17 Sep 2020
17 Sep 2020
Boosting performance in machine learning of geophysical flows via scale separation
- 1Laboratoire des Sciences du Climat et de l’Environnement, CE Saclay l’Orme des Merisiers, UMR 8212CEA-CNRS-UVSQ, Université Paris-Saclay & IPSL, 91191 Gif-sur-Yvette, France
- 2London Mathematical Laboratory, 8 Margravine Gardens, London, W68RH, UK
- 3LMD/IPSL, Ecole Normale Superieure, PSL research University, Paris, France
- 4DRF/IRFU/DEDIP//LILAS Departement d’Electronique des Detecteurs et d’Informatique pour la Physique, CE Saclayl’Orme des Merisiers, 91191 Gif-sur-Yvette, France
- 1Laboratoire des Sciences du Climat et de l’Environnement, CE Saclay l’Orme des Merisiers, UMR 8212CEA-CNRS-UVSQ, Université Paris-Saclay & IPSL, 91191 Gif-sur-Yvette, France
- 2London Mathematical Laboratory, 8 Margravine Gardens, London, W68RH, UK
- 3LMD/IPSL, Ecole Normale Superieure, PSL research University, Paris, France
- 4DRF/IRFU/DEDIP//LILAS Departement d’Electronique des Detecteurs et d’Informatique pour la Physique, CE Saclayl’Orme des Merisiers, 91191 Gif-sur-Yvette, France
Abstract. Recent advances in statistical and machine learning have opened the possibility to forecast the behavior of chaotic systems using recurrent neural networks. In this article we investigate the applicability of such a framework to geophysical flows, known to involve multiple scales in length, time and energy and to feature intermittency. We show that both multiscale dynamics and intermittency introduce severe limitations on the applicability of recurrent neural networks, both for short-term forecasts, as well as for the reconstruction of the underlying attractor. We suggest that possible strategies to overcome such limitations should be based on separating the smooth large-scale dynamics from the intermittent/small-scale features. We test these ideas on global sea-level pressure data for the past 40 years, a proxy of the atmospheric circulation dynamics. Better short and long term forecasts of sea-level pressure data can be obtained with an optimal choice of spatial coarse grain and time filtering.
- Preprint
(2575 KB) -
Supplement
(3101 KB) - BibTeX
- EndNote
Davide Faranda et al.
-
RC1: 'Review of "Boosting performance in machine learning of geophysical flows via scale separation"', Anonymous Referee #1, 14 Oct 2020
-
AC1: 'answer to RC1 Anonymous Reviewer', Davide Faranda, 18 Dec 2020
-
AC1: 'answer to RC1 Anonymous Reviewer', Davide Faranda, 18 Dec 2020
-
RC2: 'Review', Anonymous Referee #2, 04 Nov 2020
-
AC2: 'answer to RC2 Anonymous Reviewer', Davide Faranda, 18 Dec 2020
-
AC2: 'answer to RC2 Anonymous Reviewer', Davide Faranda, 18 Dec 2020
Davide Faranda et al.
Davide Faranda et al.
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
658 | 77 | 4 | 739 | 16 | 9 | 7 |
- HTML: 658
- PDF: 77
- XML: 4
- Total: 739
- Supplement: 16
- BibTeX: 9
- EndNote: 7
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1