Preprints
https://doi.org/10.48550/arXiv.2210.11529
https://doi.org/10.48550/arXiv.2210.11529
27 Jun 2024
 | 27 Jun 2024
Status: this preprint is open for discussion.

Identifying Lightning Processes in ERA5 Soundings with Deep Learning

Gregor Ehrensperger, Thorsten Simon, Georg Johann Mayr, and Tobias Hell

Abstract. Atmospheric environments favorable for lightning and convection are commonly represented by proxies or parameterizations based on expert knowledge such as CAPE, wind shears, charge separation, or combinations thereof. Recent developments in the field of machine learning, high resolution reanalyses, and accurate lightning observations open possibilities for identifying tailored proxies without prior expert knowledge.

To identify vertical profiles favorable for lightning, a deep neural network links ERA5 vertical profiles of cloud physics, mass field variables and wind to lightning location data from the Austrian Lightning Detection & Information System (ALDIS), which has been transformed to a binary target variable labeling the ERA5 cells as cells with lightning activity and cells without lightning activity. The ERA5 parameters are taken on model levels beyond the tropopause forming an input layer of approx. 670 features. The data of 2010–2018 serve as training/validation.

On independent test data, 2019, the deep network outperforms a reference with features based on meteorological expertise. SHAP values highlight the atmospheric processes learned by the network which identifies cloud ice and snow content in the upper and mid-troposphere as very relevant features. As these patterns correspond to the separation of charge in thunderstorm cloud, the deep learning model can serve as physically meaningful description of lightning.

Depending on the region, the neural network also exploits the vertical wind or mass profiles to correctly classify cells with lightning activity.

Gregor Ehrensperger, Thorsten Simon, Georg Johann Mayr, and Tobias Hell

Status: open (until 23 Aug 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'No compliance with the policy of the journal', Juan Antonio Añel, 09 Jul 2024 reply
    • AC1: 'Reply on CEC1', Gregor Ehrensperger, 15 Jul 2024 reply
      • CEC2: 'Reply on AC1', Juan Antonio Añel, 16 Jul 2024 reply
        • AC2: 'Reply on CEC2', Gregor Ehrensperger, 17 Jul 2024 reply
Gregor Ehrensperger, Thorsten Simon, Georg Johann Mayr, and Tobias Hell

Data sets

The ERA5 Global Reanalysis Hans Hersbach et al. http://dx.doi.org/10.1002/qj.3803

Model code and software

xai_lightningprocesses Gregor Ehrensperger et al. http://dx.doi.org/10.5281/zenodo.10899180

Gregor Ehrensperger, Thorsten Simon, Georg Johann Mayr, and Tobias Hell

Viewed

Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.

Total article views: 126 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
126 0 0 126 0 0
  • HTML: 126
  • PDF: 0
  • XML: 0
  • Total: 126
  • BibTeX: 0
  • EndNote: 0
Views and downloads (calculated since 27 Jun 2024)
Cumulative views and downloads (calculated since 27 Jun 2024)

Viewed (geographical distribution)

Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.

Total article views: 96 (including HTML, PDF, and XML) Thereof 96 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Jul 2024
Download
Short summary
Lightning can cause significant damages to infrastructure and pose risks to individuals. As lightning is a short and local event it is not explicitly resolved in atmospheric models. Instead, auxiliary descriptions based on meteorological expert knowledge are used to assess lightning. We used AI that successfully discovered on its own the ingredients that experts know to be essential for lightning in the well-studied region of the Alps. Additionally, it also recognized regional differences.