Articles | Volume 28, issue 1
Nonlin. Processes Geophys., 28, 111–119, 2021
https://doi.org/10.5194/npg-28-111-2021
Nonlin. Processes Geophys., 28, 111–119, 2021
https://doi.org/10.5194/npg-28-111-2021

Research article 09 Feb 2021

Research article | 09 Feb 2021

Training a convolutional neural network to conserve mass in data assimilation

Yvonne Ruckstuhl et al.

Download

Interactive discussion

Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Katja Gänger on behalf of the Authors (17 Dec 2020)  Author's response
ED: Publish subject to technical corrections (17 Dec 2020) by Alberto Carrassi
AR by Yvonne Ruckstuhl on behalf of the Authors (18 Dec 2020)  Author's response    Manuscript
Download
Short summary
The assimilation of observations using standard algorithms can lead to a violation of physical laws (e.g. mass conservation), which is shown to have a detrimental impact on the system's forecast. We use a neural network (NN) to correct this mass violation, using training data generated from expensive algorithms that can constrain such physical properties. We found that, in an idealized set-up, the NN can match the performance of these expensive algorithms at negligible computational costs.