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

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

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.

Additional Information

© 2021. American Geophysical Union. Issue Online: 09 November 2021. Version of Record online: 09 November 2021. Accepted manuscript online: 16 October 2021. Manuscript accepted: 08 October 2021. Manuscript revised: 04 October 2021. Manuscript received: 29 April 2021. This research has been supported by the National Aeronautics and Space Administration (NASA grant no. NNX17AG35G) and the National Science Foundation (NSF grant no. AGS-1550816). BAN, PCJ & JLJ were supported by NASA (grant nos. NNX15AJ23G, NNX15AH33A, 80NSSC19K0124, and 80NSSC18K0630). PJD & EJTL acknowledge support from the NSF (grant no. AGS-1650786). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. MP & SSY acknowledge the support from the National Research Foundation of Korea (NRF) funded by the Korean government (MSIT; grant no. NRF-2018R1A2B2006965). MJK was supported by an NSF Atmospheric and Geospace Sciences Postdoctoral Research Fellowship (AGS-PRF; grant no. 1524860). CDF, BBP & QP acknowledge support from the NSF (grant no. AGS-1652688) and the National Oceanic and Atmospheric Administration (NOAA grant no. NA17OAR4310012). The authors are grateful to Chuck Brock (NOAA) for the in situ measurements of aerosol microphysical properties during the ATom1–4 campaigns, Andrew J. Weinheimer, Denise D. Montzka & David J. Knapp (ARCTAS, DISCOVER-AQ^(TX, KORUS-AQ, and WE-CAN) and Thomas B. Ryerson (ATom1–4, DC3, and SEAC⁴RS) for (NOç) and (O₃) measurements, Paul O. Wennberg, John D. Crounse & Hannah M. Allen for (SO₂) measurements during the ARCTAS and ATom1–4 campaigns, Kevin R. Barry (funded by NSF grant no. AGS-1660486; WE-CAN: CCNc and SMPS), Sonia M. Kreidenweis (WE-CAN: CCNc and HR-ToF-AMS), Kathryn A. Moore (funded by the NSF Graduate Research Fellowship grant no. 006784; WE-CAN: CCNc), Darin W. Toohey & Michael Reeves (WE-CAN: UHSAS), Lauren A. Garofalo & Delphine K. Farmer (funded by the NOAA grant no. NA17OAR4310010; WE-CAN: HR-ToF-AMS), and Joel A. Thornton (WE-CAN: CIMS (SO₂) measurements). The authors are thankful to Michael Shook & Gao Chen at the NASA Langley Research Center Airborne Science Data for Atmospheric Composition (https://www-air.larc.nasa.gov/index.html) for data curation. The authors also thank the DOE ARM SGP Research Facility teams for the operation and maintenance of instruments, quality checks, and making their measurement data publicly available. Data Availability Statement. Data from the following aircraft campaigns were used in this study—ARCTAS (Jacob et al., 2010): ARCTAS Team (2020), ATom1–4 (Brock, Williamson, et al., 2019): Allen et al. (2019), Brock, Kupc, et al. (2019), Jimenez et al. (2019), and Ryerson et al. (2019), DC3 (Barth et al., 2015): DC3 Team (2013), DISCOVER-AQTX: DISCOVER-AQ Team (2014), KORUS-AQ (Jordan et al., 2020): KORUS-AQ Team (2018), SEAC4RS (Toon et al., 2016): SEAC4RS Team (2014), WE-CAN: WE-CAN Team (2019). Additional dual column CCNc measurement (Uin et al., 2017a, 2017b) data were obtained from the Atmospheric Radiation Measurement (ARM) user facility, a U.S. Department of Energy (DOE) Office of Science User Facility managed by the Biological and Environmental Research program. All data sets used in this study are publicly available and individually detailed as follows: [CCN0.18–0.86], PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the ARCTAS (Jacob et al., 2010) campaign: ARCTAS Team (2020, https://www-air.larc.nasa.gov/cgi-bin/ArcView/arctas). PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the ATom1–4 (Brock, Williamson, et al., 2019) campaigns: Allen, Crounse, Kim, Teng, and Wennberg (2019); Ryerson, Thompson, Peischl, and Bourgeois (2019); Jimenez et al. (2019); Brock, Kupc, et al. (2019, https://espo.nasa.gov/atom/archive/browse/atom/DC8). [CCN0.13–0.68], PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the DC3 (Barth et al., 2015) campaign: DC3 Team (2013, https://www-air.larc.nasa.gov/missions/dc3-seac4rs/). [CCN0.14–0.60], PNSD, PM composition and mass, [SO2], [NOx], [O3], T, and RH measured during the DISCOVER-AQTX campaign: DISCOVER-AQ Team (2014, https://www-air.larc.nasa.gov/cgi-bin/ArcView/discover-aq.tx-2013). [CCN0.6], PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the KORUS-AQ (Jordan et al., 2020) campaign: KORUS-AQ Team (2018, https://www-air.larc.nasa.gov/missions/korus-aq/). [CCN0.09–0.56], PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the SEAC4RS (Toon et al., 2016) campaign: SEAC4RS Team (2014, https://www-air.larc.nasa.gov/cgi-bin/ArcView/seac4rs). [CCN0.079–0.73], PNSD, PM1 composition and mass, [SO2], [NOx], [O3], T, and RH measured during the WE-CAN campaign: WE-CAN Team (2019, https://www-air.larc.nasa.gov/cgi-bin/ArcView/firexaq). Dual column CCNc measurement (Uin et al., 2017a, 2017b) data were obtained from the Atmospheric Radiation Measurement (ARM) user facility, a U.S. Department of Energy (DOE) Office of Science User Facility managed by the Biological and Environmental Research program, which is publicly available at the ARM Discovery Data Portal (https://www.archive.arm.gov/discovery/).

Attached Files

Published - 2021GL094133.pdf

Supplemental Material - 2021gl094133-sup-0001-supporting_information_si-s01.pdf

Files

2021GL094133.pdf

Files (4.9 MB)

Name	Size	Download all
2021GL094133.pdf md5:0f1eac68f2b23de63d07e7fdf2eb7dea	2.0 MB	Preview Download
2021gl094133-sup-0001-supporting_information_si-s01.pdf md5:86d11085159991f82e6504494e905619	2.8 MB	Preview Download

Additional details

	All versions	This version
Views	0	0
Downloads	0	0
Data volume	0 Bytes	0 Bytes