Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations

<p>Natural aerosol components such as particulate methanesulfonic acid (MSA<span class="inline-formula"><sub>p</sub></span>) play an important role in the Arctic climate. However, numerical models struggle to reproduce MSA<span class="inline-formula&qu...

Full description

Saved in:
Bibliographic Details
Main Authors: J. B. Pernov, W. H. Aeberhard, M. Volpi, E. Harris, B. Hohermuth, S. Ishino, R. B. Skeie, S. Henne, U. Im, P. K. Quinn, L. M. Upchurch, J. Schmale
Format: Article
Language:English
Published: Copernicus Publications 2025-06-01
Series:Atmospheric Chemistry and Physics
Online Access:https://acp.copernicus.org/articles/25/6497/2025/acp-25-6497-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849420376603361280
author J. B. Pernov
J. B. Pernov
W. H. Aeberhard
M. Volpi
E. Harris
E. Harris
B. Hohermuth
S. Ishino
R. B. Skeie
S. Henne
U. Im
P. K. Quinn
L. M. Upchurch
L. M. Upchurch
J. Schmale
author_facet J. B. Pernov
J. B. Pernov
W. H. Aeberhard
M. Volpi
E. Harris
E. Harris
B. Hohermuth
S. Ishino
R. B. Skeie
S. Henne
U. Im
P. K. Quinn
L. M. Upchurch
L. M. Upchurch
J. Schmale
author_sort J. B. Pernov
collection DOAJ
description <p>Natural aerosol components such as particulate methanesulfonic acid (MSA<span class="inline-formula"><sub>p</sub></span>) play an important role in the Arctic climate. However, numerical models struggle to reproduce MSA<span class="inline-formula"><sub>p</sub></span> concentrations and seasonality. Here we present an alternative data-driven methodology for modeling MSA<span class="inline-formula"><sub>p</sub></span> at four High Arctic stations (Alert, Gruvebadet, Pituffik (formerly Thule), and Utqiaġvik (formerly Barrow)). In our approach, we create input features that consider the ambient conditions experienced during atmospheric transport (e.g., dimethyl sulfide (DMS) emission, temperature, radiation, cloud cover, precipitation) for use in two data-driven models: a random forest (RF) regressor and an additive model (AM). The most important features were selected through automatic selection procedures, and their relationships with MSA<span class="inline-formula"><sub>p</sub></span> model output was investigated. Although the overall performance of our data-driven models on test data is modest (max. <span class="inline-formula"><i>R</i><sup>2</sup>=0.29</span>), the models can capture variability in the data well (max. Pearson correlation coefficient <span class="inline-formula">=</span> 0.77), outperform the current numerical models and reanalysis products, and produce physically interpretable results.</p> <p>The data-driven models selected features which can be grouped into three categories, the sources, chemical processing, and removal of MSA<span class="inline-formula"><sub>p</sub></span>, with specific differences between stations. The seasonal cycles and selected features suggest gas-phase oxidation is relatively more important during peak concentration months at Alert, Gruvebadet, and Pituffik (Thule), while aqueous-phase oxidation is relatively more important at Utqiaġvik (Barrow). Alert and Pituffik (Thule) appear to be more influenced by processes aloft than in the boundary layer. Our models usually selected chemical-processing-related features as the main factors influencing MSA<span class="inline-formula"><sub>p</sub></span> predictions, highlighting the importance of properly simulating oxidation-related processes in numerical models.</p>
format Article
id doaj-art-77920792f7964b249e905b2de4e85227
institution Kabale University
issn 1680-7316
1680-7324
language English
publishDate 2025-06-01
publisher Copernicus Publications
record_format Article
series Atmospheric Chemistry and Physics
spelling doaj-art-77920792f7964b249e905b2de4e852272025-08-20T03:31:46ZengCopernicus PublicationsAtmospheric Chemistry and Physics1680-73161680-73242025-06-01256497653710.5194/acp-25-6497-2025Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrationsJ. B. Pernov0J. B. Pernov1W. H. Aeberhard2M. Volpi3E. Harris4E. Harris5B. Hohermuth6S. Ishino7R. B. Skeie8S. Henne9U. Im10P. K. Quinn11L. M. Upchurch12L. M. Upchurch13J. Schmale14Extreme Environments Research Laboratory, École Polytechnique Fédérale de Lausanne, Sion, Switzerlandnow at: School of Earth and Atmospheric Sciences, Queensland University of Technology, Brisbane, AustraliaSwiss Data Science Center, ETH Zurich and École Polytechnique Fédérale de Lausanne, SwitzerlandSwiss Data Science Center, ETH Zurich and École Polytechnique Fédérale de Lausanne, SwitzerlandSwiss Data Science Center, ETH Zurich and École Polytechnique Fédérale de Lausanne, Switzerlandnow at: Climate and Environmental Physics, University of Bern, Sidlerstrasse 5, 3012 Bern, SwitzerlandSchroders Capital ILS, Zurich, SwitzerlandInstitute of Nature and Environmental Technology, Kanazawa University, Kanazawa, JapanCICERO, Center for International Climate Research, Oslo, NorwayEmpa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, SwitzerlandDepartment of Environmental Science/Interdisciplinary Centre for Climate Change, Aarhus University, Roskilde, DenmarkPacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration, Seattle, WA, USAPacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration, Seattle, WA, USACooperative Institute for Climate, Ocean, and Ecosystem Studies, University of Washington, Seattle, WA, USAExtreme Environments Research Laboratory, École Polytechnique Fédérale de Lausanne, Sion, Switzerland<p>Natural aerosol components such as particulate methanesulfonic acid (MSA<span class="inline-formula"><sub>p</sub></span>) play an important role in the Arctic climate. However, numerical models struggle to reproduce MSA<span class="inline-formula"><sub>p</sub></span> concentrations and seasonality. Here we present an alternative data-driven methodology for modeling MSA<span class="inline-formula"><sub>p</sub></span> at four High Arctic stations (Alert, Gruvebadet, Pituffik (formerly Thule), and Utqiaġvik (formerly Barrow)). In our approach, we create input features that consider the ambient conditions experienced during atmospheric transport (e.g., dimethyl sulfide (DMS) emission, temperature, radiation, cloud cover, precipitation) for use in two data-driven models: a random forest (RF) regressor and an additive model (AM). The most important features were selected through automatic selection procedures, and their relationships with MSA<span class="inline-formula"><sub>p</sub></span> model output was investigated. Although the overall performance of our data-driven models on test data is modest (max. <span class="inline-formula"><i>R</i><sup>2</sup>=0.29</span>), the models can capture variability in the data well (max. Pearson correlation coefficient <span class="inline-formula">=</span> 0.77), outperform the current numerical models and reanalysis products, and produce physically interpretable results.</p> <p>The data-driven models selected features which can be grouped into three categories, the sources, chemical processing, and removal of MSA<span class="inline-formula"><sub>p</sub></span>, with specific differences between stations. The seasonal cycles and selected features suggest gas-phase oxidation is relatively more important during peak concentration months at Alert, Gruvebadet, and Pituffik (Thule), while aqueous-phase oxidation is relatively more important at Utqiaġvik (Barrow). Alert and Pituffik (Thule) appear to be more influenced by processes aloft than in the boundary layer. Our models usually selected chemical-processing-related features as the main factors influencing MSA<span class="inline-formula"><sub>p</sub></span> predictions, highlighting the importance of properly simulating oxidation-related processes in numerical models.</p>https://acp.copernicus.org/articles/25/6497/2025/acp-25-6497-2025.pdf
spellingShingle J. B. Pernov
J. B. Pernov
W. H. Aeberhard
M. Volpi
E. Harris
E. Harris
B. Hohermuth
S. Ishino
R. B. Skeie
S. Henne
U. Im
P. K. Quinn
L. M. Upchurch
L. M. Upchurch
J. Schmale
Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
Atmospheric Chemistry and Physics
title Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
title_full Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
title_fullStr Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
title_full_unstemmed Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
title_short Data-driven modeling of environmental factors influencing Arctic methanesulfonic acid aerosol concentrations
title_sort data driven modeling of environmental factors influencing arctic methanesulfonic acid aerosol concentrations
url https://acp.copernicus.org/articles/25/6497/2025/acp-25-6497-2025.pdf
work_keys_str_mv AT jbpernov datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT jbpernov datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT whaeberhard datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT mvolpi datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT eharris datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT eharris datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT bhohermuth datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT sishino datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT rbskeie datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT shenne datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT uim datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT pkquinn datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT lmupchurch datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT lmupchurch datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations
AT jschmale datadrivenmodelingofenvironmentalfactorsinfluencingarcticmethanesulfonicacidaerosolconcentrations