scholarly journals Constructing a data-driven receptor model for organic and inorganic aerosol – a synthesis analysis of eight mass spectrometric data sets from a boreal forest site

2018 ◽  
Author(s):  
Mikko Äijälä ◽  
Kaspar R. Daellenbach ◽  
Francesco Canonaco ◽  
Liine Heikkinen ◽  
Heikki Junninen ◽  
...  
Keyword(s):  
Boreal Forest ◽  
Organic Aerosol ◽  
Mass Spectrometric ◽  
Chemical Components ◽  
Forest Site ◽  
Aerosol Composition ◽  
Mass Spectrometric Data ◽  
Spectrometric Data ◽  
Aerosol Mass ◽  
Inorganic Aerosol

Abstract. The interactions between organic and inorganic aerosol chemical components are integral to understanding and modelling climate and health-relevant aerosol physicochemical properties, such as volatility, hygroscopicity, light scattering and toxicity. This study presents a synthesis analysis for eight data sets, of non-refractory aerosol composition, measured at a boreal forest site. The measurements, performed with an aerosol mass spectrometer, cover in total around 9 months over the course of 3 years. In our statistical analysis, we use the complete organic and inorganic unit-resolution mass spectra, as opposed to the more common approach of only including the organic fraction. The analysis is based on iterative, combined use of (1) data reduction, (2) classification and (3) scaling tools, producing a data-driven chemical mass balance type of model capable of describing site-specific aerosol composition. The receptor model we constructed was able to explain 83 ± 8 % of variation in data, increased to 96 ± 3 % when signals from low signal-to-noise variables were not considered. The resulting interpretation of an extensive set of aerosol mass spectrometric data infers seven distinct aerosol chemical components for a rural boreal forest site: ammonium sulphate (35 % of mass), low and semi-volatile oxidised organic aerosols (27 and 12 %), biomass burning organic aerosol (11 %), a nitrate containing organic aerosol type (7 %), ammonium nitrate (5 %), and hydrocarbon-like organic aerosol (3 %). Some of the additionally observed, rare outlier aerosol types likely emerge due to surface ionisation effects, and likely represent amine compounds from an unknown source and alkaline metals from emissions of a nearby district heating plant. Compared to traditional, simplistic inorganics apportionment methods for aerosol mass spectrometer data, our statistics-based method provides an improved, more robust approach, yielding readily useful information for the modelling of submicron atmospheric aerosols physical and chemical properties. The results also shed light on the division between organic and inorganic aerosol types and dynamics of salt formation in aerosol. Equally importantly, the combined methodology exemplifies an iterative analysis, using consequent analysis steps by a combination of statistical methods. Such an approach offers new ways to home in on physicochemically sensible solutions with minimal need for a priori information or analyst interference. We therefore suggest that similar statistics-based approaches offer significant potential for un/semi supervised machine-learning applications in future analyses of aerosol mass spectrometric data.

scholarly journals Constructing a data-driven receptor model for organic and inorganic aerosol – a synthesis analysis of eight mass spectrometric data sets from a boreal forest site

2019 ◽  
Vol 19 (6) ◽  
pp. 3645-3672 ◽  
Author(s):  
Mikko Äijälä ◽  
Kaspar R. Daellenbach ◽  
Francesco Canonaco ◽  
Liine Heikkinen ◽  
Heikki Junninen ◽  
...  
Keyword(s):  
Boreal Forest ◽  
Organic Aerosol ◽  
Mass Spectrometric ◽  
Chemical Components ◽  
Forest Site ◽  
Aerosol Composition ◽  
Mass Spectrometric Data ◽  
Spectrometric Data ◽  
Aerosol Mass ◽  
Inorganic Aerosol

Abstract. The interactions between organic and inorganic aerosol chemical components are integral to understanding and modelling climate and health-relevant aerosol physicochemical properties, such as volatility, hygroscopicity, light scattering and toxicity. This study presents a synthesis analysis for eight data sets, of non-refractory aerosol composition, measured at a boreal forest site. The measurements, performed with an aerosol mass spectrometer, cover in total around 9 months over the course of 3 years. In our statistical analysis, we use the complete organic and inorganic unit-resolution mass spectra, as opposed to the more common approach of only including the organic fraction. The analysis is based on iterative, combined use of (1) data reduction, (2) classification and (3) scaling tools, producing a data-driven chemical mass balance type of model capable of describing site-specific aerosol composition. The receptor model we constructed was able to explain 83±8 % of variation in data, which increased to 96±3 % when signals from low signal-to-noise variables were not considered. The resulting interpretation of an extensive set of aerosol mass spectrometric data infers seven distinct aerosol chemical components for a rural boreal forest site: ammonium sulfate (35±7 % of mass), low and semi-volatile oxidised organic aerosols (27±8 % and 12±7 %), biomass burning organic aerosol (11±7 %), a nitrate-containing organic aerosol type (7±2 %), ammonium nitrate (5±2 %), and hydrocarbon-like organic aerosol (3±1 %). Some of the additionally observed, rare outlier aerosol types likely emerge due to surface ionisation effects and likely represent amine compounds from an unknown source and alkaline metals from emissions of a nearby district heating plant. Compared to traditional, ion-balance-based inorganics apportionment schemes for aerosol mass spectrometer data, our statistics-based method provides an improved, more robust approach, yielding readily useful information for the modelling of submicron atmospheric aerosols physical and chemical properties. The results also shed light on the division between organic and inorganic aerosol types and dynamics of salt formation in aerosol. Equally importantly, the combined methodology exemplifies an iterative analysis, using consequent analysis steps by a combination of statistical methods. Such an approach offers new ways to home in on physicochemically sensible solutions with minimal need for a priori information or analyst interference. We therefore suggest that similar statistics-based approaches offer significant potential for un- or semi-supervised machine-learning applications in future analyses of aerosol mass spectrometric data.

scholarly journals Biogenic and biomass burning organic aerosol in a boreal forest at Hyytiälä, Finland, during HUMPPA-COPEC 2010

2013 ◽  
Vol 13 (24) ◽  
pp. 12233-12256 ◽  
Author(s):  
A. L. Corrigan ◽  
L. M. Russell ◽  
S. Takahama ◽  
M. Äijälä ◽  
M. Ehn ◽  
...  
Keyword(s):  
Factor Analysis ◽  
Boreal Forest ◽  
Biomass Burning ◽  
Sampling Site ◽  
Organic Aerosol ◽  
Ftir Spectra ◽  
Oxidation Products ◽  
Forest Site ◽  
Submicron Aerosol ◽  
Aerosol Mass

Abstract. Submicron aerosol particles were collected during July and August 2010 in Hyytiälä, Finland, to determine the composition and sources of aerosol at that boreal forest site. Submicron particles were collected on Teflon filters and analyzed by Fourier transform infrared (FTIR) spectroscopy for organic functional groups (OFGs). Positive matrix factorization (PMF) was applied to aerosol mass spectrometry (AMS) measurements and FTIR spectra to identify summertime sources of submicron aerosol mass at the sampling site. The two largest sources of organic mass (OM) in particles identified at Hyytiälä were (1) biogenic aerosol from surrounding local forest and (2) biomass burning aerosol, transported 4–5 days from large wildfires burning near Moscow, Russia, and northern Ukraine. The robustness of this apportionment is supported by the agreement of two independent analytical methods for organic measurements with three statistical techniques. FTIR factor analysis was more sensitive to the chemical differences between biogenic and biomass burning organic components, while AMS factor analysis had a higher time resolution that more clearly linked the temporal behavior of separate OM factors to that of different source tracers even though their fragment mass spectrum were similar. The greater chemical sensitivity of the FTIR is attributed to the nondestructive preparation and the functional group specificity of spectroscopy. The FTIR spectra show strong similarities among biogenic and biomass burning factors from different regions as well as with reference OM (namely olive tree burning organic aerosol and α-pinene chamber secondary organic aerosol (SOA)). The biogenic factor correlated strongly with temperature and oxidation products of biogenic volatile organic compounds (BVOCs), included more than half of the oxygenated OFGs (carbonyl groups at 29% and carboxylic acid groups at 22%), and represented 35% of the submicron OM. Compared to previous studies at Hyytiälä, the summertime biogenic OM is 1.5 to 3 times larger than springtime biogenic OM (0.64 μg m−3 and 0.4 μg m−3, measured in 2005 and 2007, respectively), even though it contributed only 35% of OM. The biomass burning factor contributed 25% of OM on average and up to 62% of OM during three periods of transported biomass burning emissions: 26–28 July, 29–30 July, and 8–9 August, with OFG consisting mostly of carbonyl (41%) and alcohol (25%) groups. The high summertime terrestrial biogenic OM (1.7 μg m−3) and the high biomass burning contributions (1.2 μg m−3) were likely due to the abnormally high temperatures that resulted in both stressed boreal forest conditions with high regional BVOC emissions and numerous wildfires in upwind regions.

scholarly journals Determination of the biogenic secondary organic aerosol fraction in the boreal forest by NMR spectroscopy

2012 ◽  
Vol 12 (2) ◽  
pp. 941-959 ◽  
Author(s):  
E. Finessi ◽  
S. Decesari ◽  
M. Paglione ◽  
L. Giulianelli ◽  
C. Carbone ◽  
...  
Keyword(s):  
Nmr Spectroscopy ◽  
Boreal Forest ◽  
Functional Group ◽  
Secondary Organic Aerosol ◽  
Organic Aerosol ◽  
Spectroscopic Techniques ◽  
Aerosol Composition ◽  
Air Masses ◽  
The North ◽  
Aerosol Mass

Abstract. The study investigates the sources of fine organic aerosol (OA) in the boreal forest, based on measurements including both filter sampling (PM1) and online methods and carried out during a one-month campaign held in Hyytiälä, Finland, in spring 2007. Two aerosol mass spectrometers (Q-AMS, ToF-AMS) were employed to measure on-line concentrations of major non-refractory aerosol species, while the water extracts of the filter samples were analyzed by nuclear magnetic resonance (NMR) spectroscopy for organic functional group characterization of the polar organic fraction of the aerosol. AMS and NMR spectra were processed separately by non-negative factorization algorithms, in order to apportion the main components underlying the submicrometer organic aerosol composition and depict them in terms of both mass fragmentation patterns and functional group compositions. The NMR results supported the AMS speciation of oxidized organic aerosol (OOA) into two main fractions, which could be generally labelled as more and less oxidized organics. The more oxidized component was characterized by a mass spectrum dominated by the m/z 44 peak, and in parallel by a NMR spectrum showing aromatic and aliphatic backbones highly substituted with oxygenated functional groups (carbonyls/carboxyls and hydroxyls). Such component, contributing on average 50% of the OA mass throughout the observing period, was associated with pollution outbreaks from the Central Europe. The less oxidized component was enhanced in concomitance with air masses originating from the North-to-West sector, in agreement with previous investigations conducted at this site. NMR factor analysis was able to separate two distinct components under the less oxidized fraction of OA. One of these NMR-factors was associated with the formation of terrestrial biogenic secondary organic aerosol (BSOA), based on the comparison with spectral profiles obtained from laboratory experiments of terpenes photo-oxidation. The second NMR factor associated with western air masses was linked to biogenic marine sources, and was enriched in low-molecular weight aliphatic amines. Such findings provide evidence of at least two independent sources originating biogenic organic aerosols in Hyytiälä by oxidation and condensation mechanisms: reactive terpenes emitted by the boreal forest and compounds of marine origin, with the latter relatively more important when predominantly polar air masses reach the site. This study is an example of how spectroscopic techniques, such as proton NMR, can add functional group specificity for certain chemical features (like aromatics) of OA with respect to AMS. They can therefore be profitably exploited to complement aerosol mass spectrometric measurements in organic source apportionment studies.

scholarly journals Determination of the biogenic secondary organic aerosol fraction in the boreal forest by AMS and NMR measurements

2011 ◽  
Vol 11 (8) ◽  
pp. 22619-22662 ◽  
Author(s):  
E. Finessi ◽  
S. Decesari ◽  
M. Paglione ◽  
L. Giulianelli ◽  
C. Carbone ◽  
...  
Keyword(s):  
Boreal Forest ◽  
Functional Group ◽  
Secondary Organic Aerosol ◽  
Organic Aerosol ◽  
Spectroscopic Techniques ◽  
Aerosol Composition ◽  
Air Masses ◽  
The North ◽  
Aerosol Mass ◽  
Fragmentation Patterns

Abstract. The study investigates the sources of fine organic aerosol (OA) in the boreal forest, based on measurements including both filter sampling (PM1) and online methods and carried out during a one-month campaign held in Hyytiälä, Finland, in spring 2007. Two aerosol mass spectrometers (Q-AMS, ToF-AMS) were employed to measure on-line air mass concentrations of major non-refractory aerosol species, while the water extracts of the filter samples were analyzed by nuclear magnetic resonance (NMR) spectroscopy for organic functional group characterization of the polar organic fraction of the aerosol. AMS and NMR spectra were processed separately by non-negative factorization algorithms, in order to apportion the main components underlying the submicrometer organic aerosol composition and depict them in terms of both mass fragmentation patterns and functional group compositions. The NMR results supported the AMS speciation of oxidized organic aerosol (OOA) into two main fractions, which could be generally labelled as more and less oxidized organics. The more oxidized component was characterized by a mass spectrum dominated by the m/z 44 peak, and in parallel by a NMR spectrum showing aromatic and aliphatic backbones highly substituted with oxygenated functional groups (carbonyls/carboxyls and hydroxyls). Such component, contributing on average 50 % of the OA mass throughout the observing period, was associated with pollution outbreaks from the Central Europe. The less oxidized component showed features consistent with less oxygenated aerosols and was enhanced in concomitance with air masses originating from the North-to-West sector, in agreement with previous investigations conducted at this site. NMR factor analysis was able to separate two distinct components under the less oxidized fraction of OA. One of these NMR-factors was associated to the formation of terrestrial biogenic secondary organic aerosol (BSOA), based on the comparison with spectral profiles obtained from laboratory experiments of terpenes photo-oxidation. The second NMR factor associated with western air masses was linked to biogenic marine sources, and was enriched in low-molecular weight aliphatic amines. Such findings provide evidence of at least two independent sources \\mbox{originating} biogenic organic aerosols in Hyytiälä by oxidation and condensation mechanisms: reactive terpenes emitted by the boreal forest and compounds of marine origin, with the latter relatively more important when predominantly polar air masses reach the site. This study is an example of how spectroscopic techniques, such as proton NMR, can add functional group specificity for certain chemical features (like aromatics) of OA with respect to AMS. They can therefore be profitably be exploited to complement aerosol mass spectrometric measurements in organic source apportionment studies.

scholarly journals Biogenic and biomass burning organic aerosol in a boreal forest at Hyytiälä, Finland, during HUMPPA-COPEC 2010

2013 ◽  
Vol 13 (6) ◽  
pp. 16151-16210 ◽  
Author(s):  
A. L. Corrigan ◽  
L. M. Russell ◽  
S. Takahama ◽  
M. Äijälä ◽  
M. Ehn ◽  
...  
Keyword(s):  
Factor Analysis ◽  
Boreal Forest ◽  
Biomass Burning ◽  
Sampling Site ◽  
Organic Aerosol ◽  
Ftir Spectra ◽  
Oxidation Products ◽  
Forest Site ◽  
Submicron Aerosol ◽  
Aerosol Mass

Abstract. Submicron aerosol particles were collected during July and August 2010 in Hyytiälä, Finland, to determine the composition and sources of aerosol at that Boreal forest site. Submicron particles were collected on Teflon filters and analyzed by Fourier transform infrared (FTIR) spectroscopy for organic functional groups (OFG). Positive matrix factorization (PMF) was applied to aerosol mass spectrometry (AMS) measurements and FTIR spectra to identify summertime sources of submicron aerosol mass at the sampling site. The two largest sources of organic mass (OM) in particles identified at Hyytiälä were (1) biogenic aerosol from surrounding local forest and (2) biomass burning aerosol, transported 4–5 days from large wildfires burning near Moscow, Russia, and northern Ukraine. The robustness of this apportionment is supported by the agreement of two independent analytical methods for organic measurements with three statistical techniques. FTIR factor analysis was more sensitive to the chemical differences between biogenic and biomass burning organic components, while AMS factor analysis had a higher time resolution that more clearly linked the temporal behavior of separate OM factors to that of different source tracers even though their fragment mass spectrum were similar. The greater chemical sensitivity of the FTIR is attributed to the nondestructive preparation and the functional group specificity of spectroscopy. The FTIR spectra show strong similarities among biogenic and biomass burning factors from different regions as well as with reference OM (namely olive tree burning BBOA and α-pinene chamber secondary organic aerosol (SOA)). The biogenic factor correlated strongly with temperature and oxidation products of biogenic volatile organic compounds (BVOCs), included more than half oxygenated OFGs (carbonyl groups at 29% and carboxylic acid groups at 22%), and represented 35% of the submicron OM. Compared to previous studies at Hyytiälä, the summertime biogenic OM is 1.5 to 3 times larger than springtime biogenic OM (0.64 μg m−3 and 0.4 μg m−3, measured in 2005 and 2007, respectively), even though it contributed only 35% of OM. The biomass burning factor contributed 25% OM on average and up to 62% OM during three periods of transported biomass burning emissions: 26–28 July, 29–30 July, and 8–9 August, with OFG consisting mostly of carbonyl (41%) and alcohol (25%) groups. The high summertime terrestrial biogenic OM (1.7 μg m−3) and the high biomass burning contributions (1.2 μg m−3) were likely due to the abnormally high temperatures that resulted in both stressed boreal forest conditions with high regional BVOC emissions and numerous wildfires in upwind regions.

Application of Factor Analysis in Imaging Mass Spectrometric Data Analysis

2014 ◽  
Vol 42 (8) ◽  
pp. 1099-1103 ◽  
Author(s):  
Yi CHEN ◽  
Fei TANG ◽  
Tie-Gang LI ◽  
Jiu-Ming HE ◽  
Zeper ABLIZ ◽  
...  
Keyword(s):  
Factor Analysis ◽  
Data Analysis ◽  
Mass Spectrometric ◽  
Mass Spectrometric Data ◽  
Spectrometric Data
Sign in / Sign up

Export Citation Format

Share Document