33 citations · 57 across the 3 of their papers we have counts for
Showing physics.ao-phShow all
3 papers · 1 filter
physics.ao-ph2025
Forecast error diagnostics in neural weather models
Uros Perkan, Ziga Zaplotnik, Gregor Skok
Deep-learning (DL) weather prediction models offer some notable advantages over traditional physics-based models, including auto-differentiability and low computational cost, enabl…
physics.ao-ph2025
A unified neural background-error covariance model for midlatitude and tropical atmospheric data assimilation
Boštjan Melinc, Uroš Perkan, Žiga Zaplotnik
Estimating background-error covariances remains a core challenge in variational data assimilation (DA). Operational systems typically approximate these covariances by transformatio…
physics.ao-ph2023★ 24 cited
3D-Var Data Assimilation using a Variational Autoencoder
Boštjan Melinc, Žiga Zaplotnik
Data assimilation of atmospheric observations traditionally relies on variational and Kalman filter methods. Here, an alternative neural-network data assimilation (NNDA) with varia…