Articles | Volume 25, issue 15
https://doi.org/10.5194/acp-25-8507-2025
https://doi.org/10.5194/acp-25-8507-2025
Research article
 | 
06 Aug 2025
Research article |  | 06 Aug 2025

Identifying drivers of surface ozone bias in global chemical reanalysis with explainable machine learning

Kazuyuki Miyazaki, Yuliya Marchetti, James Montgomery, Steven Lu, and Kevin Bowman

Related authors

Quantifying biases in TROPESS AIRS, CrIS, and joint AIRS+OMI tropospheric ozone products using ozonesondes
Elyse A. Pennington, Gregory B. Osterman, Vivienne H. Payne, Kazuyuki Miyazaki, Kevin W. Bowman, and Jessica L. Neu
Atmos. Chem. Phys., 25, 8533–8552, https://doi.org/10.5194/acp-25-8533-2025,https://doi.org/10.5194/acp-25-8533-2025, 2025
Short summary
Trace gas atmospheric rivers: remote drivers of air pollutants
Mukesh Rai, Kazuyuki Miyazaki, Vivienne Payne, Bin Guan, and Duane Waliser
EGUsphere, https://doi.org/10.5194/egusphere-2025-399,https://doi.org/10.5194/egusphere-2025-399, 2025
Short summary
Assessing the relative impacts of satellite ozone and its precursor observations to improve global tropospheric ozone analysis using multiple chemical reanalysis systems
Takashi Sekiya, Emanuele Emili, Kazuyuki Miyazaki, Antje Inness, Zhen Qu, R. Bradley Pierce, Dylan Jones, Helen Worden, William Y. Y. Cheng, Vincent Huijnen, and Gerbrand Koren
Atmos. Chem. Phys., 25, 2243–2268, https://doi.org/10.5194/acp-25-2243-2025,https://doi.org/10.5194/acp-25-2243-2025, 2025
Short summary
Quantifying uncertainties in satellite NO2 superobservations for data assimilation and model evaluation
Pieter Rijsdijk, Henk Eskes, Arlene Dingemans, K. Folkert Boersma, Takashi Sekiya, Kazuyuki Miyazaki, and Sander Houweling
Geosci. Model Dev., 18, 483–509, https://doi.org/10.5194/gmd-18-483-2025,https://doi.org/10.5194/gmd-18-483-2025, 2025
Short summary
Assessment of regional and interannual variations in tropospheric ozone in chemical reanalyses
Dylan Jones, Lucas Prates, Zhen Qu, William Cheng, Kazuyuki Miyazaki, Takashi Sekiya, Antje Inness, Rajesh Kumar, Xiao Tang, Helen Worden, Gerbrand Koren, and Vincent Huijen
EGUsphere, https://doi.org/10.5194/egusphere-2024-3759,https://doi.org/10.5194/egusphere-2024-3759, 2025
Short summary

Related subject area

Subject: Gases | Research Activity: Machine Learning | Altitude Range: Troposphere | Science Focus: Chemistry (chemical composition and reactions)
High-resolution greenhouse gas flux inversions using a machine learning surrogate model for atmospheric transport
Nikhil Dadheech, Tai-Long He, and Alexander J. Turner
Atmos. Chem. Phys., 25, 5159–5174, https://doi.org/10.5194/acp-25-5159-2025,https://doi.org/10.5194/acp-25-5159-2025, 2025
Short summary
Implications of VOC Oxidation in Atmospheric Chemistry: Development of a Comprehensive AI Model for Predicting Reaction Rate Constants
Xin Zhang, Jiaqi Luo, Wenxiao Pan, Qiao Xue, Xian Liu, Jianjie Fu, Aiqian Zhang, and Guibin Jiang
EGUsphere, https://doi.org/10.5194/egusphere-2025-1241,https://doi.org/10.5194/egusphere-2025-1241, 2025
Short summary
Multi-Machine Learning Approaches to Modeling Small-Scale Source Attribution of Ozone Formation
Zheng Xiao, Yifeng Lu, and Guangli Xiu
EGUsphere, https://doi.org/10.5194/egusphere-2025-160,https://doi.org/10.5194/egusphere-2025-160, 2025
Short summary
Technical note: Towards atmospheric compound identification in chemical ionization mass spectrometry with pesticide standards and machine learning
Federica Bortolussi, Hilda Sandström, Fariba Partovi, Joona Mikkilä, Patrick Rinke, and Matti Rissanen
Atmos. Chem. Phys., 25, 685–704, https://doi.org/10.5194/acp-25-685-2025,https://doi.org/10.5194/acp-25-685-2025, 2025
Short summary
Diagnosing ozone–NOx–VOC–aerosol sensitivity and uncovering causes of urban–nonurban discrepancies in Shandong, China, using transformer-based estimations
Chenliang Tao, Yanbo Peng, Qingzhu Zhang, Yuqiang Zhang, Bing Gong, Qiao Wang, and Wenxing Wang
Atmos. Chem. Phys., 24, 4177–4192, https://doi.org/10.5194/acp-24-4177-2024,https://doi.org/10.5194/acp-24-4177-2024, 2024
Short summary

Cited articles

Altmann, A., Toloşi, L., Sander, O., and Lengauer, T.: Permutation importance: a corrected feature importance measure, Bioinformatics, 26, 1340–1347, 2010. a
Archibald, A. T., Neu, J. L., Elshorbany, Y. F., Cooper, O. R., Young, P. J., Akiyoshi, H., Cox, R. A., Coyle, M., Derwent, R. G., Deushi, M., Finco, A., Frost, G. J., Galbally, I. E., Gerosa, G., Granier, C., Griffiths, P. T., Hossaini, R., Hu, L., Jöckel, P., Josse, B., Lin, M. Y., Mertens, M., Morgenstern, O., Naja, M., Naik, V., Oltmans, S., Plummer, D. A., Revell, L. E., Saiz-Lopez, A., Saxena, P., Shin, Y. M., Shahid, I., Shallcross, D., Tilmes, S., Trickl, T., Wallington, T. J., Wang, T., Worden, H. M., and Zeng, G.: Tropospheric Ozone Assessment Report: A critical review of changes in the tropospheric ozone burden and budget from 1850 to 2100, Elementa: Science of the Anthropocene, 8, 034, https://doi.org/10.1525/elementa.2020.034, 2020. a
Bauwens, M., Compernolle, S., Stavrakou, T., Müller, J.-F., van Gent, J., Eskes, H., Levelt, P. F., van der A, R., Veefkind, J. P., Vlietinck, J., Yu, H., and Zehner, C.: Impact of Coronavirus outbreak on NO2 pollution assessed using TROPOMI and OMI observations, Geophys. Res. Lett., 47, e2020GL087978, https://doi.org/10.1029/2020GL087978, 2020. a
Betancourt, C., Stomberg, T. T., Edrich, A.-K., Patnala, A., Schultz, M. G., Roscher, R., Kowalski, J., and Stadtler, S.: Global, high-resolution mapping of tropospheric ozone – explainable machine learning and impact of uncertainties, Geosci. Model Dev., 15, 4331–4354, https://doi.org/10.5194/gmd-15-4331-2022, 2022. a, b
Boersma, K., Eskes, H., Richter, A., De Smedt, I., Lorente, A., Beirle, S., Van Geffen, J., Peters, E., Van Roozendael, M., and Wagner, T.: QA4ECV NO2 tropospheric and stratospheric vertical column data from OMI (Version 1.1), Royal Netherlands Meteorological Institute (KNMI) [data set], https://doi.org/10.21944/qa4ecv-no2-omi-v1.1, 2017. a
Download
Short summary
This study employs explainable machine learning to analyze the causes of significant biases in surface ozone estimates from chemical reanalysis. By analyzing global observations and chemical reanalysis outputs, key bias drivers, such as meteorological conditions and precursor emissions, were identified. This provides actionable insights to improve chemical transport models, observation systems, and emissions inventories, ultimately enhancing ozone reanalysis for better air pollution management.
Share
Altmetrics
Final-revised paper
Preprint