A Unified Neural Background‐Error Covariance Model for Midlatitude and Tropical Atmospheric Data Assimilation

Abstract Estimating and modeling background‐error covariances remains a core challenge in variational data assimilation (DA). Operational systems typically approximate these covariances by transformations that separate geostrophically balanced components from unbalanced inertia‐gravity modes—an appr...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Boštjan Melinc, Uroš Perkan, Žiga Zaplotnik
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	American Geophysical Union (AGU) 2026-01-01
Schriftenreihe:	Journal of Advances in Modeling Earth Systems
Schlagworte:	variational data assimilation background‐error covariances atmospheric balances tropical data assimilation neural network data assimilation latent space
Online-Zugang:	https://doi.org/10.1029/2025MS005360
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!