Preprints
https://doi.org/10.5194/egusphere-2022-147
https://doi.org/10.5194/egusphere-2022-147
 
19 Apr 2022
19 Apr 2022
Status: this preprint is open for discussion.

Classification of tropical cyclone containing images using a convolutional neural network: performance and sensitivity to the learning dataset

Sébastien Gardoll and Olivier Boucher Sébastien Gardoll and Olivier Boucher
  • Institut Pierre-Simon Laplace, Sorbonne Université/CNRS, Paris, France

Abstract. Tropical cyclones (TCs) are one of the most devastating natural disasters, which justifies monitoring and prediction on short and long timescales in the context of a changing climate. In this study, we have adapted and tested a convolutional neural network for the classification of reanalysis outputs according to the presence or absence of TCs. We use a number of meteorological variables to form TC-containing and background images from both the ERA5 and MERRA-2 reanalyses. The presence of TCs is labelled from the HURDAT2 dataset. Special attention was paid on the design of the background image set to make sure it samples similar location and time to the TC-containing image. We have assessed the performance of the CNN using accuracy but also the more objective AUC and AUPRC metrics. Many failed classifications can be explained by the meteorological context, such as a situation with cyclonic activity but not yet classified as TC by HURDAT2. We also tested the impact of interpolation and of mix and match the training and test image sets on the performance of the CNN. We showed that applying an ERA5-trained CNN on MERRA-2 images works better than applying a MERRA-2 trained CNN on ERA5 images.

Sébastien Gardoll and Olivier Boucher

Status: open (until 14 Jun 2022)

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

Sébastien Gardoll and Olivier Boucher

Sébastien Gardoll and Olivier Boucher

Viewed

Total article views: 235 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
182 47 6 235 1 1
  • HTML: 182
  • PDF: 47
  • XML: 6
  • Total: 235
  • BibTeX: 1
  • EndNote: 1
Views and downloads (calculated since 19 Apr 2022)
Cumulative views and downloads (calculated since 19 Apr 2022)

Viewed (geographical distribution)

Total article views: 229 (including HTML, PDF, and XML) Thereof 229 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 16 May 2022
Download
Short summary
Tropical cyclones (TCs) are one of the most devastating natural disasters, which justifies monitoring and prediction in the context of a changing climate. In this study, we have adapted and tested a convolutional neural network for the classification of reanalysis outputs (ERA5 and MERRA-2 labelled by HURDAT2) according to the presence or absence of TCs. We tested the impact of interpolation and of mix and match the training and test sets on the performance of the CNN.