Abundance and Emission Flux of the Anthropogenic Iron Oxide Aerosols From the East Asian Continental Outflow

Yoshida, Akitoshi; Ohata, Sho; Moteki, Nobuhiro; Adachi, Kouji; Mori, Tatsuhiro; Koike, M.; Takami, Akinori

doi:10.1029/2018jd028665

Cited by 24 publications

(35 citation statements)

References 52 publications

(102 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We designate this case as the "base case" and use only the color ratio and incandescent peak height to differentiate rBC from FeO x , as well as the additional criteria of the core scattering to differentiate anthropogenic FeO x from mineral dust with metallic inclusions. This is based on the method that has previously been used in, e.g., Moteki et al (2017), Ohata et al (2018), and Yoshida et al (2018). Figure 6.…”

Section: Base Casementioning

confidence: 99%

“…These methods are used, for example, to classify images, for text-to-speech applications, and for identifying handwritten digits and are also increasingly being applied to scientific applications, including atmospheric aerosol measurements. While machine learning approaches have been used to classify single particle aerosol mass spectra (Zawadowicz et al, 2017;Christopoulos et al, 2018) and biological aerosols detected via ultraviolet-lightinduced fluorescence (Robinson et al, 2013;Ruske et al, 2017Ruske et al, , 2018Savage and Huffman, 2018), they have not yet been applied to the problem of classifying aerosols detected via laser-induced incandescence. We review detection of incandescing aerosols with the SP2 and describe measurements on laboratory samples in Sect.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Lamb

2019

Atmos. Meas. Tech.

View full text Add to dashboard Cite

Single particle soot photometers (SP2) use laserinduced incandescence to detect aerosols on a single particle basis. SP2s that have been modified to provide greater spectral contrast between their narrow and broad-band incandescent detectors have previously been used to characterize both refractory black carbon (rBC) and light-absorbing metallic aerosols, including iron oxides (FeO x ). However, single particles cannot be unambiguously identified from their incandescent peak height (a function of particle mass) and color ratio (a measure of blackbody temperature) alone. Machine learning offers a promising approach for improving the classification of these aerosols. Here we explore the advantages and limitations of classifying single particle signals obtained with a modified SP2 using a supervised machine learning algorithm. Laboratory samples of different aerosols that incandesce in the SP2 (fullerene soot, mineral dust, volcanic ash, coal fly ash, Fe 2 O 3 , and Fe 3 O 4 ) were used to train a random forest algorithm. The trained algorithm was then applied to test data sets of laboratory samples and atmospheric aerosols. This method provides a systematic approach for classifying incandescent aerosols by providing a score, or conditional probability, that a particle is likely to belong to a particular aerosol class (rBC, FeO x , etc.) given its observed single particle features. We consider two alternative approaches for identifying aerosols in mixed populations based on their single particle SP2 response: one with specific class labels for each species sampled, and one with three broader classes (rBC, anthropogenic FeO x , and dust-like) for particles with similar SP2 responses. Predictions of the most likely particle class (the one with the highest mean probability) based on applying the trained random forest algorithm to the single particle features for test data sets comprising examples of each class are compared with the true class for those particles to estimate generalization performance. While the specific class approach performed well for rBC and Fe 3 O 4 (≥ 99 % of these aerosols are correctly identified), its classification of other aerosol types is significantly worse (only 47 %-66 % of other particles are correctly identified). Using the broader class approach, we find a classification accuracy of 99 % for FeO x samples measured in the laboratory. The method allows for classification of FeO x as anthropogenic or dust-like for aerosols with effective spherical diameters from 170 to > 1200 nm. The misidentification of both dust-like aerosols and rBC as anthropogenic FeO x is small, with < 3 % of the dust-like aerosols and < 0.1 % of rBC misidentified as FeO x for the broader class case. When applying this method to atmospheric observations taken in Boulder, CO, a clear mode consistent with FeO x was observed, distinct from dust-like aerosols.

show abstract

Section: Base Casementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Lamb

2019

Atmos. Meas. Tech.

View full text Add to dashboard Cite

show abstract

“…This is based on the method that has previously been used in e.g. Moteki et al (2017); ; Yoshida et al (2018).…”

Section: Base Casementioning

confidence: 99%

“…Bottom, right: Confusion matrix for the base case identifying all aerosols in Region 3 as anthropogenic FeOx, to provide an upper limit on anthropogenic FeOx. atmospheric observations of FeO x in East Asia as dominated by anthropogenic emissions (Moteki et al, 2017;Yoshida et al, 2018).…”

Section: Base Casementioning

confidence: 99%

“…An SP2 was used in one study to estimate the hematite content in Saharan dust measured off the West coast of Africa, and demonstrated good closure between 10 the hematite concentration associated with single dust particles and the optical properties observed in the dust plume (Liu et al, 2018). While previous work focusing on anthropogenic FeO x has relied on using the optical size of these aerosols after any volatile coatings have evaporated as an additional criteria to differentiate anthropogenic and non-anthropogenic aerosols with metallic components (Yoshida et al, 2016;Moteki et al, 2017;Yoshida et al, 2018), this method is limited to the range over which the SP2 optically sizes these aerosols (generally limited to ∼170-350 nm volume equivalent diameter 15 for FeO x ). Previous studies using the SP2 to quantify FeO x associated with anthropogenic sources have also not provided quantitive measures of classification performance for these aerosols.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Lamb¹

2019

Preprint

View full text Add to dashboard Cite

Single particle soot photometers (SP2) use laser-induced incandescence to detect aerosols on a single particle basis.Both refractory black carbon (rBC) and other light absorbing metallic aerosols, including iron oxides (FeO x ), have been characterized by the SP2, but single particles cannot be unambiguously identified from their incandescent peak height (a function of particle mass) and color ratio (a measure of blackbody temperature) alone. Machine learning offers a promising approach for improving the classification of these aerosols. Here we explore the advantages and limitations of classifying single 5 20 1 Atmos. Meas. Tech. Discuss., https://doi.

show abstract

Elemental analysis of soils along the South African National Road (N3) — a combined approach including statistics, pollution indicators, and geographic information system (GIS)

Batwa-Ismail

Moodley

Mutanga

2021

Environ Monit Assess

View full text Add to dashboard Cite

Abundance and Emission Flux of the Anthropogenic Iron Oxide Aerosols From the East Asian Continental Outflow

Cited by 24 publications

References 52 publications

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Classification of iron oxide aerosols by a single particle soot photometer using supervised machine learning

Elemental analysis of soils along the South African National Road (N3) — a combined approach including statistics, pollution indicators, and geographic information system (GIS)

Contact Info

Product

Resources

About