decrease (increase) in the shallow street canyons. This is because outward flow becomes strong and NOX flux toward the outsides of the street canyons increases, resulting in less NO titration. In the deep street canyons, outward flow becomes weak and outward NOX flux decreases, resulting in an increase (decrease) in NOX (O3) concentration.
Keywords: CFD-chemistry-coupled model; Reactive pollutants; Street canyon; Aspect ratio; VOC–NOX ratio;

Characteristics of new particle formation events in Nanjing, China: Effect of water-soluble ions by Junlin An; Honglei Wang; Lijuan Shen; Bin Zhu; Jianan Zou; Jinhui Gao; Hanqing Kang (32-40).
New particle formation (NPF) events and water-soluble ions were studied at the meteorological building on the campus of the Nanjing University of Information Science and Technology (NUIST), which is located in the western part of the Yangtze River Delta (YRD). A wide-range particle spectrometer (WPS) provided particle number size distributions between 10 nm and 10 μm, whereas water-soluble ions for particles with diameters between 10 nm and 18 μm were measured using a 13-stage Nano-MOUDI aerosol sampler and 850 professional Ion Chromatography (IC). Additionally, meteorological data, trace gas concentrations and mass concentration were recorded. Ten NPF days were captured during the measurement period from 08 July to 02 August 2012. The mean aerosol number concentration, which was primarily composed of Aitken-mode particles, i.e., with diameters of 20–100 nm, was 13,664 cm−3, which was 1.9 times larger than that on non-NPF days. The results suggest that the NPF events were only slightly affected by O3, SO2, and NO2; the primary factors affecting NPF events were meteorological factors and air mass directions. NPF events were found to be favorable during the summer in the presence of high temperatures, strong radiation, low humidity, strong winds and clean air masses originating from the southeastern coast. The mean growth rate (GR), formation rate (J10), condensational sink (CS), condensing vapor rate (Q), and condensation vapor (C) were determined to be 7.6 nm h−1, 3.7 cm−3 s−1, 2.8 × 10−2 s−1, 2.9 × 106 cm−3 s−1, and 10.5 × 107 cm−3, respectively, on NPF days. The largest effects of the studied NPF events were on the mass and water-soluble ion concentrations of Aitken-mode particles, followed by nuclei-mode particles; few contributions to accumulation- and coarse-mode particles were observed. Different water-soluble ions were observed to have distinct interactions with the NPF events. The proportions of NH 4 + , SO 4 2 − , NO 3 − , K+ and Mg2+ in nuclei- and Aitken-mode particles to the total concentrations on NPF days were substantially higher than on non-NPF days.
Keywords: Atmospheric aerosol; NPF event; Number concentration; Water-soluble ions; Yangtze River Delta;

Gridded atmospheric emission inventory of 2,3,7,8-TCDD in China by Tao Huang; Chongguo Tian; Kai Zhang; Hong Gao; Yi-Fan Li; Jianmin Ma (41-48).
Establishment of the dioxins emission inventory has been considered as a crucial step toward risk assessment and elimination of dioxins contaminations. Based on a total dioxin emission inventory in China from different emission categories in 2004, this study created a gridded emission inventory of 2,3,7,8-TCDD, the most toxic congener in dioxins, in China in 2009 with a 1/4° longitude by 1/4° latitude resolution. It was estimated that annually total 371 ± 53 g (average ± standard deviation) of 2,3,7,8-TCDD was released into the atmosphere in 2009 over China, increasing approximately by 37% compared with its emission in 2004. Differing from most developed countries where municipal waste incinerations were regarded as a major atmospheric emission source, in China ferrous and non-ferrous metal production made the largest contribution to 2,3,7,8-TCDD air emission (138 ± 16 g), followed by waste incineration (109 ± 12 g), power and heating generation (62 ± 9 g), and production of mineral products (47 ± 8 g). The rest of sources contributed approximately 3% to the total 2,3,7,8-TCDD emission in 2009. Iron and steel industries are mainly located in Beijing–Tianjin–Hebei (BTH) and Yangtze River Delta (YRD) regions, whereas waste incinerators are mainly located in Pearl River Delta (PRD) region. Higher 2,3,7,8-TCDD emissions were found in these three regions. While the BTH, YRD, and PRD accounted for only about 4% of total land area of China, they contributed approximately 14%, 15%, and 5% to the total 2,3,7,8-TCDD emission in 2009 in China, respectively.
Keywords: Dioxin; 2,3,7,8-TCDD; Gridded emission inventory; China;

Investigation of the tracers for plastic-enriched waste burning aerosols by Sudhanshu Kumar; Shankar G. Aggarwal; Prabhat K. Gupta; Kimitaka Kawamura (49-58).
To better identify the tracers for open-waste burning (OWB) aerosols, we have conducted aerosol sampling at 2 landfill sites, i.e., Okhla and Bhalswa in New Delhi. The metals such as, As, Cd, Sb and Sn, which have been observed almost negligible in remote aerosols, are found abundantly in these OWB aerosol samples (n = 26), i.e., 60 ± 65, 41 ± 53, 537 ± 847 and 1325 ± 1218 ng m−3, respectively. Samples (n = 20) collected at urban locations in New Delhi, i.e., at Employees' State Insurance (ESI) hospital and National Physical Laboratory (NPL) also show high abundances of these metals in the particles. Filter samples are also analyzed for water-soluble dicarboxylic acids (C2–C12) and related compounds (oxocarboxylic acids and α-dicarbonyls). Terephthalic acid (tPh) was found to account for more than 77% of total diacids determined in OWB aerosols. However, such a high abundance of tPh is not observed in aerosols collected at urban sites. Instead, phthalic acid (Ph) was found as the third/fourth most abundant diacid (∼3%) following C2 (>70%) and C4 (>12%) in these waste burning influenced urban aerosols. A possible secondary formation pathway of Ph by photo-degradation of phthalate ester (di-2-ethylhexyl phthalate) in plastic-waste burning aerosol is suggested. Ionic composition of OWB aerosols showed that Cl is the most abundant ion (40 ± 8% of total ions determined). The correlation studies of the potential metals with the organic tracers of garbage burning, i.e., phthalic, isophthalic and terephthalic acids show that especially Sn can be used as marker for tracing the plastic-enriched waste burning aerosols.
Keywords: Plastic waste burning aerosol; Chemical composition; Tracer; Metals; Diacids; Phthalic acid formation pathway;

The open top chamber (OTC) method was used in combination with real-time quantitative PCR and terminal restriction fragment length polymorphism (T-RFLP) techniques in the wheat field to study the influence of different levels of O3 concentrations (ambient air filtered by activated carbons, 40 ppb, 80 ppb and 120 ppb) on the quantity and community structure of methanotrophic bacteria. O3 stress can influence the potential methane oxidation rate (PMOR) and potential methane production rate (PMPR) in the farmland soil. O3 treatment of 40 ppb improved significantly the 16S rRNA gene copy number in the total methanotrophic bacteria pmoA, and type I and type II methanotrophic bacteria in the soil depth of 0–20 cm. When the O3 concentration reached 120 ppb, the 16S rRNA gene copy number in the total methanotrophic bacteria pmoA and type I methanotrophic bacteria decreased significantly as compared to the control treatment in 10–20 cm layer. The 16s rRNA gene copy number of total methanotrophic bacteria pmoA and type I and type II methanotrophic bacteria were influenced by different O3 concentration and soil depth. The T-RFLP analysis indicated that O3 stress influenced significantly the community structure of the methanotrophic bacteria in soil, causing potential threat to the diversity of methanotrophic bacteria. It seems to imply that the rise of O3 concentration could produce an impact on the carbon cycling and the methane emission of the wheat field soil by changing the community structure and diversity of methanotrophic bacteria, which then influences the global climate change.
Keywords: Community; Methanotrophic bacteria; Ozone; Real-time quantitative PCR; Soil; T-RLFP; Winter wheat;

Evolution of surface O3 and PM2.5 concentrations and their relationships with meteorological conditions over the last decade in Beijing by Ziyin Zhang; Xiaoling Zhang; Daoyi Gong; Weijun Quan; Xiujuan Zhao; Zhiqiang Ma; Seong-Joong Kim (67-75).
In this study, hourly and daily records since 2005 and correlation, regression and composite methods were used to analyze the long-term evolution of surface O3 and PM2.5 concentrations at the rural station of Shangdianzi (SDZ) and urban station of Baolian (BL) over Beijing and their relationships with meteorological conditions. The results show that the mean concentrations of PM2.5 (O3) decreased (increased) at the urban and rural stations over the last decade. The linear trends of the annual mean concentrations of PM2.5 at BL and SDZ were −31.8 ug/m3/10yr (−4.3%/yr) (p < 0.01) and −13.3 ug/m3/10yr (−2.9%/yr) (p < 0.05), respectively. In winter, the mean wind speed (W s) and relative humidity (RH) were the most closely correlated with O3 at both stations, whereas RH and sunshine hours (S) were most closely correlated with PM2.5. The correlation coefficients and explained variances in spring and autumn were generally less than those in winter and greater than those in summer. Moreover, increase in precipitation can significantly reduce the PM2.5 concentration in both urban and rural areas in Beijing, whereas trace and light precipitation more effectively decreases the O3 concentration. Concentrations of PM2.5 (O3) on haze days increased by 114% (3%) and 162% (20%) compared with that on non-haze days at the urban and rural stations, respectively. This result suggests that haze is a major manifestation of air pollution in Beijing.
Keywords: PM2.5; O3; Meteorological factors; Haze days; Beijing;

A new post-processing method for surface particulate matter (PM2.5) predictions from the National Oceanic and Atmospheric Administration (NOAA) developmental air quality forecasting system using the Community Multiscale Air Quality (CMAQ) model is described. It includes three main components:• A real-time quality control procedure for surface PM2.5 observations;• Model post-processing at each observational site using historical forecast analogs and Kalman filtering;• Spreading the forecast corrections from the observation locations to the entire gridded domain.The methodology is tested using 12 months of CMAQ forecasts of hourly PM2.5, from December 01, 2009 through November 30, 2010. The model domain covers the contiguous USA, and model data are verified against U.S. Environmental Prediction Agency AIRNow PM2.5 observations measured at 716 stations over the CMAQ domain. The model bias is found to have a strong seasonal dependency, with a large positive bias in winter and a small bias in the summer months, and also to have a strong diurnal cycle.Five different post-processing techniques are compared, including a seven-day running mean subtraction, Kalman-filtering, analogs, and combinations of analogs and Kalman filtering. The most accurate PM2.5 forecasts have been found to be produced when using historical analogs of the hourly Kalman-filtered forecasts, referred to as KFAN. The choice of meteorological variables used in the hourly analog search is also found to have a significant effect. A monthly error analysis is computed, in each case using the remaining 11 months of the data set for the analog searches. The improvement of KFAN errors over the raw CMAQ model errors ranges from 50 to 75% for MAE and from 40 to 60% for the correlation coefficient. Since the post-processing analysis is only done at the locations where observations are available, the spreading of post-processing correction information over nearby model grid points is necessary to make forecast contour maps. This spreading of information is accomplished with an eight-pass Barnes-type iterative objective analysis scheme. The final corrected CMAQ forecast over the entire domain is composed of the sum of the original CMAQ forecasts and the KFAN bias information interpolated over the entire domain, and is applied on an hourly basis.
Keywords: Analog forecast; CMAQ; Kalman-filtering; PM2.5;

On-road emissions of carbonyls from vehicles powered by biofuel blends in traffic tunnels in the Metropolitan Area of Sao Paulo, Brazil by Thiago Nogueira; Kely Ferreira de Souza; Adalgiza Fornaro; Maria de Fatima Andrade; Lilian Rothschild Franco de Carvalho (88-97).
On-road emissions of carbonyls from the current vehicle fleet of Brazil were determined in two experimental campaigns, conducted in traffic tunnels located in the Metropolitan Area of São Paulo (MASP), in southeastern Brazil. Among carbonyl species, formaldehyde and acetaldehyde were the most abundant in all sampling periods. In Brazil, heavy-duty vehicles (HDVs) run on a blend of 95% regular diesel/5% biodiesel from soy, whereas light-duty vehicles (LDVs) run on gasohol (75–80% gasoline/20–25% ethanol) or hydrous ethanol. We found that HDVs showed the highest overall carbonyl emissions, although LDVs were responsible for high emissions of acetaldehyde. In comparison with LDVs in California, which are powered by 90% gasoline/10% ethanol, LDVs in Brazil were found to emit 352% and 263% more formaldehyde and acetaldehyde.Display Omitted
Keywords: Air pollution; Aldehyde; Emission factor; Ethanol biofuel; Megacity; Vehicle emissions;

Because recent laboratory testing indicates that some low vapor pressure-volatile organic compounds (LVP-VOC) solvents readily evaporate at ambient conditions, LVP-VOCs used in some consumer product formulations may contribute to ozone formation. The goal of this study is to determine the fraction of LVP-VOCs available for ozone formation from the use of consumer products for two hypothetical emissions. This study calculates and compares the fraction of consumed product available for ozone formation as a result of (a) volatilization to air during use and (b) down-the-drain disposal. The study also investigates the impact of different modes of releases on the overall fraction available in ambient air for ozone formation. For the portion of the LVP-VOCs volatilized to air during use, we applied a multi-compartment mass-balance model to track the fate of emitted LVP-VOCs in a multimedia urban environment. For the portion of the LVP-VOCs disposed down the drain, we used a wastewater treatment plant (WWTP) fate model to predict the emission rates of LVP-VOCs to ambient air at WWTPs or at the discharge zone of the facilities and then used these results as emissions in the multimedia urban environment model. In a WWTP, the LVP-VOCs selected in this study are primarily either biodegraded or removed via sorption to sludge depending on the magnitude of the biodegradation half-life and the octanol-water partition coefficient. Less than 0.2% of the LVP-VOCs disposed down the drain are available for ozone formation. In contrast, when the LVP-VOC in a consumer product is volatilized from the surface to which it has been applied, greater than 90% is available for photochemical reactions either at the source location or in the downwind areas. Comparing results from these two modes of releases allows us to understand the importance of determining the fraction of LVP-VOCs volatilized versus disposed down the drain when the product is used by consumers. The results from this study provide important information and modeling tools to evaluate the impact of LVP-VOCs on air quality and suggest the need for future research on emissions of LVP-VOCs at the point of use.
Keywords: Consumer products; Environmental modeling; Low vapor pressure-volatile organic compounds; Ozone; Wastewater treatment plant;

We evaluated the fuel-cycle emissions of greenhouse gases (GHGs) and air pollutants (NOx, SO2, PM10, and PM2.5) of electric vehicles (EVs) in China and the United States (U.S.), two of the largest potential markets for EVs in the world. Six of the most economically developed and populated regions in China and the U.S. were selected. The results showed that EV fuel-cycle emissions depend substantially on the carbon intensity and cleanness of the electricity mix, and vary significantly across the regions studied. In those regions with a low share of coal-based electricity (e.g., California), EVs can reduce GHG and air pollutant emissions (except for PM) significantly compared with conventional vehicles. However, in the Chinese regions and selected U.S. Midwestern states where coal dominates in the generation mix, EVs can reduce GHG emissions but increase the total and urban emissions of air pollutants. In 2025, EVs will offer greater reductions in GHG and air pollutant emissions because emissions from power plants will be better controlled; EVs in the Chinese regions examined, however, may still increase SO2 and PM emissions. Reductions of 60–85% in GHGs and air pollutants could be achieved were EVs charged with 80% renewable electricity or the electricity generated from the best available technologies of coal-fired power plants, which are futuristic power generation scenarios.Display Omitted
Keywords: Electric vehicles; Generation mix; Renewable electricity; Fuel economy; Life-cycle analysis;

Monsoon onset signal in the stable oxygen and hydrogen isotope ratios of monsoon vapor by Rohit Srivastava; R. Ramesh; Naveen Gandhi; R.A. Jani; Ashutosh K. Singh (117-124).
Whereas data on deuterium (D) and oxygen (18O), effective tracers of hydrological processes, are available for global precipitation, such data on atmospheric water vapor are limited. With the advent of satellites capable of measuring D in atmospheric water vapor, mesoscale moisture transport processes (e.g. monsoons) could be detected early provided the signal is well above the noise. Our results of daily measurements of stable oxygen and hydrogen isotopic compositions of atmospheric water vapor (δ18Ovapor, δDvapor) and rain (δ18Orain, δDrain) during 2007–2008 CE over Ahmedabad, India, show that this is indeed so: (i) the onset of monsoon is marked by a dramatic decrease of ∼2.9‰ and ∼60‰, in δ18Ovapor, δDvapor, respectively, and 46‰ in deuterium excess (δD – 8. δ18O) of local atmospheric water vapor (ii) δ18Ovapor, δDvapor, δ18Orain and δDrain exhibit correlated variations during the monsoon season (mid-June to mid-September): the daily variations of δ18Ovapor and δDvapor are higher during large rain events, with sharp negative excursions from mean values. Deuterium excess of vapor and rain are indistinguishable from each other, within their natural variability, consistent with the achievement of isotopic equilibrium between them.
Keywords: Monsoon; Stable isotopes; Isotopic fractionation; Atmospheric water vapor;

Biomass fuel smoke particles (BFSPs) of rural kitchens collected during dry and wet seasons were characterized for elements, anions and carbon. The BFSPs of kitchens using varied biomass fuel types viz. cow dung stick, mixed biomass, cow-dung stick-mixed biomass and sugarcane bagasse were chosen for the study. The BFSPs from cow dung fuel stick showed higher levels of elements, anions and particulate carbon than other BFSPs. Calcium, K, Fe and Mg were the major elements found in all BFSPs, which did not vary much between the seasons. Sulphate was found to be the dominant anion present in all BFSPs followed by Clˉ and PO4 3−. Seasonal variation was pronounced in the case of abundance of anions and particulate carbon. The ratio OC/EC, often used as source signature of biomass burning, was found to be within 1.89–7.41 and 1.72–6.19 during dry and wet seasons respectively.Display Omitted
Keywords: Biomass burning; Rural kitchen; Elements; Ions; Carbon;

An experimental investigation into the atmospheric degradation of piperazine by Stephen White; Dennys Angove; Merched Azzi; Anne Tibbett; Ian Campbell; Michael Patterson (133-139).
The atmospheric degradation of piperazine was investigated using an indoor smog chamber. Experiments were carried out in the presence of nitrogen oxides (NOx), ozone or nitric acid. Piperazine reacted rapidly under all evaluated conditions: irradiated in the presence of NOx and with ozone and nitric acid in the dark. Gas phase products from the oxidation of piperazine were identified by infrared spectroscopy, DNPH cartridges followed by HPLC analysis, and by sampling chamber gas through Tenax sorbent material followed by analysis using thermal desorption GC-ITMS (gas chromatography ion trap mass spectrometry).Eight compounds were positively identified, with a further nine compounds tentatively identified using GC–MS based on molecular weight and mass spectra. Ammonia formation was observed from piperazine oxidation, and its formation was from the subsequent reactions of photooxidation products of piperazine rather than directly from the reaction of piperazine. The nitrosamine and nitramine expected from piperazine, N-nitrosopiperazine, and N-nitropiperazine, were both identified and confirmed using 15NO, with a tentative maximum yield of nitrosamine of less than 5% observed.Aerosol yields, relative to total piperazine reacted not including that which absorbed to the walls, were considerably high but were not able to be quantified absolutely due to unusual behaviour of the scanning mobility particle sizer instrument to aerosol containing amines. The reaction of piperazine with gas phase nitric acid gave rise to immediate formation of aerosol.
Keywords: Piperazine; PCC emissions; Nitrosamine; Nitramine;

Sources and temporal variations of constrained PMF factors obtained from multiple-year receptor modeling of ambient PM2.5 data from five speciation sites in Ontario, Canada by Uwayemi M. Sofowote; Yushan Su; Ewa Dabek-Zlotorzynska; Ankit K. Rastogi; Jeff Brook; Philip K. Hopke (140-150).
The Canadian National Air Pollution Surveillance (NAPS) network operates five fine particulate matter (PM2.5) speciation sites in Ontario. Data from 2005 to 2010 from these sites were subjected to constrained positive matrix factorization (PMF) to obtain factors of particulate matter pollution. Eight factors were found to be common across the sites. These constrained factors were particulate sulfate and nitrate, aged combustion aerosols, biomass/wood combustion, non-ferrous metal smelting, crustal matter, steel works, and road salt. Other factors found at these sites were largely related to local sources in their respective neighbourhoods. Diagnostic results from constraining the PMF factors could be used to select the appropriate post-PMF analyses as they reveal the factors that were expected to have regional or local sources. Interpretation of the spatial trends of factors was guided by the extent of similarity of target ratios across the five sites as determined by their coefficient of variation (C.V.). On average, particulate sulfate had PM2.5 mass contributions that ranged from 7% (Wallaceburg) to 36% (Simcoe) and nitrate species ranged from 7% (Simcoe) to 23% (Windsor). A clear low rural – high urban divide for particulate nitrate was also observed. These non-uniform impacts across the sites indicate that a single representative location is not suitable for studying PM2.5 in Ontario. Temporal variations showed greater abundance of particulate sulfate in the summer and particulate nitrate in the winter respectively. Yearly median values of the sulfate factor that is largely energy-related have been declining for most sites since peaking in 2007. A trajectory ensemble model, simplified quantitative transport bias analysis (sQTBA) was used to identify potential source regions of the constrained sulfate and nitrate factors while radial plots of the conditional probability function (CPF) were used to explore the more locally-sourced factors. The US Midwest, east coast and Canada's west were dominant regional sources.
Keywords: PM2.5; PMF; Receptor modeling; Speciation; Source apportionment;

Constraining the factor analytical solutions obtained from multiple-year receptor modeling of ambient PM2.5 data from five speciation sites in Ontario, Canada by Uwayemi M. Sofowote; Yushan Su; Ewa Dabek-Zlotorzynska; Ankit K. Rastogi; Jeff Brook; Philip K. Hopke (151-157).
Rotational ambiguity in factor analyses leads to solutions that are not always consistent with reality. The inherent non-negativity constraints in positive matrix factorization (PMF) help to prevent factor solutions from becoming overly unrealistic, but they are not sufficient to prevent unwanted rotations that could manifest in factors that should have similar compositions varying across multiple sites. The Canadian National Air Pollution Surveillance (NAPS) network operates five fine particulate matter (PM2.5) speciation sites in Ontario. Data from these sites from 2005 to 2010 were subjected to PMF to obtain factors representing sources of particulate matter. Eight factors were found to be common across these sites. These factors had profiles that varied greatly from one site to the other, suggesting that the PMF solutions were impacted by some rotational ambiguity. New features in the EPA PMF V5 program allow the use of a priori information to impose mathematical constraints that guide the evolution of the factor solutions. These constraints reduce the rotational space. In situations where major emissions sources are known and located in the neighborhood of receptors, or emissions inventories and literature source profiles exist, it is easy to use these profiles to force the factor solutions to conform to the expected signatures. In our case, reported source profiles were neither available nor applicable due to the large spatial span of potential sources and receptor sites. This work describes how such constraints can be generated and used in these complex situations. The fundamental principle explored in this work is the concept of ‘stiffness’ of PMF solutions to identify the desirable non-rotating factors.
Keywords: PM2.5; PMF; Receptor modeling; Speciation; Source apportionment;