Journal cover Journal topic
Geoscientific Model Development An interactive open-access journal of the European Geosciences Union
Journal topic

Journal metrics

IF value: 5.240
IF5.240
IF 5-year value: 5.768
IF 5-year
5.768
CiteScore value: 8.9
CiteScore
8.9
SNIP value: 1.713
SNIP1.713
IPP value: 5.53
IPP5.53
SJR value: 3.18
SJR3.18
Scimago H <br class='widget-line-break'>index value: 71
Scimago H
index
71
h5-index value: 51
h5-index51
Preprints
https://doi.org/10.5194/gmd-2020-211
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-2020-211
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Submitted as: development and technical paper 21 Aug 2020

Submitted as: development and technical paper | 21 Aug 2020

Review status
This preprint is currently under review for the journal GMD.

Combining Ensemble Kalman Filter and Reservoir Computing to predict spatio-temporal chaotic systems from imperfect observations and models

Futo Tomizawa1 and Yohei Sawada1,2,3 Futo Tomizawa and Yohei Sawada
  • 1School of Engineering, the University of Tokyo, Tokyo, Japan
  • 2Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan
  • 3RIKEN Center for Computational Science, Kobe, Japan

Abstract. Prediction of spatio-temporal chaotic systems is important in various fields, such as Numerical Weather Prediction (NWP). While data assimilation methods have been applied in NWP, machine learning techniques, such as Reservoir Computing (RC), are recently recognized as promising tools to predict spatio-temporal chaotic systems. However, the sensitivity of the skill of the machine learning based prediction to the imperfectness of observations is unclear. In this study, we evaluate the skill of RC with noisy and sparsely distributed observations. We intensively compare the performances of RC and Local Ensemble Transform Kalman Filter (LETKF) by applying them to the prediction of the Lorenz 96 system. Although RC can successfully predict the Lorenz 96 system if the system is perfectly observed, we find that RC is vulnerable to observation sparsity compared with LETKF. To overcome this limitation of RC, we propose to combine LETKF and RC. In our proposed method, the system is predicted by RC that learned the analysis time series estimated by LETKF. Our proposed method can successfully predict the Lorenz 96 system using noisy and sparsely distributed observations. Most importantly, our method can predict better than LETKF when the process-based model is imperfect.

Futo Tomizawa and Yohei Sawada

Interactive discussion

Status: final response (author comments only)
Status: final response (author comments only)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
[Login for Authors/Topical Editors] [Subscribe to comment alert] Printer-friendly Version - Printer-friendly version Supplement - Supplement

Futo Tomizawa and Yohei Sawada

Model code and software

RC_Analysis_Lorenz96 Futo Tomizawa and Yohei Sawada https://doi.org/10.5281/zenodo.3907291

Futo Tomizawa and Yohei Sawada

Viewed

Total article views: 185 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
141 41 3 185 14 4 4
  • HTML: 141
  • PDF: 41
  • XML: 3
  • Total: 185
  • Supplement: 14
  • BibTeX: 4
  • EndNote: 4
Views and downloads (calculated since 21 Aug 2020)
Cumulative views and downloads (calculated since 21 Aug 2020)

Viewed (geographical distribution)

Total article views: 147 (including HTML, PDF, and XML) Thereof 147 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Saved

No saved metrics found.

Discussed

No discussed metrics found.
Latest update: 19 Oct 2020
Publications Copernicus
Download
Short summary
A new method to predict chaotic systems from observation and process-based models is proposed by combining machine learning with data assimilation. Our method is robust to the sparsity of observation networks, and can predict more accurately than a process-based model when it is biased. Our method effectively works when both observations and models are imperfect, which is often the case in geoscience. Therefore, our method is useful to solve a wide variety of prediction problems in this field.
A new method to predict chaotic systems from observation and process-based models is proposed by...
Citation