?ARPEGE Cloud Cover Forecast Postprocessing with Convolutional Neural Network - INSU - Institut national des sciences de l'Univers Access content directly
Journal Articles Weather and Forecasting Year : 2021

?ARPEGE Cloud Cover Forecast Postprocessing with Convolutional Neural Network

Florian Dupuy
  • Function : Author
  • PersonId : 765895
  • IdRef : 159218969
Valentin Kivachuk Burdá
  • Function : Author
Mohamed Chafik Bakkay
  • Function : Author
Jean-Christophe Jouhaud
  • Function : Author
Maud-Alix Mader
  • Function : Author
Guillaume Oller
  • Function : Author


Cloud cover provides crucial information for many applications such as planning land observation missions from space. It remains, however, a challenging variable to forecast, and numerical weather prediction (NWP) models suffer from significant biases, hence, justifying the use of statistical postprocessing techniques. In this study, ARPEGE (Météo-France global NWP) cloud cover is postprocessed using a convolutional neural network (CNN). CNN is the most popular machine learning tool to deal with images. In our case, CNN allows the integration of spatial information contained in NWP outputs. We use a gridded cloud cover product derived from satellite observations over Europe as ground truth, and predictors are spatial fields of various variables produced by ARPEGE at the corresponding lead time. We show that a simple U-Net architecture (a particular type of CNN) produces significant improvements over Europe. Moreover, the U-Net outclasses more traditional machine learning methods used operationally such as a random forest and a logistic quantile regression. When using a large number of predictors, a first step toward interpretation is to produce a ranking of predictors by importance. Traditional methods of ranking (permutation importance, sequential selection, etc.) need important computational resources. We introduced a weighting predictor layer prior to the traditional U-Net architecture in order to produce such a ranking. The small number of additional weights to train (the same as the number of predictors) does not impact the computational time, representing a huge advantage compared to traditional methods.
Fichier principal
Vignette du fichier
[15200434 - Weather and Forecasting] ARPEGE Cloud Cover Forecast Postprocessing with Convolutional Neural Network.pdf (6.41 Mo) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

insu-03668380 , version 1 (15-05-2022)





Florian Dupuy, Olivier Mestre, Mathieu Serrurier, Valentin Kivachuk Burdá, Michaël Zamo, et al.. ?ARPEGE Cloud Cover Forecast Postprocessing with Convolutional Neural Network. Weather and Forecasting, 2021, 36, pp.567-586. ⟨10.1175/WAF-D-20-0093.1⟩. ⟨insu-03668380⟩
56 View
8 Download



Gmail Facebook Twitter LinkedIn More