English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Variational data assimilation in ocean models depends on the ability to modelgeneral correlation operators in the presence of coastlines. Grid-point filtersbased on diffusion operators are widely used for this purpose, but come with acomputational bottleneck - the costly estimation of normalization factors forevery model grid point. In this paper, we show that a simple convolutionalneural network can effectively learn these normalization factors with betteraccuracy than the current operational methods. Our network is tested with atwo-dimensional diffusion operator from the NEMOVAR ocean data assimilationsystem, applied to a global ocean grid with approximately one degree horizontalresolution. The network is trained on exact normalization factors estimated bya brute-force method. Knowing that convolutional networks can only modeltranslation-equivariant functions, we ensure that the normalization estimationproblem is indeed translation-equivariant. Specifically, we show how the numberof inputs of this problem can be reduced while preserving translationequivariance. Adding the distance to the coastline as an input channel is foundto improve the performance of the network around coastlines. Extensions tothree-dimensional diffusion and to higher horizontal resolutions are discussed.Removing the computational bottleneck associated with normalization opens theway to using adaptive correlation models for operational ocean dataassimilation. The code for this work is publicly available athttps://github.com/FolkeKS/DL-normalization/tree/core-features

Application of deep learning to the estimation of normalization coefficients in diffusion-based covariance models