Global fields of daily accumulation-mode particle number concentrations using in situ observations, reanalysis data, and machine learning

Abstract. Accurate global estimates of accumulation-mode particle number concentrations (N100) are essential for understanding aerosol–cloud interactions and their climate effects and for improving Earth system models. However, traditional methods relying on sparse in situ measurements lack comprehensive coverage, and indirect satellite retrievals have limited sensitivity in the relevant size range. To overcome these challenges, we apply machine learning (ML) techniques – multiple linear regression (MLR) and eXtreme Gradient Boosting (XGB) – to generate daily global N100 fields using in situ measurements as target variables and reanalysis data from the Copernicus Atmosphere Monitoring Service (CAMS) and ERA5 as predictor variables. Our cross-validation showed that ML models captured N100 concentrations well in environments well-represented in the training set, with over 70 % of daily estimates being within a factor of 1.5 of observations. However, performance declines in underrepresented regions and conditions, such as in clean and remote environments, including marine, tropical, and polar regions, underscoring the need for more diverse observations. The most important predictors for N100 in the ML models were aerosol-phase sulfate and gas-phase ammonia concentrations, followed by carbon monoxide and sulfur dioxide. Although black carbon and organic matter showed the highest feature importance values, their opposing signs in the MLR model coefficients suggest that their effects largely offset each other’s contributions to the N100 estimate. By directly linking estimates to in situ measurements, our ML approach provides valuable insights into the global distribution of N100 and serves as a complementary tool for evaluating Earth system model outputs and advancing the understanding of aerosol processes and their role in the climate system.

Authors

Institutions

Publication Details

Journal
Aerosol Research
Published
2025-11-28
DOI
https://doi.org/10.5194/ar-3-589-2025
Primary Topic
Atmospheric chemistry and aerosols
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Global fields of daily accumulation-mode particle number concentrations using in situ observations, reanalysis data, and machine learning

Markku Kulmala, Lauri Laakso, Jakub Ondráček, Shahzad Gani et al.
Aerosol Research
Atmospheric chemistry and aerosols
article

Global fields of daily accumulation-mode particle number concentrations using in situ observations, reanalysis data, and machine learning

Markku Kulmala, Lauri Laakso, Jakub Ondráček, Shahzad Gani, Don Collins, Tuomo Nieminen, N. Mihalopoulos, Kristina Plauškaitė, Mira L. Pöhlker, Pauli Paasonen, Peter Tunved, Roy M. Harrison, Kerneels Jaars, Antti Hyvärinen, Victoria A. Sinclair, Ari Laaksonen, Adam Kristensson, Daniel Holmberg, Alfred Wiedensohler, Colin D. O'Dowd, Marco A. Franco, Benjamin Bergmans, Ximeng Qi, Kai Puolamäki, Ville Vakkari, Tuukka Petäjä, Veli‐Matti Kerminen, Aino Ovaska, Rakesh K. Hooda, John Backman, Elio Rauth
article en

Abstract

Abstract. Accurate global estimates of accumulation-mode particle number concentrations (N100) are essential for understanding aerosol–cloud interactions and their climate effects and for improving Earth system models. However, traditional methods relying on sparse in situ measurements lack comprehensive coverage, and indirect satellite retrievals have limited sensitivity in the relevant size range. To overcome these challenges, we apply machine learning (ML) techniques – multiple linear regression (MLR) and eXtreme Gradient Boosting (XGB) – to generate daily global N100 fields using in situ measurements as target variables and reanalysis data from the Copernicus Atmosphere Monitoring Service (CAMS) and ERA5 as predictor variables. Our cross-validation showed that ML models captured N100 concentrations well in environments well-represented in the training set, with over 70 % of daily estimates being within a factor of 1.5 of observations. However, performance declines in underrepresented regions and conditions, such as in clean and remote environments, including marine, tropical, and polar regions, underscoring the need for more diverse observations. The most important predictors for N100 in the ML models were aerosol-phase sulfate and gas-phase ammonia concentrations, followed by carbon monoxide and sulfur dioxide. Although black carbon and organic matter showed the highest feature importance values, their opposing signs in the MLR model coefficients suggest that their effects largely offset each other’s contributions to the N100 estimate. By directly linking estimates to in situ measurements, our ML approach provides valuable insights into the global distribution of N100 and serves as a complementary tool for evaluating Earth system model outputs and advancing the understanding of aerosol processes and their role in the climate system.

Aerosol ResearchVol. 3(2)
University of California, Riverside (US), University of Jordan (JO), Finnish Meteorological Institute (FI), University of Helsinki (FI), University of Crete (GR), Applied Science Private University (JO), Stockholm University (SE), University of Eastern Finland (FI), Universidade de São Paulo (BR), Lund University (SE), Ollscoil na Gaillimhe – University of Galway (IE), University of Würzburg (DE), North-West University (ZA), Center for Physical Sciences and Technology (LT), Institut Scientifique de Service Public (BE), Leibniz Institute for Tropospheric Research (DE), National Observatory of Athens (GR), Czech Academy of Sciences, Institute of Chemical Process Fundamentals (CZ), Indian Institute of Technology Delhi (IN), Beijing University of Chemical Technology (CN), University of Birmingham (GB), Nanjing University (CN)
Openalex Percentile: Top 58%
Atmospheric chemistry and aerosols
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.