Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles

Arshad Arjunan Nair; Fangqun Yu; Pedro Campuzano-Jost; Paul J. DeMott; Ezra J.T. Levin; Jose L. Jimenez; Jeff Peischl; Ilana B. Pollack; Carley D. Fredrickson; Andreas J. Beyersdorf; Benjamin A. Nault; Minsu Park; Seong Soo Yum; Brett B. Palm; Lu Xu; Ilann Bourgeois; Bruce E. Anderson; Athanasios Nenes; Luke D. Ziemba; Richard H. Moore; Taehyoung Lee; Taehyun Park; Chelsea R. Thompson; Frank Flocke; Lewis Gregory Huey; Michelle J. Kim; Qiaoyun Peng

doi:10.1029/2021GL094133

Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles

Arshad Arjunan Nair
, Fangqun Yu
, Pedro Campuzano-Jost
, Paul J. DeMott
, Ezra J.T. Levin
, Jose L. Jimenez
, Jeff Peischl
, Ilana B. Pollack
, Carley D. Fredrickson
, Andreas J. Beyersdorf
, Benjamin A. Nault
, Minsu Park
, Seong Soo Yum
, Brett B. Palm
, Lu Xu
, Ilann Bourgeois
, Bruce E. Anderson
, Athanasios Nenes
, Luke D. Ziemba
, Richard H. Moore

Taehyoung Lee, Taehyun Park, Chelsea R. Thompson, Frank Flocke, Lewis Gregory Huey, Michelle J. Kim, Qiaoyun Peng

Atmospheric Chemistry Observations & Modeling Lab

SUNY Albany
University of Colorado Boulder
Colorado State University
Now at Handix Scientific
National Oceanic and Atmospheric Administration
University of Washington
NASA Langley Research Center
California State University San Bernardino
Aerodyne Research, Inc.
Yonsei University
California Institute of Technology
Institute of Chemical Engineering and High Temperature Chemical Processes
Swiss Federal Institute of Technology Lausanne
Georgia Institute of Technology
Hankuk University of Foreign Studies

Research output: Contribution to journal › Article › peer-review

24 Scopus citations

Abstract

Cloud condensation nuclei (CCN) are mediators of aerosol-cloud interactions, which contribute to the largest uncertainty in climate change prediction. Here, we present a machine learning (ML)/artificial intelligence (AI) model that quantifies CCN from model-simulated aerosol composition, atmospheric trace gas, and meteorological variables. Comprehensive multi-campaign airborne measurements, covering varied physicochemical regimes in the troposphere, confirm the validity of and help probe the inner workings of this ML model: revealing for the first time that different ranges of atmospheric aerosol composition and mass correspond to distinct aerosol number size distributions. ML extracts this information, important for accurate quantification of CCN, additionally from both chemistry and meteorology. This can provide a physicochemically explainable, computationally efficient, robust ML pathway in global climate models that only resolve aerosol composition; potentially mitigating the uncertainty of effective radiative forcing due to aerosol-cloud interactions (ERF_aci) and improving confidence in assessment of anthropogenic contributions and climate change projections.

Original language	English
Article number	e2021GL094133
Journal	Geophysical Research Letters
Volume	48
Issue number	21
DOIs	https://doi.org/10.1029/2021GL094133
State	Published - Nov 16 2021

Keywords

Cloud condensation nuclei (CCN)
aerosols
aircraft campaign observations
explainable artificial intelligence (xAI)
machine learning
particle size distribution (PNSD)

Access to Document

10.1029/2021GL094133

Cite this

Nair, A. A., Yu, F., Campuzano-Jost, P., DeMott, P. J., Levin, E. J. T., Jimenez, J. L., Peischl, J., Pollack, I. B., Fredrickson, C. D., Beyersdorf, A. J., Nault, B. A., Park, M., Yum, S. S., Palm, B. B., Xu, L., Bourgeois, I., Anderson, B. E., Nenes, A., Ziemba, L. D., ... Peng, Q. (2021). Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles. Geophysical Research Letters, 48(21), Article e2021GL094133. https://doi.org/10.1029/2021GL094133

@article{29a3493e041049fd8beff2959671825e,

title = "Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles",

abstract = "Cloud condensation nuclei (CCN) are mediators of aerosol-cloud interactions, which contribute to the largest uncertainty in climate change prediction. Here, we present a machine learning (ML)/artificial intelligence (AI) model that quantifies CCN from model-simulated aerosol composition, atmospheric trace gas, and meteorological variables. Comprehensive multi-campaign airborne measurements, covering varied physicochemical regimes in the troposphere, confirm the validity of and help probe the inner workings of this ML model: revealing for the first time that different ranges of atmospheric aerosol composition and mass correspond to distinct aerosol number size distributions. ML extracts this information, important for accurate quantification of CCN, additionally from both chemistry and meteorology. This can provide a physicochemically explainable, computationally efficient, robust ML pathway in global climate models that only resolve aerosol composition; potentially mitigating the uncertainty of effective radiative forcing due to aerosol-cloud interactions (ERFaci) and improving confidence in assessment of anthropogenic contributions and climate change projections.",

keywords = "Cloud condensation nuclei (CCN), aerosols, aircraft campaign observations, explainable artificial intelligence (xAI), machine learning, particle size distribution (PNSD)",

author = "Nair, \{Arshad Arjunan\} and Fangqun Yu and Pedro Campuzano-Jost and DeMott, \{Paul J.\} and Levin, \{Ezra J.T.\} and Jimenez, \{Jose L.\} and Jeff Peischl and Pollack, \{Ilana B.\} and Fredrickson, \{Carley D.\} and Beyersdorf, \{Andreas J.\} and Nault, \{Benjamin A.\} and Minsu Park and Yum, \{Seong Soo\} and Palm, \{Brett B.\} and Lu Xu and Ilann Bourgeois and Anderson, \{Bruce E.\} and Athanasios Nenes and Ziemba, \{Luke D.\} and Moore, \{Richard H.\} and Taehyoung Lee and Taehyun Park and Thompson, \{Chelsea R.\} and Frank Flocke and Huey, \{Lewis Gregory\} and Kim, \{Michelle J.\} and Qiaoyun Peng",

year = "2021",

month = nov,

day = "16",

doi = "10.1029/2021GL094133",

language = "English",

volume = "48",

journal = "Geophysical Research Letters",

issn = "0094-8276",

publisher = "John Wiley and Sons Inc.",

number = "21",

}

Nair, AA, Yu, F, Campuzano-Jost, P, DeMott, PJ, Levin, EJT, Jimenez, JL, Peischl, J, Pollack, IB, Fredrickson, CD, Beyersdorf, AJ, Nault, BA, Park, M, Yum, SS, Palm, BB, Xu, L, Bourgeois, I, Anderson, BE, Nenes, A, Ziemba, LD, Moore, RH, Lee, T, Park, T, Thompson, CR, Flocke, F, Huey, LG, Kim, MJ & Peng, Q 2021, 'Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles', Geophysical Research Letters, vol. 48, no. 21, e2021GL094133. https://doi.org/10.1029/2021GL094133

TY - JOUR

T1 - Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles

AU - Nair, Arshad Arjunan

AU - Yu, Fangqun

AU - Campuzano-Jost, Pedro

AU - DeMott, Paul J.

AU - Levin, Ezra J.T.

AU - Jimenez, Jose L.

AU - Peischl, Jeff

AU - Pollack, Ilana B.

AU - Fredrickson, Carley D.

AU - Beyersdorf, Andreas J.

AU - Nault, Benjamin A.

AU - Park, Minsu

AU - Yum, Seong Soo

AU - Palm, Brett B.

AU - Xu, Lu

AU - Bourgeois, Ilann

AU - Anderson, Bruce E.

AU - Nenes, Athanasios

AU - Ziemba, Luke D.

AU - Moore, Richard H.

AU - Lee, Taehyoung

AU - Park, Taehyun

AU - Thompson, Chelsea R.

AU - Flocke, Frank

AU - Huey, Lewis Gregory

AU - Kim, Michelle J.

AU - Peng, Qiaoyun

PY - 2021/11/16

Y1 - 2021/11/16

N2 - Cloud condensation nuclei (CCN) are mediators of aerosol-cloud interactions, which contribute to the largest uncertainty in climate change prediction. Here, we present a machine learning (ML)/artificial intelligence (AI) model that quantifies CCN from model-simulated aerosol composition, atmospheric trace gas, and meteorological variables. Comprehensive multi-campaign airborne measurements, covering varied physicochemical regimes in the troposphere, confirm the validity of and help probe the inner workings of this ML model: revealing for the first time that different ranges of atmospheric aerosol composition and mass correspond to distinct aerosol number size distributions. ML extracts this information, important for accurate quantification of CCN, additionally from both chemistry and meteorology. This can provide a physicochemically explainable, computationally efficient, robust ML pathway in global climate models that only resolve aerosol composition; potentially mitigating the uncertainty of effective radiative forcing due to aerosol-cloud interactions (ERFaci) and improving confidence in assessment of anthropogenic contributions and climate change projections.

AB - Cloud condensation nuclei (CCN) are mediators of aerosol-cloud interactions, which contribute to the largest uncertainty in climate change prediction. Here, we present a machine learning (ML)/artificial intelligence (AI) model that quantifies CCN from model-simulated aerosol composition, atmospheric trace gas, and meteorological variables. Comprehensive multi-campaign airborne measurements, covering varied physicochemical regimes in the troposphere, confirm the validity of and help probe the inner workings of this ML model: revealing for the first time that different ranges of atmospheric aerosol composition and mass correspond to distinct aerosol number size distributions. ML extracts this information, important for accurate quantification of CCN, additionally from both chemistry and meteorology. This can provide a physicochemically explainable, computationally efficient, robust ML pathway in global climate models that only resolve aerosol composition; potentially mitigating the uncertainty of effective radiative forcing due to aerosol-cloud interactions (ERFaci) and improving confidence in assessment of anthropogenic contributions and climate change projections.

KW - Cloud condensation nuclei (CCN)

KW - aerosols

KW - aircraft campaign observations

KW - explainable artificial intelligence (xAI)

KW - machine learning

KW - particle size distribution (PNSD)

UR - https://www.scopus.com/pages/publications/85118854969

U2 - 10.1029/2021GL094133

DO - 10.1029/2021GL094133

M3 - Article

AN - SCOPUS:85118854969

SN - 0094-8276

VL - 48

JO - Geophysical Research Letters

JF - Geophysical Research Letters

IS - 21

M1 - e2021GL094133

ER -

Machine Learning Uncovers Aerosol Size Information From Chemistry and Meteorology to Quantify Potential Cloud-Forming Particles

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this