TY - GEN
AU - Wei Pan
AU - Massimo Bonavita
AU - Marcin Chrust
AU - Elias Hólm
AB - Accurate representation of flow-dependent background-error covariance is essential for modern ensemble–variational data assimilation systems, but estimating ensemble covariance from sufficiently large ensembles remains computationally expensive. In this work, we propose a machine-learning approach to emulate ensemble-derived background-error variance fields from severely under-sampled ensembles. Conditional generative machine-learning models are trained to map variance estimates derived from 5-member ensembles to those obtained from operational 50-member ECMWF Ensemble Data Assimilations (EDA). The trained models are deployed within a hybrid ensemble4D-Variational data assimilation system and evaluated over an independent three-month period in an operational numerical weather prediction environment.<br />
Results show that the machine-learning-emulated variance fields reproduce the structure and analysis impact of the full 50-member system, yielding comparable observation-space diagnostics and forecast error characteristics while using only a small fraction of the ensemble size. To our knowledge, this represents the first demonstration in an operational environment of machine-learning emulation of ensemble-derived background-error variance statistics within a fully cycled numerical weather prediction data assimilation framework.
BT - Technical Memoranda
CY - Reading
DA - 02/2026
DO - 10.21957/0b7e4d4426
M1 - 936
N2 - Accurate representation of flow-dependent background-error covariance is essential for modern ensemble–variational data assimilation systems, but estimating ensemble covariance from sufficiently large ensembles remains computationally expensive. In this work, we propose a machine-learning approach to emulate ensemble-derived background-error variance fields from severely under-sampled ensembles. Conditional generative machine-learning models are trained to map variance estimates derived from 5-member ensembles to those obtained from operational 50-member ECMWF Ensemble Data Assimilations (EDA). The trained models are deployed within a hybrid ensemble4D-Variational data assimilation system and evaluated over an independent three-month period in an operational numerical weather prediction environment.<br />
Results show that the machine-learning-emulated variance fields reproduce the structure and analysis impact of the full 50-member system, yielding comparable observation-space diagnostics and forecast error characteristics while using only a small fraction of the ensemble size. To our knowledge, this represents the first demonstration in an operational environment of machine-learning emulation of ensemble-derived background-error variance statistics within a fully cycled numerical weather prediction data assimilation framework.
PB - ECMWF
PP - Reading
PY - 2026
T2 - Technical Memoranda
TI - Data-Driven Emulation of Background-Error Variance in Variational Data Assimilation
UR - &nbsp;
ER -