A22B-01
On the Interpretation of Oxygenated Organic Aerosols (and their Subtypes) Arising from Factor Analysis of Aerosol Mass Spectrometer Data
Zhang et al. (ES&T 2005; ACP 2005) first performed factor analysis (FA) of Aerodyne Aerosol Mass
Spectrometer (AMS) complete organic aerosol (OA) mass spectra. This study showed that an oxygenated
organic aerosol (OOA) factor accounted for 2/3 of the OA mass at an urban site in Pittsburgh and strongly
linked OOA to secondary organic aerosols (SOA). Many subsequent studies and the application of more
powerful solution algorithms such as Positive Matrix Factorization (PMF) to the same FA problem have
demonstrated the importance of OOA at most locations (e.g. Volkamer et al., GRL, 2006; Zhang et al., GRL,
2007; Lanz et al., ACP, 2007 and ES&T, 2008; Ulbrich et al., ACPD, 2008). Multiple studies have also
identified several subtypes of OOA (e.g. OOA-1 and OOA-2). This type of analysis offers new insights
because it provides some chemical resolution on the total OA mass with high time and size resolution, and
bypasses the limitations of techniques that only analyze tracers and which may favor more reduced species.
However the chemical resolution is limited and careful interpretation of the FA output is required, including
the use of database spectra, time series of external tracers, tracer ratios, back-trajectory analyses, size-
distribution analyses, etc. This presentation will address the interpretation of total OOA and its subfactors
across a large range of locations in urban, suburban, rural, remote, and forested areas, and will compare
with the results of other source apportionment techniques. Based on data from multiple datasets we conclude
that (1) anthropogenic SOA in and downwind of urban areas is an important source of OOA; (2) motor
vehicles, meat cooking, and trash burning are unlikely to be sources of primary OOA; (3) SOA from biogenic
and biomass burning precursors are also clear sources of OOA; (4) primary biomass burning OA (P-BBOA)
typically shows significant differences with ambient OOA factors; (5) heterogeneous oxidation of urban POA
may give rise to OOA although it can be at most a small contributor to ambient concentrations due to its
limited mass emission in industrialized countries. Subtypes of OOA in multiple ambient datasets are shown to
arise from (a) continued oxidative processing of emissions with increasing photochemical age; (b) partitioning
of semivolatile SOA species under conditions of lower temperature and increased humidity; and (c) different
source regions affecting a receptor site. The interpretation and limitations of FA techniques, which currently
need to assume a constant mass spectrum for each component, for cases in which the real spectra are
evolving due to atmospheric aging will be discussed.
http://cires.colorado.edu/jimenez/ams.html
A22B-02
To What Extent Can Biogenic SOA be Controlled?
Anthropogenic pollution facilitates conversion of naturally emitted volatile organic compounds (VOCs) to the particle phase, enhancing biogenic secondary organic aerosol (SOA) in the atmosphere. 'Biogenic' SOA is a product of natural precursor VOCs as well as anthropogenic constituents. Anthropogenic pollution increases concentrations of oxidants (O3, OH, NO3) that convert biogenic VOCs to semi-volatile, condensable, SOA-contributing species. Furthermore, increases in concentrations of primary organic matter facilitate condensation of semi-volatile biogenic species to the particle phase. Therefore, a portion of biogenic SOA can be attributed to anthropogenic emissions and may be controllable. Direct measurement of this component is not possible, but it can be estimated through modeling. To test the contribution of anthropogenic pollution to biogenic SOA, CMAQ model simulations were conducted for the continental U.S. (August 18 –- September 4) using a recently developed SOA module that includes several biogenic SOA precursors: isoprene, sesquiterpenes, monoterpenes. Biogenic SOA formation pathways include oxidation of volatile species to semi-voltaile products followed by gas-to-particle partitioning, oligomerization and in-cloud processes. CMAQ simulations were conducted with and without anthropogenic emissions. The relative contributions of individual pollution classes (i.e., NOx, VOCs, NH3, SO2 and primary organic particulate matter) were also evaluated by removing the individual components from the emissions, one at a time. Model results demonstrate a strong influence of anthropogenic pollution on predicted 'biogenic' SOA concentrations. In these simulations, approximately 40% of biogenic SOA in the eastern U.S. can be controlled. Total biogenic SOA was most sensitive to primary organic particulate matter emissions, suggesting reductions in PM emissions could reduce biogenic SOA by ~20% . Individual biogenic SOA species exhibit different responses to anthropogenic emission reductions. Cloud- produced SOA, which has biogenic and anthropogenic VOC precursors was most sensitive to NOx reductions.
A22B-03
Organic Aerosol Composition as Measured by Complementary In-Situ Techniques
Multiple in-situ techniques have been developed over the past several years to investigate the chemical composition of organic matter found in atmospheric particles. Organic matter typically comprises, by mass, between 20% to 80% of PM2.5 (particulate matter with diameters < 2.5 um). The chemical makeup of the organic fraction can aid in determining the origins and transformations of atmospheric particles that alter the Earth's water and energy cycle and are detrimental to human health. Three novel in-situ instruments were deployed during the Study of Organic Aerosol in Riverside (SOAR) campaign in southern California during the summer of 2005, the Aerodyne high-resolution aerosol mass spectrometer (AMS), the aerosol time-of-flight mass spectrometer (ATOFMS), and the thermal desorption aerosol gas chromatograph (TAG). The AMS system scans the complete electron ionization (EI) mass spectrum of atmospheric aerosols and is capable of determining total PM1 organic aerosol mass. Organic components can be extracted by analyzing the entire high-resolution AMS organic spectrum with factor analysis techniques such as Positive Matrix Factorization (PMF). The ATOFMS acquires the complete mass spectrum of single particles using laser ablation. The TAG instrument, using GC separation prior to EI-MS analysis, is capable of measuring individual organic compounds in the particle phase with 1-hour time resolution. Here we compare the temporal variation of common mass fragments as measured by the two bulk-sampling aerosol instruments (i.e., AMS and TAG), as well as compare organic aerosol sources/components from PMF analyses of TAG compounds and AMS spectra. The single-particle ATOFMS data is analyzed through an ART-2a clustering algorithm, and its results will also be compared to the PMF results. Finally, we explore the possibility of combining these measurement and analytical techniques to better define the sources and atmospheric transformation of organic particles.
A22B-04
Origins and Composition of Fine Atmospheric Carbonaceous Aerosol in the Sierra Nevada Mountains, California
We present hourly in-situ measurements of organic speciation of fine atmospheric aerosol (PM2.5) collected above a ponderosa pine plantation in the Sierra Nevada Mountains, California. The measurements were obtained using thermal desorption aerosol GC-MS (TAG) as part of the Biosphere Effects on AeRosols and Photochemistry EXperiment (BEARPEX) in late summer 2007. The early part of the campaign was characterized by hot and dry conditions and frequent influence from forest fire events. The later part was generally cooler and more humid with sporadic precipitation, which lead to significant increases in biogenic emissions. The hourly time resolution of TAG provides far greater information on diurnal variations than more traditional filter samples. We observe higher concentrations of anthropogenic marker compounds (e.g., phthalic and benzoic acids) in the early afternoon with the arrival of the pollution plume from Sacramento. Higher concentrations of biogenic oxidation marker compounds (e.g., pinonaldehyde and nopinone) are observed at night, suggesting local oxidation of primary biogenic emissions in the shallow nocturnal boundary layer. Observations via electrospray ionization high-resolution time-of-flight mass spectrometry (ESI-HR-TOFMS) of several nitrooxy-organosulfate products from terpene oxidation provides evidence for nighttime oxidation by nitrate radicals within the forest canopy. Radiocarbon (carbon-14) analysis at the site shows that >70% of the carbon present in the PM2.5 aerosol is modern indicating the importance of biogenic sources on aerosol composition at this site. We use the TAG data in conjunction with gas phase volatile organic carbon (VOC) and aerosol mass spectrometer (AMS) measurements to assign and investigate the relative contributions of different source categories (e.g., biomass burning, primary biogenic emissions, secondary biogenic oxidation, etc.) to the total observed organic aerosol mass during the BEARPEX campaign.
A22B-05
Evaluating Simulations of Primary Anthropogenic and Biomass Burning Organic Aerosols using Aerosol Mass Spectrometer Data and Positive Matrix Factorization Analysis
Most model predictions of organic matter are currently underestimated because the processes contributing to secondary organic aerosol (SOA) formation and transformation are not well understood. Since research associated with developing a better framework to improve the representation of specific gas-to-particle partitioning processes controlling SOA based on new measurements and theoretical relationships is on- going, this study seeks to determine whether 3-D models can adequately predict concentrations of primary organic aerosols (POA). If one assumes POA is non-volatile, then errors in POA predictions will results from uncertainties in the emission inventories and errors in transport and mixing processes. The WRF-chem model is used to predict POA in the vicinity of Mexico City during the 2006 MILAGRO field campaign. Particulate matter emission rates were obtained from urban and regional Mexican emission inventories and from biomass burning estimates derived from MODIS "hotspot" and vegetation databases. Organic aerosol predictions are evaluated using data from Aerodyne Aerosol Mass Spectrometer (AMS) instruments deployed at four ground sites and on two research aircraft and from Sunset Laboratory OCEC instruments deployed at two ground sites. Positive Matrix Factorization (PMF) has recently been applied to derive components of organic aerosols including: hydrocarbon-like organic aerosol (HOA), oxidized organic aerosol (OOA), and biomass burning organic aerosols (BBOA). The temporal variation of HOA is often similar to primary emissions of other species in urban areas. PMF analysis is currently available for three of the ground sites and for some of the aircraft flights. We found that the predicted POA was consistently lower than the measured organic matter at the ground sites, which is consistent with the expectation that SOA should be a large fraction of the total organic aerosol mass. A much better agreement was found when predicted POA was compared with HOA+BBOA, suggesting that the emission rates were reasonable overall. A similar finding was found using the AMS instruments on the aircraft on days with relatively low biomass burning. On days with a significant number of fires, the predicted POA was greater than the total observed organic matter as the aircraft flew directly downwind of the biomass burning sources, suggesting that biomass burning emissions were too high or that there were errors in way the model treated plume rise or horizontal mixing of point sources. We will discuss how uncertainties in the emissions contribute to differences between the observed and simulated organic aerosol concentration at each of the sites.
A22B-06
Evaluation of New Secondary Organic Aerosol Models for a Case Study in Mexico City
Recent field studies have found large discrepancies in the measured vs. modeled SOA mass loadings in both
urban and regional polluted atmospheres. The reasons for these large differences are unclear. Here we
revisit a case study of SOA formation in Mexico City described by Volkamer et al. (2006) and show that the
increase in OA/CO during photochemistry is consistent with results from several groups during MILAGRO,
during a period when the impact of regional biomass burning is minor or negligible. Then we use the case
study to evaluate three new SOA models: 1) the update of aromatic SOA yields from recent chamber
experiments (Ng et al., 2007); 2) the formation of SOA from glyoxal (Volkamer et al., 2007); 3) and the
formation of SOA from primary semivolatile and intermediate volatility species (P-S/IVOC) (Robinson et al.,
2007). We also evaluate the effect of reduced partitioning of SOA into POA (Song et al., 2007). Traditional
SOA precursors (mainly aromatics) by themselves still fail to produce enough SOA to match the observations
by a factor of ~7. The new low-NOx aromatic pathways with very high SOA yields make a negligible
contribution in this urban environment as the RO2 + NO reaction dominates the fate of the RO2
radicals. Glyoxal contributes several μg m-3 to SOA formation, with similar timing as the
measurements. P-S/IVOC concentrations are estimated as in equilibrium with measured POA, and these
species introduce a large amount of carbon that was not in models before. With the formulation in Robinson
et al. (2007) these species have a high SOA yield, and this mechanism can close the gap in SOA mass
between measurements and models in our case study. However, this model has many parameters that are
not well constrained and much experimental work is needed for its realistic assessment. The volatility of
model SOA is too high when compared to the measurements, while the O/C of all model SOA is somewhat
lower than the observations. The effects of dilution and aging are predicted to be strong and counteract each
other as the urban air is advected and aged downwind from the city. Finally, the model is not very sensitive
to the assumptions of partitioning phases/activities under the conditions of this case study.
Ng, N.L., et al.: Secondary organic aerosol formation from m-xylene, toluene, and benzene, Atmos. Chem.
Phys., 7, 3909-3922, 2007.
Robinson, A.L., et al.: Rethinking Organic Aerosols: Semivolatile Emissions and Photochemical Aging,
Science, 315, 1259-1262, 2007.
Song, C., et al.: Effect of hydrophobic primary organic aerosols on secondary organic aerosol formation from
ozonolysis of α-pinene, Geophys. Res. Letters, 34, L20803, doi:10.1029/2007GL030720, 2007.
Volkamer, R., et al.: Secondary organic aerosol formation from anthropogenic air pollution: Rapid and higher
than expected, Geophys. Res. Lett., 33, L17811,doi: 10.1029/2006GL026899, 2006.
Volkamer, R., et al.: A missing sink for gas-phase glyoxal in Mexico City: Formation of secondary organic
aerosol, Geophys. Res. Lett., 34, L19807,doi:10.1029/2007GL030752, 2007.
A22B-07
The Effect of Temperature and Oxidation on the Relative Importance of Primary and Secondary Organic Aerosol in Aged Urban Air Masses
Recent laboratory experiments have suggested that unidentified semi-volatile organic compounds (SVOCs) from urban vehicle gas/particle emissions may contribute to the aerosol particle phase. These SVOCs and potentially others likely possess vapour pressures in a range whereby dilution downwind of sources and/or temperature changes may cause these compounds to rapidly condense on or evaporate from existing OA. This process can potentially affect the relative proportions of SOA and POA to total OA, while altering the aerosol carbon and oxygen content. However, few ambient studies have been able to quantify the condensable SVOC aerosol fraction. The current study presents results from a recent field campaign where data on aerosol organic carbon and oxygen from exact mass measurements with a HR-ToF-AMS are combined with VOC ratios to investigate the evolution of aerosols downwind of an urban region. The ratio of total carbon to total oxygen (C/O) in the aerosols, and the elemental composition of the OA are utilized to estimate the amount of SOA mass which can be added via photo-chemistry during transport from nearby urban areas. HR-ToF-AMS exact mass measurements are also used to provide evidence of temperature dependent condensation/evaporation of SVOCs and to determine the relative importance of SOA, POA and SVOC condensable material to the total OA at the site. Results indicate that oxygenated and non-oxygenated POA dominate the OA. However, non-volatile SOA and condensed SVOC material does comprise a significant portion of the total OA in urban influenced air masses at this site. The results imply that the temperature dependent partitioning of primary emitted semi-volatile gases should be included into regional/global models for an improved understanding of OA and the role of urban areas.
A22B-08
Sources and Transformations of Organic Aerosol Particles Measured With a new Single Particle Technique During the MILAGRO 2006 Field Campaign in Mexico City
In March 2006, the MILAGRO (Megacity Initiative Local and Global Research Observations) field campaign took place in and around the Mexico City Metropolitan Area (MCMA). The main objective of the campaign was to understand the chemical evolution and influence of the MCMA plume on a regional to global scale. For this purpose, an extensive network of stationary and mobile sampling platforms was used to monitor the meteorological conditions, gas phase composition, and aerosol particles at different locations within the MCMA. Single particle mass spectrometry has become an important part of the suite of instrumentation necessary to characterize the ambient aerosol. Single particle data reveal important details about ambient aerosol mixing state, sources, and rates of atmospheric transformations. Results from the initial field deployment of a new single particle technique - single particle thermal vaporization mass spectrometry will be presented. The instrument (known as a LS-C-ToF-AMS) combines a light scattering module with the existing time-of-flight version of the aerosol mass spectrometer (ToF-AMS) developed by Aerodyne Research, Inc. A unique feature of the LS-C-ToF-AMS is that it measures both single particle and ambient ensemble size and non-refractory chemical composition. The results were obtained at the T1 ground site located 40 km northeast of the city center. During 72 hours of continuous sampling, the LS-C-ToF-AMS measured 12,853 submicron sized single particles while simultaneously collecting standard ensemble mass loading, chemistry, and size distribution information. The single particles were characterized by their optical diameter, vacuum aerodynamic diameter, and chemical composition (NH4, NO3, SO4, Chl, HOA, and OOA) obtained from mass spectra using standard AMS analysis techniques. The single particle results indicate that most of the observed particles in the MCMA were internal mixtures of particulate material from multiple sources. With the exception of the sulfate and chloride contributions, the single particle mass and chemical composition was governed in large part by condensation of secondary photochemical nitrate and oxidized organic material. Primary emissions of HOA particles from local combustion sources and particles with high chloride mass fractions potentially from biomass burning were also identified. In the early morning, while the boundary layer was low and local traffic emissions were high, the single particle data showed that the ambient aerosol ensemble was an external mixture of HOA organic particles and more processed particles of mixed composition. During the day, due to photochemical processing, the composition of the single particles became dominated by more highly oxygenated organic components. Single particle results suggest that oxygenated organics condense onto the surfaces of all pre-existing particles during each afternoon. The transformations in the non-refractory organic composition of the single particles will be described in the context of the regional-scale meteorology, atmospheric processing, and different organic aerosol sources influencing the T1 site.