Title: Using machine learning to derive cloud condensation nuclei number concentrations from commonly available measurements

Abstract

Cloud condensation nuclei (CCN) number concentrations are an important aspect of aerosol–cloud interactions and the subsequent climate effects; however, their measurements are very limited. We use a machine learning tool, random decision forests, to develop a random forest regression model (RFRM) to derive CCN at 0.4 % supersaturation ([CCN0.4]) from commonly available measurements. The RFRM is trained on the long-term simulations in a global size-resolved particle microphysics model. Using atmospheric state and composition variables as predictors, through associations of their variabilities, the RFRM is able to learn the underlying dependence of [CCN0.4] on these predictors, which are as follows: eight fractions of PM_2.5 (NH₄, SO₄, NO_3, secondary organic aerosol (SOA), black carbon (BC), primary organic carbon (POC), dust, and salt), seven gaseous species (NO_x, NH₃, O₃, SO₂, OH, isoprene, and monoterpene), and four meteorological variables (temperature (T), relative humidity (RH), precipitation, and solar radiation). The RFRM is highly robust: it has a median mean fractional bias (MFB) of 4.4 % with ≈96.33 % of the derived [CCN0.4] within a good agreement range of -60%<+60% and strong correlation of Kendall's τ coefficient ≈0.88. The RFRM demonstrates its robustness over 4 orders of magnitude of [CCN0.4] over varying spatial (such as continentalmore » to oceanic, clean to polluted, and near-surface to upper troposphere) and temporal (from the hourly to the decadal) scales. At the Atmospheric Radiation Measurement Southern Great Plains observatory (ARM SGP) in Lamont, Oklahoma, United States, long-term measurements for PM_2.5 speciation (NH₄, SO₄, NO₃, and organic carbon (OC)), NO_x, O₃, SO₂, T, and RH, as well as [CCN0.4] are available. We modify, optimize, and retrain the developed RFRM to make predictions from 19 to 9 of these available predictors. This retrained RFRM (RFRM-ShortVars) shows a reduction in performance due to the unavailability and sparsity of measurements (predictors); it captures the [CCN0.4] variability and magnitude at SGP with ≈67.02 % of the derived values in the good agreement range. This work shows the potential of using the more commonly available measurements of PM_2.5 speciation to alleviate the sparsity of CCN number concentrations' measurements.« less

Authors:

Nair, Arshad Arjunan  ^[1];

• Search DOE PAGES for author "Nair, Arshad Arjunan"
• Search DOE PAGES for ORCID "0000-0003-2530-7757"

Yu, Fangqun  ^[1]

• Search DOE PAGES for author "Yu, Fangqun"
• Search DOE PAGES for ORCID "0000-0001-8862-4835"

---

+ Show Author Affiliations

1. State Univ. of New York (SUNY), Albany, NY (United States)

Publication Date:

Thu Nov 05 00:00:00 EST 2020

Research Org.:

Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Atmospheric Radiation Measurement (ARM) Data Center

Sponsoring Org.:

USDOE Office of Science (SC), Biological and Environmental Research (BER); National Science Foundation (NSF); National Aeronautics and Space Administration (NASA); New York State Energy Research and Development Authority

OSTI Identifier:

1725776

Grant/Contract Number:  

AC05-76RL01830; AGS-1550816; NNX17AG35G; 137487

Resource Type:

Accepted Manuscript

Journal Name:

Atmospheric Chemistry and Physics (Online)

Additional Journal Information:

Journal Name: Atmospheric Chemistry and Physics (Online); Journal Volume: 20; Journal Issue: 21; Journal ID: ISSN 1680-7324

Publisher:

Copernicus Publications, EGU

Country of Publication:

United States

Language:

English

Subject:

54 ENVIRONMENTAL SCIENCES