{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T12:12:42Z","timestamp":1775736762494,"version":"3.50.1"},"reference-count":38,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,1,15]],"date-time":"2025-01-15T00:00:00Z","timestamp":1736899200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Big Data"],"abstract":"<jats:p>Atmospheric ozone chemistry involves various substances and reactions, which makes it a complex system. We analyzed data recorded by Switzerland's National Air Pollution Monitoring Network (NABEL) to showcase the capabilities of machine learning (ML) for the prediction of ozone concentrations (daily averages) and to document a general approach that can be followed by anyone facing similar problems. We evaluated various artificial neural networks and compared them to linear as well as non-linear models deduced with ML. The main analyses and the training of the models were performed on atmospheric air data recorded from 2016 to 2023 at the NABEL station Lugano-Universit\u00e0 in Lugano, TI, Switzerland. As a first step, we used techniques like best subset selection to determine the measurement parameters that might be relevant for the prediction of ozone concentrations; in general, the parameters identified by these methods agree with atmospheric ozone chemistry. Based on these results, we constructed various models and used them to predict ozone concentrations in Lugano for the period between January 1, 2024, and March 31, 2024; then, we compared the output of our models to the actual measurements and repeated this procedure for two NABEL stations situated in northern Switzerland (D\u00fcbendorf-Empa and Z\u00fcrich-Kaserne). For these stations, predictions were made for the aforementioned period and the period between January 1, 2023, and December 31, 2023. In most of the cases, the lowest mean absolute errors (MAE) were provided by a non-linear model with 12 components (different powers and linear combinations of NO<jats:sub>2<\/jats:sub>, NO<jats:sub>X<\/jats:sub>, SO<jats:sub>2<\/jats:sub>, non-methane volatile organic compounds, temperature and radiation); the MAE of predicted ozone concentrations in Lugano was as low as 9 \u03bcgm<jats:sup>\u22123<\/jats:sup>. For the stations in Z\u00fcrich and D\u00fcbendorf, the lowest MAEs were around 11 \u03bcgm<jats:sup>\u22123<\/jats:sup> and 13 \u03bcgm<jats:sup>\u22123<\/jats:sup>, respectively. For the tested periods, the accuracy of the best models was approximately 1 \u03bcgm<jats:sup>\u22123<\/jats:sup>. Since the aforementioned values are all lower than the standard deviations of the observations we conclude that using ML for complex data analyses can be very helpful and that artificial neural networks do not necessarily outperform simpler models.<\/jats:p>","DOI":"10.3389\/fdata.2024.1469809","type":"journal-article","created":{"date-parts":[[2025,1,15]],"date-time":"2025-01-15T06:12:48Z","timestamp":1736921568000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Analysis and prediction of atmospheric ozone concentrations using machine learning"],"prefix":"10.3389","volume":"7","author":[{"given":"Stephan","family":"R\u00e4ss","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Markus C.","family":"Leuenberger","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,1,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1016\/j.grj.2017.08.001","article-title":"The application of machine learning for evaluating anthropogenic versus natural climate change","volume":"14","author":"Abbot","year":"2017","journal-title":"Geo. Res. J"},{"key":"B2","volume-title":"Multilayer Perceptrons","author":"Almeida","year":"2020"},{"key":"B3","unstructured":"Ballaman\n              R.\n            \n            \n              Weber\n              R.\n            \n            \n              Emmenegger\n              L.\n            \n            \n              H\u00fcglin\n              C.\n            \n            \n              Reimann\n              S.\n            \n          \n          Nationales Beobachtungsnetz f\u00fcr Luftfremdstoffe (NABEL) Messkonzept 2020\u20132030\n          \n          2020"},{"key":"B4","volume-title":"The Shallow and the Deep: A Biased Introduction to Neural Networks and Old School Machine Learning","author":"Biehl","year":"2023"},{"key":"B5","doi-asserted-by":"publisher","first-page":"180","DOI":"10.3390\/atmos13020180","article-title":"Machine learning in weather prediction and climate analyses\u2014applications and perspectives","volume":"13","author":"Bochenek","year":"2022","journal-title":"Atmosphere"},{"key":"B6","unstructured":"A guide to recurrent neural networks and backpropagation\n          \n          1\n          10\n          \n            \n              Boden\n              M.\n            \n          \n          Dallas Project\n          2\n          2002"},{"key":"B7","volume-title":"Deep Learning with Python","author":"Chollet","year":"2021"},{"key":"B8","doi-asserted-by":"publisher","first-page":"656","DOI":"10.1001\/jamapediatrics.2013.1329","article-title":"Missing data and multiple imputation","volume":"167","author":"Cummings","year":"2013","journal-title":"JAMA Pediatr"},{"key":"B9","doi-asserted-by":"publisher","first-page":"119807","DOI":"10.1016\/j.jenvman.2023.119807","article-title":"Predicting dissolved oxygen level using young's double-slit experiment optimizer-based weighting model","volume":"351","author":"Dong","year":"2024","journal-title":"J. Environ. Manage"},{"key":"B10","doi-asserted-by":"publisher","first-page":"3999","DOI":"10.5194\/gmd-11-3999-2018","article-title":"Challenges and design choices for global weather and climate models based on machine learning","volume":"11","author":"Dueben","year":"2018","journal-title":"Geoscient. Model Dev"},{"key":"B11","unstructured":"Data From: Data query NABEL (National Air Pollution Monitoring Network)\n          \n          2024"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1091","DOI":"10.1080\/1073161X.1993.10467187","article-title":"Atmospheric chemistry of tropospheric ozone formation: scientific and regulatory implications","volume":"43","author":"Finlayson-Pitts","year":"1993","journal-title":"Air & Waste"},{"key":"B13","volume-title":"An Introduction to Statistical Learning: With Applications in R","author":"Gareth","year":"2013"},{"key":"B14","doi-asserted-by":"publisher","first-page":"426","DOI":"10.1016\/j.procs.2010.12.071","article-title":"Definition of artificial neural networks with comparison to other networks","volume":"3","author":"Guresen","year":"2011","journal-title":"Procedia Comput. Sci"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"B16","unstructured":"Huff\n              K.\n            \n          \n          The Effects of Volcanic Sulfur Dioxide on the Ozone Layer\n          \n          1996"},{"key":"B17","unstructured":"Adadelta"},{"key":"B18","unstructured":"Adam"},{"key":"B19","unstructured":"Nadam"},{"key":"B20","unstructured":"Rmsprop"},{"key":"B21","doi-asserted-by":"publisher","first-page":"69","DOI":"10.3137\/ao.460104","article-title":"Stratospheric ozone chemistry","volume":"46","author":"McConnell","year":"2008","journal-title":"Atmosph.-Ocean"},{"key":"B22","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-662-68313-2","volume-title":"Computer Science in Sport: Modeling, Simulation, Data Analysis and Visualization of Sports-related Data","author":"Memmert","year":"2024"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1008","DOI":"10.1007\/s10661-024-13209-6","article-title":"Assessment and prediction of water quality index (wqi) by seasonal key water parameters in a coastal city: application of machine learning models","volume":"196","author":"Mo","year":"2024","journal-title":"Environ. Monit. Assess"},{"key":"B24","volume-title":"Convolutional Neural Networks","author":"Montesinos L\u00f3pez","year":"2022"},{"key":"B25","doi-asserted-by":"publisher","first-page":"8074","DOI":"10.1007\/s11356-017-9239-3","article-title":"The effects of ozone on human health","volume":"25","author":"Nuvolone","year":"2018","journal-title":"Environ. Sci. Pollut. Res. Int"},{"key":"B26","doi-asserted-by":"publisher","first-page":"162434","DOI":"10.1016\/j.scitotenv.2023.162434","article-title":"The underlying mechanisms of pm2.5 and o3 synergistic pollution in east china: photochemical and heterogeneous interactions","volume":"873","author":"Qu","year":"2023","journal-title":"Sci. Total Environm"},{"key":"B27","doi-asserted-by":"publisher","first-page":"3106","DOI":"10.3390\/electronics12143106","article-title":"A review on dropout regularization approaches for deep neural networks within the scholarly domain","volume":"12","author":"Salehin","year":"2023","journal-title":"Electronics"},{"key":"B28","doi-asserted-by":"publisher","first-page":"275","DOI":"10.1029\/1999RG900008","article-title":"Stratospheric ozone depletion: A review of concepts and history","volume":"37","author":"Solomon","year":"1999","journal-title":"Rev. Geophys"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1423","DOI":"10.1007\/s00477-023-02628-5","article-title":"Robust augmented estimation for hourly pm 2.5 using heteroscedastic spatiotemporal models","volume":"38","author":"Song","year":"2024","journal-title":"Stochast. Environm. Res. Risk Assessm"},{"key":"B30","doi-asserted-by":"publisher","first-page":"109237","DOI":"10.1016\/j.cie.2023.109237","article-title":"Combined forecasting tool for renewable energy management in sustainable supply chains","volume":"179","author":"Sun","year":"2023","journal-title":"Comp. Indust. Eng"},{"key":"B31","doi-asserted-by":"publisher","first-page":"100296","DOI":"10.1016\/j.bdr.2021.100296","article-title":"NGCU: a new rnn model for time-series data prediction","volume":"27","author":"Wang","year":"2022","journal-title":"Big Data Res"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1138","DOI":"10.3390\/en11051138","article-title":"Short-term load forecasting with multi-source data using gated recurrent unit neural networks","volume":"11","author":"Wang","year":"2018","journal-title":"Energies"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-016-0043-6","article-title":"A survey of transfer learning","volume":"3","author":"Weiss","year":"2016","journal-title":"J. Big Data"},{"key":"B34","doi-asserted-by":"publisher","first-page":"2267","DOI":"10.1016\/j.csda.2010.04.005","article-title":"Avoiding bias due to perfect prediction in multiple imputation of incomplete categorical variables","volume":"54","author":"White","year":"2010","journal-title":"Comput. Statist.Data Analys."},{"key":"B35","doi-asserted-by":"publisher","first-page":"101843","DOI":"10.1016\/j.apr.2023.101843","article-title":"Ozone variability driven by the synoptic patterns over china during 2014\u20132022 and its implications for crop yield and economy","volume":"14","author":"Yan","year":"2023","journal-title":"Atmos. Pollut. Res"},{"key":"B36","doi-asserted-by":"publisher","first-page":"e2022G","DOI":"10.1029\/2022GL101113","article-title":"Tropospheric ozone during the last interglacial","volume":"49","author":"Yan","year":"2022","journal-title":"Geophys. Res. Lett"},{"key":"B37","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1038\/s41586-019-1277-1","article-title":"Isotopic constraint on the twentieth-century increase in tropospheric ozone","volume":"570","author":"Yeung","year":"2019","journal-title":"Nature"},{"key":"B38","doi-asserted-by":"publisher","first-page":"8549","DOI":"10.5194\/acp-24-8549-2024","article-title":"Characteristics and sources of nonmethane volatile organic compounds (nmvocs) and o3noxnmvoc relationships in zhengzhou, china","volume":"24","author":"Zhang","year":"2024","journal-title":"Atmosph. Chem. Phys"}],"container-title":["Frontiers in Big Data"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2024.1469809\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,15]],"date-time":"2025-01-15T06:12:53Z","timestamp":1736921573000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2024.1469809\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,15]]},"references-count":38,"alternative-id":["10.3389\/fdata.2024.1469809"],"URL":"https:\/\/doi.org\/10.3389\/fdata.2024.1469809","relation":{},"ISSN":["2624-909X"],"issn-type":[{"value":"2624-909X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,15]]},"article-number":"1469809"}}