Machine learning uncovers aerosol size information from chemistry and meteorology to quantify potential cloud-forming particles

Nair, Arshad Arjunan; Yu, Fangqun; Campuzano‐Jost, Pedro; DeMott, Paul J.; Levin, Ezra J. T.; Jimenez, José L.; Peischl, J.; Pollack, Ilana B.; Fredrickson, Carley D.; Beyersdorf, A. J.; Nault, Benjamin A.; Park, Minsu; Yum, Seong Soo; Palm, Brett B.; Li, Xu; Bourgeois, Ilann; Anderson, Bruce; Nenes, Athanasios; Ziemba, L. D.; Moore, Richard H.; Lee, Taehyoung; Park, Taehyun; Thompson, C. R.; Flocke, Frank; Huey, L. G.; Kim, Michelle; Peng, Qiaoyun

doi:10.21203/rs.3.rs-244416/v1

Cited by 3 publications

(1 citation statement)

References 42 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a recent study, Nair et al. (2021) showed that machine learning can extract aerosol size information from aerosol composition and additionally from atmospheric chemical and meteorological variables. In this study, we employ outputs from long‐term (30‐year) simulations of a global size‐resolved (sectional) aerosol microphysics model and a machine‐learning tool to develop a Random Forest Regression Model (RFRM) for PNC.…”

Section: Introductionmentioning

confidence: 99%

Use of Machine Learning to Reduce Uncertainties in Particle Number Concentration and Aerosol Indirect Radiative Forcing Predicted by Climate Models

Luo

Bauer

et al. 2022

Geophysical Research Letters

Self Cite

View full text Add to dashboard Cite

All cloud droplets formed in the atmosphere start with tiny particles that act as cloud condensation nuclei (CCN). These aerosols can alter cloud properties and precipitation (Albrecht, 1989;Twomey, 1977) and thereby indirectly influence the Earth's radiation budget and climate change. The radiative forcing (RF) associated with aerosolcloud interactions (aci) remains the largest source of uncertainty in climate prediction. According to the Intergovernmental Panel on Climate Change's Fifth Assessment Report (IPCC AR5, 2013), RF aci of anthropogenic aerosols was estimated to be −0.55 W•m −2 with "low" level of confidence. A number of post-IPCC AR5 global climate modeling studies still show large discrepancy in the values of RF aci , ranging from ∼ −0.35 W•m −2 (Nazarenko et al., 2017), to −0.7 W•m −2 (Rotstayn et al., 2014) to −1.08 W•m −2 (Bauer et al., 2020), to −1.28 W•m −2 (Tonttila et al., 2015), to −1.54 W•m −2 (Bauer et al., 2020), and to −2.19 W•m −2 (Zhang et al., 2016). In order to confidently interpret past and accurately project future climate change, it is essential to reduce RF aci discrepancy among different models.RF aci depends strongly on the response of number concentrations of particles that can act as CCN to anthropogenic emissions (Albrecht, 1989;Twomey, 1977). The increase in cloud drops with particle number concentration (PNC) has been confirmed by many aircraft measurements (e.g., Ramanathan et al., 2001). PNC exhibits significant spatial and temporal variability due to the non-linear dependence of new particle formation and growth

show abstract

Section: Introductionmentioning

confidence: 99%

Use of Machine Learning to Reduce Uncertainties in Particle Number Concentration and Aerosol Indirect Radiative Forcing Predicted by Climate Models

Luo

Bauer

et al. 2022

Geophysical Research Letters

Self Cite

View full text Add to dashboard Cite

show abstract

Factors influencing ambient particulate matter in Delhi, India: Insights from machine learning

Patel

Bhandari

Gani

et al. 2023

Aerosol Science and Technology

View full text Add to dashboard Cite

Parameterization of size of organic and secondary inorganic aerosol for efficient representation of global aerosol optical properties

Zhu,

Martin,

Croft

et al. 2023

Atmos. Chem. Phys.

View full text Add to dashboard Cite

Abstract. Accurate representation of aerosol optical properties is essential for the modeling and remote sensing of atmospheric aerosols. Although aerosol optical properties are strongly dependent upon the aerosol size distribution, the use of detailed aerosol microphysics schemes in global atmospheric models is inhibited by associated computational demands. Computationally efficient parameterizations for aerosol size are needed. In this study, airborne measurements over the United States (DISCOVER-AQ) and South Korea (KORUS-AQ) are interpreted with a global chemical transport model (GEOS-Chem) to investigate the variation in aerosol size when organic matter (OM) and sulfate–nitrate–ammonium (SNA) are the dominant aerosol components. The airborne measurements exhibit a strong correlation (r=0.83) between dry aerosol size and the sum of OM and SNA mass concentration (MSNAOM). A global microphysical simulation (GEOS-Chem-TOMAS) indicates that MSNAOM and the ratio between the two components (OM/SNA) are the major indicators for SNA and OM dry aerosol size. A parameterization of the dry effective radius (Reff) for SNA and OM aerosol is designed to represent the airborne measurements (R2=0.74; slope = 1.00) and the GEOS-Chem-TOMAS simulation (R2=0.72; slope = 0.81). When applied in the GEOS-Chem high-performance model, this parameterization improves the agreement between the simulated aerosol optical depth (AOD) and the ground-measured AOD from the Aerosol Robotic Network (AERONET; R2 from 0.68 to 0.73 and slope from 0.75 to 0.96). Thus, this parameterization offers a computationally efficient method to represent aerosol size dynamically.

show abstract

Machine learning uncovers aerosol size information from chemistry and meteorology to quantify potential cloud-forming particles

Cited by 3 publications

References 42 publications

Use of Machine Learning to Reduce Uncertainties in Particle Number Concentration and Aerosol Indirect Radiative Forcing Predicted by Climate Models

Use of Machine Learning to Reduce Uncertainties in Particle Number Concentration and Aerosol Indirect Radiative Forcing Predicted by Climate Models

Factors influencing ambient particulate matter in Delhi, India: Insights from machine learning

Parameterization of size of organic and secondary inorganic aerosol for efficient representation of global aerosol optical properties

Contact Info

Product

Resources

About