the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Deep learning-based downscaling of PM2.5, NO2 and O3 over Europe across emission-reduction scenarios
Abstract. Atmospheric pollutants pose significant health risks in Europe, where many cities exceed WHO and EU air quality standards. To support informed air quality policies, simulations of air pollution levels are required. They are generally provided by Chemical Transport Models, but these can be computationally expensive at high spatial resolution. In this context, deep learning methods have emerged as alternatives to full simulations. Using low and high resolution EMEP simulations, we investigate how different training strategies affect the robustness of a Residual Channel Attention Network for downscaling hourly particulate matter, nitrogen dioxide and ozone ground concentration fields from a 0.1°×0.05° to a 0.02°×0.01° resolution over Europe. The geographic extent and temporal sampling of the training data appear to be key determinants of the performance of the downscaling process: training the network on days that are evenly distributed throughout the year generally yields a better performance than training it on the same number of days but restricted to the winter season, and, similarly, the choice of a larger spatial domain during training increases the model performance in comparison with a smaller spatial domain. The optimised model achieves better performance in reconstructing fine scale spatial gradients than a bilinear interpolation of low resolution simulations. It generalises to multiple emission reduction scenarios when the link between the perturbed emissions and the target pollutant is strong and direct, while weaker and more indirect responses are less reliably reconstructed. An optimal compromise between temporal resolution and downscaling accuracy is also identified, corresponding to approximately one day of temporal averaging. These results highlight the reliability of neural network-based downscaling techniques, to support their integration into operational air quality assessment and scenario planning tools as complements to full simulations.
- Preprint
(14491 KB) - Metadata XML
-
Supplement
(1350 KB) - BibTeX
- EndNote
Status: open (until 25 Dec 2026)
-
CEC1: 'Comment on egusphere-2026-5777 - No compliance with the policy of the journal', Juan Antonio Añel, 03 Nov 2026
reply
-
AC1: 'Reply on CEC1', Anthony Rey-Pommier, 04 Nov 2026
reply
Dear chief editor, thank you for your comment. We have revised the "Code and data availability" section, adding links and DOIs for the EMEP code, CAMS-REG-ANT v8.0, ECMWF IFS data, and the proxy dataset archive for emission transformation. All are now openly accessible and cited in the following revised paragraph. All new versions of the manuscript will contain this updated section.
Code and data availability
The SRNN code, configuration files, pretrained model checkpoint, input data for the scenario with halved emissions ("-50%ALL") on a reduced domain and period, and evaluation notebooks are archived at https://doi.org/10.5281/zenodo.23376831 (Rey-Pommier et al., 2026). The repository provides the resources required to reproduce the downscaling of NO2 concentrations and evaluate model performance. The EMEP MSC-W chemical transport model (version 5.5) used to produce the air quality simulations is openly available at https://github.com/metno/emep-ctm and archived at https://doi.org/10.5281/zenodo.14507729 (EMEP MSC-W, 2024). A full technical description is given in Simpson et al. (2012). The CAMS-REG-ANT anthropogenic emission inventory is available from the ECCAD catalogue at https://eccad.sedoo.fr/#/catalogue (https://permalink.aeris-data.fr/CAMS-REG-ANT, last access: 4 November 2026) under version number 8.0 with reprocessed time series from 2005 to 2022 (Kuenen et al., 2022). The ECMWF IFS meteorological fields are available in open access from the ECMWF Open Data portal at https://www.ecmwf.int/en/forecasts/datasets/open-data (https://doi.org/10.21957/open-data, last access: 8 September 2026) (ECMWF, 2024). The code implementing the proxy-based downscaling methodology of Bessagnet et al. (2023) is archived with the underlying spatial proxy datasets at https://doi.org/10.5281/zenodo.7803933 (Bessagnet et al., 2022).
The references appearing in this paragraph are the following:
Rey-Pommier, A.; Pisoni, E.; Beauchamp, M.; Bessagnet, B.; de Meij, A.; Zauli Sajani, S.; Thunis, P. Deep learning-based downscaling of PM2.5, NO2 and O3 over Europe across emission-reduction scenarios. Zenodo, 2026, https://doi.org/10.5281/zenodo.23376831.EMEP MSC-W. Open Source EMEP/MSC-W model v5.5 (202412). 2024, https://doi.org/10.5281/zenodo.14507729.
Simpson, D.; Benedictow, A.; Berge, H.; Bergström, R.; Emberson, L.D.; Fagerli, H.; Flechard, C.R.; Hayman, G.D.; Gauss, M.; Jonson, J.E.; et al. The EMEP MSC-W chemical transport model–technical description. Atmospheric Chemistry and Physics 2012, 12, 7825–7865, https://doi.org/10.5194/acp-12-7825-2012
Kuenen, J.; Dellaert, S.; Visschedijk, A.; Jalkanen, J.-P.; Super, I.; Denier van der Gon, H. CAMS-REG-v4: A state-of-the-art high-resolution European emission inventory for air quality modelling. Earth System Science Data 2022, 14, 491–515.
European Centre for Medium-Range Weather Forecasts (ECMWF). ECMWF Open Data. 2024, https://doi.org/10.21957/open-data.
Bessagnet, B.; Pisoni, E.; de Meij, A.; Létinois, L.; Thunis, P. A simple and fast method to downscale chemistry transport model output fields from the regional to the urban/district scale. Environmental Modelling & Software 2023, 164, 105692.575
Bessagnet, B.; Pisoni, E.; de Meij, A.; Létinois, L.; Thunis, P. A simple and fast method to downscale chemistry transport model output fields from the regional to the urban/district scale. Zenodo, 2022, https://doi.org/10.5281/zenodo.7803933.
Citation: https://doi.org/10.5194/egusphere-2026-5777-AC1 -
CEC2: 'Reply on AC1', Juan Antonio Añel, 04 Nov 2026
reply
Dear authors,
Thanks for your reply. Unfortunately, it does not solve some of the issues pointed out before. The sites linked to access the CAMS-REG-ANT anthropogenic emission inventory and the ECMWF IFS meteorological fields are not suitable repositories for scientific publication and long-term preservation of the assets. Namely:
- They do not appear to have a published policy for data preservation over many years or decades (some flexibility exists over the precise length of preservation, but the policy must exist).
- They do not appear to have a published mechanism for preventing authors from unilaterally removing material. Archives must have a policy which makes removal of materials only possible in exceptional circumstances and subject to an independent curatorial decision,
- They do not appear to issue a persistent identifier such as a DOI or Handle for each precise dataset.
If we have missed a published policy which does in fact address this matter satisfactorily, please post a response linking to it. If you have any questions about this issue, please post them in a reply.
Therefore, please, store the used data in an acceptable repository and reply to this comment with the necessary information (link and permanent handler).
Juan A. Añel
Geosci. Model Dev. Executive Editor
Citation: https://doi.org/10.5194/egusphere-2026-5777-CEC2
-
CEC2: 'Reply on AC1', Juan Antonio Añel, 04 Nov 2026
reply
-
AC1: 'Reply on CEC1', Anthony Rey-Pommier, 04 Nov 2026
reply
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 308 | 103 | 29 | 440 | 53 | 27 | 28 |
- HTML: 308
- PDF: 103
- XML: 29
- Total: 440
- Supplement: 53
- BibTeX: 27
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Dear authors,
Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".
https://www.geoscientific-model-development.net/policies/code_and_data_policy.html
In your "Code and Data Availability" statement (I note that the name of the section is incorrect in the manuscript, and must be changed) you do not include links and citations to repositories that contain all the code and data used in your work. Namely they are missing the repositories for the full EMEP code, the complete CAMS anthropogenic emission inventories, the raw ECMWF IFS meteorological datasets, and the raw underlying GIS spatial proxy datasets.
This is not acceptable according to the policy of the journal, and your manuscript should have not been accepted for Discussions because of it. The fact that it happened was an editorial oversight, and we apologize for it. Our policy clearly states that all the code and data necessary to replicate a manuscript must be published openly and freely to anyone in an acceptable repository at the submission time. The GMD review and publication process depends on reviewers and community commentators being able to access, during the discussion phase, the code and data on which a manuscript depends, and on ensuring the provenance of replicability of the published papers for years after their publication.
Therefore, we are granting you a short time to solve this situation, as we cannot have manuscripts under discussion that do not comply with our policy. You have to reply to this comment in a prompt manner with the information for the repositories containing all the code and data that you use to produce and replicate your manuscript. The reply must include the link and permanent identifier (e.g. DOI). Also, any future version of your manuscript must include the modified Code and Data Availability section with the new information.
I must note that if you do not fix this problem, we cannot continue with the peer-review process or accept your manuscript for publication in GMD.
Juan A. Añel
Geosci. Model Dev. Executive Editor