Preprints
https://doi.org/10.5194/egusphere-2023-551
https://doi.org/10.5194/egusphere-2023-551
11 May 2023
 | 11 May 2023
Status: this preprint is open for discussion.

Exploiting radar polarimetry for nowcasting thunderstorm hazards using deep learning

Nathalie Rombeek, Jussi Leinonen, and Ulrich Hamann

Abstract. Severe convective weather events, such as hail, lightning and heavy rainfall pose a great threat to humans and cause a considerable amount of economic damage. Nowcasting convective storms can provide warning signals and mitigate the impact of these storms. Dual-polarization weather radars are a crucial source of information for nowcasting severe convective events; nevertheless, they are most often not considered in nowcasting. These radars provide signatures of different hydrometeors. This work presents the importance of polarimetric variables as an additional data source for nowcasting thunderstorm hazards using an existing neural network architecture with convolutional and recurrent layers. This network has a common framework, which enables nowcasting of hail, lightning and heavy rainfall for lead times up to 60 min with a 5 min resolution. The study area is covered by the Swiss operational radar network, which consists of five operational polarimetric C-band radars. Results indicate that including polarimetric variables and quality indices improve the accuracy of nowcasting heavy precipitation and lightning, with the largest improvement found for heavy precipitation.

Nathalie Rombeek et al.

Status: open (until 29 Jun 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Nathalie Rombeek et al.

Data sets

Data archive for Exploiting radar polarimetry for nowcasting thunderstorm hazards using deep learning Nathalie Rombeek, Jussi Leinonen, and Ulrich Hamann https://doi.org/10.5281/zenodo.7760740

Model code and software

GitHub repository c4dl-polar Nathalie Rombeek, Jussi Leinonen, and Ulrich Hamann https://github.com/MeteoSwiss/c4dl-polar/

Nathalie Rombeek et al.

Viewed

Total article views: 129 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
92 33 4 129 3 1
  • HTML: 92
  • PDF: 33
  • XML: 4
  • Total: 129
  • BibTeX: 3
  • EndNote: 1
Views and downloads (calculated since 11 May 2023)
Cumulative views and downloads (calculated since 11 May 2023)

Viewed (geographical distribution)

Total article views: 129 (including HTML, PDF, and XML) Thereof 129 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 04 Jun 2023
Download
Short summary
Severe weather, such as hail, lightning and heavy rainfall can be hazardous to humans and property. Dual-polarization weather radars provide crucial information to forecast these events by detecting precipitation types. This study analyzes the importance of dual-polarization data for predicting severe weather during the next 60 min using an existing deep learning model. The results indicate that including these variables improves the accuracy of predicting heavy rainfall and lightning.